KINA25921ENN_002.pdf

8/13/2019 KINA25921ENN_002.pdf

http://slidepdf.com/reader/full/kina25921enn002pdf 1/176

Component test facility

for a 700 °C power plant (Comtes700)

Research and

Innovation EUR 25921 EN

8/13/2019 KINA25921ENN_002.pdf


EUROPEAN COMMISSION

Directorate-General for Research and Innovation

Directorate G — Industrial Technologies

Unit G.5 — Research Fund for Coal and Steel

E-mail: [email protected]

[email protected]

Contact: RFCS Publications

European Commission

B-1049 Brussels

8/13/2019 KINA25921ENN_002.pdf


European Commission

Research Fund for Coal and Steel

Component test facility for a 700 °C power plant

(Comtes700)

G. Schmidt, K. Foß, C. Hölzel, D. RossbergerAlstom Power Systems GmbH — Augsburger Straße 712, 70329 Stuttgart, GERMANY

O. KroghBurmeister & Wain Energy A/S — Lundtoftegardsvej 93A, 2800 Kgs. Lyngby, DENMARK

S. A. Jensen, J. L. MadsenDONG Energy A/S — Nesa Allé 1, 2820 Gentofte, DENMARK

P. BillardEDF R & D — Site des Renardieres Ecuelles, 77818 Moret-sur-Loing, FRANCE

G. Gierschner, C. UllrichE.ON New Build & Technology GmbH — Alexander-von-Humboldt-Straße 1, 45896 Gelsenkirchen, GERMANY

L. GhiribelliENEL Ingegneria e Innovazione SpA — Via Andrea Pisano 120, 56122 Pisa PI, ITALY

K.-G. Tak, F. KlaukeHitachi Power Europe GmbH — Schifferstraße 80, 47059 Duisburg, GERMANY

S. HuysmansLaborelec GDF Suez — Rodestraat 125, 1630 Linkebeek, BELGIUM

R. Mohrmann, A. MöbiusRWE Technology GmbH — Huyssenallee 2, 45128 Essen, GERMANY

H. Edelmann

Siemens AG — Freyeslebenstraße 1, 91058 Erlangen, GERMANY

C. Stolzenberger, S. PolenzVGB PowerTech e.V. — Klinkestraße 27–31, 45136 Essen, GERMANY

Grant Agreement RFCP-CT-2004-000031 July 2004 to 31 December 2011

Final report

Directorate-General for Research and Innovation

2013 EUR 25921 EN

8/13/2019 KINA25921ENN_002.pdf


LEGAL NOTICE

Neither the European Commission nor any person acting on behalf of the Commission isresponsible for the use which might be made of the following information.

The views expressed in this publication are the sole responsibility of the authors and do notnecessarily reflect the views of the European Commission.

More information on the European Union is available on the Internet (http://europa.eu).

Cataloguing data can be found at the end of this publication.

Luxembourg: Publications Office of the European Union, 2013

ISBN 978-92-79-29379-5doi:10.2777/98172

© European Union, 2013Reproduction is authorised provided the source is acknowledged.

Printed in Luxembourg

P -

Europe Direct is a service to help you find answersto your questions about the European Union

Freephone number (*):

00 800 6 7 8 9 10 11

(*) Certain mobile telephone operators do not allow access to 00 800 numbers or these calls may be billed.

8/13/2019 KINA25921ENN_002.pdf


TABLE OF CONTENTS

1 Final Summary ...................................................................................................................... 5

1.1 Objectives ............................................................................................................................. 5

1.2 Results obtained and their usefulness ................................................................................... 5

2 Scientific and Technical description of the Results ............................................................ 19

2.1 Work Package 1: Engineering ............................................................................................ 19

2.2 Work Package 2: Material & Manufacturing ...................................................................... 40

2.3 Work Package 3: Valves & Measuring Devices ................................................................. 58

2.4 Work Package 4: C & I ....................................................................................................... 71

2.5 Work Package 5: Erection and dismantling ........................................................................ 85

2.6 Work Package 6: Operation ................................................................................................ 92

2.7 Work Package 7: Evaluation ............................................................................................. 118

2.8 Work Package 8: Coordination ......................................................................................... 153

2.9 Exploitation and Impact of the Research Results ............................................................. 157

3 List of tables and figures ................................................................................................... 165

4 Glossary ............................................................................................................................ 170

3

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


1 FINAL SUMMARY

1.1 OBJECTIVES

The objective is to design, manufacture, erect and operate a Component Test Facility (CTF) to test high

temperature durable new materials needed to realise a coal-based power plant with efficiencies above

50 %.The host plant of the CTF is the German power plant Scholven F owned by E.ON. It is a coal-fired

power plant with steam conditions of 535 °C and maximum 230bar, a maximum output of 740 MW and

an average operation time of more than 5,000h a year. These are the general conditions of a testing pe-

riod of approx. 20,000 hours of operation with the following focuses:

• manufacturing, bending and welding – also dissimilar – of the materials used,

• long-term creep behaviour linked with real plant dimensions and other design features,

• in-plant monitoring,

• determination and evaluation of residual service life,

• in-service inspections,• operational testing of Ni-based alloys for tubes, pipes and valves,

• gaining information on magnetite layer on the ribs of internal ribbed tubes,

• operational behaviour of all components,

• fluegas corrosion and steam oxidation behaviour of the materials and

• erosion effects due soot blowing.

The results of the CTF flow back into engineering and design of components to optimise them for the

next phase of a 700 °C power plant, the erection and operation of a demonstration plant power plant in

Europe.

The project is divided into eight work packages and covers the following parts: engineering (WP 1),

material and manufacturing (WP 2), valves and measuring devices (WP 3), control and instrumentation

(WP 4), erection (WP 5), operation (WP 6), evaluation (WP 7), and coordination (WP 8).

1.2 RESULTS OBTAINED AND THEIR USEFULNESS

1.2.1 Work Package 1: Engineering

The CTF had to be designed to fit within the steam generator of unit F of the host Plant. One target was

that the plant operation should not be affected. The heating surfaces of the evaporator and superheater

were designed to reach a temperature of 700°C in a wide load range. The connecting and auxiliary

components such as pipes, headers, valves, spray attemperator and drainages were designed to the spe-

cific temperature to transport, cool down and launch steam to a superheater or reheater. The pressure

part was designed for a lifetime of 200,000 hours according to TRD. The requirements according Pres-

sure Equipment Directive guideline Module G (PED) for the pressure-loaded parts such as steam pip-

ing, superheater or evaporator had to be fully implemented. The permit for the design was given from

TÜV.

The evaporator consisted of 44 parallel tubes with a tube-fin-tube arrangement and made from T24,

HCM and Alloy 617B. The material for the inlet header was 13 CrMo44 and the material for the outlet

header was P92. Internally riffled tubes were designed in the T24 section in order to improve the heat

transfer properties. A FEM analysis fixed a maximum average wall metal temperature for stress calcula-

5

8/13/2019 KINA25921ENN_002.pdf


tion. The average evaporator outlet temperature was in operation around 50°C, lower than expected,

caused by the geometry and heat flux.

The superheater was calculated for an arrangement in the convective pass of host superheater 2. The

chosen materials for the two panels were Sanicro 25, HR3C and DMV310 N (austenitic steel) and Alloy

617B and Alloy 740 (nickel based alloys). Two tubes of Alloy 263 were designed and integrated later.

The inlet header was adjusted to P92, the outlet header to Alloy 617B.The HP steam pipe was calculated for Alloy 617B and a lifetime of 200,000h with an OD of 220mm

and a wall thickness of 50mm.

Four spray attemperators were designed. One was designed for a temperature of 610°C to control the

main steam temperature before the superheater. Two others were designed for a temperature of 700°C

to reduce steam temperature to 540°C before re-entry into the final host superheater. All three were

similarly designed for horizontal operation. The OD was 220mm and the wall thickness was reduced to

30mm. Missing drainages lead to a two-phase flow in the high temperature region. The fourth attem-

perator was integrated in the HP-bypass valve to cool down the steam to the host reheater.

The draining system was divided into three sections depending on the temperature level and was improved during operation to avoid a two-phase flow.

The insulation layer thickness was calculated for a maximum surface temperature of 60°C. It was nec-

essary to switch to a microporous insulation material where the required space was not available.

The operation mode of the CTF was flexible to follow the host plant operation constantly. The follow-

ing modes were possible:

normal operation at full load of the main steam generator,

part load operation at part load with sliding pressure,

controlled HP-bypass operation where the steam was expanded and cooled in the HP-bypass

valve and discharged to the hot reheater system and

start-up operation with special attention to sufficient cooling of the evaporator surface.

Boiler cleaning was not planned during commissioning because it was not possible to blow out in any

direction.

1.2.2 Work Package 2: Material & Manufacturing

The chosen materials were ordered in general according to EN 10204 with a 3.1 A certificate. Only

minor material components were ordered with a 3.1 B certificate. Some materials like Alloy 740, Alloy

617B, P92 or Sanicro 25 had to be ordered via a PMA because they were not specified in codes and

standards at that time.

The material was processed according to common standards such as EN 10216. To minimise any trials,

material specifications have been compiled for all materials before ordering. Additional tests were con-

sidered for new materials, if necessary. All manufacturing steps, like bending and welding and qualifi-

cation procedures followed the EN standards. The Notified Body TÜV was responsible for interim cer-

tificates and the final approval of the CTF as a whole.

The superheater showed indications at all Alloy 617B tubes after manufacturing. The tubes had to be

peeled due to overleaps with a depth of up to 1.6mm on the outer surface from 44.5 x 10mm to 41 x

8.5mm. The flaws were not found during longitudinal UT and standard Eddy Current testing, but by a

specific Eddy Current Circograph combined with UT for wall thickness measurement. The weldability

of the material was good (TIG welding). RT was used for volumetric testing instead of UT as planned.

6

8/13/2019 KINA25921ENN_002.pdf


The dissimilar P92/Alloy 617B stubs of the superheater inlet header were welded using P91 filler metal,

whereas Alloy 617B filler metal had been specified. All stubs had to be welded again.

Alloy 617B evaporator tubes showed surface defects and had to be ground. Flaws were similar to those

observed at superheater tubes. The circumferential welds of HCM12, 13CrMo4-5 and P92 had to be

heat treated to keep the hardness low and achieve good ductility. Tube-fin-tube welding and fin-fin-

welding were carried out by SAW with the suggested wire/flux combinations and without pre-heating.Hardness measurements showed expected values for chosen evaporator material.

The first forgings of thick-walled components displayed a larger grain size (-3) the than specified grain

size (2-4) in one half of each part. A new heat treatment procedure was developed achieving a grain size

in total up to 1, which was acceptable according to investigations. The temperature of the solution heat

treatment and re-heating up to hot working temperature level was reduced in the forging process. The

holding time at solution annealing temperature was also reduced. Heat treatment was done by step-

heating to achieve holding time compensating temperature differences between the core area and the

outlying area below solution temperature. The high toughness of Alloy 617B required special tools for

machining and low cutting speeds were applied. On average machining consumed five times more time

compared to ferritic steels. TIG welding of thick-walled components was performed without filler metal

(pure fusion of base material) and combined TIG/SMAW welding with filler without heat treatment. RT

was applied 100% after the root or first filler pass. PT was performed 100 % at 1/3, 2/3 and 3/3 of the

wall thickness. PT during welding of nickel based alloys implied that the metal temperature had to be

cooled down from the interpass temperature of 150°C to a level lower than 50°C before use. This pro-

cedure increased manufacturing time. Especially the time frame for manufacturing of circumferential

butt welds was time consuming.

1.2.3 Work Package 3: Valves & Measuring Devices

The valve design was applied according the relevant codes and the high steam temperature of 710°Cdemanded the use of Ni-based Alloy 617B for all pressure retaining parts. The design followed in sev-

eral cases a monobloc design that led to increased material volume and machining time compared to

alloyed steel but was necessary from the point of view of structural strength.

Check and swing check valves were designed similar to the design which is used for conventional pow-

er plants. Gate valves had a two-part design and the sealing surfaces on the seat rings and wedges were

clad with Stellite 6 material. The surface of the stem shaft has been coated. The packing temperature

had to be reduced by prolongation of the stem. The stuffing box at the upper end of the cover was de-

signed with cooling fins at the outer surface to use graphite gaskets with stainless steel caps. The globe

valves had a monobloc design and followed the concept of standard high pressure valves. The valve

neck was prolonged in order to reduce the temperature of the stuffing box. The Safety Relief Valve was

made out of Alloy 617B in the high temperature parts and a steel material in the conventional tempera-

ture zones. A spring loaded safety relief function with an additional pneumatic load actuator was

adapted to avoid overpressure. Screw bolts had a comparatively high elongation length because the

intermediate flange was designed especially long to avoid overheating of the spring. The start-up valve,

the control valve and the HP-bypass valve were designed as angle valves. The stuffing box was placed

at a sufficient distance and equipped with cooling fins to avoid oxidation of the graphite sealing materi-

al. The HP-bypass valve was designed with the aid of numerical methods to optimise temperature dis-

tribution, flow dynamics and wall thickness. It was carried out with one integrated water injection noz-

zle to cool down the steam of the host power plant reheater temperature.

7

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


1.2.4 Work Package 4: C & I

The control and instrumentation system was designed according to EN codes and standards (e.g. EN

60584-2, Accuracy of temperature sensors and EN 50156-1, Safety regulations for DCS and BMS Sys-

tems). The control system was linked to the existing host power plant to visualise all information from

the sensors in the control room and to protect and control the CTF and the host plant. The recorded datawere forwarded to the COMTES700 partners for further analysis (lifetime). A remote control system

was established.

More than 50 Type S thermocouples (PtRh-Pt) were installed to measure steam or inner and outer wall

temperatures. This type of thermocouple was chosen due to its tolerance of +/-1K in the range from 0 to

1,000 °C. Gold plates at the tip end were mounted for fast and accurate measurement. The electrical

insulation material was MgO (purity above 99.4 %) with a high resistivity of more than 1.0 *1012

Ohm*cm at 700°C in comparison with 5.0*107 Ohm*cm for TiO2. A polynomial calculation of tem-

perature values calculated from thermo voltage demanded much higher accuracy than typical for stand-

ard solutions. Compensation cables were avoided to increase the accuracy. Due to the increased heat

conduction the length of the sheath tube had to be extended to ensure that temperatures at the connec-tion head remained below 120°C. After four years of operation and more than 160 starts the thermo-

couples showed no thermal drift. It could be assumed they would reach the same operation time again.

A self-calibrating thermocouple was tested alternative to type S thermocouples, see WP7.

The pressure measurement was tailor-made due to standard design, although materials were not appli-

cable for this temperature range. The pressure gauge valve was made of Alloy 617B and designed with

cooling fins and heat shield to reduce the sealing temperature. The measurement worked satisfactorily

and no problems occurred.

The flow measurement was designed as a Venturi flow measurement. Standard materials could not be

used at 700°C. The manufacturing material was Alloy 617B but it had to be taken into considerationthat the expansion factor rose from 14.4*10-6/K at 600°C to 15.1*10-6/K at 700°C and could lead to

failure of the steam flow measurement. Steam flow failures of up to 5 % can appear if the wrong expan-

sion factor was chosen. The measurement worked excellently during operation.

1.2.5 Work Package 5: Erection

The CTF had to be installed within the steam generator of unit F of the host plant. The unit was com-

missioned in 1979 and produces electricity with a net output of 676 MWe and thermal output of 1,860

MWth by burning hard coal. The installation had to take place within the tight schedule of an ordinary

overhaul period of the power plant. After installation, the unit could not be operated without the CTF.

Despite the very ambitious time schedule, the power plant was able to go into commercial operation in

July 2005 with a delay of only two weeks.

The service exposed materials are of a very high scientific and technological value. Many material

samples were taken not only for evaluation purpose of the project but also in order to test the material in

other European research projects.

The call for bids for erection of the CTF was issued in December 2004. In March 2005 a consortium

was awarded the erection (steel works, pipes, valves, evaporator and superheater). The start date with

regard to components outside the boiler was in April 2005 and the start date with regard to the more

time critical components located inside the boiler was in May 2005. It was planned to perform the pres-

sure test end of June.

9

8/13/2019 KINA25921ENN_002.pdf


The design of the evaporator suspension had to be revised due to a bulge in the membrane wall of the

existing boiler, which was the exact planned location of the test evaporator, and the test evaporator had

to be lifted twice.

Some components were delivered quite late and the design of the piping and steel structures had to be

revised because of collisions with the existing pipes. This caused a delay of 12 days. Non-functionality

of the control technology and a leakage of the hydraulic actuator of the stop valve caused another twodays of delay.

In 2006, installations in the host plant as well as minor modifications of devices and safety aspects were

completed.

It was decided to dismantle the CTF on 11 August 2009 and the dismantling started immediately. Dis-

mantling was finished end of March 2010.

All components which were located inside the boiler had to be removed as fast as possible to start-up

the power plant again. Pipe connections to the steam generator were cut and open pipe or stub ends

were closed with caps. The remaining 700°C piping system, including valves outside the boiler was

dismantled in the first quarter of 2010.

The dismantled components were marked and transported to the storage hall. The material was stored

for approx. two years in the hall and was available for further sampling. The remaining material was

scrapped at the end of 2011 after the final samples had been taken.

Finally, more than 100 material samples were taken from evaporator, superheater, headers and the

steam piping system. A number of valves were sent to the valve manufacturers for investigation.

1.2.6 Work Package 6: Operation

The operation of the CTF started in July 2005. As it was closely connected to the host plant, this couldnot be operated without the CTF. In 2007 the first problem with a crack in a thick-walled Alloy 617B

component occurred. In the following years these problems increased. Several repairs had to be carried

out and a high number of laboratory investigations led to deeper understanding of the material.

Finally, it was decided to stop the CTF after more than 20,000 hours of operation in August 2009, as

initially planned, and to dismantle the CTF. In the meanwhile plans were revised to extend the opera-

tion period with approx. two years in order to gain additional experience; however this could not be

realised. The decision was taken due to problems with the conventional part of the facility and the

700°C part of the facility. Unpredictable time and cost for further repair works led to the decision.

The operation of the CTF demonstrated that full-scale 700°C components made of Ni-based alloys can

be operated in power plants. It also showed, however, that improvement of the material handling with

regard to manufacture, installation and maintenance is necessary. Challenges with thick-walled alloy

components still exist and additional tests in future test rigs are necessary.

Several macro cracks with partial steam leakages occurred and a high number of microcracks at repair

welds were detected:

- Cracks occurred in areas with a high degree of additional stresses, such as attemperators with

two-phase flow.

- Cracks were present in the area of stub welds and circumferential welds.

- Welds from the erection time and repair welds were affected.

10

8/13/2019 KINA25921ENN_002.pdf


- Two circumferential repair welds in the thick-walled 700°C area without heat treatment, neither

before nor after welding, caused the most severe cracks, which were detected after some months

of daily start/stop cycles.

- Heat treated repair welds (980°C/3 h) showed micro cracks (dye penetrant test indications) but no

macro cracks.

- Cracks started at the outer surface and crack propagation was interdendritic and intergranular.

The operation showed that a post-weld heat treatment at 980°C for thick-walled Alloy 617B compo-

nents has a good effect on the weld. This kind of heat treatment can be applied to erection welds and

repair welds. It was not possible to develop and to verify a final repair concept for thick-walled Alloy

617B components during the operation time. Open questions have to be solved in follow-up investiga-

tions and projects. Promising ideas are already available.

The experience from the CTF is being considered in subsequent projects such as ENCIO and HWT II.

Post-weld heat treatment at 980°C is considered for thick-walled Alloy 617B components where the

steam is cooled down with colder steam or a combination of water and flowing steam in order to

achieve good dispersion of the water and thereby avoid a two-phase flow.

22,400 hours of operation were achieved between 2005 and 2009. The facility was operated for 13,000h

at steam temperatures above 680 °C. A maximum steam temperature of 725°C was reached in the outer

tubes of the test superheater. Unit F was operated with many starts and stops during the operation peri-

od thereby making it possible to test the components under cyclic loading conditions, which was origi-

nally not expected. 576 starts were counted until the end of operation.

Some problems occurred with the control technology and the valves in the first months of operation

which caused minor non-availabilities. Valve problems were mainly caused by leaky gaskets and actua-

tors and could be solved.

Apart from one tube damage in the superheater end of 2006, no incidents of damage to materials appeared in the first two years of operation. Cracks on thick-walled Ni-based alloys were detected for the

first time in spring 2007. Further cracks were detected in the following years. Cracks and problems with

repair welds led to a long outage period in the last year of operation.

Modifications of the facility were carried out during the 2008 summer outage. Additional drain pipes

were introduced and the slope of some pipes was optimised. Problems with condensate, especially at

spray attemperators, led to these measures.

The first samples from the superheater were taken in the 2007 summer outage. The final sampling was

carried out after the dismantling in 2009 and 2010.

Several defects at thick-walled Alloy 617B components outside the boiler had to be repaired, whereasthin-walled components suffered only one failure. Cracks occurred in the highly stressed spray attem-

perators 2 and 3, where feed water reduced the steam temperature from 700 to 540°C. Small indications

were detected at repair welds of Alloy 617B components. Repair welds were made in parts where mate-

rial was cut out and fitting pieces had to be installed.

The following failures were observed:

Superheater

o Leakages were detected at two tubes in the area of welds connecting Alloy 617B with

Sanicro25 in October 2006. The crack was initiated by stress relaxation cracking in Alloy

617B, induced by manufacturing related issues like hardness, grain structure, etc., and op-erational loads. A successful repair weld was applied.

11

8/13/2019 KINA25921ENN_002.pdf


Spray attemperator 2

o PT indications were observed on stubs of injection nozzle assemblies in August 2008. La-

boratory investigation showed that the cracks ran interdentritically with carbides and pores

on the grain boundaries in front of the crack tip. Indications were ground out and weld

build-up was made using TIG welding. Heat treatments were avoided due to the various

thermal expansions of the attemperator assembly. Cracks were detected in the weld metal

in February 2009.

o Indications were observed at a bracket in September 2008. The bracket and the affected

part of the attemperator were removed and metallographic investigation confirmed the in-

dications as cracks. The affected attemperator part was replaced with a fitting piece with-

out the bracket. During this repair strong deformations at the protective pipe were detected.

o Visual inspection showed a crack of the protective pipe close to the injection nozzle as-

sembly. The end part was cut out and replaced with the end part of spray attemperator 3.

Heat treatment was carried out only on the side connecting the attemperator to the steam

pipe.

o In this fitting piece crack propagation was detected in February 2009. In order to ensure

continued commercial operation of the host plant, spray attemperator 2 was immediately

replaced with spray attemperator 3.

Spray attemperator 3

o The regular outage in 2006 revealed strong deformations and cracks of the protective pipe

of spray attemperator 3. The pipe was replaced and the attemperator was connected to the

steam pipe with two circumferential welds without heat treatment. One of the repair welds

failed in April 2007.

o In March 2007, a weld failure at the Alloy 617B counter bearing of a spray injection noz-

zle assembly had to be ground out and was welded again. New leakage was detected in

February 2008.o In April 2007, a deflection of the turbine valve was detected and the pipe, which includes

spray attemperator 3, was slightly bent. Water, injected through a preheating pipe before

the turbine valve, had accumulated at the bottom of the turbine valve and spray attempera-

tor 3 area impacting weld failures. Two welds were replaced with fitting pieces applying

prWHT.

o In February 2008, the turbine valve was inclined and the pipe, which includes spray attem-

perator 3, was obviously deflected. Cracked welds at the protective pipe and water puddles

were observed. One crack was located in the HAZ of an electrode welded circumferential

weld in the middle part of the attemperator, the other at a near stub weld. Both welds were

manufactured during erection. Further PT indications were detected at stubs installed tocentre the protective pipe. Replacement with three fitting pieces was carried out using elec-

trode welding. While the base metal was not heat treated, the main pipe was annealed at

980°C for 3h in a furnace. PT and UT showed indications at stub welds after annealing.

Cracks were propagating interdendritically in the weld metal. Crack flanks showed strong

oxidation with accumulation of Al, interdendritic areas and grain boundaries displayed

significant accumulation of carbides containing Mo. The cracks propagated into the base

material and the crack flanks were covered with Mo-Cr carbides. Indications disappeared

after grinding and welds were built-up with filler metal. The new circumferential repair

weld showed indications at the stubs. Cracks of 1-2mm length propagated interdendritical-

ly in the weld metal and intergranularly in the HAZ of the base material. The interdendriticareas of the weld metal as well as the grain boundaries of the base material displayed mas-

12

8/13/2019 KINA25921ENN_002.pdf


sive ‘saucer -type’ accumulation of carbides containing Mo and Cr. The repair welds were

ground, cut out and TIG built-up using a filler metal with a lower boron content. Two

welds were filled with TIG orbital. This welding technique was applied with good results.

Attemperator 3 was annealed again in a furnace before installation of the protective pipe in

early 2009.

Circumferential repair welds of steam pipes

o In 2007 all thick-walled circumferential welds of the 700°C steam pipe part were checked

with mechanised UT. Manufacturing failures were found, which were not detected using

RT during erection. Four welds were replaced with fitting pieces in 2008 using electrode

welding and PWHT at 980°C/3 h. After some months of operation PT indications (UT

showed acceptable indications) had to be eliminated again by grinding, cutting and built-up

welding.

o The available repair concept was not appropriate. Different kinds of optimisations were

tested, which led to better results, but local indications were still found quite frequently

during the works. The works were extremely time and cost intensive. All repair works

were finished in May 2009 after more than two months of standstill.

1.2.7 Work Package 7: Evaluation

Most of the findings within Work Package 7 are a result of a comprehensive scope of material investi-

gations carried out by several laboratories. The scope of investigations was defined in an evaluation

matrix describing the samples and the relevant investigations. All relevant component types and materi-

als were covered in this matrix. The scope included investigations of oxidation, corrosion and micro-

structure with microscopy, hardness test, impact test and other mechanical tests.

In addition to the material investigations; investigations of consumed lifetime and evaluation of relevant

design assumptions have been carried out. A limited scope of other activities have also been investigat-ed and reported.

Due to failures in thick-walled components further studies were initiated, leading to additional results

regarding failure analysis, repair procedures and non-destructive testing.

The membrane wall was manufactured using T24, HCM12 and Alloy 617B and showed no cracks or

leakage during operation. The materials performed as expected with respect to oxidation and corrosion

considering the alloy compositions and operating temperature. High scatter in corrosion data for T24

exposed at metal temperatures above 546ºC suggests that the operating temperature should be limited

for this material without further investigation. Investigations of welds showed similar oxidation and

corrosion characteristics as base metals. Hardness measurements added results regarding hardness pro-

files across welds in conformity with the expected profiles for the investigated materials. Decarburisa-

tion was also observed as expected in cases where non-matching filler metal was used. Only minor im-

perfections were observed in the investigations of the welds.

The results have proven that it is partly possible to design, manufacture and operate a membrane wall in

a 700ºC component test facility; in view of the fact that no damage, no excessive oxidation or corrosion

and no considerable microstructural changes were observed. Results indicate that the maximum operat-

ing temperature for the steel alloys should be considered. It should be noted that design and operational

loads may have deviated from a full scale plant.

The superheater was manufactured using DMV310N, HR3C, Sanicro 25, Alloy 617B and Alloy 740.

During operation a tube leakage occurred in Alloy 617B adjacent to a dissimilar weld. The crack was

initiated as stress relaxation cracking, induced by cold deformation, grain structure and stresses. A re-

13

8/13/2019 KINA25921ENN_002.pdf


pair concept including annealing heat treatment was applied successfully. Investigation of the oxidation

provided results describing oxide morphology, oxide thicknesses and the internal oxidation zones of Ni-

based alloys. The lowest oxide thickness on austenitic steels was observed for Sanicro 25 and on Ni-

based alloys the oxide layer thickness was comparable for Alloy 617B and Alloy 740. Alloy 740 did,

however, show the most pronounced internal oxidation along grain boundaries. Investigation of corro-

sion provided results describing corrosion layers, corrosion rates and presence of internal corrosion. The

majority of the investigated materials showed minor thickness reductions, comparable to the deviation

from nominal thickness for the as-received samples. Locally, large thickness reductions for Sanicro 25

and Alloy 617B suggest that a more detailed study should be initiated in the critical temperature ranges

and considering the coal sulphur content. Microstructural investigations added results regarding grain

sizes and precipitates, and in combination with creep tests it was suggested that the version of Alloy

740 used may suffer from microstructural instability. Investigations of welds generally showed similar

oxidation and corrosion characteristics as base metals; although in some cases with local effects near

fusion line. Hardness measurements added results regarding hardness profiles across welds. Only minor

imperfections were observed in the investigations of welds.

The results have proven that it is partly possible to design, manufacture and operate a superheater in a700ºC component test facility. The tube leakage which occurred during operation showed that the as-

received condition of Alloy 617B tubes and the manufacture can affect the susceptibility for stress re-

laxation cracking; and it led to the development of a repair concept which was applied successfully. The

observations of local high corrosion rates for some alloys indicate that it is very likely that the operation

time of the superheater will be less than the design lifetime; and further studies of corrosion in critical

temperature ranges should be considered.

The thick-walled components was manufactured using mainly Alloy 617B, and with some use of P92

and low alloy steel. The planned microstructural investigations of thick-walled Alloy 617B added re-

sults regarding grain size distribution, precipitates, and characteristics at grain boundaries and in the

matrix. Hardness tests and mechanical tests confirmed that service exposed materials have increased

hardness and tensile strength but with lower ductility. Increased hardness on inner surfaces compared to

bulk hardness was also observed. Thin chromium oxide caused local depletion of Cr and dissolution of

precipitates. In some cases shallow oxidation pits and grain boundary attack were observed. Only minor

imperfections were observed in the investigations of the welds.

The first crack in thick-walled components was observed after approximately 15,000 hours of opera-

tion, and in the following period additional cracks were observed. This led to further detailed investiga-

tion of the welds and components concerned. In most cases, cracks were reported to nucleate in the

weld and to propagate in an interdendritic and intergranular manner to the base material. It was suggest-

ed that residual stresses and operational stresses promoted the failure mechanism.

The results have proven it partly possible to design, manufacture and operate a piping system in a 700ºC

component test facility. Code requirements were the background for not applying PWHT, but results

suggest that PWHT should be applied for the welding of Alloy 617B components. The horizontal spray

attemperator design was used due to limited space, but in combination with insufficient drainage this

may have contributed to the observed failures.

The observed failures in thick-walled Ni-based components led to the necessity of repair procedures. In

the first attempts the weld sequence for the reduction of residual stresses was optimised, initially within

the range of qualification of the welding procedures used during manufacture, followed by improve-

ment in the selection of welding processes and filler materials. It also became obvious that heat treat-ments, both pre-welding and post-welding, should be investigated and applied to reduce re-cracking

14

8/13/2019 KINA25921ENN_002.pdf


susceptibility. Heat treatment studies included microstructural studies with focus on precipitates and

precipitate behaviour at various temperatures. An experimental test of site heat treatment for solution

annealing documented the applicability of the developed equipment and the thermal cycle of the pipe

material. A weld test programme was then initiated to develop an optimised weld repair procedure,

based on selected combinations of welding processes and heat treatments. Preliminary results indicate a

good potential for the development of a repair concept.

During manufacture of thick-walled Alloy 617B components for the CTF; volumetric testing was only

applied for the first 1/3 of each weld, due to limitations in the available techniques for volumetric test-

ing. After the observations of cracks in thick-walled components it became obvious that mechanised

ultrasonic testing should be introduced for inspection purposes in the following operation period.

Mechanised ultrasonic testing became qualified and verified during the project and was applied success-

fully for the inspection of a large number of welds in the piping system. With this method a number of

welds with indications were identified and replaced. Verification was done by comparing non-

destructive test results with destructive laboratory investigations, and the results were in good agree-

ment. The mechanised ultrasonic test method can be recommended for further application of thick-

walled Ni-based components for the power generation industry and other industries.

Calculations of consumed lifetime and design review were performed for a selected position in the Al-

loy 617B part of the superheater, for selected components in the main steam piping system and for

spray attemperators 2 and 3. The results from calculations of consumed lifetime and design review have

generally shown moderate consumed lifetime in the operation period. Only for the spray attemperators

the lifetime consumption was found to be high as a result of the two-phase flow thereby confirming the

importance of an improved attemperator concept and a sufficient drainage system. The results have also

identified the necessity for further corrosion studies in selected temperature ranges for the superheater

and the importance of post-weld treatment and the extent of NDT for the reduction of notch factors and

thereby the consumed lifetime for thick-walled components.

Three additional technologies were tested; self-calibrating thermocouples, compact insulation material

and nanoceramic coating. The self-calibrating thermocouples were promising.

1.2.8 Work Package 8: Coordination

The COMTES700 continued the idea of the 700°C power plant started in 1998 with the AD700 phases

1 and 2 projects. The project has impacted on European and national research projects, namely ENCIO

and GKM HWT II, and has challenged China, India and Japan to plan their own 700°C technology re-

search programmes.

Coordination was required for the relationships between six contractors and eighteen co-financing part-ners (suppliers and utilities) from eight European countries. Several groups were established to manage

major tasks (Steering Committee, Project Management Group, Temporary Support Group Financial,

Editorial Group, WP 6 Operation, WP 7 Evaluation).

By joining forces, the installation of the CTF could be ensured with a delay of only two weeks. The

project was extended for two years exploiting operation experience for a future 700°C demonstration

power plant. The unexpected termination of CTF operation was compensated by an increase in the in-

vestigation efforts. Additional resources needed were covered by acceptance of a new partner, know-

how access agreements and additional funding of all partners. These activities required the establish-

ment of a complex agreement structure. Requests from other 700°C projects for investigation material

were generally approved against return of results.

15

8/13/2019 KINA25921ENN_002.pdf


In order to enhance operation experience, the COMTES700 parties successfully applied for a project

extension of two years. Simultaneously, operation problems forced the host plant to shut down the pro-

duction unit, which was not separable from the CTF. As time and costs for repairs were unpredictable,

the host plant decided to dismantle the CTF. Consequently, all project resources were focussed on the

evaluation part and a comprehensive programme of additional laboratory investigation was established

to exploit the operating experience.

Various consortium internal projects were supported with material free of charge, or materials were

provided to third parties. Generally, the COMTES700 parties appreciated and supported other projects

in favour of the advancement of 700°C technologies. These activities are planned to be supervised by a

“700°C Group”.

The total budget increased from EUR 15.3 mio. up to more than EUR 26 mio. The CTF caused outages

of the host power plant unit for several weeks. Related costs were covered by RWE and E.ON. The

budget increase was balanced by the project partners, acceptance of a new party and know-how access

agreements. Complex payments between the parties had to be arranged and controlled. Tax questions

had to be settled.

The COMTES700 results were communicated to all partners in workshops and a web-based document

file system was established and maintained. The public was informed by a project web page and more

than 40 publications.

1.2.9 Conclusion and possible applications and patents

Key components have been designed, manufactured, erected and operated in the host power plant in

order to demonstrate if the necessary technical solutions to realise a 700°C demonstration power plant

are available.

An evaporator panel made up of three sections with three different materials: T24, HCM12 andAlloy 617B.

Two superheater panels for the final superheating up to 700 C. Tested materials are HR3C,

DMV310N, Sanicro25, Alloy 617B and Alloy 740.

Valves including a HP-bypass valve.

Thick-walled components such as headers and spray attemperators. All components for 700 C

were made of Alloy 617B.

Control and instrumentation devices.

A turbine control valve was simultaneously tested outside the scope of this project.

1.2.9.1 Evaporator

Since the obtained outlet temperature was approx. 50C below the planned value, the project has not

demonstrated that the required materials are found. In addition, the tested panel was not subjected to

external static loads as will be the case in a coil section of a tower boiler.

For practical reasons a membrane wall had to be erected without PWHT on site. However, all of the

three tested materials have shown shortcomings:

The section made of T24 performed without problems, but many problems with this material

has been experienced lately as described in VGB news 2011 volume 11.

16

8/13/2019 KINA25921ENN_002.pdf


In the course of the project, the creep strength values for HCM12 were reduced with the result

that this material is no longer a relevant material for membrane wall applications.

Tests of the Alloy 617B evaporator panel were performed. However, it is questionable due to

the result from the membrane wall in the MARCKO700 project which demonstrated a very

short lifetime when subjected to external loads.

Therefore, it must be concluded that further testing of materials for membrane walls is needed.

1.2.9.2 Superheater

Corrosion rates that were found locally for Alloy 617B tubes result in lifetime estimates as low as

50,000h. These parts would have to be exchanged depending on the chosen design.

Even though actual lifetimes of the final stages of a superheater are often found to be less than the gen-

eral design lifetime of 200,000h, the cost of replacing parts of the superheater is very high due to high

material cost. Longer lifetimes are needed in order to make a 700°C power plant economically feasible.

1.2.9.3 Thick-walled components

Due to the susceptibility to stress relaxation cracking, post-weld heat treatment of thick-walled compo-

nents in Alloy 617B becomes mandatory for thick-walled sections. Since the superheater sections in this

material performed satisfactorily without PWHT, a thickness limit above which PWHT is required re-

mains to be determined.

Developed procedures for performing repair welds using prWHT on service exposed materials in Alloy

617B must be further tested to demonstrate that proper solutions have been found. Testing of the long

term properties of the most promising procedures are underway.

NDT methods and acceptance levels for in-service evaluation of thick-walled components must be de-

veloped further. Mechanised UT seems to be a promising technique.

In spray attemperators special care must be taken to reduce thermal stresses as well as avoiding two-

phase flow. Two-stage injections of steam and water, respectively, are suggestions for further develop-

ment.

1.2.9.4 Valves

Thick-walled valves, e.g. HP-bypass valves, showed crack formation starting on the inside of the valve

bodies. Since the cracks were undetectable by NDT in the in-service state it is questionable if the tested

valves could be operated safely for long periods.

Therefore, in order to reduce the risk of both fatigue and creep cracking, it is vital to test available im-

proved designs where local stress concentrations due to thickness changes are reduced, and reduce tem-

perature differences across the wall thickness during load changes (including start and stop sequences).

17

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


2 SCIENTIFIC AND TECHNICAL DESCRIPTION OF THERESULTS

2.1 WORK PACKAGE 1: ENGINEERING

2.1.1 Work Package Objectives

Design of the whole Component Test Facility (CTF)

- basic thermal calculation

- adaption to host plant

Design of the several components of the CTF:

- evaporator

- superheater

- headers

- attemperators

- steam pipe

- valves

Permission from the authorities for operation of the CTF

2.1.2 Comparison of Initially Planned Activities and Work Accomplished

The operational experience showed that no major deviations with scientific relevance occurred.

2.1.3 Description of Activities and Discussion

2.1.3.1 CTF Design

2.1.3.1.1 Description of the CTF

The CTF was installed in the steam generator of unit F of the E.ON Power Plant Scholven (bituminous

coal-fired once through steam generator), Germany. Unit F was commissioned in 1979 and produces

electricity with a net output of 676 MW and thermal output of 1,860 MW by burning hard coal. The

unit produces 2200 t steam per hour, 40 t of this were taken for the CTF and heated up to 700°C. The

steam was cooled down and returned to the main boiler at the end of the facility.

The CTF was erected in 2005 and started up 14 days later than scheduled (15 July 2005) due to unexpected problems with the production of the components. The CTF was planned to operate until 2009.

The installation of such a test facility within a commercially operating power plant poses many chal-

lenges and any operational problems experienced by the CTF would immediately affect the commer-

cially operating power plant and might lead to full stop. With this reason, even from the start of the

project, reliability was an important criterion for the CTF.

19

8/13/2019 KINA25921ENN_002.pdf


Basically, the following components were installed:

evaporator (44 pipes, manufactured from T24, HCM 12 and Alloy 617B; dimensions: approx. 8m

x 9m,

superheater (2 x 16 pipes as superheater batch conducted, from Sanicro 25, HR3C, Alloy 740 and

Alloy 617B, length: 17m each,

HP header,

safety valve, HP-bypass valve and start-up valve,

turbine inlet valve (no EU-sponsorship), and

connective and thick-walled piping.

The whole chain from engineering and manufacturing to operational behaviour, creep properties, steam

oxidation and corrosion was covered.

The test evaporator of the CTF was installed at a height of between 56.5 and 66.2 meters in front of the

boiler membrane wall (Figure 1).

Figure 1: Construction of the test heat exchange surfaces and location in the boiler

Both of the superheater batches were suspended in between already existing superheater surfaces of the

unit F on a supporting tube structure.A new platform at a height of 76m within the boiler was constructed for a spray cooler, the turbine inlet

valve, the HP-bypass valve, gate valves and other valves.

The components of the CTF, which were exposed to 700 °C, were manufactured from Alloy 617B. In

addition to Alloy 617B some other materials were tested (Alloy 740, Alloy 263, Sanicro 25,

DMV310N, HR3C, T24 and HCM12).

A turbine control valve was also installed in the CFT. Investigation of this valve was not covered by the

COMTES700 project. The CFT only served as a host for this valve.

The water/steam process is shown in the P&I-diagram 9.21602/00010-0001 which is based on the flow

diagram of the description of the CTF in the proposal submission forms. Figure 2

20

8/13/2019 KINA25921ENN_002.pdf


Figure 2: Isometric drawing of the CTF

The cold steam for the CTF was extracted from the SH1-inlet header (390°C) of the main steam genera-

tor. The steam flow was measured (26NA45F001-003) and controlled (26NA45S002) with respect to

the steam temperature downstream from the first heating surface (evaporator 26NA45). A 100% bypass

valve (26NA45S004) with fixed minimum flow position ensured minimum flow through the CTF. Thesteam was heated in the first heating surface (evaporator 26NA45) to the controlled outlet temperature

of 610°C. On the way to the next heating surface (superheater 26NA65) the first injection stage

(26NA55Z015) was arranged. The steam temperature was reduced in order to control the superheater

outlet temperature. While passing through the superheater surface the steam reached the live steam

temperature (705°C) of the CTF.

In the piping system downstream from the superheater surface the following components were installed:

- start-up injection station (26NA75Z015)

- HP-bypass valve (26NA77S003)

- safety valve (26NA77S094)- start-up control valve (26NA79S002)

- turbine valve (26NA76S005)

- final injection (26NA76Z020)

The CTF conformed to the task described in the work package description of WP1 in the proposal sub-

mission forms.

The start-up injection station (26NA75Z015) was only in operation during start-up and while switching

from normal operation mode to bypass operation mode for controlled temperature increase in the down-

stream piping system.

The turbine valve (26NA76S005) and a long HP steam pipe (26NA76Z010) made of Alloy 617B wereintroduced in the CTF downstream from the start-up injection for test purposes.

21

8/13/2019 KINA25921ENN_002.pdf


The HP-bypass valve was used for test purposes or when the available pressure difference of the SH

system of the main steam generator did not ensure sufficient cooling mass flow in part load operation

mode.

In normal operation mode in the final injection stage (26NA76Z020) the steam temperature of the CTF

was further reduced to meet the live steam parameters of the main steam generator (538°C). Before

being discharged to the SH system of the main steam generator (SH4 outlet), the steam flow was meas-ured (26NA76F001) in order to balance mass flows and detect leakages.

The calculated mass flow rates and the steam temperatures of the CTF when the main steam generator

operated at full load were as Table 2 following:

Table 2: Calculated data of the CTF at full load of the boiler

LocationPressure

bar

Temperature

°C

Mass flow rate

kg/s

Inlet conditions downstreamSH1 222 390 11,57

Inlet evaporator 219 390 11,57

Outlet evaporator 216 610 11,57

Downstream 1st injection 216 580 12,00

Inlet superheater 215 580 12,00

Outlet superheater 212 705 12,00

Downstream final injection 210 536 14,59

Downstream HP-bypass valve 41 538 13,43

If the steam leaving the CTF at the superheater outlet (12kg/s, 705°C and 212bar) was expanded via a

turbine this could generate 16MWe. The calculated values of the pressure drop at full load of the CTF

can be seen in Table 3.

Table 3: Pressure drop of the CTF

No. Designation Quantity Length di Oper. Pres. Oper. Temp. Pressure loss

[m] [mm] [bar] [°C] [bar]

1. Inlet Pipe (from SH1 Inlet) 1 ca.100 123,9 222,0 390,0 0,6*

Control Valve 390,0 1,5

Isolating Valves + Flow Meter 1 390,0 0,52. Evaporator 44 ca. 50 19,5 219,4 390,0 - 610,0 3,0

3. Pipe incl. Spray Cooler 1 1 ca. 40 155,1 216,4 610,0 - 580,0 0,4

4. SH 32 ca. 50 24,5 216,0 580,0 - 705,0 2,6

5. Pipe incl. Spray Cooler 2 / 3 1 ca. 60 119,1 213,4 705,0 - 536,0 2,0

6. Pipe to SH4 Outlet 1 ca. 60 163,1 211,4 536,0 0,9

Isolating Valves + Flow Meter 2 536,0 0,5

After Isolating Valves 210,0 536,0

Total 12,0

* included a level difference of above 50m

According to measurements, the available pressure drop of the main steam generator between SH1 inletand SH4 outlet resulted in approx. 14bar.

22

8/13/2019 KINA25921ENN_002.pdf


2.1.3.1.2 Modes of Operation

The CTF was arranged in parallel to the superheater system of the main steam generator. For sufficient

cooling of the test surfaces the steam mass flow through the CTF had to be high enough. Since the

available pressure difference for the CTF was given by the operation of the main steam generator at

lower loads the cooling of the surfaces had to be assured by discharging the steam to the hot reheat sys-

tem of the main steam generator. Normal operation

The main steam generator was operated at full load in once-through operation mode. The CTF was sup-

plied with cold steam from the SH1 inlet header. The steam flowed via the evaporator, the first injection

and the superheater stage with control of the outlet temperatures. Having passed the piping upstream

from the final injection, the steam was cooled down to the main steam generator live steam conditions

and discharged to the SH4 outlet header.

In this operation mode the HP-bypass was closed as well as the isolation valve to the auxiliary steam

supply (26RQ65S502), the safety valve (26NA77S094) and the start-up control valve (26NA76S002).

The start-up injection (second injection stage 26NA75Z015) was out of operation.

Controlled HP-bypass operation

Steam supply and temperature control were the same as for normal operation mode. The steam was

expanded and cooled in the HP-bypass valve and discharged to the hot RH system.

Part load operation

The main steam generator operated at part load in sliding pressure mode. The temperatures of the steam

downstream evaporator and superheater stage were controlled (610°C downstream evaporator and

705°C downstream superheater). When further decreasing the boiler load and the resulting pressure

difference did not ensure a sufficient cooling the operation mode had to be changed from normal opera-

tion mode to controlled bypass operation mode.

Start-up operation

Attention had to be paid to ensure sufficient cooling of the evaporator surface during start-up operation.

For the T24 grade a max. temperature limit of approx. 600°C (metal temperature) had to be respected.

Two start-up modes were foreseen:

- cold start-up (after inspection or long standstills)

- warm start-up after weekend or over-night shut-down, restart after emergency shut-down

Cold start-up

The main steam generator and the CTF were at zero pressure. The start-up procedure should be carried

out with the lowest burner level at a firing capacity of approx. 5%. This resulted in a flue gas tempera-

ture of approx. 600°C below the evaporator panel of the CTF in the furnace chamber.

The cooling of the CTF was realised with auxiliary steam when the pressure in the start-up vessel of the

main steam generator was below the level of the auxiliary steam line (16bar). The control valve

(26NA45S002) and the isolating valve upstream turbine (26NA76S503) valve were closed, the isolating

valve in the auxiliary steam pipe (26RQ65S504) and the start-up control valve (26NA75Z015) were

open. The auxiliary steam quantity was controlled in order to reach the set point of 610°C at the evapo-

rator outlet while respecting the allowable temperature gradients of the installed surfaces, their headers,

the HP steam pipe and the turbine valve. The auxiliary steam travelled through the superheater stageand was discharged to the atmosphere via the start-up control valve.

23

8/13/2019 KINA25921ENN_002.pdf


The steam supply was switched to the main steam generator (extraction from SH1 inlet header) when

the pressure in the start-up vessel of the main steam generator was high enough. The control valves in

the main line and in the bypass line (26NA45S002/S004) were opened and the start-up control valve

and the auxiliary steam isolating valve were closed. The steam from the main steam generator was heat-

ed and discharged via the HP-bypass valve the hot RH pipe.

In case that the available pressure difference was too small it could be necessary to discharge the steamto the atmosphere.

In case of increasing pressure difference the operation mode could be switched from bypass operation

to normal operation mode.

The firing capacity should not be increased until the outlet temperature of the evaporator was below or

equal to 610°C.

Warm start-up

The main steam generator and the CTF were in standby operation mode. A start-up procedure with the

lowest burner level was recommended. Cooling of the CTF was realised with steam from the main

steam generator. The isolating valves in the auxiliary steam pipe and the start-up control valve were

closed, while the control valve in the feeding pipe and the HP-bypass valve were open. The isolating

valve upstream the turbine valve was closed. The steam was heated in the evaporator and the superheat-

er and was discharged via the HP-bypass valve to the hot reheat system of the main steam generator.

It could be necessary to discharge the steam to the atmosphere in case of insufficient pressure differ-

ence.

The increase of the firing rate was limited to sufficient cooling of the evaporator surface. The steam

outlet temperature should not exceed 610°C.

In case of sufficient available pressure difference the operation mode could be switched to normal oper-ation mode.

Switch from HP-bypass operation to normal operation and vice versa

The steam pipe downstream the turbine valve was kept warm by means of HP-steam from the main

steam generator (approx. 490°C).

The steam pipe downstream the HP-bypass control valve was kept warm by means of RH-steam from

the main steam generator (approx. 470°C).

When switching from normal operation mode to bypass mode or vice versa the steam temperature in the

discharge piping was reduced to approx. 535°C with the start-up injection. Then the temperature was

increased by a rate of 10-15K/min. up to the set value of 705°C with the start-up injection. When the set

value was achieved, the injections in the HP-bypass valve and the final injection reduced the steam

temperature to the set values for the main steam generator.

All details of the operation modes were summarised in an operating manual.

24

8/13/2019 KINA25921ENN_002.pdf


Figure 3: P&I diagram of the CTF

25

8/13/2019 KINA25921ENN_002.pdf


2.1.3.2 Components Design

The CTF consisted of the following components:

- evaporator with headers

- superheater with headers

- HP steam pipe

- attemperators

- valves

TÜV Nord was chosen as the Nominated Body who was to check the design of the pressure part. The

pressure part was designed (according to TRD) for a life cycle of 200,000h.

The materials for headers, pipes and tubes, the codes for stress values and the tolerances are summa-

rised in Table 4:

Table 4: Materials for headers, pipes and tubes and the applied codes

26

8/13/2019 KINA25921ENN_002.pdf


2.1.3.2.1 Evaporator with Headers

The evaporator surface of the CTF consisted of 44 parallel, smooth tubes with a tube-fin-tube arrange-

ment. They formed one surface panel which was installed close to the left wall of the furnace chamber

between level +57m and + 67m near the SH1 panels of the main steam generator. In order to simulate

the situation of the evaporator wall in a furnace chamber, the rear side of the panel was isolated permit-

ting single edge heat radiation. Taking into consideration the occurring temperatures three differentmaterials were selected (T24, HCM12, Alloy 617B). Additionally in the first section (T24 material) two

parallel tubes were partly made as riffled tubes with reduced wall thicknesses.

The evaporator tubes were fed by an inlet header and discharged the steam to an outlet header mounted

outside of the furnace chamber. These headers were fed unilaterally.

The geometrical data are summarised in Table 5.

Table 5: Geometrical data of the evaporator

Section No of paralleltubes

Outer diametermm

Wall thicknessmm

Pitchmm

Material

Wall tubes 1 42 33,7 7,1 48 T24Wall tubes 1 2 33,7 5,6 48 T24

Wall tubes 2 44 33,7 7,1 48 HCM12

Wall tubes 3 44 33,7 7,1 48 Alloy 617B

Inlet header 1 168,3 28 - 13CrMo4-5

Outlet header 1 219,1 45 - P92

The data for stress calculation (according to TRD) are shown in Table 6.

Table 6: Data from stress calculation

Section Operation

pressure / barg

Design pres-

sure / bar g

Operation tem-

perature / °C

Design temper-

ature / °C

Wall tubes 1 217,4 247,4 505 575

Wall tubes 2 216,6 246,6 560 630Wall tubes 3 216 246 610 680

Inlet header 219 249 410 425

Outlet header 216 246 610 625

The evaporator panel is shown in Figure 4.

Figure 4: Outline of the evaporator panel

27

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


Figure 5: Calculated wall temperatures for T24

Figure 6: Calculated wall temperatures for HCM12 and Alloy 617B

The FEM calculations show the temperature distribution for the cross section of the evaporator pipe.

The parameters for the resulting temperature distribution are the outer heat flux, internal heat transfer,

steam temperature and geometrical data (OD, wall thickness, fin geometry) and material parameters

(thermal conductivity). Sections with the same colour represent areas with the same temperature and

heat flux range as indicated in the legend. Important temperatures resulting from the FEM analysis are

the max. metal temperature found for thermal durability of the applied material and the max. mid walltemperature found for stress calculation. Due to the location and the geometrical properties the parame-

ters steam temperature, outer heat flux and inner heat transfer vary.

29

8/13/2019 KINA25921ENN_002.pdf


2.1.3.2.2 Selected Design for T24 Riffled Tubes for Wall Tubes 1

The geometry of the T24 internally riffled tubes used in the section “Wall Tubes 1” of the CTF evapo-

rator heating surface was designed by Siemens Figure 7.

The riffled tubes were installed when the evaporator tubing was not inclined in order to improve the

heat transfer properties and to avoid local overheating under subcritical two-phase operation conditions.

Compared to the design of the smooth tubes the design of the internally riffled tubes had to be based on

the following specific parameters:

- equivalent ID

- equivalent wall thickness.

These parameters used for the geometry identification of riffled tubes could be explained as follows:

- : rib lead angle

- : rib side angle

- da: tube OD

- deq : equivalent ID

- seq : equivalent wall thickness (= 0.5* ( da – deq ))

Figure 7: Geometry of riffled tubes

For T24 riffled tubes used for the evaporator heating surface of the CTF the following tube data was

selected in Table 9:

Table 9: Geometrical data of the riffled tubes

Item Unit Value

Riffled Tube Outside Diameter mm 33,7

Equivalent Internal Diameter mm 20,1

Equivalent Wall Thickness mm 6,8

Min. Wall Thickness mm 5,6

Rib Lead Angle ° 30Rib Side Angle ° 55

Rib Profil - B *)

*) According to data from the supplier V&M Tubes

30

8/13/2019 KINA25921ENN_002.pdf


2.1.3.2.3 Superheater with Headers

The superheater surface was arranged in the convective pass of the main steam generator in the area of

the SH2 surface and consisted of two assemblies. The tubes of the surface had a U-shape and were in-

stalled between two adjacent SH2 assemblies. The surface was supported by hanger tubes. The trans-

versal pitch of the original SH2 surface was split in the area where the CTF superheater surface was

arranged. The geometrical data are summarised in Table 10.

Table 10: Geometrical data of the superheater

Section No of paralleltubes

Outer diametermm

Wall thicknessmm

Transversal pitchmm

Material

Wall tubes 1 32 44,5 10 720 Alloy 174,HR3C,

DMV310N

Wall tubes 2 32 44,5 10 720 Alloy 617B,

Alloy 740

Inlet header 1 168,3 45 - P92

Outlet header 1 219,1 50 - Alloy 617B

The data for stress calculation (according to TRD) is shown in Table 11.

Table 11: Data for stress calculation

Section Operation pressure

/ bar g

Design pressure /

bar g

Operation temper-

ature / °C

Design tempera-

ture / °CWall tubes 1 214 244 635 685

Wall tubes 2 212 242 705 755

Inlet header 215 245 580 585Outlet header 212 242 705 720

The arrangement of the superheater in the convective pass is shown in Figure 8.

31

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


Table 12: Thermal design data of the superheater

2.1.3.2.4 HP Steam pipe

A 59.3m long steam pipe was installed in the CTF, downstream from the start-up injection with the

following technical data:OD 219.1mm

Wall thickness 50mm

Material Alloy 617B

Operation pressure 212bar g

Design pressure 242bar g

Operation temperature 705°C

Design temperature 710°C

2.1.3.2.5 Attemperators

Four attemperators were installed in the CTF:

1. first injection E1 downstream from the evaporator

2. second injection E2 downstream from the superheater (start-up injection)

3. final injection E3 downstream from the turbine valve

4. injection in the HP-bypass valve

The calculation of the mass flows is summarised in Table 13-15.

33

8/13/2019 KINA25921ENN_002.pdf


Table 13: Calculated flows of the CTF

Location Pressure Temp. Mass Flow Spray Inject.

bar °C kg/s kg/s

Inlet pipe 222 390 11,57

Evap. inlet 219 390 11,57

Evap. outlet 216 610 11,57

After injection 1 216 580 12 0,43

SH inlet 215 580 12

SH outlet 212 705 12

After injection 3 210 536 14,59 2,59

After HP-bypass valve 41 538 13,43 1,43

Table 14: Design data final injection

Final Injection

Steam at superheater outlet p = 212 bar, t = 705 °C h1 = 3814 kJ/kg

Steam mass flow from superheater m = 12 kg/s

Injection water mass flow m2 = m (h1 - h3)/(h3 - h2) = 2,59 kg/s

Injection water conditions p = 250 bar, t = 265 °C h2 = 1158 kJ/kg

Steam after injection p =210 bar, t = 536 °C h3 = 3342 kJ/kg

Steam mass flow after injection m4 = 14,59 kg/s

Table 15: Design data HP-bypass injection

HP-bypass Injection

Steam at superheater outlet p = 212 bar, t = 705 °C h1 = 3814 kJ/kg

Steam mass flow from superheater m = 12 kg/s

Injection water mass flow m2 = m (h1 - h3)/(h3 - h2) = 1,43 kg/s

Injection water conditions p = 250 bar, t = 265 °C h2 = 1158 kJ/kg

Steam after injection p = 41 bar, t = 538 °C h3 = 3531 kJ/kg

Steam mass flow after injection m4 = 13,43 kg/s

The E1-3 injection consisted of two stages, one stage could be isolated. For E2 and E3 the first stage(Spray injection A) had two nozzles and the second stage (Spray injection B) had three nozzles (see

Figure 9).

Figure 9: Outline of attemperator arrangement

34

8/13/2019 KINA25921ENN_002.pdf


Figure 10: Pressure drop of nozzles vs. flow capacity of attemperator

Figure 11: Details of spray attemperator 3

Due to arrangement circumstances the spray attemperators were installed in horizontal positions. This

caused severe problems due to attemperator leakages which could not be drained properly by the drain-

ing system. The drainage system was optimised during the plant operation period Figure 11.

35

8/13/2019 KINA25921ENN_002.pdf


2.1.3.2.6 Drains

The drainage system was divided into three sections according to the maximum operation temperature

ranges. Every section was connected to an individual lance of the flash tank Figure 12. An additional

lance on the flash tank was installed for venting.

Figure 12: Flash tank

2.1.3.2.7 Insulation

The insulation of the complete CTF was specified by Alstom and ordered by E.ON. Special attention

was given to the max. operation temperatures of the applied materials and the need for exponentially

increased layer thicknesses due to the high temperatures. Due to the fact that the CTF had to be de-

signed to fit the free space in the existing boiler, the required spaces was not always available, andtherefore a special microporous insulation material had to be applied.

2.1.3.2.8 Valves

This subject is discussed in WP3.

2.1.3.2.9 Design Feedback

The early operation of the CTF resulted in defects defect in the liquid tightness of all installed valves.

The reason for this was found in magnetite particles which were released from the inner surface of the

pipes and tubes.

Persistent leakage of the HP-bypass spray cooler necessitated a change in the arrangement of the pipefrom the HP-bypass to the hot reheat due to severe distortion of the affected pipe.

36

8/13/2019 KINA25921ENN_002.pdf


There were constant problems at all spray coolers due to leakage and insufficient drainage of the sec-

tion. The horizontal spray cooler arrangement had to be adapted to suit the existing plant layout which

made it difficult to ensure proper drainage.

2.1.3.3 Commissioning

Generally the commissioning of the CTF differed not much from the commissioning of conventional

plants, the commissioning tasks were basically identical. The following were carried out:

Checks for completeness and correct erection

signal check for each gauge and motor,

trial runs of all drives and motors,

check of measurements for plausibility,

logic tests,

preparation of plant or equipment for inspection by authorities and

set-up of closed control loops.

Optimisation of closed and open control loops and check of operational parameters

trial run and hand over to host/client.

2.1.3.3.1 Course of Commissioning

Table 16: Course of CommissioningDATE

Plan AchievedFirst commissioning engineer on site June 06th 2005 June 6th 2005End of mechanical erection with last actuator in place June 22nd 2005 July 10th 2005Authority (TÜV)test of safety/protective functions July 13th 2005First steam to CTF July 13th 2005 July 14th 2005First ignition of boiler after implementation of CTF July 14th 2005 July 14th 2005First commercial operation after implementation of CTF July 14th 2005 July 15th 2005First time reached 705°C at SH 2.5 outlet July 18th 2005First operation of CTZ to SH 4 (normal mode) Aug. 5th 2005Manual operation of CTF in three shifts with pre-setting

of closed control loops until

Aug. 11th 2005

Commissioning of open loop controls Aug./Sept. 2005End of “trial run” and “take over” by host July 21st 2005 Oct. 14th 2005

37

8/13/2019 KINA25921ENN_002.pdf


2.1.3.3.2 Delay for Conventional Reasons

The delayed end of erection of the CTF was partly caused by supply problems.

E.g. the hydraulic drive for the shut-off gate in front of the turbine valve was delivered just three

days prior to start of the plant.

Due to late hand-over from erection to commissioning, it was only possible to carry out cold

erection of most of the equipment before going over to steam.

Sequences to be commissioned in parallel with commercial operation of boiler, with priority of

safe operation.

Lack of support by control systems supplier (rare presence on weekends and unusual working

hours; often change of people).

Break-down of communication between existing control system and new control system.

Unavailability of boiler (PA-fan damage; load programme changes by load dispatcher)

2.1.3.3.3 Delay due to Scientific Reasons

Within the commissioning time further obstructions and delays were generated by:

On-site machining of Ni-based materials was underestimated.

E.g. in black steels the reaming of thermowells takes one shift for 10 pieces. In Ni-based material

only 3 to 4 pieces could done at the same time.

The turbine control valve was not properly designed and was leaking heavily.

Different thermal expansion values for casing and inside material led to low tension of the head

seal causing leakage of the valve. The steam was lead directly out the boiler house thereby caus-

ing sound emissions above 120 dB(A). Therefore normal operation was possible during restricted

daytime hours only.

The boiler cleaning concept was not available/skipped during the project thereby causing much

debris to be caught in drain valves or safety valve (see next chapter).

2.1.3.3.4 Boiler Cleaning

The boiler was not cleaned during the commissioning since the plant could not be blown out in any

direction.

Blow-out from the conventional boiler through the black material piping into the stainless steel

parts of the CTF to the atmosphere was not possible without contamination of the stainless steel

piping.

Reverse blowing from the SH 4 to the SH 1 inlet headers would have caused similar problems,

especially after the boiler had been overhauled too.

Partial blow-out only was not an option, since it would have required several long standstill peri-

ods to build and rebuild the provisional piping for each stage of the blowing procedure.

For the next steps of the COMTES700 programme it is very import to take the ability for easy cleaning

into account and to organise the material selection accordingly.

38

8/13/2019 KINA25921ENN_002.pdf


2.1.4 Conclusions

2.1.4.1 General

The realisation and the operation of the CTF proved a partly successful design, manufacturing, con-

struction and integration of equipment and components made of new materials (e.g. heating surfaces

and valves in alloys) into an existing system.

The technical difficulties resulting from different materials and different operation temperatures were

mostly successfully solved. The different operation temperatures occurred for example when joining the

CTF evaporator to the main evaporator resulting in significant differences in operation temperatures and

temperature ranges.

Complex automatic operation of the CTF was successfully implemented.

During the operation, sufficient cooling of the CTF components was continuously ensured without any

impact on the firing system.

The design, manufacturing and operation of the CTF showed partially successful co-operation by dif-

ferent engineering groups from different companies (e.g. Alstom engineering, E.ON engineering etc.).

Generally, the assumptions that were made in the design process were confirmed by the operation of the

CTF.

Damage occurred at the injections where the protective pipe suffered cracks and deformations due to

the high temperature differences. This has to be considered in the design for commercial application.

At present, all the tools are available which are needed to design a commercial plant with the high

steam temperatures (according to EN norms and standards).

Due to the arrangement and the geometrical facts of the evaporator surface the temperature range of the

exhaust steam was much wider than expected. Consequently, the heat pick up of the evaporator surfacewas determined by the tube with the highest outlet temperature. The resulting average temperature in

the steam pipe downstream the evaporator surface was lower than the design temperature.

2.1.4.2 Actual Applications

There is no commercial 700°C plant today. Part of the CTF was used as reference for the new projects,

ENCIO and HWT II. The design parts are in general applied in conventional USC power plants.

2.1.4.3 Technical Potential for the Use of the Results

The intensified use of FEM analysis for the design is a prerequisite.

39

8/13/2019 KINA25921ENN_002.pdf


2.2 WORK PACKAGE 2: MATERIAL & MANUFACTURING


Ordering materials

Proof of weldability and bendability

Development of testing technologies for the new materials

Manufacturing of the following components:

o evaporator

o superheater

o headers

o attemperators

o steam pipe

Proof of manufacturability

Qualification through inspecting bodies (e.g. TÜV)


Since it was the first time that components from Ni-based raw materials were to be manufactured, the

manufacturing of the thick-walled components, such as headers and steam piping, proved to be very

time-consuming and prohibitive. Alloy 617B tubes for the superheater part had to be peeled from the

initial dimensions of OD 44.5x10mm to OD 41x8.25mm, since overleaps were found after bending. At

the evaporator tubes of Alloy 617B (OD 33.7 x 7.1mm), flaws were discovered too. After the tubes had

been welded to the fins, 100% penetration testing was carried out which revealed 130 crack indications.These had to be ground until no more flaws were detected while maintaining the minimum wall thick-

ness. It also proved very time-consuming to do the NDT on welds of Ni-based alloys, since UT is too

imprecise on materials with an austenitic structure. Therefore, a PT/RT testing sequence had to be exe-

cuted. At the SH-inlet header, the dissimilar P91/Alloy 617B stubs to the header made of Alloy 617B

had been welded using P91 filler metal. After Alloy 617B was specified, the stubs had to be reconnect-

ed using the correct filler metal. During solution annealing in the manufacturing process of the thick-

walled Alloy 617B raw material (for pipe), the grain size exceeded the specified grain size of the mate-

rial specification. Therefore, the solution annealing process had to be changed from an one-step to a

two-step heating process in order to achieve even temperature distribution up to the core of block in the

first step and then heat up to the solution annealing temperature.

40

8/13/2019 KINA25921ENN_002.pdf



2.2.3.1 Ordering Material

This chapter discusses the materials used for several parts of the CTF.

Evaporator

The evaporator part included the heating surface with the materials T24, HCM12 and Alloy 617B. Ta-

ble 17 gives an overview of the different materials and the corresponding supplier.

Table 17: Dimensions and suppliers of the evaporator materialsMaterial Dimension in mm Supplier Certificate

Tubes T24 33,7 x 7,1 V&M 3.1A

Riffled tubes T24 33,7 x 7,1 V&M 3.1A

Fin material T24 6,00 x 15,1 Guelde 3.1B

Tubes HCM12 33,7 x 7,1 Sumitomo 3.1AFin material HCM12 6,00 x 15,1 Guelde 3.1B

Tubes Alloy 617B 33,7 x 7,1 DMV PMA

Fin material Alloy 617B 6,00 x 15,1 Guelde PMA

The materials were ordered with a 3.1A certificate accord-

ing to EN 10204 with the exception of the fin materials

which were ordered with a 3.1B certificate. A grade 3.1A

acceptance certificate means that it was issued and con-

firmed by an expert listed in the official regulations, in

accordance with these and the associated technical rules.

Alloy 617B is excluded since a Particular Material Ap-

praisal (PMA) was applied. The PMA was made according

to the corresponding or equivalent VdTÜV data sheet. Be-

cause of the melting loss during welding it was especially

important for the modified version of Alloy 617 to show

aluminium content not lower than 0.8 mass-%.

Figure 13 as an example, shows the 3.1A certificate for the

ordered HCM12 by Sumitomo.

Figure 13: 3.1A certificate of HCM12 evaporator tube

41

8/13/2019 KINA25921ENN_002.pdf


Piping

The steam piping includes the inlet and outlet of the evaporator and superheater besides the piping to

the bypass, superheater and reheater and the start-up and safety valve. Table 18 gives an overview of

the different materials and the corresponding suppliers.

Table 18: Dimensions and suppliers of steam piping materialsMaterial Dimension in mm Supplier Certificate

SH1 – Evaporator inlet 13CrMo4-4 168,3 x 22,2 Buhlmann 3.1AEvaporator outlet P92 219,1 x 45 Saarschmiede PMA

SH-Inlet P92 219,1 x 32 Saarschmiede PMA

SH-Outlet – Spray Cooler 3

Alloy 617B 219,1 x 50 Saarschmiede PMA

Pipe to HP-Bypass Alloy 617B 219,1 x 50 Saarschmiede PMAStart-Up Valve / SafetyValve

Alloy 617B 133,0 x 25 Saarschmiede PMA

Pipe to SH4-Outlet P91 219,1 x 28 V&M 3.1APipe to RH2-Outlet 10CrMo9-10 244,5 x 20 Buhlmann 3.1A

As for the evaporator parts, if applicable a 3.1A certificate was ordered. P92 and Alloy 617B were ex-

cluded since a PMA was applied Figure 14. As an example, shows some pages of the PMA for Alloy

617B.

Figure 14: PMA of Alloy 617B

42

8/13/2019 KINA25921ENN_002.pdf


Superheater

The superheater was manufactured by HPE using the high strength/corrosion resistant materials

Sanicro 25, HRC3/DMV310N, Alloy 617B and Alloy 740. Table 19 gives an overview of the different

materials and the corresponding suppliers.

Table 19: Dimensions and suppliers of the superheater materials

Material Dimension in mm Supplier Certificate

Tubes Sanicro 25 44,5 x 10 Sandvik PMA

Tubes HR3C 44,5 x 10 Sumitomo 3.1A

Tubes DMV310N 44,5 x 10 DMV 3.1ATubes Alloy 617B 44,5 x 10 DMV PMA

Tubes Alloy 740 44,5 x 10 Special Metals PMA

Filler Materials

The ordered filler materials and the respective suppliers for the evaporator parts are listed in Table 20.Figure 15 shows the 3.1B certificate of the Alloy 617B welding rod (S Ni6617 mod).

Table 20: Dimensions and suppliers of filler materials

Material Dimension in mm Supplier CertificateTIG welding rod T24 OD2,4 Boehler 2.2

SAW electrode T24 OD2,0 Boehler 2.2

SAW flux T24 - Boehler 2.2

TIG welding rod Alloy 617B. OD 2,4 Boehler/Thyssen 3.1B

SAW electrode Alloy 617B. OD 2,0 Boehler/Thyssen 3.1BSAW flux Alloy 617B. - Boehler/Thyssen 3.1BSMAW electrode Alloy 617B. OD 2,5 Boehler/Thyssen 3.1B

SMAW electrode Alloy 617B. OD 3,2 Boehler/Thyssen 3.1B

Orbital electrode Alloy 617B. OD 1,2 Boehler/Thyssen 3.1B

TIG welding rod P92 OD 2,4 Boehler/Thyssen 3.1B

SMAW electrode P92 OD 2,5 Boehler/Thyssen 3.1B

SMAW electrode P92 OD 3,2 Boehler/Thyssen 3.1B

43

8/13/2019 KINA25921ENN_002.pdf


Figure 15: 3.1B certificate of Alloy 617B welding rod

44

8/13/2019 KINA25921ENN_002.pdf


Table 21 gives an overview of the chemical composition of novel materials.

Table 21: Chemical composition of novel materials applied for the CTF

45

8/13/2019 KINA25921ENN_002.pdf


Adherence to Formal Requirements

Installation

For the installation of the CTF into Scholven Unit F, several formal procedures based on code require-

ments had to be taken into consideration. Since Scholven is an existing plant, the requirements accord-ing to PED Directive, guideline Module G (PED) for the pressure-loaded parts such as steam piping,

superheater or evaporator, and TRD201 had to be fully implemented.

Since several novel materials were im-

plemented in the CTF which were not

specified at that time in the codes and

standards, the implementation was han-

dled according to given standards such

as EN 10216. To minimise any trials,

material specifications were compiled

for all materials before ordering. It has been considered whether additional test-

ing of new materials is necessary.

The materials specifications and the

welding lists and the drawings had to be

approved by the Notified Body before

ordering or manufacturing.

Figure 16: Interim certificate of the Notified Body (RWTÜV)

Approval Process

The Notified Body has followed the project from the early planning phase for design approval, overissue of the interim certificates for the different parts of the CTF until final approval of the erected facil-ity (Figure 16).

Manufacturing Process

For all manufacturing steps, like bending and welding, qualification procedures (WPQR and BPQR)

according to EN standard and the VGB-guideline have been followed.

46

8/13/2019 KINA25921ENN_002.pdf


Bending

The evaporator tubes with the dimension Ø 33.7 x 7.1mm were cold bent with a radius of 60mm which

corresponds to an R/D ratio of 1.78. According to the relevant VdTÜV material data sheets, stress relief

of T24 and HCM 12 and solution annealing following by mechanical testing of Alloy 617B were car-

ried out. For this, all requirements have been fulfilled. The hardness deviation between unbent and bent

tubes is lower in conformity with DIN EN 12952-5. In addition, surface crack testing was carried out.The thick-walled components were bent by inductive bending.

Figure 17: Cover sheet of WPQR HCM12/Alloy 617B BW

Welding

For the welding, WPQRs were carried out to meet the requirements of the guidelines. Tube-fin-tube

welding and fin-fin-welding were carried out by SAW with the suggested wire/flux combinations and

without pre-heating.

The requirements in accordance with VdTÜV451-68/1 have been fulfilled. The hardness testing showed

the expected hardness values for T24, HCM12 and Alloy 617B

The MT and PT at the membrane walls showed no indications or failures.

Figure 17 shows the WPQR cover sheet of the dissimilar butt weld HCM12/Alloy 617B. Also as re-

gards homogenous welds of HCM12, dissimilar welds were post-weld heat treated.

2.2.3.2 Manufacturing

Attention was paid to the manufacture of the thick-walled base materials. The final machining of the

inner surface was of utmost importance to ensure that no imperfections resulted in a corroding surface.

The machining of the final surface of Alloy 617B, which consumed more time and lowered the durabil-

ity of the cutting tools compared to ferritic steels can be seen in Figure 18.

47

8/13/2019 KINA25921ENN_002.pdf


For the forging of the thick-walled pipes, two steps werecarried out to ensure that the grain size was homogenousover the whole wall thickness and within the specifiedgrain size range as provided by DIN EN ISO 643 (grainsize 2÷4). In one particular case, a nonconformity report(NCR) by Saarschmiede occurred, where the grain size of

the manufacturing lot reached up to -3 for Alloy 617B(see Figure 19 for reference). These two pipes (OD 219.1x 50mm), showed the mentioned grain size in one half ofeach tube, while the other half showed the specified grainsize. Also a square block from Sempell intended for thevalve body showed grain sizes larger than specified.

Figure 18: Final machining of inner surface of Alloy 617B

The NCR by Saarschmiede referred to thePMA from 8 July 2003 which was carried

out by TÜV Rheinland and the material

specification no. 9.21602/00 189-70013

and -7018. The PMA refers to the creep

rupture strength in the temperature range

600 – 750°C of Alloy 617 with the speci-

fied chemical composition according to

VdTÜV data sheet 485. The deviation of

grain size is according to enclosures in e-

mail from Saarschmiede from 22 Novem- ber 2004.

Figure 19: Alloy 617B. with grain size number G= -3

48

8/13/2019 KINA25921ENN_002.pdf


The test reports by Saarschmiede have been accepted due to fulfilled mechanical technological tests

(impact test values) and an additional performed superficial fuse test which lead to satisfying results.

To ensure that even grain sizes with the specified value is achieved, two steps are needed for the forging

process:

- forging of the half of the pipe,

- reheating up to hot working temperature with a long holding time,

- forging of the second half of the pipe and

- solution heat treatment.

This resulted in:

- No further deformation on the half of the pipe which was forged in the first manufacturing step

- Two heat treatments with long holding times at a relatively high temperature caused no growth

of grain size instead of only one heat treatment.

In short, a long holding time at a rela-

tively high temperature during solu-

tion heat treatment and during the

heat-up phase to hot working tem-

perature ensures equal temperature

distribution between outlying area and

core area.

Figure 20 shows the forging of the

first half of an Alloy 617B pipe (OD

219.1 x 50mm) in two steps.

Figure 20: Forging of Alloy 617B (OD219.1 x 50mm)

To ensure that the materials with a non-conformgrain size do not pose a risk, TIG welding with-out filler metal (pure fusion of base materialFigure 21) and combined TIG/SMAW weldingwith filler at the materials were performed. The

stress caused by shrinkage of the weld seemedstrange given the range of the yield and shouldgive more precise information if the material canavoid risks like cracks in the microstructure.

Figure 21: TIG fusion welding without filler metal

49

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


Figure 24: New heat treatment method for Alloy 617 components

To provide better understanding of which grain sizes can be expected with a given grain size number G,

Table 22 gives an overview of the number G-7 up to G13. The values have been adopted from ASTM E

112 and EN ISO 643, since ASTM just shows the calculated values from 0 onwards ( – 1 according to

the EN ISO standard). EN ISO shows values from – 7 onwards. The logarithmic equation for calculating

the G-values differs slightly for both standards, the difference is ΔG=0.05. This difference is negligible,

since the grain size cannot be determined more accurately than half a unit, even under the most favour-able conditions.

Table 22: Grain size no G according to EN ISO 643 (G-7 ÷ G-1) and ASTM E 112 (G0 ÷ G13)

Grain sizenumber

G

Average graindiameter

(mm)

Grain sizenumber

G

Averagegrain

diameter(µm)

Grain sizenumber

G

Averagegrain

diameter(µm)

-7 4000 0 359 7 32

-6 2828 1 254 8 23-5 2000 2 180 9 16

-4 1414 3 127 10 11

-3 1000 4 90 11 7,9

-2 707 5 64 12 5,6

-1 500 6 45 13 4

T e m p e r a t u r e

Time

t1 t2

T1

T2

Regular solution heat treatment

- Faster heating up to solution

annealing temperature

- Longer holding time at solution

annealing temperature

51

8/13/2019 KINA25921ENN_002.pdf


All Alloy 617B tubes supplied by DMV

with dimensions of OD 44.5 x 10mm for the

superheater had to be mechanically peeled

to OD 41 x 8.5mm due to indications on the

outer surface of the tubes. The indications

were found during penetration testing after

bending of some tubes. Subsequently, some

straight tubes were tested too and showed

the same flaws. The indications have been

identified as overleaps with a depth of up to

1.6mm, which were not found in the lon-

gitudinal UT and Standard Eddy Current

testing since the morphology of the defects

were semi-circular or arrow-shaped. DMV

re-tested the surface of the peeled tubes

using a specific Eddy Current Circograph

together with UT to measure the wall thick-

ness. In addition, PT was performed on all

tubes to verify that the tubes were free of

indications.

Figure 25: and Figure 26: show the resultsafter dye penetrant testing before peeling.

Figure 26: Microsection of DMV 617 SH-Tube

The right side of the SH outlet header was manufactured by Babcock & Wilcox Volund ApS for BWE.

The left side was manufactured by Alstom Neumark, where the two parts were also welded together.

The high toughness of the Alloy 617B material required special tools for machining and low cutting

speeds which resulted in a narrow time window for the whole project. Because of this, the time frame

for the erection plan was also affected (see WP5 for more information).

Figure 27: Superheater outlet header

Figure 25: Indication of DMV 617 SH-Tube

52

8/13/2019 KINA25921ENN_002.pdf


In the manufacturing of the SH inlet header, the dissimilar P91/Alloy 617B stubs for the Alloy 617B

header were welded with P91 filler metal, whereas the design specified the Alloy 617B filler metal. For

this reason all stubs had to be reconnected with the correct filler metal.

A WPS and related WPQR were developed for welding

of Alloy 617B. The weldability of the material was found

to be good for TIG welding. However, for manual metalarc welding a high proportion of the weld material had to

be scrapped due to problems with the cover falling off

during the welding process or drops forming at the elec-

trode ends that switched off the arc.

For the NDT it was found that due to the austenitic mate-

rial, the heavy attenuation and an unacceptable signal to

noise ratio ultrasonic testing could not be used as origi-

nally specified. For this reason a radiographic examina-

tion was made instead for volumetric testing. Dye pene-

trant testing was used for surface defects. All welds were

found to be acceptable.

It is believed that with special equipment and specially

trained operators it should be possible to use ultrasonic

testing for this type of components.

Overall, the scope of manufacturing included the inlet and outlet headers made of P92 for the evapora-

tor and for the superheater, which were made of Alloy 617B. The thick-walled parts also included the

spray attemperator and the steam piping, including a branch piece (Alloy 617B).All components were manufactured and documented in accordance with the WPQRs, the welding lists

and the design approval by the Notified Body. Figure 27 – 27 show some of the manufactured parts of

the thick-walled components.

Figure 29: Evaporator outlet header (P92)

Figure 28: Spray Attemperator

53

8/13/2019 KINA25921ENN_002.pdf


As described in the “Evaporator” section, PWHT had to be applied to 13CrMo4-5 and P92 materials of

the headers.

Manufacturing of the evaporator panel Figure 30 was carried out in accordance with VdTÜV data sheet

451-68/1 (manufacturing of gas-tight welded tube walls). The butt welding, tube-fin welding and fin-fin

welding were performed according to the required standards and documented for future purposes. No

post-weld heat treatment was applied to the welds of T24 since the material has been developed in sucha way that no heat treatment has to be done after welding.

The circumferential welds of HCM12, either homogenous or dissimilar, had to be heat treated to keep

the hardness low and achieve a good ductility. This also applies to the inlet and outlet headers

(13CrMo4-5 and P92) and the associated stub welds.

Figure 30: Manufactured panel with T24/HCM12/Alloy 617B.

Issues arose at the Alloy 617B tubes. After PT,

surface defects were found Figure 31. The fail-

ures that occurred (130 in total) were ground

since the evaporator part has already been manu-

factured. The remaining wall-thickness wasmeasured but since some of the tubes had a thin-

ner wall thickness after grinding than permitted

(for a design calculation of 100,000h), one quad-

ruplet (tubes no. 15 ÷ 18) had to be replaced.

These were the same flaws as for the superheater

tubes.

Figure 31: Indication found after PT

54

8/13/2019 KINA25921ENN_002.pdf


Different scopes of NDT have been followed in the manufacturing and erection of the components. The

following table gives an overview of the scope of testing of the different materials and combinations.

Table 23: NDT conducted on CTF parts

Evaporator +Superheater

Circumferential butt welds ferrit-ic/martensitic alloyshomogenous / dissimilar

RT100%PT25%Hardness testing per heat treatment lot

Circumferential butt welds nickel based alloys homogenous/dissimilar

RT100%PT100%

Tube-Fin welds of ferritic and nickel based alloys

PT100%

Thick walledcomponents

Circumferential butt weld ferritic al-loys

100% UT100% MTHT per heat treatment lot, max 10%for furnace heat treatment

Circumferential butt weld nickel basedalloys

100% RT after root pass with first

filler pass100% PT after each layer

Circumferential butt weldferritic/nickel based alloys

UT100%MT100%HT per heat treatment lot, max 10%for furnace heat treatment

Attachments (e.g. stubs) to headernickel based alloys

PT100% after each layer

Attachments (e.g. stubs) to headerferritic alloys

MT100%HT per heat treatment lot

As it can be seen from Table 23, NDT of welded parts is time consuming. Especially the time frame for

manufacturing of circumferential butt welds of the thick-walled pipes needs to be considered. Since RT

cannot be applied for the most part to these components, ultrasonic testing is the right choice. Only for

ferritic materials, however, since nickel based alloys show an anisotropic behaviour. Penetration testing

during welding of Ni-based alloys Figure 32 means that after each layer the metal temperature has to be

cooled down to a level lower than 50°C before applying the dye penetrant. Since the dwell time in-

creases from ferritic over austenitic materials to nickel based alloys this adds to the total time.

55

8/13/2019 KINA25921ENN_002.pdf


Figure 32: PT testing of Alloy 617B V-weld (OD219.1 x WT50mm) and while welding withTIG/SMAW

2.2.4 Conclusions

2.2.4.1 GeneralThe evaporator components were manufactured without major deviations or non-conformances for the

T24 and the HCM12 part. Material properties were kept as specified, with the only exemption being as

mentioned in 1.2.1 chapter for two pipes and one valve body block forged in Alloy 617B type material

where the grain size deviated from the specified grain size.

Welding procedures, quality assurance and quality control proved to be ready for processing Alloy

617B thereby attaining quality welding seams of very few NDT indications. Only one tube sections at

the evaporator part needed to be replaced after testing. This replacement came with indications of the

Alloy 617B tubes which had to be gr ound to an acceptable wall thickness after indications during

penetration testing. The Alloy 617B superheater tubes, showing the same flaws (overleap/tears), were peeled to a smaller diameter (from OD 44.5mm to OD 41mm), still maintaining the minimum

calculated wall thickness for an operation time of 100,000h.

At the time of planning and manufacturing until erection of the CTF, the relaxation cracking

phenomenon of Alloy 617B was not known and therefore no post-weld heat treatment (stabilising

annealing) was applied. Also it was found to be really time-consuming and prohibitive to manufacture

the thick-walled pipes as massive forged pieces with subsequent drilling in order to achieve the final

dimensions. Since manufacturing was carried out according to state of the art for conventional power

plants in 2004, it proved insufficient and the gained knowledge has to be transferred to future

applications.


Alloy 617B is part of subsequent test rigs, e. g. HWT I in Mannheim. Pipes made from Alloy 617B will

also be installed in the subsequent test plant, HWT II. Thus, testing of the material goes on either as

tube of heating surfaces or boiler external piping. All related test programmes will add new insight to

the COMTES700 test results and form the next steps towards implementation of a 700°C technology.

The knowledge learnt about relaxation cracking of welds of thick-walled Alloy 617B pipes transfers to

HWT II and to other projects such as ENCIO. At ENCIO, aged pipes from the CTF will be used to gain

knowledge about the cracking behaviour under different heat treatment and post-weld heat treatment

conditions (stabilising annealing at 980°C for 3h).

56

8/13/2019 KINA25921ENN_002.pdf


Wyman-Gordon proved in 2009 on behalf of Siemens that it is indeed possible to extrude thick-walled

Ni-based alloys such as Alloy 617B, where two pipes had been manufactured.


Especially, knowledge gained about manufacturing of thick-walled components which are made of

Alloy 617B helped avoiding repetitions of a grain size enlargement due to wrong heat treatment proce-

dures. The result also needs to be expanded to cover austenitic materials in general. Moreover, the

knowledge and skills obtained in relation to methods and techniques regarding how to join such really

divers materials for the heating surface of the evaporator are among the major achievements.

It proved insufficient to use longitudinal UT and Standard Eddy Current testing to find the overleaps or

tears of the Alloy 617B tubes, since the morphology of the defects were semi-circular or arrow-like.

According to DMV, the flaws can be detected using a specific ET Circograph by means of a rotating

probe where the UT testing still has to be maintained to measure the wall thickness.

As regards the use of thick-walled pipes, either austenite or Ni-based, the NDT takes a long time to

perform and also the manufacturing of these components is time-consuming. These aspects have to be

considered for future planning to organise the time schedules for manufacturing of the materials, manu-

facturing of the components and erection. Improvements of manufacturing methods, including welding,

are in the pipeline and will pave the way for commercial application. The cracking of thick-walled Al-

loy 617B welds mentioned above and the need for heat treatment are also achievements.

57

8/13/2019 KINA25921ENN_002.pdf


2.3 WORK PACKAGE 3: VALVES & MEASURING DEVICES


Design of valves

Proof of concept

Order of materials

Manufacturing of valves

Qualification through inspecting bodies (e.g. TÜV)

Monitoring of valves

One of the main tasks of work package 3 (WP3) was to define the design parameters and the minimum

size of a HP-bypass valve for later scale-up. The following parameters were fixed:

Operating pressure higher than 200bar

Operating temperature equal to 700°C

Possible material selection for valve body and valve stem is Alloy 617B. The approval of valves

for 700°C by the authorities is expected to be limited to the mechanical design

Design steam mass flow 25 to 30kg/s for the HP-bypass


In the course of the Comtes700 project the valve-manufacturers delivered 182 valves such as HP-

bypass valve, steam control valve, start-up valve and gate, globe, check valves as well as a safety valve

for the application in the test rig installed in the coal-fired power plant Scholven, unit F. All valves weredesigned according to its design data respectively and were disassembled for further detailed visual

inspections after finishing the operation as scheduled.

The whole production procedure including the design and manufacturing of the valves could be per-

formed according to the initially planned way. No major deviations between planned and done work

could observed. Nevertheless, especially, Alloy 617B related design and manufacturing requirements

had to be taken into account.


At the start of the project, two valve manufacturers were chosen. They both were the partner companyin the AD700-project and exhibit required expertise for making high-performance valves in the temper-

ature range up to 700°C. In the COMTES700 project, it was decided to share out the manufacturing

regarding to valve types. It was confirmed by the valve manufacturers the using only 50% of the valves’

mass flow capacity would be sufficient to make all valve tests reliable and all conclusions regarding the

operational behaviour of the valve could be drawn. Therefore, in order to reduce investment costs of the

whole Component Test Facility (CTF) in Scholven, it was decided to design the CTF for a realistic

steam flow of 12kg/s. The detailed start-up and operation of the CTF can be found in the report “Oper a-

tion” for work package 6 (WP6). The valve manufacturers considered the following test runs as neces-

sary:

58

8/13/2019 KINA25921ENN_002.pdf


(i) 100h-500h operation time at full load with the bypass station serving as a control valve. During

this operation period the valve was closed occasionally to check the attrition and shut-off prop-

erties.

(ii) 10-20 times operation in bypass mode at max. steam conditions at valve inlet in order to have a

high level of the heat up and cool down rate.

(iii) The bypass station should be installed in the test rig in such a manner that it would be possibleto check the valve and exchange the valve body seat, if relevant, during normal operation of the

host plant is possible.

(iv) Both valve manufacturers agreed that one HP-bypass valve is sufficient for test purposes.

(v) For integration of the bypass station in the test rig the following details were considered: (1)

tempering of the steam inlet and outlet connection lines, (2) tightness of the valve could be

monitored during operation by temperature increase at the outlet or by structure (3) Pneumatic

drive is possible due to the small valve size and cheaper.

(vi) Steam velocity in the HP-inlet piping up to 50 m/s, spray cooler separate, downstream of the

valve and insulation same as piping.

2.3.3.1 Design

The valve design showing the basic design elements and taking load arising from the 700°C steam tem-

perature into account is given in the following. The design of the valves followed the basic design pro-

cedures and rules for high pressure and temperature valves. All relevant codes (e.g. the PED directive,

97, 97/23/EG; TRD 110, and also EN 12516-2:2004, Industrial valves – casing strength – Part 2: Calcu-

lation method for steel valve casings) have been applied. Due to the comparatively high design tem-

perature and design pressure conditions of the steam (T ~ 710°C, p ~ 215bar) almost all pressure retain-

ing parts had to be made of the Ni-based Alloy 617B. The other valve parts, which were not highlytemperature loaded, were made of alloyed steel.

A main design issue was the economical use of the high-grade and therefore high price of the Alloy

617B. From a structural strength point of view the valve body should be made of a single piece without

any weld joints. This monobloc design would lead to comparatively increased material and machining

costs. Therefore, it was decided to make use of pre-manufactured parts as far as it seemed possible and

cost effective. In consequence welding of Alloy 617B was a main issue in connection with the valve

design as well as during manufacturing. The valves have been manufactured according to PED, module

H without approval by a notified body.

2.3.3.1.1 Gate valvesThe valve bodies of the gate valves were made in a way as shown in Figure 33. The body (1) was a two-

part design, made of a lower valve body part and an upper valve body neck. Both parts were made of

forged rectangular raw pieces and connected by a weld joint. For joining of the body parts butt welding

was chosen. For isolation of the steam flow a two-piece wedge plate design (4) was chosen. To avoid

overpressure within the neck the valve has an overpressure nozzle (3). The overpressure was controlled

by a rupture disc device. The seat rings (2), the wedge (4), the cover (6) and the segment ring and thrust

ring were also made of Alloy 617B. The sealing surfaces on the seat rings and wedges were cladded

with Stellite 6 material.

For cover pressure sealing most of the gate valves had graphite gaskets with stainless steel caps (7). In

one case it was decided to use a silver plated metallic gasket. In order to have a very tight seal as well as

sufficiently low friction and wear between the stem shaft and the cover, it was decided to use graphite

59

8/13/2019 KINA25921ENN_002.pdf


packing rings (9) for all gate valves. In consequence the packing temperature had to be reduced by pro-

longation of the stem/cover protruding. Additionally, the stuffing box at the upper end of the cover was

designed with cooling gills at the outer surface. For the stuffing box design a so-called safety stuffing

box gland (8) using disc springs in order to control the packing pressure was chosen.

The stems of the gate valves (5) were made of the high temperature resistant steel 1.4922. The surface

of the stem shaft was treated in an HVOF (High Velocity Oxygen Fuel) thermal spray process with Cr-carbide plating and diamond grinding. This procedure was chosen for the stem material and stems sur-

face treatment because of sufficiently low friction and wear between the stem surface, the gland pack-

ing and the guiding area within the cover and since this was considered to be the predominant technical

requirement.

The connections between the valve body and the actuator were made of 17%-Cr steel. A rod design (10)

was chosen for the valve yoke in order to minimise the heat transfer towards the actuator. Depending on

the valve purpose different actuators such as hand-wheel, electric and hydraulic actuators were mount-

ed.

Figure 33: Cross sectional drawing of gate valves GH 251 and DN125 (KKS no. 26NA76S004)

2.3.3.1.2 Globe valves

The valve bodies of the globe valves were in a monobloc design as shown in Figure 34. The inlet and

outlet geometries of the valve body (1) as well as the geometry of the valve seating are comparable to

those of the VA500 standard HP-pressure valve series. The seating surface of valve body was clad with

Stellite 6. The stem seating surface did not have a hard surface layer. The valve neck was prolonged in

order to reduce the temperature of the stuffing box. Additionally, the valve body was provided with

cooling fins between the lower valve part and the stuffing box, so that graphite packing rings could be

used for the stuffing box sealing (4). The Belleville spring loaded stuffing box gland bolts (3) made sure

that packing shrinkage could be compensated for and thus leakage of the stuffing box was avoided. For

almost all globe valve stems high temperature resistant austenitic steel was used. Only in two cases a

17%-Cr steel was applied. The stem seating surface had no surface hard cladding. Valve yokes (5) were

made of F91 steel and manufactured by die forging. For stem lifting a standard valve head (6) according

to DIN EN ISO 5210 was used.

60

8/13/2019 KINA25921ENN_002.pdf


Figure 34: Cross sectional drawing of the globe valves VHA510 and DN50 (KKS no. 26NE22S521)

2.3.3.1.3 Safety relief valve

In order to avoid system overpressure a spring loaded safety relief valve with an additional pneumatic

load actuator was installed in the test loop. This valve was designed according to TRD 421, AD 2000

and DIN EN ISO 4126-5. A cross sectional drawing is shown in Figure 35. The pneumatic load actuator

(10) is controlled by the control unit STE4.

The Alloy 617B was utilised for the temperature loaded, pressure retaining inlet nozzle only (1). In

order to keep the costs low the upper blow-off body (2) was made of steel 1.4910. The inlet nozzle was

inserted into the upper valve body by a thread. The tightness between inlet nozzle and blow-off body

has been obtained by a weld joint. A negative effect of the mismatch of the thermal elongation of nozzle

and valve body materials was reduced by a ring-shaped groove at the lower side of the valve body. AFEM stress analysis was made in order to analyse the stress distribution between the inlet nozzle and

the upper valve body part. This showed that the stress level was within an acceptable range. The guid-

ing surfaces at the lifting support (5) and the guide (4) were clad with Stellite 6. Inititally, disc (3) was

made of the austenitic material 1.4980. As the project progressed it was decided later to change over to

a disc made of Stellite 6.

61

8/13/2019 KINA25921ENN_002.pdf


Figure 35: Cross sectional drawing of safety valve SOH (KKS no. 26NA77S094)

The intermediate flange (6) was given a particularly long design so that non-permissible overheating of

the pressure spring (8) could be avoided, thereby ensuring correct functioning of the safety valve. Dueto the especially long intermediate flange (6), the elongation of the screw bolts was relatively long. This

is advantageous when the valve opens and escaping steam heats up the upper part of the valve. The

spring material was 1.8159, which is a standard material for spring loaded safety valves in conventional

power plants.

2.3.3.1.4 Swing check valves

The basic design of this valve is similar to the design of the steel swing check valves used for example

in conventional steam power plants. Design rules and calculations were applied in the same manner.

The seat surfaces of the valve body (1) and on the disc (2) had no surface hardfacing. The purpose of

seat surface hardfacing for the steel swing check valves was to avoid corrosion. Due to the sufficientlyhigh corrosion resistance of Alloy 617B seat cladding was not needed.

The main design features of the check valve can be seen in Figure 36. The bearings of the swing axis

shaft (4) are positioned within a holding ring. The holding ring, disc (2) and disc lever (3) can be dis-

mounted as a hole. Thus a protruding of the shaft outside the valve body is not needed. For covering of

the valve body a pressure valve cover (7) was used. All parts of the valve were made of Alloy 617B

except for the shaft, screws and pins. For pressure sealing a graphite gasket with stainless steel caps was

used (7).

62

8/13/2019 KINA25921ENN_002.pdf


Figure 36: Cross sectional drawing of swing check valve KRH400 and DN80 (KKS no. 26NA76S508)

2.3.3.1.5 Check valves

The in-principle design of this high temperature check valve is similar to the design of the check valves

used in conventional power plants. A cross sectional drawing is given in Figure 6. The main valve parts

such as valve body (1) and valve cover (4) were made of forged raw blocks. All valve parts were madeof Alloy 617B, except for the guiding bush (8), bolts (5) and nuts (6). Bolts and nuts were made of a

high strength austenitic steel material. For cover sealing a bolted bonnet with a camprofile metallic

gasket (3) was chosen.

Figure 37: Cross sectional drawing of check valve VR500 and DN50 (KKS no. 26NE22S522)

2.3.3.1.6 Start-up valve, control valve and HP-bypass valve

All three valves were designed as angle valves (see Figure 38-40). The Stem and the seat were coated

with boron to achieve a high temperature anti-abrasive protection. To avoid burning out (oxidation) of

the graphite sealing material, the stuffing box was placed at a suitable distance from the bonnet. Therewere also cooling vanes mounted around the stuffing box. Distance and active area as well as the shape

of the cooling vanes were tested in a special test rig (1:1 scale). The test rig was heated to 710°C with

electric heaters. The sealing gasket between bonnet and body were metal gaskets in a "C" profile shape.

The seats were screwed in and the plug consisted of a cylindrical part with an additional cage. To avoid

overheating of the coupled actuators, a yoke assembly was designed in order to achieve a low heat

transfer rate. High mechanical stability was also considered. The valves had one or two separate

buttweld ends. In the outlet of the HP-bypass, there was a spray attemperator with an integrated water

injection nozzle. Two of the valves had electrical actuators; one valve (HP-bypass) had a pneumatic

actuator. The design work was carried out with the aid of numerical methods like Computational Fluid

Dynamics (CFD) and the Finite Element Method (FEM) to optimise temperature distribution, flowdynamic and wall thickness.

63

8/13/2019 KINA25921ENN_002.pdf


Figure 38: Left: Cross sections of start-up valve (KKS no. 26NA76S002); Rright: control valve (KKSno. 26NA76S006)

Figure 39: HP-bypass valve (KKS no. 26NA77S003)

64

8/13/2019 KINA25921ENN_002.pdf


2.3.3.2 Commissioning and Manufacturing

Most of the construction material needed was made of forged raw material, Alloy 617B. Alloy 617B

was purchased from the company Saarschmiede. The valves were manufactured at the Sempell or HO-

RA manufacturing sites.

Due to the mechanical and machining properties of Alloy 617B the manufacturing processes were ra-

ther challenging. The high mechanical strength of the material, especially at room temperature, and thehigh ductility require dedicated machining parameters. Special cutting tools and cutting edge geome-

tries were needed. Although the machining of Alloy 617B was optimised machining time lasated be-

tween 4 and 8 times longer compared to machining of high grade steel materials. Also cutting tool wear

and therefore costs were considerably higher. The available machining and tooling technology was a

prerequisite to manufacture all valve components according to the quality requirements and within the

pre-calculated machining times and costs.

The welding of Alloy 617B was an important major issue. It is known that Alloy 617B tends to generate

hot cracks during welding. Therefore excessive heat input should be avoided and submerged arc weld-

ing with its high heat input could not be applied. The butt weld joints were made by multilayer welding

using TIG welding and manual electrode welding. In order to avoid overheating an interpass tempera-

ture of 100°C could not be exceeded. Additionally NDT of the root, 1/3, 2/3 of the weld volume and the

top layers had to be carried out. At the end of the welding process all buttwelds were X-ray tested. The

butt welding and hard-facing procedures of Alloy 617B were qualified by performing welding proce-

dure qualifications procedures. It should be mentioned that the filler material type covered electrode

(UTP6170 Co mod.) and TIG welding rod (UTP A 6170 Co mod.) were used for the welding proce-

dure.

2.3.3.3 Function and material behaviour under 700°C temperature conditions

The experiences gained during the operation of the plant together with the results of the visual inspec-tion and test of movability of the valves showed that the functionality of the valves was given. After

termination of the test loop operation period extensive and detailed investigation of a variety of valves

were carried out.

The material investigations of the gate valve bodies revealed that two gate valves, which were exposed

to high thermal transients, showed cracking within the weld between lower valve body and valve body

neck. Cracks were limited to the welded area. Uninfluenced base material showed no cracking (Figure

40).

In this context the operational load of the valves has to be taken into account. The numbers of cold and

warm starts as well as the temperature transient levels seem to exert essential influence on crackingwithin the welds. Additionally, the residual stresses within the weld joint resulting from the welding

process have to be considered.

Beside the investigation of the integrity of the pressure retaining valve parts, it was of interest the con-

dition of the isolating components like wedge plates, discs, seats and gaskets. Hardfacing material Stel-

lite 6 was used mainly for surface cladding of metallic seating surfaces as well as for guiding surfaces.

The investigations showed that in general the used hardfacing layers tolerate the operational loads. Re-

markable abrasive wear of seat rings and wedge plates had to be noticed only in the main steam isola-

tion valve (Figure 41).

At first view it might be concluded that wear resistance of Stellite is not sufficient for 700°C conditions.For a definite evaluation of the wear it has to be taken into account, however, that the main steam isola-

65

8/13/2019 KINA25921ENN_002.pdf


tion valve was exposed to extremely high temperature transients and high temperature differences dur-

ing operation. Thus deformation of the planarity of the seat ring and the wedge must be assumed. If

applicable and where appropriate for flow isolation under high duty conditions, globe valves are rec-

ommended.

Another important result was that graphite and metallic gaskets can be used successfully in 700°C ap-

plications. Neither the graphite nor the metallic gaskets which were chosen for pressure sealed coverswere inappropriate. Especially the chosen graphite gaskets with metallic edge protectors showed no

oxidation or corrosive degradation. The same result was achieved for graphite gland packings.

Figure 40: Overview of casing neck weld of gate valve. Red arrows show cracks after dye penetranttesting (KKS no. 26NA76S004-01)

Figure 41: Adhesion damage of seat ring (KKS no. 26NA76S004-01)

As regards the design of the HP-bypass valve there were some problems with the C-ring gasket at the

bonnet sealing. Due to leakage problems the C-ring gasket had to be replaced in a semiannual overhaul.

To avoid similar problems in future, improvements in other designs were made by Hora. The wear of

the internal parts was within normal variations, without any noticeable issues.

After finishing the operation of the CFT, the HP-bypass valve, the steam control valve and the start-up

valve were disassembled for further investigation. These control valves were made from Alloy 617B.

Each valve has been dismantled in HORA’s workshop. The dismantling of the valves piece-by-piece

took place without any problems. The trims and the screws were in good or even very good condition,

given the operational lifetime. The surface of the seats of the steam control valve and the start-up valve

66

8/13/2019 KINA25921ENN_002.pdf


were well preserved. Only the surface of the seat of the HP-bypass valve showed slight but reasonable

indications of erosion. In the lower area of the throttle bores of the start-up valve and HP-bypass valve,

indications of erosion were found. Also, the control edge of the start-up valve piston showed slight

erosion, and the control edge of the HP-bypass valve showed clear obvious indications of erosion

(Figure 42). The safeguarding welded joints of the perforated discs in the outlet of the HP-bypass valve

showed circular, visible cracks following the entire welded joint. At the opposite outer surface of the

valve body there seemed to be visible crack-like indication at valve body surface. In addition to the

visual inspection a dye penetration test was also carried out in the presence of HPE, E.ON and Hora.

The dye penetration testing showed no indications of cracks.

Figure 42: Seat of the HP-bypass after 22,000 hours of operation

In view of these findings the parties jointly decided to focus further investigations on the HP-bypass

valve only. Therefore, after completion of the NDT, the valve body was cut into pieces that already

showed visual crack formation.

The two parts cut out of the HP-bypass valve were subjected to radiographic testing (source: Linear

Accelerator). Figure 43 shows the HP-bypass before the detailed investigation. The first part was the

one with the crack-like indications and the second part was the one with the welded end of the inlet

nozzle (seam “A”). The radiographs were evaluated by a HORA expert. Neither the radiographs of the

first part nor the one of the second part showed any indication of cracks. The crack-like indications atthe outer surface of the valve body could not be detected by radiographs as cracks. Figure 44 shows the

cross section of the HP-bypass valve with cracks, at thread pitches in the section (red arrows) and at

transitions of the cross section (blue arrows). The light optical micrograph gives a detailed view of a

crack propagated from a thread root.

67

8/13/2019 KINA25921ENN_002.pdf


Figure 43: Dissmanteld HP-bypass valve body for detailed investigation

Figure 44: Cross section of the HP- bypass valve (seam ‘A’) with cracks (arrows) and the appropriatemicroscopic view on a crack propagated from a thread root

Before testing the surfaces of the welded end, the part was machined in order to obtain parallel surfaces

which are a precondition for ultrasonic testing. The mechanised ultrasonic testing was carried out at

Müller und Medenbach GmbH.

After 22,000 hours of operation at intermittent temperature stresses of up to approx. 720°C, the HP-

bypass valve was disassembled when the CTF was shut down, and the valve was subjected to a

thorough examination. The examination included, amongst others, NDT of the material. On completion

of the NDT process the valve body was cut in the planes that already showed visual crack formation.

This revealed that the outlet of the HP-bypass valve in the section of the safeguarding welded joints of

the perforated discs was particularly conspicuous (viewed in the direction of the flow, it is located

upstream from the spray evaporator nozzle). It should be noted that sharp-edged tools must be avoided

for the machining of the hardened surface.

The hardness detected in the area with visual cracks decreases continuously from approx. 280HV10 forthe outer diameter to as little as 260HV10, until approx. 12mm before the inner diameter and then it

seam ’ A’

68

8/13/2019 KINA25921ENN_002.pdf


increases for the remaining cross section of the inner diameter to 350HV10 (Figure 45). This steep but

steady increase in hardness over a distance of 12mm cannot be explained by hardening during the shape

cutting process, since the significant increase in hardness is found to take place all through the cross

section of the inner wall over a distance of 12mm. A conservative explanation could be that this effect

is the result of the stress induced γ’-precipitation.

Figure 45: Hardness gradient in valve cross section

2.3.4 Conclusions

2.3.4.1 General

The CTF suffered no major outages due to valve problems which mean that the main function of the

valves in the CTF lived up to their intended purpose. The process demonstrated that effects could be

gained from the welding procedure and the heat treatment procedure before and after welding.

The results and experience gained in the project show that the chosen basic valve designs, including the

additional 700°C related design features, were adequate to tolerate the loads arising from the operation.

In addition to this, it is also evident from the project that weld joints of thick-walled components under

certain conditions tend to develop relaxation cracks. A possible counter measure that can be taken to

overcome this is to apply a post-weld heat treatment procedure. Another solution is to design a mon-

obloc valve body.

The valve sealing components such as graphite gaskets and packings, were sufficiently resistant to the

extremely high steam temperatures. Actuators could be applied to the valves so that no heat induced

damage could take place.

Machining and welding of Alloy 617B are complicated processes which required machining times 4-8

times longer compared to steel materials. Therefore, it would be reasonable for future R&Dprojects to

include investigations of this in order to reduce the manufacturing efforts. Especially the manufacturing

processes which led to near net shape raw pieces, such as forging or Hot Isostatic Pressing (HIP), are of

topical interest. One other important point in future R&D activities could the thermomechanical behav-

iour of Alloy 617B components.

69

8/13/2019 KINA25921ENN_002.pdf


There are still some concerns about the operating safety of thick-walled Alloy 617B valves that are

frequently subjected to operationally unavoidable heating and cooling cycles. Careful operating

practices, limited design adaptations, technological processes (heat treatment) or a combination of

various measures should be taken into consideration with respect to the application of a HP-bypass

valve in future 700°C applications.

It has to be investigated if the heat-up and cool-down rates during start-ups and shut-downs can bedecelerated in order to maintain an operational pattern suitable for the materials/components that are

designed accordingly. It should be investigated if the Alloy base material can be improved in order to

reduce the operations-related tendency to hardening by employing metallurgic and/or technological

measures. Redesign measures have to be taken, such as reduction in material accumulation (which

cannot be avoided completely), the use of the largest possible cross-over radii and smooth transitions

from one wall thickness to another, in order to prevent premature crack formation.


After the completion of the design and manufacturing of the COMTES700 valves the two valve manu-

facturers designed high performance valves made of Alloy 617B for the project 50+ performed by

E.ON. Here the net efficiency climbed up to 50% in a coal-fired power plant mainly by developing new

materials. Future projects would benefit from experience acquired in the framework of the

COMTES700 project.

The manufacturers HORA and Sempell plan to deliver a variety of valves for later projects, such as

ENCIO (partially funded by RFCS) and GKM HWT II (partially funded by BMWi (German ministry)),

will focus on the 700°C material main issues such as proof of thick-walled components, investigations

to qualify pipe manufacturing, welding procedure standards, repair and erection concepts and modelling

with validation of material and component behaviour.


Based on the existing know-how from the manufacturing of HP and high-temperature valves additional

dedicated knowledge of the design and manufacturing of valves for 700°C applications was developed

in the process. Assumptions and/or simulation results which had been carried out during the design

phase could be proven by the experience and the results from the test rig operation.

70

8/13/2019 KINA25921ENN_002.pdf


2.4 WORK PACKAGE 4: C & I


1. Select and develop instrumentation to monitor temperature, pressure and flow and control of the

CTF and more sophisticated parameters for future plants with the highest accuracy.

2. Implement advanced closed loop control and safe operation. Work out strategies for monitoring

and detecting wear and tear of components.

3. Collecting data, storage and visualisation for fault detection, stress and consumed lifetime calcu-

lation.

4. Control flow and temperature within a narrow band corresponding to future boiler operating con-

ditions.

The objectives required the application of hook-up equipment which was able to measure steam and

pipe material temperatures of 700 °C or more and tolerate the temperature differences during start-up,

shut-down and load changes.For the engineering and delivery of the C&I equipment the following tasks had to be covered:

o process engineering, in cooperation with other WPs,

o development and adaptation of measuring devices and corresponding hook-up, e.g. temperature

pockets,

o workshop tests,

o integration into DCS and boiler protection system,

o supervision and coordination of erection ,

o commissioning of the CTF including optimisation of the C&I system,

o data storage, fault detection, data export for evaluation in WP7,

o development of risk assessment and development of field test procedures and

o clarifications with the Notified Body.

Lessons learnt and suggestions gathered for future development were carried on with the design part-

ners involved.

2.4.2 Comparison of Initially Planned Activities and Work Accomplishment

2.4.2.1 Separate Projects integrated During Project Execution

Turbine Control Valve

Additional temperature measurements and an operation programme had to be integrated into the DCS

System. One temperature gradient limitation control loop was added in the operation programme. A

silencer had to be installed afterwards for the turbine control valve.

“Fixed Point” Temperature Measurement System

Sheath tubes had to be welded to the pipes to enable tests at approx. 600°C and 700 °C. For control of

the measurement system a separate PC was installed in the control room.

71

8/13/2019 KINA25921ENN_002.pdf


2.4.2.2 Commissioning Disturbances

Plant optimisation of the control loops needed longer time than anticipated. Tests were often only

allowed during start-up and shut-down sequences of the steam generator.

Torque set points from control actuators needed to be readjusted. Lubrication intervals had to be

reduced. The types of grease had to be harmonised.


WP4 stipulated that:

the control system was to be integrated in the normal control system of the plant;

data logging could either be implemented in a separate system or using the plant’s data acquisi-

tion system;

data to be measured were:

o flow,

o pressure,

o steam temperature,

o inner and outer wall temperatures together with mid wall temperatures at strategic posi-

tions,

o hanger positions; and

data were to be measured and recorded during the whole operation period.

The scope comprised definition of relevant measuring points and control strategies (in connection with

WP1) and specification and procurement of transmitters and control equipment.

The host facility assumed the responsibility for supervising the erection and commissioning.

Definition of measurements and means to implement the control strategy can be seen from the P&I

diagram of the CTF.

The control strategy was finally expressed in the operating manual which refered to the engineering

documents

o P&I Diagram,

o List of Measuring Points,

o

List of Electrical Consumers,o System Description and

o Functional Plans and Control Specifications.

These documents served as input for the detailed description and specification of the operation and

control of the CTF.

72

8/13/2019 KINA25921ENN_002.pdf


2.4.3.1 Engineering of Hook-up and Special Components

For this WP the project team had to select and develop instrumentation to enable temperature, pressure

and flow monitoring and control of temperature and minimum steam flow in the CTF with the best ac-

curacy that is technically possible. The major challenge was related to the very high temperatures in the

context of state-of-the-art power plant technology.

The following high temperature measurements were required for the CTF:

o 13 thermocouples in Alloy 617B protective tube for 705 °C operation temperature,

o 8 thermal elements for wall differential temperature for 705 °C operation temperature,

o 30 sheath tube thermocouples for wall panel for 705 °C operation temperature,

o 44 sheath tube thermal elements for wall panel for 620 °C operation temperature,

o 3 pressure transmitter to measure the operation pressure in 705°C pipes and

o 2 pressure switches to measure the operation pressure in 705°C pipes for control of the high tem-

perature safety valve had to be installed.

Several other sheath tube thermal elements were installed to monitor valve design tests.

Remark:

The thermocouples which were to be applied for operation in the range of up to 705°C had to be de-

signed for 745°C.

2.4.3.1.1 Temperature measuring

Type S thermocouples with accuracy class 1 (tolerance +/-1K in the range 0 – 1,000°C) were selected to

measure the temperature.

The associated thermo wells were made from piping material. Gold plates were mounted at their tips tomaximise the thermal conductivity for fast and accurate readings.

The electrical insulation material chosen for the temperature sensors in the CTF was MgO. For future

applications above 705°C it is recommended to select materials with higher electrical resistance against

temperatures above 700°C.

As of the investigation date, no manufacturer insulating material has been found of that can live up to

all requirements appropriate for this test plant. Therefore, mats made of various material coats had to be

used to achieve a higher insulation thickness and thereby meet the challenging requirements of the de-

sign and engineering of a steam generator running above 700 °C.

Further investigations focused on HfO (Hafnium oxide) materials instead of MgO for future projects.

Due to the increased heat conduction at higher parameters, the length of the protection pipes had to be

extended to ensure that temperatures at the connection heads were less than 120°C (see Figure 46).

73

8/13/2019 KINA25921ENN_002.pdf


Figure 46: Outer temperature profile of a thermocouple

From left to right Figure 46 shows extended protection pipes of the measuring device. The change in

colours clearly indicates the temperature drop of the external parts as they protrude from the thermal

insulation. The hot spots at the connection head are caused by thermal conduction through the thermal

sensor.

Floating cold reference point compensation was used to avoid additional inaccuracy which can arise

because of the compensation cable. The temperature was calculated using the digital control system.During project design phase a dedicated high quality transmitter was not commercially available. High

speed calculation consumes remarkable resources from the control system.

Zoom technology was utilised to reduce the digitalisation fault to less than 0.2K for appropriate spray-

water control.

The response time setting of the spray-water control was optimised to a loop time below 20 sec under

full load conditions.

74

8/13/2019 KINA25921ENN_002.pdf


2.4.3.1.2 Pressure Measuring

A tailor-made hook-up design (see Figure 47) had to be carried out for pressure measurements under

high temperature conditions. Standard design materials were not available. Special solutions were re-

quired for every single component. Figure 48 shows a drawing of the pressure gauge valve Alloy 617B

and Figure 49 shows the valve installed for permanent testing in the CTF.

Figure 47: Pressure measurement equipment as used for the CTF. First shut-off valves (see WP3Valves), impulse pipes and transmitter shut-off valve made of Alloy 617B

1) Pressure tap welded to steam pipe Alloy 617B

2) Primary insulation valves Alloy 617B3) Impulse pipe Alloy 617B4) Pressure gauge valve Alloy 617B (see Figure 48)5) Pressure gauge support6) Sealing washer7) Transmitter8) T-connect

9) Drain valve

75

8/13/2019 KINA25921ENN_002.pdf


Figure 48: Design drawing of pressure gauge stop valve tailor-made for the CTF

Figure 49: Pressure gauge stop valve built in for long-term design testing under high temperatureconditions (insulation dismantled). The white insulation is high temperature insulation microtherm, the

other is a standard insulation material. 2.4.3.1.3 Flow Measuring Device for High Temperature Conditions

Standard design materials were available for operation temperatures up to 540°C. Solutions for higher

temperatures required regulatory approval.

In the temperature range concerned the expansion factors of the respective materials needed particular

attention. The specific expansion factors vary in a band from 12.2 *10-6/K to 17 *10-6/K.

Surfaces otherwise identical at room temperature may differ by approx. 2.3% at 700 °C. Other instances

show the alteration of specific expansion factors over temperature, e.g. the expansion factor for Alloy

617 that fluctuates from 14.4 * 10-6 /K at 600 °C up to 15.1 * 10-6 /K at 700°C.

76

8/13/2019 KINA25921ENN_002.pdf


Choosing the wrong specific expansion factors could result in flaws in the steam flow readings of up to

5%. Figure 50 is shows the flow nozzle design, and Figure 51 shows the nozzle ready for insulation.

Figure 50: Flow measuring venture nozzle design as chosen for the CTF

For proper flow calculation based on thermal expansion above 500 °C the accurate material value is

used instead of the simplified material class.

Figure 51: External view of the installation position of the venture flow nozzle

2.4.3.2 Commissioning

Sensors and measuring loops were calibrated with the accuracy as is standard for accredited institutes

with their particular equipment Figure 52 and 53. Control loop optimisation was required in a wide

operating range considering the complex interlocks with boiler operation and the specifics of the CTF.

The polynomial calculation of temperature values out of received “mV” signals demands much higher

accuracy than is typical for standard solutions. A larger number of reference points than normal had to

be calculated for accurate definition of the calibration curves. The reference points were positioned in

specific areas to achieve higher accuracy.

77

8/13/2019 KINA25921ENN_002.pdf


Figure 52: Readings during commissioning of the closed loop control

The diagram was printed during commissioning with the closed loop control as part of defining thedynamic characteristics.

Figure 53: Evaporator and superheater temperture control monitor screen shot

The screen shot shows the temperatures of all wall panel tubes in the test evaporator and superheater.

Temperature variations result from the pipe position in the panel.

78

8/13/2019 KINA25921ENN_002.pdf


2.4.3.3 DCS and Boiler Protection System Integration

2.4.3.3.1 Control Strategy

The outlet temperature of the test evaporator was primarily controlled by 2 x 100 % flow control valves.

The outlet temperature of the superheater was primarily controlled by spray cooler 1 between the

evaporator and the superheater in the CTF. In case of operational disturbances or cold start-up

conditions the HP-bypass valve would increase the steam flow to imitate the max. temperature at the

evaporator or superheater outlet as the third stage start-up control valve would open. 705 °C live steam

would be cooled down by spray cooler 3 or the HP-bypass valve.

Spray attemperator 2 operated during start-up und shut-down sequences. Under extremely hazardous

conditions the boiler tripped.

Flow and temperatures were continuously calculated in the DCS based on raw data. In future plants,

calculations can be realised using a tailor-made transmitter. Figure 54 shows the tolerance band at

different operating conditions

Figure 54: Temperature differentials vs. the temperature band

The interpolation points are optimised to increase the accuracy between 640°C and 780°C. The blue

line represents the accuracy of the reference temperature.

The set-up of data collection and recording to allow subsequent calculations by the COMTES700

partners are illustrated in Figure 55.

-2,00

-1,00

0,00

1,00

2,00

3,00

4,00

0,0 100,0 200,0 300,0 400,0 500,0 600,0 700,0 800,0

Temp. °C

D i f f e r e n z P o l y n . - S

t ü t z s t . ° C

Tpolyn.-Tstuetzst. Tpolyn.-Tvergleichsmessst.

79

8/13/2019 KINA25921ENN_002.pdf


Figure 55: Control system architecture

The specification and procurement of the transmitters and control equipment were to expand into ranges

so far not common to power plant technology. The range of temperature measurements and operation

temperature would change from 450 to 650°C in actual plants to 600°C to 800°C in the CTF. For future

pressure measurements the feed water pressure will rise up to 800bar under hazardous conditions and

up to 520bar for live steam. An additional target for the temperature measurement is to keep the

resolution below 0.2K.

A major task was to provide proof of the applicability of materials which can be utilised to manufacture

hook-ups e. g. fittings, protection sleeves, shut-off valves etc. for operation under high temperature

condition.

Temperature sensors were monitored under long-term operating conditions, such as Type S thermocou-

ples (PT Rh PT) and were confirmed to deliver the expected quality attributes (1K sustained accuracy

up to 1000°C, 4 years (24,000 h) operation with no measurable deviations). During the four years the

CTF started 180 times a year.

For base-load coal-fired power plants and warm standby power plants it can be expected that the live

time of sensors can be double. For quick start boilers new materials are to be selected for the sensorcoatings or other noble elements are to be referred. Materials, other than noble metals, are subject to

several metallurgical restrictions.

Self-calibrating thermocouples with build in reference point still require further investigation into ways

of adopting their designs to extreme operating condition (see WP7).

Field tests of new electrical insulation materials for thermocouples are strongly recommended. The

purity of the electrical insulation material is very important for high temperature measurements. Since

compensation cables add to the measuring deviations the use of these cables in power plants for exact

measurements must be avoided. New types of transmitters for temperature measurements appear on the

market which are particularly ideal for noble metal thermocouples and facilitate external floating pointcompensation.

80

8/13/2019 KINA25921ENN_002.pdf


The achievable control precision is highly dependent on the degree of resolution that the temperature

measurements can detect. For high temperature applications the resolution must fall below 0.2K.

Impulse pipes and fittings for pressure measurements need to adopt the new boiler design data. Codes

and standards such as VGB regulations need to be updated and they must be extended to include new

materials capable of tolerating 750°C (live steam temperature) under increased pressures up to 800bar

(max. feed pump pressure under fault conditions). Some transmitter connections are actually limited to413bar.

The following codes and standards were taken into account:

o Machinery Directive 98/37/EG (same as currently used version)

o Pressure Equipment Directive 97/23/EG (same as currently used version)

o DN EN 12952, boiler codes

o DIN EN 50156-1, safety regulations for DCS and BMS systems

o EN 60584-2, accuracy of temperature sensors

o IEC60751, industrial platinum resistance platinum temperature sensors

o DIN 19211, levelling vessel

o DKD_r_5_3

o DIN 53763, protection tube

o VGB R 123, collection of recommendation for measurements as a guideline

2.4.3.3.2 Monitoring

Long-term monitoring and data logging enable fatigue and ageing calculations based on data drawn

from the process control system of the host unit and from the CTF. Specific attention has to be paid to

events and incidents that are signs of abnormal operation and of relevance to component lifetime, such

as trips, load shedding and excessive parameter ramps.

Process data were continuously collected and recorded for process control and subsequent scien-

tific/technical assessment by the COMTES700 partners in accordance with the defined scopes of work.

The host plant made the recorded data available on CD every six months to the COMTES700 partners

for analysis and stress calculations. This provided input for later calculation of lifetime consumption.

Alstom managed the data collected by Provia.

81

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


2.4.3.3.4 Engineering

A pressure gauge stop valve designed for temperatures of up to 740°C and made of Alloy 617 was test-

ed. The valve passed the test for type approval.

Type S thermocouples were adapted for use above 705 °C.

The melding points of aluminium (660.323 °C) and silver (961.78°C) were initially used for calibration.

When testing the Type S thermo sensors at the beginning and end of test programme, thermal drift

could not be detected during the four years of operation with more than 180 starts a year.

Avoidance of soiling effects was identified as a major point of concern for the design of the calibration

equipment. For measurement above 700°C, the specific electrical resistance of the material drops sharp-

ly (see Figure 58). Traces of elements like ZrO3 or TiO2 strengthen this effect. For measurements

above 700°C the purity of MgO must be above 99.4%. Iron oxide must be very low since starts to de-

compose above 750°C.

Figure 58: The specific insulation factor of insulation materials at different temperatures

2.4.3.3.5 Commissioning

Achievements are:

proof of applicability and functionality of valve leakage detection sensors at spray-water attem-

perators and implementation and operation of closed loop control under fluctuating heat transfer rates.

Commissioning and plant operation motivates for further research and developments in the fields of

valve disc tightness and avoidance of leakages under 700oC conditions,

700oC steam cooler design and

pipe routing including venting / draining system optimisation.

The CTF was commissioned and made operational under normal operating conditions and controlled

process parameters.

83

8/13/2019 KINA25921ENN_002.pdf


2.4.4 Conclusions

2.4.4.1 General

Today’s norms and standards need to be revised to allow for the high temperature and pressure applica-

tions. With particular purposes in mind new norms under EN code will have to be worked out and is-

sued. The revision of VGB R 123C, collection of recommendations for measurements, as a guideline is

currently under way.

The following were identified as items that deserve additional consideration:

The first operation experience with the new electrical insulation material (HfO) applicable for tempera-

ture ranges around and above 700 °C were gained and from this options and guidelines for design can

be defined.

The project enabled development of transmitters that worked well with Type S thermocouples and

floating cold reference temperature solutions provided by several suppliers.

Non precious thermo elements e.g. type N for existing and actually built power plants are being tested.

The accuracy required according to DIN EN12952-3 part 6.1.5 can be reached. The long-term stabilityof Type N remains to be proven.

Different materials for different fix points and design of fix point sensors were tested to cover the whole

range of operating conditions for actual und future power plants. Particular attention was thereby paid

to long lasting properties such as long-term stability of measuring signals (see WP7).


Reference is made to the subsequent test rig installation, GKM 725°C HWT, focusing on material tests

with dynamic temperature exposure of tubing and piping applicable for power units of up to 50% effi-

ciency. Another project testing material above 700°C is the ENCIO project. Future investigations will

also have to focus on cost reduction.


Material combinations allowing improved long-term stability and not using noble metal became feasi-

ble.

Higher accuracy can make it possible to operate the steam generator with 2K higher steam temperatures

using the same pipes. Higher operation temperatures at the turbine inlet will increase the thermo dy-

namic efficiency of the process.

84

8/13/2019 KINA25921ENN_002.pdf


2.5 WORK PACKAGE 5: ERECTION AND DISMANTLING


Installation of the Component Test Facility (CTF) in host plant

View of “structural conditions as received”

Removal of the CTF following completion of the test programme


In general, the planned activities could be transferred completely into reality. Some problems during

manufacturing and installation of the facility and an ambitious erection schedule led to two weeks’ de-

lay of the CTF. It has to be mentioned that the host plant could not be operated without the CTF since

the test facility was very closely connected with it. The consequence was two weeks’ non-availability of

the host power plant.

2.5.3 Description of Activities and Discussion –

Erection

The CTF was installed in the steam generator of unit F of the E.ON Power Plant Scholven (bituminous

coal-fired once through steam generator), Germany. Unit F was commissioned in 1979 and produces

electricity with a net output of 676 MW and thermal output of 1,860 MW by burning hard coal. The

unit produces 2200 t steam per hour, 40 t of this were taken for the CTF and heated up to 700°C. The

steam was cooled down and returned to the main boiler at the end of the facility.

The CTF was erected in 2005 and started up 14 days later than scheduled (15 July 2005) due to unex-

pected problems with the production of the components. The CTF was planned to operate until 2009.

The installation of such a test facility within a commercially operating power plant poses many chal-lenges and any operational problems experienced by the CTF would immediately affect the commer-

cially operating power plant and might lead to full stop. With this reason, even from the start of the

project, reliability was an important criterion for the CTF.

Basically, the following components were installed:

evaporator (44 pipes, manufactured from T24, HCM 12 and Alloy 617B; dimensions: approx. 8m

x 9m,

superheater (2 x 16 pipes as superheater batch conducted, from Sanicro25, HR3C, Alloy 740 and

Alloy 617B, length: 17m each,

HP header,

safety valve, HP-bypass valve and start-up valve,

turbine inlet valve (no EU-sponsorship), and

connective and thick-walled piping.

The test evaporator of the CTF was installed at a height of between 56.5 and 66.2 meters in front of the

boiler membrane wall (Figure 1).

Both of the superheater batches were suspended in between already existing superheater surfaces of the

unit F on a supporting tube structure.

85

8/13/2019 KINA25921ENN_002.pdf


A new platform at a height of 76m within the boiler was constructed for a spray cooler, the turbine inlet

valve, the HP-bypass valve, gate valves and other valves.

The components of the CTF, which were exposed to 700 °C, were manufactured from Alloy 617B. In

addition to Alloy 617B some other materials were tested (Alloy 740, Alloy 263, Sanicro 25,

DMV310N, HR3C, T24 and HCM12).

Figure 59: The companies involved in the erection phase of the CTF

On 10 November 2004 the erection meeting was held and the tender for the erection was prepared.

E.ON issued the call for bids for erection of the CTF on 23 December 2004 and a lockable storage hallwas erected on the parking lot between Unit F and G. On 15 March 2005 the erection (steel works,

pipes, valves, evaporator and superheater) of the CTF was assigned to a consortium consisting of EHR

(EssenerHochdruck-Rohrleitungsbau GmbH) and EAS (E.ON Anlagenservice GmbH). The start date of

the CTF erection for the components outside the boiler was 23 April 2005 and the start date for the

components located inside the boiler was 23 May 2005. It was planned to perform the pressure test at

the end of June 2005. On 23 April 2005 the erection of the CTF started as planned. The steelworks and

piping were almost completed on time due to the increased effort of the involved companies (Figure

59).

At the end of June 2005 most of the steelworks, piping, insulation etc. was completed. The design of theevaporator suspension had to be revised due to a bulge in the membrane wall of the existing boiler,

which was exactly at the place where it was the plan to locate the test evaporator, which had to be lifted

twice.

However, some of the components were delivered quite late and the design of the piping and steel struc-

tures had to be revised because of collisions with the existing pipes. This caused a delay of 12 days.

Non-functionality of control technology and a leakage of the hydraulic actuator of stop valve caused

another two days of delay. This resulted in a delay of the first firing of the CTF of approx. two weeks.

The CTF was first fired on 14 July 2005 instead of 1 July 2005 (Table 24).

During 2006, installation of the CTF in the host plant and the necessary minor modifications of devicesand safety aspects were completed.

86

8/13/2019 KINA25921ENN_002.pdf


Table 24: Erection and start-up of the major components

2.5.4 Description of Activities and Discussion – Dismantling

It was decided to dismantle the CTF on 11 August 2009 and the dismantling started immediately (see

work package 6). The work was finished by the end of March 2010.

All components which were located inside the boiler had to be removed immediately in order to start-up

the power plant without the CTF as fast as possible. All pipe connections to the steam generator were

disconnected and the open pipe or stub ends were closed with caps.

The remaining 700°C piping system including the valves outside the boiler was dismantled in the first

quarter of 2010.

A storage hall was erected at the beginning of the dismantling activities. The dismantled components

were marked and transported to the storage hall.

The dismantling took place in 2 phases.

2.5.4.1 1st phase (August – September 2009)

The Scholven Unit F had to go on steam as fast as possible without the CTF after the decision of dis-

mantling as mentioned in the chapter on WP6 (operation). Since the CTF was very closely connected

with the power plant, unit F could not go into operation without CTF directly after the dismantling deci-

sion. Comprehensive works had to be done:

The evaporator and the superheater would have been destroyed if the power plant had gone into

operation without the CTF. Therefore these components, including the headers, had to be re-

moved immediately in order to avoid damage to the power plant and to make an investigation of

the material afterwards possible.

All pipes that connected the facility with the power plant had to be disconnected. The open pipe

or stub ends at the steam generator had to be closed with caps.

The control and communication system of the facility had to be disconnected from the power

plant.

The evaporator was cut in 4 segments before the removal. Scaffolding was installed at the entire front

side of the component up to a height of approx. 50 m. Furthermore, a platform inside the boiler had to

be installed (Figure 60).

87

8/13/2019 KINA25921ENN_002.pdf


Figure 60: Scaffolding in front of the evaporator inside the boiler

The two superheater panels (Figure 61) were not removed in one piece. A comprehensive scaffolding

system was installed inside the boiler up to a height of approx. 70 m. The superheater was removed by

cutting it in approx. 500 parts. Every single tube part was marked with a numbering system in order tomake a sampling afterwards possible. In a next step, the tubes were severed and taken out of the boiler.

Figure 61: Left: superheater tubes inside the boiler. Right: marks on superheater

Figure 62: Left: superheater outlet header before dismantling. Right: end of 700°C steam pipe at the endof the first dismantling phase

88

8/13/2019 KINA25921ENN_002.pdf


The following components were removed in the first dismantling phase:

evaporator,

superheater,

headers,

first parts of the steam pipes in connection with the headers, and

spray attemperator 1 and 2.

The control and communication system of the facility was disconnected from the host plant. The power

plant was able to go on steam without problems at the end of dismantling phase 1.

2.5.4.2 2nd phase (February – March 2010)

The 700°C piping system, including the valves outside the boiler, was dismantled in the first quarter of

2010. All parts that could be of interest for future examination or use were marked before (Figure 63)

being severed.

Figure 63: Left: Determination of cutting points. Right: Pipe parts with identification marks aftersevering

The 700°C steam piping system was removed and transported to the storage hall within the power plant

site as well as all Ni-based alloy valves. The material was stored for approx. two years in the hall and

was available for further sampling. The remaining material was scraped at the end of 2011 after the

final samples had been taken.

Findings during the dismantling

Swarf was found in all pipes due to the cutting operations which were done during the dismantling.

Deposits were detected in one instance in preheating pipe KKS no. 26NA71Z010 before valve KKS no.

26NA76S507 (Figure 64).

89

8/13/2019 KINA25921ENN_002.pdf


Figure 64: Deposits in a pre-heating pipe detected during the dismantling

A crack was detected in the spray injection nozzle directly after the HP-bypass valve Figure 65. The

HP-bypass valve was picked up by a valve manufacturer for investigation.

Figure 65: Crack in spray injection nozzle after HP-bypass valve detected during the dismantling

Sampling

Many material samples were taken from the CTF during the operating phase and several investigations

were carried out (Figure 66-67).

Figure 66: Left: Superheater tubes in shelves. Right: Evaporator

90

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


2.6 WORK PACKAGE 6: OPERATION


Start up of the Component Test Facility (CTF)

Operation of the CTF

Monitoring of operational data


The task of Work Package 6 was start-up and operation of the CTF. The operation started in July 2005

and lasted until August 2009. After the first start-up the COMTES700 group had to realise that the pro-

ject package “Operation” involved more work than initially planned. This work package played an im-

portant role in the project. Problems in relation to C&I and to valves could be solved after some months

of operation.

In 2007, the first problem with a crack in a thick-walled Alloy 617B component occurred. In the follow-ing years an increased incidence of these problems was found. The group members had to face un-

envisaged challenges. Several repairs were carried out and a high number of laboratory investigations

led to a high increase of knowledge and a deeper understanding of the material.


The COMTES700 partners had to face many challenges during the operation time as mentioned above.

Therefore, the following description will only deal with activities or problems, which are in relation

with the operation of the Ni-based components at 700 °C.

2.6.3.1 Operational Data

22,400 hours of operation could be achieved between 2005 and 2009 and the facility was in fact in ser-

vice for 13,000h at temperatures above 680 °C (Figure 68).

Figure 68: Hours of operation

0 h

1.000 h

2.000 h

3.000 h

4.000 h

5.000 h

6.000 h

7.000 h

8.000 h

9.000 h

1 0

5 0

9 0

1 3 0

1 7 0

2 1 0

2 5 0

2 9 0

3 3 0

3 7 0

4 1 0

4 5 0

4 9 0

5 3 0

5 7 0

6 1 0

6 5 0

6 9 0

7 3 0

13.000 h

above 680°C

Long outages

92

8/13/2019 KINA25921ENN_002.pdf


Unit F was operated with many starts and stops during its operating period. This made it possible to test

the components under cyclic loading conditions; 576 starts were counted (Figure 69) in total during this

period.

Figure 69: Number of starts

A maximum steam temperature of 725°C was achieved in the outer tubes of the test superheater (Figure

70: ).

Figure 70: Temperature distribution in the superheater

2.6.3.2 Main Activities and Non-availabilities during Operation

Some problems during the manufacturing and installation of the facility and an ambitious erection

schedule led to a start-up delay of approx. two weeks (see chapter on erection).

There were some problems with the control technology and the valves in the first months of operation

which caused short periods of non-availability. The problems with the valves were mainly caused by

leaky gaskets and the actuators. These problems could be solved (see chapter 3.3.3.3).

The first samples from the superheater were taken during the 2007 summer outage.

Apart from damage to a single tube in the superheater towards the end of 2006, there was no damagefor the first two years of operation. Cracks on thick-walled Ni-based alloys were detected for the first

0

50

100

150

200

250

300

350

400450

M a y - 0 5

D e c - 0 5

J u l - 0 6

J a n - 0 7

A u g - 0 7

F e b - 0 8

S e p - 0 8

M a r - 0 9

O c t - 0 9

N u m b e r o f s t a r t s

434hot starts

122

warm starts

20

cold starts

Operation fro m

July 2005

until

August 2009

5 9 0

6 1 0

6 3 0

6 5 0

6 7 0

6 9 0

7 1 0

7 3 0

5 9 0

6 1 0

6 3 0

6 5 0

6 7 0

6 9 0

7 1 0

7 3 0

SH Outlet

SH Inlet

700°C

average

° C

up to 725 °C

Design of the superheater: Hitachi Power Europe

93

8/13/2019 KINA25921ENN_002.pdf


time in spring 2007. Further cracks were detected in the following years. Cracks and problems with

repair welds led to a long outage period in the last year of operation.

Modifications of the facility were carried out during the 2008 summer outage. Additional drain pipes

were implemented and the slope of some pipes was optimised. Problems with condensate especially at

spray attemperators led to these measures.

The final sampling was carried out after the dismantling in 2009 and 2010. Figure 71

Figure 71: Main activities and non-availabilities

0

1

2

3

4

5

6

7

J u n .

0 5

D e z .

0 5

J u n .

0 6

D e z .

0 6

J u n .

0 7

D e z .

0 7

J u n .

0 8

D e z .

0 8

J u n .

0 9

D a y s o f N o n - A v a i l a b i l i t i e s

1 4

Problems

with gaskets and

control technology

Material

defects

5

Extension

of

standstill

73

Non availabiliti es

J u n .

0 5

D e z .

0 5

J u n .

0 6

D e z .

0 6

J u n .

0 7

D e z .

0 7

J u n .

0 8

D e z .

0 8

J u n .

0 9

Main activi ties

E r e c t i o n

M o d i f i c a t i o n

o f t h e f a c i l i t y

D i s m a n t l i n g

a n d f i n a l

s a m p l i n g

F i r s t s a m -

p l i n g

C o m p r e h e n -

s i v e r e p a i r s

94

8/13/2019 KINA25921ENN_002.pdf


2.6.3.3 Failures at Test Components

During the operation of the CTF some defects were identified at thick-walled components outside the

boiler and had to be repaired.

The thin-walled components were not susceptible to defects. Only one failure can be reported from the

superheater.Many defects were detected in the thick-walled 700°C part of the CTF, which was made of Alloy 617B.

Cracks occurred in the very highly stressed spray attemperators 2 and 3, where feed water reduced the

steam temperature from 700°C to 540°C.

Small indications occurred in parts where repair welds were conducted at Alloy 617B components.

Repair welds were placed in parts where material was cut out and fitting pieces had to be installed.

These incidents of damage (Figure 72) are described in detail in the following text and in the chapter on

evaluation.

Figure 72: CTF parts with failures

2.6.3.4 Superheater Failure

The thin-walled components inside the boiler were not susceptible to damage. In four years of opera-tion, only a single damage occurred at the superheater.

Damage was detected at two tubes in the area of welds. These welds connected Alloy 617B with

Sanicro 25.

Steam from

test super-

heater

700°C

700°C

700°C700°C

540°C540°C

Pipes to start-up and

safety valves

Turbine

Valve

Steam to hot

reheat pipeSteam to

superheater 4

outlet header

Hp – bypass valve

700°C

Steam from

test super-

heater

700°C

700°C

700°C700°C

540°C540°C

Pipes to start-up and

safety valves

Turbine

Valve

Steam to hot

reheat pipeSteam to

superheater 4

outlet header

Hp – bypass valve

700°C

Repair welds

with indications at

fitting pieces

Facility design: Alstom

S p r a y

i n j e c

t i o n c

o o l e r

3

S p r a y

i n j e c

t i o n c

o o l e r

2

Failures in high

stressed spray

injection coolers

spray attemperator 3


95

8/13/2019 KINA25921ENN_002.pdf


Figure 73: Failure in the superheater

Leakages were detected in the test superheater in the 9th and 10th tubes in October 2006. The defects

were located at the transition area within the tube of Sanicro25 to Alloy 617B. To determine the cause

of the failures, the tubes were sent to the Hitachi Power Europe laboratory for examination. The two

tubes were marked as “Tube 1” and as “Tube 2” (see Figure 73 picture). The primary defect occurred intube 1 (Figure 74). Tube 2 was eroded by escaping steam from tube 1.

No expansions or macroscopic deformations were detected. In the tube 2 (Alloy 617B) an intergranular

crack was detected approx. 3 – 5mm next to the fusion line of the circumferential weld to the material

Sanicro 25 at approx. 180° of the tube circumference.

Figure 74: Intergranular crack at tube 1

Crack initiation nextto the weld

Eroded by

escaping steam

Tube 1

Tube 2

super heater

turbine

valve

700°C

536°C

210 bar

hp- system

410°C / 44 bar hot reheat system

attemperator 2

hp bypass

attemperator 3

96

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


2.6.3.5 Damages at Spray Attemperator 2

Spray attemperator 2 was installed after the test superheater. The attemperator controlled the tempera-

ture only in a small range. The attemperator did not reduce the temperature from 700 to 540°C in nor-

mal operation mode. It reduced the temperature to 540°C only in special operation modes if problems

with the following spray attemperators occurred (Figure 76-77).

Figure 76: Location of spray attemperator 2 in the CTF

Figure 77: Spray attemperator 2

spray injection noz-zle assembly 1

spray injection noz-zle assembly 2

flow direction

connection to outlet

header of superheater

pipe to hot reheat

pipe of host power

plant

pipe to turbinevalve and spray

injection cooler 3

hp bypass valve

spray injection cooler 2spray attemperator 2

super heater

turbine

valve

700°C

536°C

210 bar

hp- system


attemperator 2

hp bypass

attemperator 3

98

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


2.6.3.7 Failure of Stub Welds in February 2009

Cracks in the weld metal were detected with dye penetrant tests at both stubs of the spray injection as-

sembly of spray attemperator 2 in February 2009 see Figure 80. The cracks were located in the areas

where repairs took place in 2008 (see 3.6.3.6). The stubs were not heat treated in 2008 after the TIG

build-up welding.

Figure 80: Cracks at the stubs of spray injection nozzle assemblies and extraction of boat-shapedsamples

Boat-shaped samples were taken for laboratory investigation. They were sent to the Ruhr-Universität

Bochum. The results were as follows:

The cracks in the weld ran interdentritically and were massively covered with oxides. Covering with

carbides (carbides containing Mo and Cr) of the interdentritic areas and the grain boundaries of the top

layers was comparatively low. Hot cracks developed in the manufacturing process were possibly the

primary damage to the weld metal. This was not clearly verifiable due to heavy oxidation of the crack

surface. Just before the crack tip, the material showed carbides and pores on the grain boundaries.

1 2

100

8/13/2019 KINA25921ENN_002.pdf


2.6.3.8 Indications at a Support of Spray Attemperator 2 in September 2008

Mechanised ultrasonic (UT) and dye penetrant tests (PT) showed indications at a bracket of spray at-

temperator 2 (Figure 81) in September 2008 during a regular outage (Figure 82-83). The bracket was

removed and an UT was carried out again. Cracks in circumferential direction were detected. The indi-

cations had a depth of up to 20mm. Metallographic investigation in the laboratory of VGB basically

confirmed the UT results.

Figure 81: Spray attemperator 2 support construction with bracket

Figure 82: Indications at bracket of spray attemperator 2

Bracket after detaching

UT test after detaching of bracket

Indications in area of bracket

support constructionbracket


101

8/13/2019 KINA25921ENN_002.pdf


Figure 83: Results of the UT of spray attempeartor at bracket

A V-groove weld was required according to the drawing. In reality there was no v-groove and the weld

was not fully penetrated.

The pipe system in this part was calculated using “Rohr2” by Alstom after detecting the cracks. The

attemperator rotated in operation 1.15° around the x-axis. A design without the bracket seemed to be

better. Therefore the bracket was not installed again.

The attemperator part with the cracks was detached in order to undertake investigations and was re-

placed with a new fitting piece.

Deformations at the protective pipe were detected after having cut off the crack-affected part (Figure

84). A pin which was installed at the end of the attemperator led to strong deformations of the protec-

tive pipe.

Figure 84: Condition of protective pipe, seen after cutting off crack-affected pipe section

Visual inspection of the internal attemperator parts was made. A crack in the protective pipe close to the

injection nozzle assembly was detected (Figure 85). Replacement of the protective pipe could not becarried out in the facility. Dismantling of the complete attemperator to replace the protective pipe would

collision with

end pin

Indication 1: 19 mm depth



Indication 4: between 5 and 9 mm depth



Indication 7: between 15 and 20 mm depth

1

2

3

4

5

102

8/13/2019 KINA25921ENN_002.pdf


have taken too much time considering the host power plant was not able to operate without attemperator

2. It was accepted just to cut off the end part of the protective pipe and to replace it with a new part. The

new part was 60mm shorter in order to avoid further collision with the end pin.

Figure 85: Results of the visual inspection, crack of protective pipe in the area of injection nozzle

assembly

To manufacture a new fitting piece would be too time consuming. Therefore, the end part of spray at-

temperator 3 was cut off and used as a fitting piece (Figure 86). This was possible due to the identical

design of attemperator 2 and 3.

Local heat treatment of the fitting piece welds was carried out only on the side which connects the at-

temperator with the steam pipe and where no protective pipe is located inside the attemperator. Heat

treatment of the other fitting piece weld was too risky due to the protective pipe inside the attemperator.

Figure 86: Different cuttings during the repair works

damaged area

103

8/13/2019 KINA25921ENN_002.pdf


2.6.3.9 Failure of Circumferential Weld in February 2009

A steam leakage was detected at the end of the Alloy 617B spray attemperator 2 in February 2011. The

crack propagation was fast and unit F had to be shut down. A crack appeared in the area of a circumfer-

ential weld of the attemperator 2 (Figure 87). The crack reached from the 3 o’clock to the 11 o’clock

positions. In addition, cracks in the longitudinal direction started from the main crack.

Figure 87: Damage affected area before and after removal of insulation

The effected weld belongs to a fitting piece which was installed during the regular outage in 2008 (see

3.6.3.8).

The filler metal (electrodes) was the same as the one which was used during the erection of the CTF and

in earlier repairs.

However, there is one important difference to other repair welds in the CTF: Heat treatment was not

carried out, neither before nor after welding. Heat treatment was too risky due to the protection pipeinside the attemperator. There was only one repair weld in the 700°C area which was not heat treated

either before welding nor after welding. This repair weld from 2006 was located at the beginning of

spray attemperator 3 and cracked in 2007 (see 3.6.3.10ff.).

The crack affected area in the spray attemperator was cut off (Figure 88). Samples were sent to the

Ruhr-Universität Bochum and the following results were reported:

The large crack started in the area of the weld and ran in an intercrystalline way through the base mate-

rial, the grain boundaries were slightly covered with carbides. The cracks in the weld ran interdendriti-

cally. The cracks were heavily oxidised. The covering with the carbides (carbides containing Mo andCr) of the interdendritic areas in the weld was even more considerably discernible than the grain bound-

aries of the base material. In immediate proximity of the weld, the heat affected zone exhibited separa-

tions of the grain boundaries.

There were no pores on the grain boundaries. Hot cracks generated in the production process could be

the primary damage. Yet this was not clearly verifiable due to the heavy oxidation of the crack surfaces.

The cracks of the base material may have been caused by internal tensions after welding.

104

8/13/2019 KINA25921ENN_002.pdf


Figure 88: Investigation of the crack

2.6.3.10 Installation of Attemperator 3 at the Position of Spray Attemperator 2

Spray attemperator 2 was replaced with the repaired spray attemperator 3 after the damage suffered end

of February 2009. Repair of the attemperator 2 would have been too time consuming.

105

8/13/2019 KINA25921ENN_002.pdf


2.6.3.11 Damages at spray attemperator 3

Spray attemperator 3 was installed at the end of the 700°C part of the CTF (Figure 89). The Alloy 617B

attemperator reduced the steam temperature from 700°C to 536°C. The wall-thickness of the main Al-

loy 617B pipe was 35mm. The cooled steam was led back into the water steam cycle of the host power

plant. Inside the pipe was a protective pipe (Figure 90).

Figure 89: Location of spray attemperator 3

Figure 90: Spray attemperator 3

spray injection

nozzle assembly 1 flow directionspray injection

nozzle assembly 2

Circumferential inlet weld

pre

heating

pipe

turbine

valve

hp

bypass

valve

pipe from

super heater

outlet header

pipe to super

heater 4

outlet headerof host plant

(536°C)

s p r a y i n j e c

t i o n c

o o l e r

3


super heater

turbine

valve

700°C

536°C

210 bar

hp- system


attemperator 2

hp bypass

attemperator 3

106

8/13/2019 KINA25921ENN_002.pdf


2.6.3.12 Defective Weld at Spray Injection Nozzle Assembly

A steam leakage was detected at spray attemperator 3 in March 2007. A weld at the Alloy 617B counter

bearing of a spray injection nozzle assembly was defective (Figure 91). The weld was ground out and

welded again. All stub welds of the attemperator were dye penetrant tested without negative results. It

was assumed that the quality of the failed weld was insufficient.

Figure 91: Defective weld at counter bearing of spray injection nozzle assembly

The power plant staff detected a leakage at the repaired weld in February 2008. Probably missing clear-

ance at the counter bearing of the spray attemperator nozzle assembly led to failure of the weld. The

defective weld was cut off for laboratory (Figure 92) investigation.

Figure 92: Sample for laboratory investigation

Metallurgical investigation of the stub weld showed a crack formation in the weld metal with crack

portions that ran in the circumferential direction and radially (Figure 93). The crack shape was inter-

crystalline and without deformation and ran along the crystal boundaries. The measured material hard-

ness varied between 300 and 315HV in the weld metal, and between 290 and 310HV in the HAZ and

base material.

protective pipe

counter bearing of

spray injection nozzle

assembly

defective weld seam

minimum clearance

of 6 mm

steam

107

8/13/2019 KINA25921ENN_002.pdf


Figure 93: Metallographic investigation

2.6.3.13 Cracked Circumferential Weld and Inclination of Turbine Valve in 2007

A steam leakage was detected at the inlet weld of spray attemperator 3 (Figure 94) in April 2007. The

weld showed a crack in the circumferential direction.

The defective weld was an electrode repair weld and not an erection weld. The attemperator was re-

moved in the regular 2006 outage due to strong deformations and cracks at the protective pipe. A new

protective pipe was installed and the attemperator was integrated into the facility at the end of the

works. The attemperator was connected to the steam pipe with two circumferential welds (wall-

thickness: 35mm). Heat treatment before or after welding was not applied.

One of these two welds failed as mentioned above after 9 months of operation.

Figure 94: Crack of spray attemperator 3 inlet weld

In addition, deflection of the turbine valve was detected in April 2007, and the pipe, which includes

spray attemperator 3, was slightly bent (Figure 95). Water at the bottom of the turbine valve and the

spray attemperator 3 area could have led to these deformations. Probably the water was injected through

a preheating pipe before the turbine valve. Most probably, the water affected the weld failure.

108

8/13/2019 KINA25921ENN_002.pdf


Figure 95: Deflection of turbine valve

The investigations took place in the laboratories of Hitachi and Alstom.

Hitachi made a number of microsections, SEM and analysis samples were taken (Figure 96). The fol-

lowing was detected:

Between the 12.00 o’clock and the 5.00 o’clock positions cold deformations in the HAZ were de-tected on the inner surface of the pipe.

Across the axis 12.00 o’clock / 6.00 o’clock a linear misalignment of up to 3.9mm was detected

on the inner surface.

Starting at the 12.00 o’clock position intercrystalline cracks were detected in the HAZ on the in-

ner surface of the pipe.

In the weld metal interdendritic cracks and in the 12.00 o’clock position hot cracks were also de-

tected at localised areas.

In the HAZ reheat cracks were detected. In the base metal grain sizes up to G= -3 were detected.

The surface hardness of the base metal was up to 447HV0.2.

Around the 6.00 o’clock position transcrystalline cracks started from inside the pipe. Moreover,

the formation of a crack network was detected in this area.

Precipitates at the grain boundaries in the base metal showed higher Cr, Mo and C contents. The

precipitates were not due to operation but were present already from the start. Possibly, a slight

particle growth could be stated.

The base metal showed oxidised, intercrystalline cracks which had concentrations of Co, Ni, Al,

Cr and Fe In the weld metal typical hot cracks as well as coarse, interdendritic cracks were detected.

pre

heating

pipe

turbine

valve

hp

bypass

valve

pipe from

super heater

outlet header

pipe to superheater 4

outlet header

of host plant

(536°C) s p

r a y i n j e c

t i o n c

o o l e r 3


109

8/13/2019 KINA25921ENN_002.pdf


Figure 96: Laboratory investigation of crack at spray attemperator 3 inlet weld

It can be derived from the test results that welding was carried out under stresses and high heat input.

During operation high additional stresses probably acted in the longitudinal pipe direction. In the lower

pipe area (6.00 o’clock position) a temperature cyclic stress was present. The base metal showed risk

factors of relaxation crack formation. The crack characteristics in the base metal and in the weld metal

indicate relaxation cracks.

A further weld before the turbine valve was cut out in order to get an impression of the conditions of

other welds in the damaged area. This weld was investigated in the laboratory of Alstom. The conclu-

sion of the investigation was:

Initiation of shrinkage cracks during the welding process at root notch based on high heat input

and impeded thermal expansion (stretcher strains at root). Initiation of cracks was supported by

impeded thermal expansion.

Additional fatigue cracks in base and weld metal especially at 6 o’clock position.

Fatigue cracks can be associated with water input.

Propagation of shrinkage cracks due to fatigue.

Non-destructive tests such as radiographic tests and dye penetrant tests were carried out in the area of

the turbine valve and the spray attemperator.

The two welds were replaced with two fitting pieces. Heat treatment with 980 °C for 3h at the four pipe

ends was carried out before welding in order to improve the behaviour of the service exposed material.

Some changes were introduced in the process control technique in order to avoid the injection of water.

Pos. 2Pos. 1

Pos. 1: 12 o’clock, root /HAZ Pos. 2: 12 o’clock, root /deformations, recrystallisations

Pos. 3: 12 o’clock, top layer, hot crack

Pos. 4: oxidized crack in the root area

Pos. 4Pos. 3

110

8/13/2019 KINA25921ENN_002.pdf


2.6.3.14 Repeated Deformations and Cracks in the Main Pipe of Spray Attemperator 3 in2008

It was detected in 2008 that the turbine valve was inclined and obviously the pipe, which includes spray

attemperator 3, was deflected. The same incident occurred in 2007 as described above. The assumed

reason for the deflection in 2007 was water, which was injected through the preheating pipe before the

turbine valve. This led to some changes in the process control technique. Obviously water was still be-ing injected. Furthermore, endoscopy of the attemperator showed cracked welds at the protective pipe

inside the attemperator. Also water puddles, deformations of the protective pipe at the bottom area and

deformations of the protective pipe in the injection nozzle assembly parts were observed (Figure 97).

Figure 97: Old protective pipe. Left: Pipe broken in 5 segments. Right: deformation at injection nozzleassembly part of the attemperator

Many improvements were carried out in the regular 2008 outage in order to avoid further induction of

water:

- A second shut-off valve was installed in the spray-water injection system of spray attemperator 3.

- A new drainage was installed behind the attemperator.

- One of the existing drainages was extended in order to drain attemperator 3.

- The slope of the steam pipe and the slope of attemperator 3 were changed.

Spray attemperator 3 was cut and taken out for repair before the regular 2008 outage. The facility was

in bypass operation due to the missing attemperator from this moment.

Two cracks in the Alloy 617B main pipe of the attemperator were observed during the works (Figure

98). One crack was located in the heat affected zone of an electrode welded circumferential weld, which

was located in the middle part of the attemperator. The weld connected the two segments of the attem-

perator main pipe. The other crack was located at a stub not far away from the defective circumferential

weld. Both welds were made during the erecting of the CTF in 2005.

111

8/13/2019 KINA25921ENN_002.pdf


Figure 98: crack at circumferential weld and stub weld of the Alloy 617B main pipe

The crack affected parts were cut out and sent to VGB’s laboratory for investigation:

The crack started from the weld area.

The macroscopic, surface crack shape showed a slight offset and no visible secondary cracks.

Surface imperfections, being a potential crack cause, were not present.

Across the wall thickness, there was a crack with a slightly branched shape that ran vertically

with respect to the surface up to close to the inner pipe surface.

The fractographic examination of an opened crack plane showed only intercrystalline rupture

portions with low deformation.

The crack surface was completely oxidised.

Microscopically, the major portion of the crack ran in the base material or in the HAZ, whereas

only a small part of the crack ran in the weld metal. There was a purely transcrystalline crack

shape without deformation present; it ran along the crystal boundaries in the weld seam area.

The crack flanks were all oxidised. The oxide layer decreased clearly in thickness from the outer

surface towards the inner surface.

Pore formation could be observed that ran on ahead of the crack formation.

In the weld metal, the material microstructure showed a markedly dendritic structure with carbide

coverage of the crystal boundaries and cloudy fine precipitations in the grain. The precipitationswere chiefly carbides with a high molybdenum content of the type M6C. The measured material

hardness was between 320 and 365HV.

The material microstructure of the base material and the HAZ showed a similar structure and

consisted of austenite with reticularly precipitated carbides on the grain boundaries and very fine-

ly spread precipitated carbides in the grain. The precipitations were mainly carbide with a high

chromium content of the type M23C6. The measured material hardness was between 290 and

320HV.

Depletion of the carbide-forming elements chromium and molybdenum in the area close to the

grain boundaries could not be found with the employed examination methods.

112

8/13/2019 KINA25921ENN_002.pdf


2.6.3.15 Further Tests and Indications at the Alloy 617B Main Pipe

Further indications were detected with PT at some stubs in addition to the detected two cracks (Figure

99). The stubs were installed in order to centre the internal protective pipe.

Figure 99: Further indications at stubs

The findings at the stubs were ground. Some findings disappeared in a dept of <1mm. Other findings

did not disappear in a depth of 4 – 5mm. Therefore, replacement of the crack affected parts with two

fitting pieces was necessary.

2.6.3.16 Installation of Fitting Pieces at the Alloy 617B Main Pipe and Annealing

Three fitting pieces with partly new stubs were installed in the spray attemperator by electrode welding.

Heat treatment of the base metal was not carried out before welding. The main pipe of the attemperator

was annealed for 3h at 980 °C in a furnace after replacing the crack affected parts with fitting pieces.

Finally, all welds were subjected to dye penetrant and mechanised ultrasonic tests.

2.6.3.17 Indications in the Stub Welds after Heat Treatment

Indications were found at stub welds of the spray attemperator injection assemblies after the annealing

process (Figure 100). These welds were not made during the repairs. They were welded during the erec-

tion of the attemperator and were in service for approx. 19,000h. An important fact is that the attem-

perator was completely tested two months before the heat treatment. No indications were found in these

areas at that time. This means that there could be a relationship between the heat treatment and the

cracks. It is also possible that the cracks were already present before the heat treatment but too small to

be detected. Small boat samples were investigated in the laboratories of Ruhr-Universität Bochum. The

indications disappeared after grinding. The welds were built-up with filler metal.

The laboratory of Bochum found out that the investigated cracks occured interdendritically in the weld

metal. The crack flanks showed strong oxidation with accumulation of aluminium. The interdendritic

areas as well as the grain boundaries of the weld metal displayed significant accumulation of carbides

containing Mo. The cracks propagated into the base material, and the crack flanks were covered with

Mo- Cr carbides.

113

8/13/2019 KINA25921ENN_002.pdf


Figure 100: Indications at stubs after the heat treatment

2.6.3.18 Indications in the Area of Circumferential Welds after Heat Treatment

Indications at the new circumferential repair welds occurred in addition to the indications at the stubs

after the heat treatment (Figure 101).

Figure 101: Indications at circumferential welds after heat treatment

Many indications were located on the fusion line between the filler metal and the base material. TheRuhr-Universität Bochum made an investigation based on a boat-shaped sample of one weld. The la-

boratory came to the following results:

The findings were cracks approx. 1 – 2mm long which progressed interdendritically in the weld metal

and intergranularly in the heat-affected zone of the base material. The interdendritic areas of the weld

metal and the grain boundaries of the base material displayed massive ‘saucer -type’ accumulation of

carbides containing Mo and Cr. Possibly the deterioration of the weld metal was a consequence of hot

cracks which occurred during welding. Due to the strong etching of the crack surfaces this could not be

verified unmistakably.

repair

weld

114

8/13/2019 KINA25921ENN_002.pdf


It could not be clarified if a higher boron content of the filler metal had a negative influence on the

weld. Anyway, it was decided to use the not modified Alloy 617B filler metal with a lower boron con-

tent for further repairs.

The repair welds were ground and cut out (Figure 102) and TIG built-up with filler metal. Two welds

were filled with TIG orbital. This welding technique was applied for the first time in the CTF. It was

possible to achieve good results.

Figure 102: Cutting of indications

2.6.3.19 Final Heat Treatment of the Complete Spray Attemperator 3 and Installation in theCTF

The complete attemperator 3 and all small built-up welds were annealed again before the installation of

the protective pipe at the end of the repairs in a furnace.

A big crack appeared at spray attemperator 2 at the end of February 2009. Since the CTF and the host

power plant were not able to operate without spray attemperator 2 and the repair of the attemperator

would have taken too much time, attemperator 2 was replaced with the repaired spray attemperator 3.

2.6.3.20 Circumferential Repair Welds of Steam Pipes

Some circumferential repair welds were conducted during the operation time of the CTF. It was realised

that the welding of service exposed Alloy 617B material is very challenging.

All thick-walled circumferential welds of the 700 °C steam pipe part were checked with mechanised

ultrasonic testing (UT) in 2007 due to the problems that occurred with thick-walled Alloy 617B welds,

as mentioned in previous chapters. It was the first time that UT was applied at these welds. Manufactur-

ing defects, which could not be detected with radiographic tests during the erection time, were found

with UT. Four welds were removed and replaced by electrode welding with fitting pieces. Heat treat-

ment after welding at 980 °C for 3h was carried out. These eight repair welds from 2008 were tested inMarch 2009 after a few months of operation.

Indications at the welds were found at all welds. Only acceptable indications with UT testing were

found. However, unacceptable dye penetrant indications (Figure 103) had to be eliminated by grinding

and cutting and a build-up welding was necessary.

115

8/13/2019 KINA25921ENN_002.pdf


Figure 103: Indications at repair weld

The indications at the fitting piece welds showed that the present repair concept was not appropriate.

Different kinds of optimisations were tested (described in the chapter 3.7.4.1.4) such as the change from

electrode welding to TIG welding in order to reduce the heat input or the application of a pre-weld heat

treatment at the service exposed base material in order to increase the ductility of the material. These

optimisations led to better results, but local indications were still found quite frequently during the

work. Several indications had to be ground out. The procedure were extremely time and cost intensive.

All repairs were finished in the beginning of May after more than two months of standstill. A final re-

pair concept could not be elaborated during the operation period of the CTF, this had to be done during

the evaluation phase of the project.

2.6.3.21 End of Operation

End of operation was scheduled in the COMTES700project after more than 20,000 hours of operation

in August 2009. Finally, it was decided to stop operation in 2009 as initially planned and to dismantlethe CTF.

The extension of the operation time of the CTF for approx. two years planned in the meanwhile in order

to gain additional experience could not be realised. The decision was taken due to problems with the

conventional part of the facility and the 700 °C part of the facility as described above. Very time and

cost consuming repair work lead to the decision to stop operation in August 2009 and to dismantle the

facility.

116

8/13/2019 KINA25921ENN_002.pdf


2.6.4 Conclusions

2.6.4.1 General

The operation of the CTF showed that components of Ni-based alloys are in principle feasible for a

700°C power plant. However, challenges with thick-walled alloy components still exist and therefore

additional tests in future test rigs are necessary. Several macrocracks with partial steam leakages oc-curred and a high number of microcracks appeared at repair welds (dye penetrant test indications).

The macrocracks can be summarised as follows:

Cracks occurred in areas with a high ratio of additional stresses, such as attemperators where a

two-phase flow was detected .

Cracks were present in the area of stub and circumferential welds.

Welds from the erection time and repair welds were affected.

There were only two circumferential repair welds in the thick-walled 700°C area without heat

treatment (980°C for 3h) either before or after welding. These two welds caused the most severecracks of the facility. The cracks were detected after some months of daily start/stop cycles.

Heat treated repair welds (980°C) showed microcracks (dye penetrant test indications) but no

macrocracks.

Cracks started at the outer surfaces.

Interdendritic and intergranular crack propagation was detected.

The operation demonstrated that postweld heat treatment at 980 °C for thick-walled Alloy 617B com-

ponents has a good effect on the weld. This kind of heat treatment can be applied to erection welds and

repair welds.It was not possible to develop and to verify a definite repair concept for thick-walled Alloy 617B com-

ponents during the operation. Open questions have to be solved in follow-up investigations and pro-

jects. Promising ideas in order to find answers to the open questions are already available.


The experience from the COMTES700 project is being considered in subsequent projects, such as EN-

CIO and HWTII. Post-weld heat treatment at 980°C is considered for thick-walled Alloy 617B compo-

nents and the steam is not cooled down using only water. It will be cooled down with colder steam or a

combination of water and flowing steam in order to achieve good dispersion of the water and therebyavoid a two-phase flow.


The operation of the CTF demonstrated that full-scale 700°C components made of Ni-based alloys can

be operated in power plants. It also showed, however, that improvement of the material handling with

regard to manufacture, installation and maintenance is necessary.

The operational experiences from the CTF will be important to follow-up research projects and the first

700°C demonstration plant.

117

8/13/2019 KINA25921ENN_002.pdf


2.7 WORK PACKAGE 7: EVALUATION


Assessment of consumed lifetime based on actual service data and finite element calculations

Check of materials condition after two years based on component inspection, samples and rele-

vant metallurgical investigations (to allow preparations to start for the final full-scale demonstra-

tion plant)

Check of materials condition after four years based on component inspection, samples and rele-

vant metallurgical investigations

Review of design and design assumptions based on the previous investigations.


Based on the work package objectives, a sampling plan and an evaluation matrix were developed to

define the investigations to be made. The evaluation matrix was adapted as the project progressed.

Failures on thick-walled components experienced during operation gave rise to a comprehensive study

of failures and repair concepts.


Samples were removed from the CTF after 11,600, 19,200 and 22,400 hours of operation (fire in boil-

er). The samples were investigated using various methods, including light optical microscopy (LOM),

scanning electron microscopy (SEM) with energy dispersive analysis (EDS), transmission electron mi-

croscopy (TEM), hardness measurements, impact, and other mechanical tests. Furthermore, creep test-

ing was performed on selected alloys.

The investigations were based on an evaluation matrix prepared before the commissioning of the CTF.

Samples from evaporator base materials and welds, superheater base materials and welds, headers and

steam piping including bends and branch pieces and spray attemperator were investigated by different

laboratories (see Table 25).

The majority of the results from superheater and evaporator were reported in the intermediate report.

This report includes the remaining results from the evaporator and superheater samples, as well as in-

vestigations of thick-walled components. Unless otherwise described the results are based on investiga-

tion of samples after 22,400 hours of operation.

Table 25: List of laboratories.

Alstom Power Systems laboratory

Centro SviluppoMateriali laboratory

DONG Energy Power laboratory

Electricite de France laboratory

Hitachi Power Europe laboratory

Laborelec laboratory

118

8/13/2019 KINA25921ENN_002.pdf


2.7.3.1 Evaporator

The additional results are summarised from deliverables.

2.7.3.1.1 Evaporator Base Material

The main results from evaporator investigations are summarised in Table 26, including hardness

(HV10) measurements.

Pit corrosion areas on the outer side of the tubes were found for all the tested materials. The dominant

elements in the corrosion layers were O, S, Cr. Significant thickness reductions (> 0.3mm) were found

for HCM12 and Alloy 617B.

It was confirmed that, as in previous measurements, the highest steam oxidation thickness was found

for the HCM12 samples. The dominant elements in the steam oxidation layers were O, Cr, Fe. In par-

ticular, non-uniform scaling was observed in the T24 rifled tube. A significant increase of hardness

compared to non-exposed material was observed only for Alloy 617B, from 170 to peak values of ap-

prox. 290HV10.

Table 26: Summary of results for evaporator tubes. Hardness values refer to the highest reported valuein the references.Sample Material Avg.

Metal

Temp.

(°C)

Area Fireside Corro-

sion

Max.

thick.

reduction

(µm)

Steam Oxidation Max.

oxide

thick.

(µm)

Hardness

HV10

Hardness

HV10

E1 T24 526 Metal Pit-corrosion 160 36 217 213Corrosionlayer

Enriched O, S,Cr

Enriched O, Cr, Fe

E2 T24 562 Metal Pit-corrosion inlarge areas andlocally

130 48 219 206

Corrosionlayer Enriched O, S,Cr Enriched O, Cr, FeE3 - E4 HCM12 557 Metal Pit-corrosion in

large areas andlocally, interdis- persed with fineoxides

57 - 350 105 218 216

Corrosionlayer

Enriched O, S,Cr

Enriched O, Cr, Fe

E5 Alloy617B

556 Metal(tubewith fin)

Pit-corrosion inlarge areas

280 19 289 258

Corrosion

layer

Enriched O, S,

Cr

Enriched O, Cr, Fe,

AlFin Hot cracks inHAZ

E6 Alloy617B

584 Metal Pit-corrosion inlarge areas andlocally

130 23 281 250

Corrosionlayer

Enriched O, S,Cr

Enriched S, Cr, Fe,Al

ER1 T24 -rifled

502 Metal Pit-corrosion("front inter-crystalline")

100 Max. sp. 27 µm 27 210 206

Corrosionlayer

Enriched O, S Non-uniform scal-ing of fine oxides.

Enriched O, Cr, Fe,Zn

119

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


In one case (transition between HCM12 and weld metal) an imperfection was found on the fusion line

(230µm). Oxidation levels comparable to those found on base metals were observed.

Table 28: Summary of results for evaporator dissimilar welds. Hardness values refer to the highestreported value in the references, except for microhardness that refers to average values in the reference.

Sam-

pleMaterial

Avg.Metal

Temp.

(°C)

Area Fireside Corrosion

Max.thick.

Reduc-

tion

(µm)

Steam

Oxida-

tion

Max.oxide

thick.

(µm)

Hard-

ness

(HV1)

Hard-

ness

(HV10)

Microhard-

ness

(HV0.2)

DW1 T24583

(cover)Base T24

Pit-corrosion in largeareas and locally

48 205 - 238

542(root)

Transi-tion

255 240 284

S Ni6617

(mod.)

Weldmetal

3 233 219 258

Corrosionlayer /

HAZ

Enriched O, CrEn-

riched

O, Fe

- - -

Transi-tion

large scaling. EnrichedO, Cr, Al, Si, Fe

220 277 276 276

HCM12Base

HCM12

Pit-corrosion in largeareas and locally. En-

riched O, Cr, Al, Si, Fe35 214 - 223

DW2 HCM12573

(cover)Base

HCM12

Pit-corrosion in largeareas and locally. En-

riched O, Cr

En-richedO, Cr,

Fe

113 214 - 226

538(root)

Transi-tion

230 µm imperfectionfusion line. Enriched O,

S, Cr, Mo65 274 276 307

S Ni

6617(mod.)

Weldmetal

En-richedO, Cr, Ni

3 300 249 303

Corrosionlayer /HAZ

- - -

Transi-tion

320 275 284

Alloy617B

BaseAlloy 617

Pit-corrosion in largeareas and locally

En-richedO, Cr,

Fe

5 357 - 318

121

8/13/2019 KINA25921ENN_002.pdf


2.7.3.2 Superheater

The results are summarised from deliverables.

The main results are summarised in

Table 29.

The average grain size was between 50 and 120µm (average grain size according to ASTM classifica-

tion between 5.3 and 2.9), with the lower values for Sanicro 25 and the higher for Alloy 617B. The

hardness in transverse sections was lower for the austenitic steels (230-250HV10), and higher for nickel

based alloys (approx. 290-310HV10 for Alloy 617B and up to 390HV10 for Alloy 740).

In the longitudinal sections across the welds, the highest values were reached in the weld material and

the HAZ, with peaks of 340-350HV10 for Alloy 617B and 390-400HV10 for Alloy 740. Microhardness

was also measured in the grains; the values showed a very high scatter, ranging from minimum values

of 100 up to maximum of 500HV0.5, up to 600 – 700HV0.5 for Alloy 740.

Concerning fireside corrosion and steamside oxidation, the predominant elements in the layers are S, O,

Cr and Fe, Cr, O respectively, confirming the results found for the evaporator. The best fireside corro-

sion resistance is achieved by Alloy 740, showing negligible layer thickness compared to approx. 50 –

60µm for Alloy 617B and higher values for austenitic steels.

122

8/13/2019 KINA25921ENN_002.pdf


Table 29: Summary of results for SH similar/dissimilar welds. Welds made with type Alloy 617B filler

metal. Hardness values refer to the highest reported value in the references.

Sample Material Area

Thickness

range

(mm)

Grain size

min-max

(µm)

ASTM

grain

class,

average (-)

Hardness,

tranverse

section

(HV10)

Hardness,

longitudinal

section

(HV10)

Micro

hardness

across GB

(HV0.005)

DW8 (SH22) HR3C Metal 9.7 - 10.1 50.9 - 53.2 5.2 235 270HAZ - - - - 300Weldmetal

- - - - 350

HAZ - - - - 320Alloy 617 Metal 6.5 - 8.8 78.0 - 141.9 3.3 317 330

DW9 (SH24)DMV310

NMetal 9.2 - 9.9 64.1 - 97.5 4.1 234 260

HAZ - - - - 320Weldmetal

- - - - 340

HAZ - - - - 340Alloy 617 Metal 8.0 - 8.6 82.1 - 120.0 3.5 316 350

DW10

(SH26)SAN25 Metal 9.4 - 9.8 42.2 - 57.8 5.3 254 270 450

HAZ - - - - 280 -Weldmetal

- - - - 340 -

HAZ - - - - 340 -Alloy 617 Metal 7.5 - 8.5 90.0 - 93.6 3.7 318 330 500

DW11

(SH19)Alloy 740 Metal

60.6 - 79.9(longitud.)

4.5HAZWeldmetalHAZ

Alloy 740 MetalAD09 Alloy 617 Metal 8.2 - 8.7 82.1 - 111.5 3.4 298

AD09 longi-tudinal Metal 80.7 - 105.4(longitud.) 3.5

SH8 Alloy 617 Metal

8.2 - 8.7 79.3 - 99.7 3.7 327

290 420HAZ 300 -Weldmetal

320 -

HAZ - -Alloy 617 Metal - 420

SH9 Alloy 617 Metal

78.0 - 91.8 3.8 389

- -HAZ - -Weldmetal

350 -

HAZ 400 -Alloy 740 Metal 9.9 - 10.3 390 700

SH6 SAN25 Metal 9.5 - 9.9 55.1 - 63.3 4.9 254 270 450HAZ - - - - 280 -Weldmetal

- - - - 300 -

HAZ - - - - 330 -Alloy 617 Metal 7.9 - 8.8 101.8 - 133.8 2.9 284 290 500

Microhardness measurements confirmed findings from the intermediate report. TEM investigations of

microstructures revealed the presence of precipitates, both intergranular and intragranular. In Alloy 740

γ’-phase was found, coarsened in the aged material, together with needle-like η-phase at the boundary

grain, enriched in Ti and Nb.

123

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


The grains of all samples appeared dotted in SEM, possibly due to precipitation of small γ' particles (see

Figure 104 right). In all samples few large primary precipitates of TiN/Ti(C, N) particles were observed

(orange arrow in Figure 104 right). Many globular particles rich in Cr and Mo were observed in grains

and along grain boundaries, dispersed irregularly throughout the structure (black arrows in SEM image

and dots in LOM image). These are most likely M23C6. In some samples, precipitates with similar com-

position were also present as elongated particles along grain boundaries. In addition, small Mo-rich

precipitates (most likely M6C) were observed at grain boundaries (white arrows Figure 104). Small

needle-like precipitates containing more Mo and less Ni than the matrix were identified in few samples.

Generally, grain boundaries contained more Cr, Fe and less Ni, Mo than the matrix.

The mechanical properties of the virgin (V10) and the service exposed steam pipe (SP2) differed; the

service exposed sample was harder, stronger, more brittle, and less ductile than the virgin sample (see

Table 31). The mechanical properties of bends (BP2 and SPB2) were similar in extrados and intrados.

Notice that the change in mechanical properties was more significant for bends than for straight pipe,

suggesting that deformation induced during bending might result in precipitation of larger amount of

precipitates. Most samples were slightly harder at the surfaces than in bulk.

Table 31: Mechanical properties of various samples. RT = room temperature, HT = 700°C

R p0.2 (MPa) R m (MPa) El (%) Red. area Z (%) Impact (J) Hardness(HV10)RT HT RT HT RT HT RT HT

V10 351 234 747 558 54 66 50 54 237 174

SP2 442 318 864 636 41 45 45 55 105 218

BP1 539 893 21 21 66 252

BP2 558 996 31 35 71-83 253

SPB2 562 983 21 32-38 73-83 268

A thin Cr 2O3 layer, enriched with Fe at the surface, developed on the steam side of the components. Its

thickness was approx. 4 – 6µm. Below this layer an internal oxidation zone of approx. 9 – 18µm was ob-

served. This zone contains Al2O3 (see Figure 105 left). The alloy adjacent to the oxide layer is depleted

in Cr and the reticular precipitates had dissolved (see Figure 105 middle). Shallow pits are observed on

the steam side of samples DE1 (see Figure 106) and SPB2.

Figure 105: Steam side oxide. Left: BSE-SEM image of H3. Middle: LOM image of SPB2 afteretching. Right: LOM image of BP1

125

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


2.7.3.3.3 Welds

The results are summarised from deliverables.

Table 33 summarises the investigated dissimilar welds. Three of the welds are between P92 and Alloy

617B but exposed at different temperatures.

Table 33: Investigated dissimilar welds

Weld Materials Operation temperature

DW3 Evaporator outlet stub weld Alloy 617B - P92 ≤ 540°CDW4 Superheater inlet stub weld P92 - Alloy 617B ≤ 540°CDW7 Circumferential pipe weld Alloy 617B- P92 ~ 536°CDW8 Circumferential pipe weld Alloy 617B-10CrMo9-10 ~ 536°C

Figure 108 shows hardness measurements performed at the root and top layer across these welds. No

relation between temperature and hardness profile is observed. A hardness increase is observed in P92close to the transition zone for all welds. A hardness decrease is observed in the root layer of DW4 ad-

jacent to this increase. The root layer is harder than the top layer for DW4, whereas root and top layer

show similar hardness for DW3 and DW7. In the root layer of DW7, a large increase in hardness is

observed in Alloy 617B close to the transition zone. All welds show decreasing hardness in Alloy 617B

with increasing distance to the transition zone.

Figure 108: Hardness measurements performed across dissimilar welds between P92 and Alloy 617B. r= root layer, t = top layer

127

8/13/2019 KINA25921ENN_002.pdf


Figure 109 shows the hardness profile measured across the dissimilar weld between 10CrMo9-10 and

Alloy 617B. A peak in hardness is observed at the transition zone between 10CrMo9-10 and weld met-

al.

Figure 109: Hardness measurements performed across the dissimilar weld between10CrMo9-10 and Alloy 617B (DW8). r = root layer, t = top layer

LOM investigation of DW7 suggests transformation of P92 into ferrite in a zone close to the transition

line (see Figure 110). Microscopy investigations showed that the interface between weld metal and base

metal was well-defined for base metals P92 and 10CrMo9-10, whereas a gradual transition is observed

between Alloy 617B and the weld metal. A gradual change in composition is observed in an approx.

10µm thick zone of the weld metal closest to the ferritic/martensitic base metals (see Figure 111); for

DW7 and DW8 particles are observed in this zone. A transition zone is observed on the 10CrMo9-10

side of the interface between weld metal and base metal (DW8).

Figure 110: Left: The HAZ of P92. Right: Alloy 617B (right) in dissimilar weld, DW7

BM WM

P92 BM

WM

128

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


2.7.3.3.4 Failures

During the first two years of operation no serious damage was detected, with an exception of one leak-

age in a superheater close to a weld between Sanicro 25 and Alloy 617B. The investigation of the dam-

aged thin walled boiler tube showed hardness values up to 400HV10. During manufacture the base

material was cold deformed and this reduced its ductility. The ductility is important for the reduction of,

and to tolerate, the residual stresses being present due to the welding process. Details of the failure andthe cause analysis are described in Chapter 3.6.3.4, Superheater Failure. A repair concept for thin

walled material was developed, which included pre-welding heat treatment at 980°C for 3h. After the

repair that solved this problem, no further damage in the thin walled tubes was experienced.

After approx. 15,000 hours of operation in the test rig, a crack was observed in a thick-walled Alloy

617B component (see Figure 94, WP6). Further cracks were observed afterwards. The areas which were

affected by cracks were investigated by metallographic investigation. Detailed investigation of the dam-

aged welds revealed an intergranular/interdendritic crack propagation in all cases. In most cases the

crack seemed to nucleate in the weld. After the nucleation the crack propagated to the base material. A

typical crack path and the associated microstructure are shown in Figure 113.

Figure 113: Crack that occurred during operation of the CTF. Left: Typical crack path close to a stub.Right: Associated microstructure

The stub welds and the circumferential connection welds were affected. Because of the special design

of the coolers that were necessary in the CTF, this was more likely to induce additional stresses in these

areas. The detected damage to the thick-walled components made repairs necessary. Therefore, thecrack affected welds were cut off and replaced by fitting pieces.

In contrast to the erection, the first attempt to carry out a repair weld showed many indications in the

dye penetration tests directly after welding (see Figure 114). From this moment on it became obvious

that the repair process of the thick-walled pipes and the service exposed pipes made of Alloy 617B pre-

sented a special challenge. For this reason the welding concept for repair welds had to be developed.

Acceptance criteria for indications level 1 (linear indications ≤ 2mm accepted) according to EN

1289:1998 were used to evaluate the weld quality.

130

8/13/2019 KINA25921ENN_002.pdf


Figure 114: Indications that occurred directly after welding of service exposed Alloy 617B in the areaof welds: a) crack in the weld, b) PT-indications during non-destructive testing, c) crack in the heataffected zone

2.7.3.3.5 Repair Concept

The first measure was to optimise the weld build-up and welding parameters within the range of quali-

fication of the WPS used during manufacturing and erection. The aim of the changes was to reduce the

residual stresses created during manufacture of the weld.

The Ni-based alloys showed significant shrinkage during cooling. Consequently high residual stresses

in the welded area were found. To reduce the stresses it was decided to modify the build-up. By weld-

ing alternately the quarters of the circumferential weld, stresses should be contributed more evenly. In

addition to this measure, the originally used electrode welding was changed to manual TIG welding

which can reduce heat input and at the same time reduce the Si content in the weld. Si reduces the melt-

ing point of the alloy and therefore widens the melting range. This enlarges the susceptibility of hotcracks in the weld during the solidification phase. Therefore, it was desired to reduce the content as

much as possible. Using shielded metal arc welding for this alloy, the electrode’s Si content cannot be

fully eliminated as it is needed for the electrode cover stability.

Changing the weld build up and the welding method led to a slight improvement of quality. However,

the result was still unacceptable due to indications in the weld and the HAZ.

As a next step, heat treatment at 980°C was performed before welding on the service exposed material.

This was done to improve the mechanical properties of the material. The overall idea was to increase

the material ductility in order to tolerate and reduce stresses induced by the welding process.

In addition to the heat treatment the base material was buffered before connecting the pipes. This was

done with the intention of reducing the exposure on the heat affected base metal during the welding

process. The welding results improved significantly because of these measures. But there were still a

large number of indications still remained, however, that had to be removed. After a very long time and

intensive repairs, all repair welds were post-weld heat treated at 980°C for 3h and the CTF could be

restarted again.

In view of the many problems during the repair it was decided to recheck all repair welds by non-

destructive testing after another 500h of operation, in August 2009. All welds that had been repaired

and tested showed several indications that had not been found before recommissioning of the CTF. Up

until today it has not been possible to finally determine the time of initiation of the cracks. There is astrong belief that small imperfections were present already before the recommissioning and grew be-

131

8/13/2019 KINA25921ENN_002.pdf


cause of the service conditions. Figure 115 compares an indication that was detected before and after

the 500h of operation. The comparison in Figure 115 shows the general tendency of growth of small

imperfections.

Figure 115: Left: Indications before operation. Right: Indications after 500 hours of operation

Parallel to the repair on site mechanical testing as well as metallographic investigations were carried out

on the service exposed material. This was done to understand the observed significant difference in the

materials behaviour caused by a pre-weld heat treatment (prWHT) at 980°C for 3h.

Therefore, the mechanical behaviour of the heat treated and non-heat treated, service exposed base ma-

terial was compared in a three point bending relaxation test. In this test the samples were deformed at

the testing temperature. The piston displacement was kept constant (4mm) during the test and the sam-

ples reaction during the relaxation phase (150 h) was observed (see Figure 116).

Figure 116: General test set-up of bending test

This investigation showed a significantly different behaviour of the heat treated and the non-heat treated

sample and supported the findings that were gained on site. The not heat treated sample showed a dra-

matic load drop caused by a macrocrack after a short period of time. The crack could easily be seen by

visual inspection (Figure 117a). In contrast to this result the heat treated samples exhibited a much bet-

ter behaviour. After 150h the test was aborted since no load drop was detected. The visual inspection

did not show any macrocracks (Figure 117c). The subsequent metallographic analysis showed only

some small microcracks that were acceptable.

132

8/13/2019 KINA25921ENN_002.pdf


a) c)

b) d)Figure 117: Result of the 3-point bending test: a) service exposed sample, b) load-time diagram forsample out of a), c) service exposed + 980°C for 3h, d) load time diagram for the sample out of c)

In order to understand the different behavioural pattern of the material under the different heat treatment

conditions, TEM investigations were carried out for both conditions. Additionally, the microstructure of

the service exposed material, heat treated at solution annealing temperature (1160°C for 1h), was inves-

tigated by TEM.The service exposed material revealed two types of precipitations that were responsible for the harden-

ing behaviour of the alloy. One type has a roundish appearance and an average size of approx. 70nm.

Test results showed that this type of precipitation was the intermetallic gamma prime phase. The second

type of precipitation in the material has a much sharper shape and is significantly smaller in size Figure

118a). Test results show that this type of precipitation a carbide type M23C6.

To understand the material’s behaviour due to heat treatment at 980°C for 3h that was observed during

the repair of the CTF, the mechanisms in the microstructure of the heat treated service exposed material

were investigated, too. In this material condition only one type of precipitation was found. Analysis

revealed that these precipitations presented the same contribution as the carbides before heat treatment.The carbides did not dissolve but slightly changed morphology because of the heat treatment. The

gamma prime phase that was found in the service exposed condition cannot be found any longer (Figure

118b). Carbides present in the grain boundaries grew slightly in size.

Because of the 980°C heat treatment before welding was not satisfactory it was decided to perform heat

treatment at 1160°C for 1h on service exposed material in the laboratory to gain new experience for a

new future repair concept. Figure 118c shows the resulting microstructure. Generally, it can be said that

that the carbides present could not be dissolved by this heat treatment but their number and size

changed significantly. Compared to the heat treatment at 980°C the carbide’s average size grew by a

factor of 4 (Figure 119).

133

8/13/2019 KINA25921ENN_002.pdf


a) b) c)Figure 118: Microstructure of service exposed Alloy 617B after different heat treatments (22,000h at700°C): a) service exposed, b) service exposed + 980°C for 3h, c) service exposed + 1160°C for 1h

Figure 119: Medium size of carbides in service exposed and heat treated Alloy 617B

Weld tests were carried out to develop an effective repair concept for service exposed and thick-walled

Alloy 617B material. Pipe material with a diameter of 219.1mm and a wall-thickness of 50mm was

taken from the 700°C part of the facility after dismantling. The material was in operation for 22,400 h.

Six welds were conducted in total (see Table 34). Welds 1 and 2 show the influence of the weld tech-

nique on the weld result, and welds 3, 4, 5 and 6 the influence of heat treatment with 980 °C and 1160

°C on the service exposed base material before welding. On-site heat treatment using heating mats was

demonstrated successfully. Heat treatment at 980 °C after welding was applied in any case.

medium size of carbides

0

20

40

60

80

100

120

140

160

180

200

i n n m

after 980°C

after 1160°C

100nm100nm 100nm

134

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


Thermocouples TE13 to TE18 describe the temperature distribution in the pipe outside the insulated

area. TE 18 has a distance of 600mm to the heating mats. 132 °C was measured as the maximum at this

location.

The pipe was cooled down as rapidly as possible after a holding time of one hour at 1160°C (see Figure

124). Two spray-water cooling devices were manufactured for this task.

The insulation cap at the end of the pipe was removed at the beginning of cooling process and also thefirst layer of the insulation at the outer pipe surface was removed. The rest of the insulation was re-

moved subsequently during cooling.

Figure 124: Rapid cooling down process after the heat treatment with water

The thermocouples at the outer surface could not be used for evaluation of the cooling process, since

these measurements are directly affected by the cooling water. The best and most conservative infor-

mation about cooling rates can be derived from TE11 and TE12. These thermocouples were fixed at the

inner pipe surface and therefore not affected by cooling water. Figure 125 shows the temperatures of the

two measurements during cooling. The cooling rate of the upper thermocouple, TE11, is higher than the

cooling rate of TE12. The spray-water device was mostly applied from above. This is the reason for the

temperature difference. A temperature of 700°C was achieved in approx. 11 minutes for TE11 and in

approx. 15 minutes for TE12. This corresponds with an intermediate cooling rate of 42K/min for TE11

and 31K/min. for TE12. The maximum cooling rate achieved at the inner surface was 53K/min.

137

8/13/2019 KINA25921ENN_002.pdf


Figure 125: Measured temperatures during cooling in the heat treatment test

A comparison of an electrode weld with a TIG orbital weld is shown in Figure 126. The gap size of the

TIG orbital welds is only 12mm compared to a gap size of 26mm at the electrode welds. The reduction

of the weld gap size leads to a reduction of internal stresses due to reduced shrinking effects and there-

fore to an improved result.

Figure 126: Comparison between electrode welds and TIG orbital welds

The results of the metallographic investigation of the welds are shown in Figure 127. TIG orbital tech-

nique leads to better results. The welding tests on service exposed material without heat treatment be-

fore welding showed quite large widening of the grain boundaries in the heat affected zone. Heat treat-

ment of the service exposed material before welding had a positive influence on the results. The widen-

ing effects could be reduced significantly with heat treatment at 980 °C. There could be no defects with

heat treatment at 1160 °C.

400

500

600

700

800

900

1000

1100

1200

0 2 4 6 8 10 12 14 16

minutes

t e m p e r a t u r e i n ° C

TE11

TE12

26 mm 12 mm

5 0 m m

Electrode weld TIG Orbital weld

138

8/13/2019 KINA25921ENN_002.pdf


Figure 127: Metallographic investigation of the six test welds

The best results could be achieved with the parameters applied for welds 4, 5 and 6. These welds will be

reproduced and tested in a weld test rig at MPA Stuttgart. The welds will be heated up to 700 °C and

stressed with internal pressure and external forces.

Furthermore, the repair concept will be verified in the ENCIO project, where repair welds produced

with the CTF material will be tested in a power plant test rig.

2.7.3.3.6 Non-destructive Testing

During the manufacture and erection phase of the CTF, the welded pipe to pipe connections of thick-

walled components (wall thickness up to 50mm) were investigated by a standard procedure:

after production of the first 1/3 of weld; radiographic test (RT) and dye penetrant test (PT)

after production of the second 1/3 of weld; dye penetrant test (PT)

after production of the final 1/3 of weld; dye penetrant test (PT)

The RT and PT investigations were consistent with the standards (PT: EN 571-1+ EN 1289 / RT: EN

1435 + EN 12517-1). It is obvious that this standard procedure does not fully exclude the presence of

volumetric defects. Therefore, mechanised ultrasonic testing was introduced in the operation phase of

the CTF.

Fluorescent dye penetrant test and mechanised ultrasonic test methods (MUT) were intensively used

during the operational phase of the CTF. For the latter, qualification and verification were performed.

For the former, acceptance criteria were defined according to EN ISO 23277. In most of the cases

where the dye penetrant test resulted an indication, the surface was ground. Figure 128 gives an exam-

ple of a mechanised ultrasonic test set-up.

400 m400 m 400 m400 m

Electrode Weld:

Service exposed

+ 980°C/3h Pre-Weld-Heat-Treatment

Electrode Weld:

Service exposed

without Pre-Weld-Heat-Treatment

Electrode Weld:

Service exposed


400 m400 m

Weld 1 Weld 3 Weld 5

TIG Orbital Weld:

Service exposed


TIG Orbital Weld:

Service exposed

without Pre-Weld-Heat-Treatment

TIG Orbital Weld:

Service exposed


400 m400 m400 m400 m400 m400 m

Weld 2 Weld 4 Weld 6

139

8/13/2019 KINA25921ENN_002.pdf


Figure 128: Test set-up for mechanised ultrasonic investigations

In 2008, 46 components or welds were investigated by mechanised ultrasonic testing. Six welds, which

were in service since the erection phase, were removed and replaced.

The mechanised ultrasonic testing was performed in accordance with EN ISO 22825. VGB guideline

VGB-R 516, following ENIQ methodology, was used for the qualification of the test method. The qual-

ification component was manufactured from Alloy 617B with the reflector positions given in Figure

129.

Figure 129: Cross section of qualification component with reflector positions (notches and drilled holes)

Detailed results from the examination zones and the UT inspection techniques were described in a

presentation at the COMTES700 Workshop IV.

Verification of mechanised ultrasonic testing was performed by destructive metallographic investiga-tions on two welds after they had been removed from the CTF. Figure 130 compares the result of non-

destructive testing (left in Figure 130) with destructive testing (Figure 130 right). The results from

mechanised ultrasonic testing correspond very well with the destructive investigations carried out by

VGB.

140

8/13/2019 KINA25921ENN_002.pdf


Figure 130: Comparison of mechanised ultrasonic testing (left) and micro-section (right)

The numbers in Figure 130 and 131 refer to the failure types and positions listed in Table 35.

Table 35: Positions and failure types with reference to Figure 130 and 131.Position Failure type

1 Hot crack2 Slag line3 Slag lines4 Slag / lack of fusion5 Slag

Figure 131: Microsections with higher magnification

141

8/13/2019 KINA25921ENN_002.pdf


2.7.3.4 Consumed Lifetime and Design Review

Based on the activities and results within WP6 Operation and WP7 Evaluation; calculations of con-

sumed lifetime and design review were performed for selected components and positions in the super-

heater, main steam piping and spray attemperators.

2.7.3.4.1 SuperheaterThe highest observed corrosion rate in the superheater was in relation to localised corrosion in Alloy617B superheater tubing; reported as a corrosion attack of 3.3mm over the full operation period in theintermediate report.

In order to estimate the consumed lifetime and the predicted lifetime, a simple calculation model wasapplied. Assuming a pressure of 210bar, metal temperature of 700ºC and a constant corrosion rate of1.5mm for 10,000h the accumulated fraction of consumed creep lifetime was calculated as a function ofoperation time. The result is shown in Figure 132. At the end of the operation of the CTF, the consumedlifetime was in the range of 0.6-2.2%, assuming material properties between average and a lower scatter

band of creep rupture strength. However, the predicted lifetime with this corrosion rate was as low as42,000 – 46,000h with the same boundary conditions. Considering that a 100,000h design life was veri-fied after surface machining of the Alloy 617B superheater tubes, these results strongly indicate that thedesign life could not be achieved with the observed corrosion.

Figure 132: Consumed lifetime in Alloy 617B superheater tubing

Design of superheaters with additional corrosion allowance up to for example 10mm is not considered

technically feasible and frequent replacement does not seem economically feasible. Further studies ofthe corrosion behaviour of appropriate Ni-based superheater tubing are therefore recommended.

142

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


Table 37: Total consumption of lifetime based on EN 12952-4:2011 No. Component Total consumed lifetime for

not welded components andgrinded welds with 100%

NDT [%]

Total consumed lifetime forreduced NDT and/or notgrinded welds [%]

1 Bend (R650; D219.1) 0.9 0.92 Branch connection 6.5 16.13 Branch connection 4.6 12.04 Branch connection 4.6 14.65 Bend (R200; D33) 1.8 1.86 Support lug 7.6 11.77 Bend (R400; D133) 2.7 2.78 Branch connection 8.5 18.5

A similar analysis of the consumed lifetime was also made using the methodology from EN 12952-3.

Since EN 12952-3 is generally used for design of components it includes additional safety margins, and

as expected it resulted in higher fractions of consumed lifetime.

2.7.3.4.3 Spray Attemperators

With reference to WP6 Operation the majority of damages occurred at the spray attemperators. For this

reason a design review focusing on failure assessment of these components was carried out. The results

are summarised from deliverables.

Indications were observed during visual inspections of spray attemperators 2 and 3 of the CTF piping

system. Indications consisted of cracks and inelastic deformations of the piping which occurred during

operation between 2005 and 2009. In order to assess computationally the accumulation of the compo-

nent damage, thermomechanical finite element analyses of the piping system and the spray attempera-

tors under operational loads were performed (see Figure 133). The required tensile data on service ex-

posed material were derived from deliverables.

Figure 133: CAE geometry of spray attemperator

The thermomechanical analyses of the transient behaviour and steady state behaviour showed relatively

high values of the radial and axial expansion of the protective pipe compared to the main pipe in both

load cases. Due to the occurrence of the high temperature gradients it can be assumed that the protective

pipe is suffering from extensive plastic deformation caused by blocking of the axial and radial supportstubs of the main pipe.

144

8/13/2019 KINA25921ENN_002.pdf


Based on the non-homogenous temperature distribution at steady state conditions with insufficient

drainage, high thermal loading at the bottom of the main pipe led to moderate plastic deformation. The

equivalent stresses slightly exceed the yield strength. Therefore it can be concluded that a repeatable

interaction of condensed water at the inner surfaces of the incomplete drained main pipe results in an

intensive material fatigue damage.

Due to the relatively low values of the nominal equivalent (von Mises) stress (< 60 MPa) in the global piping system under steady state operational conditions, widespread creep damage across the pipe cross

section was not expected.

To simulate the far-field global time-dependent forces and moments at the boundaries of the spray at-

temperator, the global beam model from the piping system analysis was now combined with the three-

dimensional solid structure of the spray attemperator. The aim of the lifetime analysis was to demon-

strate quantitatively the impact of insufficient drainage on the lifetime consumption. Therefore the life-

time approximation compares the fatigue damage due to normal transient start-up conditions and the

fatigue damage resulting from insufficient drainage. The computed lifetime of spray attemperator 3

under long-term thermal design conditions was definitely reduced based on a two-phase fluid flow at all

verified positions compared to normal/regular transient operation. Here insufficient drainage reduced

the fatigue lifetime mainly at the bottom of the pipe.

In addition an extended lifetime analysis based on EN 12952 under realistic load conditions of the spray

attemperator structure was carried out. The measured values of the outer surface temperatures of spray

attemperator 2 installed in May 2009 were incorporated into the FEM model as the thermal boundary

conditions. Weld manufacturing issues, such as grinding of welds and the extent of NDT, influenced

distinctly on the fatigue strength of the analysed locations. Lifetime analysis according to EN 12952-3

showed that for a load cycle with a two-phase fluid flow, the analysed circumferential welds at posi-

tions 2 and 3 consumed theoretically more than 100% of fatigue life. Positions 2 and 3 are shown in

Figure 134. Lifetime assessment results according to EN 12952-4 of spray attemperator 2 showed alsothat the transient load intensified by the two-phase fluid flow as well as the steady state load accelerated

the accumulation of the material damage. The maximum calculated total material damage reached 84%

at the circumferential weld at position 3.

Figure 134: Analysed part positions of spray attemperator 2

The cracks and plastic deformation found at the spray attemperators of the CTF main steam pipe system

were quantitatively verified to be caused by the two-phase fluid flow. It is believed that the two-phase

fluid flow is caused by leaking valves and insufficient draining systems. Temperature measurements at6 o’clock and 12 o’clock positions at critical positions could have indicated this situation.

145

8/13/2019 KINA25921ENN_002.pdf


2.7.3.5 Other activities

2.7.3.5.1 Self-calibrating Thermocouples

While the increase in steam temperature allowed better utilisation of fuel, exceeding the maximum tem-

perature significantly decreased power plant lifetime with regard to material properties. Therefore self-

calibrating thermocouples based on miniature fixed-point cells, see Figure 135 were developed allowingmore precise measurements and thus enabling a mode of operation closer to the temperature limit. In

the temperature range from 535 to 545°C the usual 5K safety interval to design temperature could be

reduced to 1K. Since 2005, industrial measuring devices based on this technique have been used in

power plants.

Figure 135: Fixed-point thermocouple principle

From August 2006 to December 2009 an accompanying project was carried out aiming to developfixed-point cells for self-calibrating thermocouples applicable at 700°C to be tested in the CTF.

For the alloys chosen for these fixed-point thermocouples (bismuth-platinum (BiPt) and germanium-

palladium (GePd)) suitable container materials had to be found at short notice.

Samples of BiPt fixed-point thermocouples were manufactured with conventional Al2O3-ceramic cells.

In the first field test, started in November 2007, a thermal compound used so far for insulating and fix-

ing of the heating conductor material in the sensors turned out not to be applicable above 620°C. An

alternative solution achieved sufficient stability also at a burn-in temperature of 750°C. The melting

point temperature determined for BiPt amounted to 715.7°C. After a short period of operation, however,

BiPt showed a strong tendency towards supercooling. Solidification after melting sometimes started

below 690°C. At the same time, the melting plateau was reduced such that it could only be observed

under calorimetrical conditions in the laboratory. The BiPt alloy proved not to be applicable on a short-

term basis.

For GePd fixed-point cells, novel thin-walled Si3N4-ceramic containers were developed. The first sam-

ples, which still suffered from some geometrical defects, were available in August 2008. Therefore, the

field test could start until December 2008 allowing only approx. 1,000 hours of operation. However, the

materials used, just as the fixed-point temperature of 735.0°C, showed promising stability. Continuation

of the test run for a minimum of 5,000h appeared necessary but was not possible.

146

8/13/2019 KINA25921ENN_002.pdf


2.7.3.5.2 Compact Insulation for Steam Pipes

From December 2007 to August 2008, an innovative insulation material was tested in the CTF aiming

at a reduction of insulation thickness and weight and a decrease of the insulation surface temperature.

Compared to conventional materials, the test insulation reduced the thickness from 295 to 200mm and

the weight reduction from 60 to 36kg/m. The surface temperature could be decreased from 70 to 50°C.

For the field test approx. 3m of the insulation of an Alloy 617B live steam pipe (outer diameter

219.1mm, wall thickness 50mm) installed at level 73.8m was replaced by the new material. The maxi-

mum live steam temperature was at 705°C with an ambient temperature of approx. 40°C.

Due to maintenance the test insulation had to be removed in August 2008. The insulation cracked and

could not be used again. As a result, the stability of the material was increased with glass fabric on the

half-shell surface. The improved material was not tested.

2.7.3.5.3 Nanoceramic Coatings

From October 2008 to August 2009, an Alloy 617B superheater tube of the CTF protected against slag-

ging with a nanoceramic coating material was tested.

The tube located at the bottom of the superheater outlet was sand-blasted with silica sand for a length of

3m up to a surface quality of SA 2.5. The coating was stirred up according to the manufacturer’s speci-

fication before application. It was applied with a coating gun with a layer thickness of 180µm (in dry

condition approx. 85µm). The test tube was subjected to 3,200 hours of operation, 86 hot starts, 21

warm starts, 8 cold starts and a maximum steam temperature of approx. 685°C.

Investigation of a sample taken after operation demonstrated total disappearance of the initial coating.

The test was not successful.

2.7.4 Conclusion

2.7.4.1 General

2.7.4.1.1 Evaporator

During operation, no cracks or leakages were observed on any of the membrane wall materials.T24 has

shown good corrosion and oxidation properties at temperatures up to 546 °C (average metal tempera-

ture). Due to the scatter of data at higher temperatures, it is not possible to reach a conclusion as regards

the corrosion behaviour of T24. In the high temperature range (550-590 °C), the steam side oxide thick-

ness of HCM12 was not higher than expected for a martensitic steel at these temperatures. Alloy 617B

showed superior corrosion and oxidation behaviour compared to HCM12, as expected for a Ni-basedalloy.

Fireside corrosion was observed on similar welds of T24, HCM12 and Alloy 617B in the evaporator.

Liquation cracks were observed in Alloy 617B tube-to-fin welds. Decarburisation near the fusion line

was observed for HCM12 on both tube and fin welds, as expected with a Ni-based filler metal. A resid-

ual gap between Alloy 617B and HCM12 fin and tube was observed.

Hardness measurements across dissimilar welded evaporator tubes showed a decrease in hardness from

HAZ to weld metal. Decarburisation took place on HCM12 to Alloy 617B and T24 to HCM12 dissimi-

lar welds as expected for dissimilar welds.

147

8/13/2019 KINA25921ENN_002.pdf


Areas with pit corrosion were found on all investigated samples and with thickness reductions larger

than 0.3mm for HCM12 and Alloy 617B.

The highest steam oxidation thickness was found on the HCM12 samples as expected considering the

service temperature. Significant increase of hardness during service was observed for Alloy 617B.

Significant higher hardness was observed in the weld metal and HAZ compared to the base material for

some T24 welds.

2.7.4.1.2 Superheater

HR3C, DMV310N, and Sanicro 25 developed duplex oxides during steam oxidation. They consisted of

an outer layer of Fe-oxide on top of Fe-Ni oxide, and an inner oxide layer containing all elements. The

thinnest oxide layer formed on Sanicro 25.On the Ni-based tube inner side, a zone with internal oxida-

tion of Al, Cr, and Ti was observed below a thin external Cr-oxide layer for Alloy 617B and Alloy 740.

The thickness of the internal oxidation zone was comparable for the two alloys in the range of 30 to

40µm.

Corrosion layers covered with deposit were formed on all samples. More deposits were observed on the

fluegas side than on the rear side of the removed tubes. The deposits consisted of Ca/Na/Ka-Al silicates,

Fe oxides and small amounts of Ca sulphates. A duplex layered structure was observed for HR3C,

DMV310, and Sanicro 25. The inner part of the corrosion layer contained Cr, S, and O for all samples.

Layers of Ni-S were observed within the inner layer of Sanicro 25 and the Ni-based alloys. Internal

corrosion was observed ahead of the corrosion layer.

The thickness reductions for HR3C, DMV310N, and Alloy 740 were comparable to the deviation from

the nominal thickness for the non-exposed samples. Large local thickness reductions were observed for

some of the Sanicro 25 (up to 1.7mm compared to the nominal thickness) and the Alloy 617B (up to

3.3mm compared to the nominal thickness) samples.Precipitates were formed along the grain boundaries of all samples during exposure resulting in in-

creased hardness. Furthermore, precipitates were observed in grains near the corrosion layer for HR3C

and DMV310N, whereas small gamma prime particles developed within the bulk grain throughout the

sample for Alloy 617B and Alloy 740. In addition, needle shaped precipitates developed in Alloy 740.

As expected, for most materials the precipitates were coarser and more numerous the higher the expo-

sure temperature.

Creep testing of Sanicro 25 and Alloy 617B showed gradual evolution of the creep rate with stress,

whereas a sudden drop in creep strength was observed for Alloy 740 which was considered an effect of

microstructural instabilities.Hardness peaks of more than 400HV10 were measured inside the HAZ of similar tube welds of Alloy

740. Local corrosion attacks of the weld metal in Alloy 740, DMV310N and Sanicro 25 were observed.

For DMV310N and HR3C increased corrosion was observed close to the weld metal. A small weld

imperfection was observed for Sanicro 25 and a small crack within the root pass was observed for

DMV310N, both related to the manufacturing process.

A hardness increase was measured on the steam side of HR3C to Alloy 617B and Sanicro 25 to Alloy

617B dissimilar superheater tube welds. This increase was not observed at the fireside. Hardness values

above 430HV10 were measured for HR3C to Alloy 617B tube welds, and in the HAZ of Alloy 740

welds hardness values above 450HV10 were measured.

148

8/13/2019 KINA25921ENN_002.pdf


Corrosion attacks of up to 150µm were observed for DMV310N to Alloy 617B and up to100µm for

Sanicro 25 to Alloy 617B welds. Weld metal attacks were also observed for HR3C to Alloy 617B and

Alloy 617B to Alloy 740 welds.

A repair concept, mainly based on a three hour annealing heat treatment process on service exposed

Alloy 617B before welding was applied successfully.

Average grain size and hardness have been measured and reported, also including microhardness fromlongitudinal sections and investigation of microstructures. The results confirm the conclusions of the

intermediate report.

Based on the measured local corrosion rates of Alloy 617B up to 3.3mm for 22,400h, it is very likely

that the operation time of the superheater will be less than the 200,000h design life. The local corrosion

attacks may partly be contributed to an increasing S content in the fuel during the operation period.

The best corrosion behaviour is observed for Alloy 740.

2.7.4.1.3 Thick-walled Components

For the thick-walled Ni-based alloy components investigated, a bimodal grain size distribution was

observed. Reticular M23C6 precipitates were observed in all service exposed Alloy 617B. Grain bounda-

ries contained more Cr and Fe and less Ni and Mo than the matrix.

Service exposed samples were harder, stronger, more brittle, and less ductile than the virgin samples.

Similar mechanical properties were observed for extrados and intrados of bends. Increased hardness at

the inner surface compared to the bulk was measured.

A thin layer of Cr 2O3 developed on the steam side resulting in depletion of Cr and dissolution of reticu-

lar precipitates. Shallow oxidation pits and deep grain boundary attack were observed on a few samples.

Similar properties were observed for the spray attemperators as for header and piping.

The hardness across welds has been measured and reported, as well as EDS line scans across transition

lines. LOM investigations of welds close to the transition line revealed only minor weld imperfections.

2.7.4.1.4 Repair Concept for Thick-walled Components

Three potential concepts for repair welding of thick-walled Alloy 617B were developed and tested with

welding tests. The three most promising concepts are mechanised TIG with prWHT at 980°C, mecha-

nised TIG with prWHT 1160°C and electrode welding with prWHT at 1160°C. All three cases included

PWHT. The potential has been documented by microstructural investigations which have shown prom-

ising results.

All three concepts will be further tested in a laboratory component test and also in the continuing test

project, ENCIO.

On-site heat treatment at temperatures up to 1160°C using heat mats was demonstrated successfully.

Heat treatment at 1160ºC is likely to become the preferred solution for weld repairs.

149

8/13/2019 KINA25921ENN_002.pdf



Mechanised ultrasonic test methods were successfully applied, qualified and verified for thick-walled

Alloy 617B components of the CTF. Therefore, these test methods can be recommended for further

application of Ni-based materials for thick-walled components in power plant applications.

2.7.4.1.6 Consumed Lifetime and Design Review

Calculations of consumed lifetime and design review were performed for a selected position in the Al-

loy 617B part of the superheater, for selected components in the main steam piping system and for

spray attemperators 2 and 3.

Based on the highest observed corrosion rate in the superheater, the calculation showed that even

though only 0.6-2.2% of the lifetime was consumed at the end of operation, a total lifetime as low as

42,000-46,000h was predicted. Improved corrosion resistance in some temperature ranges is considered

necessary to have a feasible concept for the superheater.

Lifetime calculations of selected main steam pipe system components showed strong dependence ofnotch factors which are affected by the post-weld treatment and the extent of NDT. Moderate total life-

time consumption was calculated assuming ground welds and 100% NDT in the range of 0.9-8.5%.

Assuming notch factors corresponding to non-ground welds or reduced NDT somewhat higher lifetime

consumption was calculated, in the range of 0.9-18.5%. Even though the consumed lifetime in this case

is moderate considering the planned operation time, it also indicates that the consumed lifetime could

have reached 100% in the design lifetime.

The indications found at spray attemperators of the CTF main steam pipe system were quantitatively

verified to be caused by a two-phase fluid flow. Consumed lifetime up to 84% was calculated in a spe-

cific circumferential weld of spray attemperator 2. It is believed that the two-phase fluid flow is caused

by leaking valves and insufficient draining systems. Temperature measurements at the 6 o’clock and 12o’clock positions at critical positions would have indicated this situation.

2.7.4.1.7 Other Activities

In addition to the planned scope, three other technologies were tested, namely self-calibrating thermo-

couples, compact insulation material and nanoceramic coatings. Of these, the self-calibrating thermo-

couples showed promising potential.


No commercial power plants use the 700°C steam cycle tested in this project. However, several of the

materials are used in 600/610°C conventional USC power plants in operation or under construction.

Test results from the COMTES700 project are scheduled to be applied in the ENCIO test facility and

HWT II.

A general interest in 700ºC coal-fired power plants is emerging in Asia.

150

8/13/2019 KINA25921ENN_002.pdf



2.7.4.3.1 Evaporator

The present results have proven that it is possible to design, manufacture and operate a membrane wall

in a 700 °C component test facility. This is based on the facts that there were no damage, no excessive

steam oxidation or corrosion, and no considerable microstructural changes were noted. The design andevaluation of the material performance covered bends, similar, dissimilar and tube-to-fin welds which is

relevant for a full-scale demonstration power plant. The evaluation included T24 rifled tubes which

showed oxidation behaviour comparable to normal seamless tubes.

Due to the membrane wall design in the CTF, the operation loads may have deviated from that of a full-

scale plant. The design life was 200,000h compared to the actual operation time of 22,400h. The possi-

ble effect of these details is not included in the positive evaluation of the evaporator material perfor-

mance.

HCM12 performed as expected in the present evaluation. However the allowable stress of this material

has been reduced in various codes and standards. No substitute candidate material has been tested in theCTF.

2.7.4.3.2 Superheater

The present results have proven that it is possible to design, manufacture and operate a superheater in a

700 °C component test facility. This evaluation is based on substantial investigations of oxidation, cor-

rosion behaviour, similar and dissimilar welds. A superheater leakage was observed, identified as stress

relaxation cracking induced by a combination of manufacturing and operation issues. Sanicro 25

showed the best oxidation behaviour of the high alloy austenitic steels. Alloy 617B and Alloy 740

showed comparable internal oxidation. However, both Sanicro 25 and Alloy617B in some cases suf-fered from local corrosion attacks, which very likely will lead to lifetimes less than the 200,000h design

life of these materials. Therefore, further clarification of this issue is recommended before using the

results for other applications.

The design of the superheater is comparable to a typical superheater design for a full-scale demonstra-

tion plant.

Due to the observed failure, a repair procedure was developed for the repair of the superheater tubes,

including a pre-welding heat treatment step. This procedure was successfully applied.

2.7.4.3.3 Thick-walled Components

The present results have shown that it is possible to design, manufacture and operate a piping system

for use in a 700°C test facility. Since no PWHT was required by the codes; no post-weld heat treatment

was originally used. Due to limited space, a horizontal spray attemperator design was chosen, with the

risk of a two-phase flow. In addition, sufficient drainage was not applied. It is believed that this con-

tributed to the failure of several welds. Currently a repair weld concept for thick-walled Alloy 617B is

in progress.

Only one candidate material for thick-walled components was tested in the CTF.

151

8/13/2019 KINA25921ENN_002.pdf


2.7.4.3.4 Repair concept

A proven and well tested repair concept for repair weld of thick-walled Alloy 617B is crucial for the

realisation of a full scale 700°C power plant. Three promising repair concepts have been developed.

The three concepts will be tested in laboratory component tests and in-service tests before the technical

potential is finally verified.

On-site heat treatment at temperatures up to 1160°C, using heat mats, was demonstrated successfully.The demonstration of heat treatment also included development of specialised equipment for coolingand documentation of the heat treatment cycle by monitoring the temperatures with numerous thermo-couples. Heat treatment at 1160ºC is likely to become the preferred solution for weld repairs.


Mechanised ultrasonic test methods were successfully applied, qualified and verified for thick-walled

Alloy 617B components of the CTF. Mechanised ultrasonic testing can be applied for NDT of welds

and for monitoring of crack growth. These test methods can be recommended for further use of Ni-

based materials for thick-walled components in power plant applications.

152

8/13/2019 KINA25921ENN_002.pdf


2.8 WORK PACKAGE 8: COORDINATION


Organise the management of the project effectively to secure project objectives, budget and time

schedule.

Coordinate the project according to the provisions in the contract with RFCS.

Establish the contracts between the utilities and the manufacturers, manage the financing and

make sure that IPRs are protected.


The COMTES700 project experienced only one deviation that had an impact on the project, extending

it for an additional two years in comparison with the original time schedule. This extension should en-

sure an operational period of more than 20,000h and sufficient time for evaluation. This demand for an

extension was the result of several CTF outages. After procuring permission for extended operation of

the CTF it had to be stopped due to unpredictable costs of sudden indications at Alloy 617B repair

welds. Instead of more operation time the focus was on expanded investigation for better material un-

derstanding and a reliable repair procedure. Other additional investigative projects will run longer than

the COMTES700 project completion (31 December 2011). In addition, it is planned to supervise these

activities with a new “700°C group”.


2.8.3.1 Coordination

The coordinator’s task consisted of project management, arrangements with six contractors and eight-een co-financing partners (suppliers and utilities) from eight European countries. Several groups were

established to support the coordinator in his management tasks.

The COMTES700 Steering Committee (CSC) had the function of supervising the project’s exe-

cution, Collaboration Agreement, contract and troubleshooting referred to by the coordinator. 16

meetings were held.

The Project Management Group’s (PMG) function included performance management of and be-

tween the parties, ensuring project work results in one integrated R&D project. 12 meetings were

held.

The Temporary Support Group Financial (TSGF) had the function of analysing the financial doc-

umentation of each contractor and to give payment recommendations to the CSC accordingly. 11

meetings were held.

The Editorial Group (EG) produced the final report. 4 meetings were held.

WPs 6 Operation and 7 Evaluation were supported by the participation of the coordinator in each meet-

ing.

153

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


Damage behaviour of critical pipe line components made of creep-resistant Ni-based super alloy

(Alloy 617B) under practical conditions, testing of components and simulation

Alloy 617B membrane wall test

Alloy 617B superheater outlet header as reserve material for ENCIO

Impact of ageing on the charpy impact resistance of Alloy 617B on virgin and service exposedmaterial

Investigation on two Alloy 617B gate valves

Local repair and heat treatments on T24, HCM12 and Alloy 617B membrane walls

Investigation on service exposed Sanicro 25 superheater samples

Providing COMTES700 material (superheater tubes and pipes) to the MACPLUS project

These activities are planned to be supervised by a new “700°C group”.

5. Workshops

The project partners were informed regularly through workshops about the technical status and the re-

sults of the project. The coordinator organised five workshops strengthening the 700°C network:

I. Workshop (January 2005) focused on the technical concept of COMTES700.

II. Workshop (October 2005) focused on the erection.

III. Workshop (November 2007) focused on operation experience.

IV. Workshop (April 2011) focused on investigation results from evaporator and superheater.

V. Workshop (April 2012) will focus on investigation results from thick-walled components and ad-

ditional investigations as well as final conclusions.6. Financial Management

The total budget increased from EUR 15.3 mio. to more than EUR 26 mio. The Component Test Facili-

ty caused outages of Scholven F lasting several weeks. These costs were covered by RWE and E.ON.

The budget increase was balanced by the project partners. Complex payments between the parties had

to be arranged and controlled. Tax questions with regard to the RFCS funding had to be settled.

7. Project Documentation

A public website and a web based document filing system accessible only for the project partners were

established and maintained for the COMTES700 project.

2.8.3.2 Reporting

Supported by the contractors the coordinator managed the 6-monthly reports, the midterm technical-

and financial report, the final technical and financial report incl. the external audits.

155

8/13/2019 KINA25921ENN_002.pdf


2.8.4 Conclusions

2.8.4.1 General

The main conclusion is that the COMTES700 project fulfilled its goals. By joining forces, the installa-

tion of the CTF could be ensured with a delay of only two weeks. The project was extended for two

years securing the necessary information for the next step – a 700°C demonstration power plant. Theunexpected termination of the CTF operation was compensated for by an increase in the investigation

efforts. Additional funds were covered by new partners, know-how access agreements and additional

funding of all partners. This activity was linked with a complex agreement structure. Requests from

other 700°C projects for COMTES700 material for investigations were generally approved upon return

of results. The COMTES700 results were communicated not only in WP7 Evaluation but also to all

partners in workshops. Additional costs, such as outages and repairs, were paid by two major utilities

and all project partners. The public was informed about the project by a project website and more than

40 publications.


The COMTES700 project was the continuation of the idea to develop a 700°C power plant that started

in 1998 with the AD700 phase 1 and 2 projects. The AD700 programme bundled together more than 40

participants from 10 European power generators, power plant equipment manufacturers and material

suppliers. COMTES700 attracted additional European power generators and continued the European

700°C network. The research activities lead to European and national research projects; one of them

being ENCIO – successor of COMTES700 (started July 2011 and runs for a 6-year period). The net-

work has inspired China, India and Japan to plan their own 700°C technology research programmes.


The coordination principles of COMTES700 are adapted to the successor project ENCIO.

156

8/13/2019 KINA25921ENN_002.pdf


2.9 EXPLOITATION AND IMPACT OF THE RESEARCH RE-SULTS

2.9.1 Actual applications and technical and economic potential for the use of

the results

Sustainable technologies for coal-fired power plants have to be developed to optimise environmental protection and to preserve valuable resources. The power plant live steam temperature should be

brought up to 700°C (Advanced USC) compared to 600°C (USC) - the best commercial plants currently

available. Therefore, the application of Ni- based super alloys is necessary. Compared to today’s Eur o-

pean coal fired power plants, this will increase the average efficiency of around 36% up to 50%. Thus,

almost a third of CO2 emissions are avoided and fuel is saved in the same amount.

COMTES700 was a demonstration project aiming at 700°C live steam temperatures that generally uti-

lised available design codes and design practice, and selected materials identified within the preceding

THERMIE/AD700 “Advanced 700°C PF Power Plant” project. Development of new materials was not

in the scope of COMTES700.

2.9.1.1 Learning Points from the Project

The COMTES700 test facility was designed and manufactured according to the European harmonised

standards. Design practice for the design of the test facility was generally corresponding to the practice

used for USC power plants; with the necessary adaptations to integrate the test facility to the host unit.

The integration required relative high attemperator cooling capacity, necessity of horizontal spray at-

temperators, which in combination with an insufficient draining system caused some operational chal-

lenges.

Materials for membrane walls: Even though all membrane wall materials tested in COMTES700 (T24,

HCM12, Alloy 617B) generally performed well, some reservations regarding all three materials werestated.

T24 has become one of the most widely used materials for membrane walls in USC boilers which are

under fabrication or commissioning in Europe the recent years. Damages have occurred in these boilers.

Investigation of causes and measures are ongoing.

Evaluations of creep rupture strength for HCM12 issued after the COMTES700 engineering and manu-

facturing, have resulted in a reduction of the allowable design stress in several national standards. The

reductions are mainly caused by a microstructural instability during high temperature service. The al-

ternatives are to strictly limit the maximum metal temperature for this material or to replace it with a

different steel based material grade. Reduction of maximum metal temperature will lead to an increased proportion of nickel alloys in the boiler design, and thereby increase investment costs. Introduction of a

different steel based grade would likely include the use of T91, T92 or other 9-12Cr steels, with the

challenges of mandatory post weld heat treatment of all welds on that part of the membrane wall.

Equipment and procedures for post weld heat treatment of site welds during the erection of martensitic

membrane walls are considered to be the largest challenge. There are currently no known manufactur-

ing and operation experience with 9-12Cr membrane walls (other than HCM12), and reduction of risks

may require further testing and demonstration.

157

8/13/2019 KINA25921ENN_002.pdf


Results from another project with simulated service loads have shown that tube-fin-welds in membrane

walls of Alloy 617B can fail [1]. Based on these results it has been questioned whether the membrane

wall test in COMTES700 have fully demonstrated the ability to operate an Alloy 617B membrane wall

for a 700ºC boiler. Further investigations and testing of Alloy 617B membrane wall samples has been

initiated. These investigations may result in improvements of material selection, fabrication procedures

and membrane wall design.

Materials for superheaters: Results have shown that the tested superheater materials (HR3C,

DMV310N, Sanicro 25, Alloy 617B, Alloy 740, Alloy C-263) generally performed well with limited

oxidation and corrosion, but Sanicro 25 and Alloy 617B did suffer from high corrosion rates in some

positions. The high corrosion rates are believed to be related to the fuel composition and the specific

temperature range, and all tested superheater materials are considered susceptible to the corrosion

mechanism, even though not observed on all materials and samples. Results from another project [2]

did not show corrosion rates in the same range. Further studies of corrosion rates at various tempera-

tures and fuel compositions are recommended to optimise superheater materials selection. Alternatives

include restrictions in fuel compositions or partial replacements of superheater sections within the unit

design lifetime.

Concern was raised regarding the microstructural stability of the tested version of Alloy 740. Structural

stability improvement of Alloy 740 is claimed to include increased Al/Ti ratio and reduced Si content,

which may have an effect on the corrosion properties. Testing of the modified material would be rec-

ommended.

Alloy C-263 was introduced late in the project as an additional candidate material for superheaters. The

operational time with this material was however too short to conclude on the feasibility. Further testing

would be recommended.

Header and piping components: The steepest learning curve in this project was realized with the opera-

tional and repair experiences with the thick walled components in Alloy 617B. It was recognised that post weld heat treatment shall be applied to reduce the susceptibility to stress relaxation cracking, non-

destructive examination is crucial for the inspection of fabrication quality and for integrity assessment

of service exposed components as well as the necessity of a workable and verified weld repair concept

due to the ageing effects with Alloy 617B.

Mechanised ultrasonic testing was successfully qualified and verified in COMTES700 for the relevant

component thicknesses, but realization of an advanced USC power plant will require larger component

thicknesses and further qualification and verification would thus be required. Design guidelines for

nickel based components should include appropriate acceptance levels.

Extensive work on weld repair concepts is ongoing, and the outcome of the concept development andverification are very important for the overall feasibility of thick walled components in Alloy 617B.

Finally it could be considered to introduce Alloy C-263 as material for header and piping, as it would

contribute with higher design stress. Introduction of C-263 would require extensive investigations, in-

cluding materials qualification, studies of stress relaxation cracking susceptibility, verification of repair

concepts and demonstration in general. Major activities have already been initiated within the ENCIO

project [3].

1 Material Qualification for the 700/720°C Power Plant (MARCKO 700), VGB Research project 281,http://www.vgb.org/en/research_project281.html

2 Esbjerg Test Rig, VGB Research project 260, http://www.vgb.org/en/research_project260.html3 European Network of Component Integration and Optimisation (ENCIO), VGB Research project 355,

http://www.vgb.org/en/research_project355.html

158

8/13/2019 KINA25921ENN_002.pdf


Valves: Generally valves operated without major problems. Thick-walled valves, e.g. HP-bypass

valves, showed crack formation starting on the inside of the valve bodies. Since the cracks were unde-

tectable by NDT in the in-service state it is questionable if the tested valves could be operated safely for

long periods. Therefore, in order to reduce the risk of both fatigue and creep cracking, it is vital to test

available improved designs where local stress concentrations due to thickness changes are reduced, and

reduce temperature differences across the wall thickness during load changes (including start and stop

sequences). Furthermore, the design solutions should allow an adequate level of NDT capability. This

will be done in ENCIO and HWT II.

Turbine components: A cast Alloy 625 turbine control valve was included in the steam cycle of

COMTES700, as an activity in a separate project (COMTES700 Turbine Valve [4]). Additional demon-

stration of cast thick-walled Alloy 625 including welds is necessary and covered by the ENCIO-project.

Key components of an advanced USC power plant have been successfully designed, manufactured,

erected and operated for about 22.400 hours in order to demonstrate if the necessary technical solution

to realise a 700°C demonstration power plant are available. The project did not succeed in demonstrat-

ing that all necessary materials and designs can currently be operated without economical risks. Much

new knowledge has been gained but critical issues still need to be solved to reduce risks for the imple-

mentation of a large scale demonstration plant.

COMTES700 impacted a number of follow-up projects on research and on demonstration level. Several

projects directly related to COMTES700-results or – materials are ongoing, monitored by a follow-up

group. Accompanying research have been started on national level, e.g. within the framework of the

German COORETEC-initiative. Most important projects for the further development of 700°C technol-

ogies are the European demonstration project ENCIO and the German demonstration project HWT II.

All these projects will be completed by 2018 allowing a final conclusion on the risks of a 700°C power

plant.

The following brief summary indicates which issues should be avoided or to proceed with:

Explore and test alternative candidate materials for membrane wall materials

Improve corrosion properties for superheater materials

Apply PWHT for thick walled A617B components

Verify weld repair concepts for thick walled A617B components

Further development of NDT methods and acceptance levels

Minimise thermal stresses in component and process design

Improved valve design

Reduce manufacturing costs

2.9.1.2 Potential for use of the results

An engineering study carried out in parallel to COMTES700 determined the specific investment costs

of a commercial power plant with a gross capacity of 1,100 MW, which is based on COMTES700’s

technology, at 2,000 €/kW [5]. These costs are definitely above of the costs of state of the art coal-fired

power plants and have to be significantly reduced before achieving commercial viability. Cost reduc-

tions are to be expected from optimisations of the design concept, but main drivers would be increased

creep strength of alternative materials and optimisation of manufacturing methods.

4 COMTES700-Turbine Valve, VGB Research project 268, http://www.vgb.org/en/research_project268.html5 Pre-engineering study NRW Power Plant 700°C (NRWPP700), Final report, Translation February 2011,

p. 108, http://www.vgb.org/en/research_project297.html

159

8/13/2019 KINA25921ENN_002.pdf


Successful development of a new generation of martensitic steels with higher creep strength properties

than P92 would allow to raise the working environment of 9-12% chromium steels, thus enabling to

reduce the use of expensive Ni-based alloys in advanced USC power plants. Appropriate European

research projects are underway [6].

Since it was the first time that power plant components from Ni-based raw materials were to be manu-

factured, manufacturing in COMTES700 was basically carried out according to the state of the art. This proved to be very time-consuming and prohibitive. Manufacturing methods for components made from

Ni-based materials were standing at the beginning of the learning curve in 2004 which implies a huge

potential for clear reductions of investment costs. Demonstration projects like HWT II and ENCIO now

include promising new approaches of manufacturing and will provide practical experience with these

(forging, HIP metallurgy, welding procedures, heat treatment, …).

The economic potential of advanced USC power plants is to be read against the background of an in-

creasing global demand for coal. Reduction of CO2-emissions from the use of coal as well as savings in

coal is of high importance for emerging economies owning huge own coal resources like India and Chi-

na. In consequence of the Fukushima accident, not only Japan is looking for a sustainable use of coal in

Asia. No wonder that these countries are interested in 700°C power plant technology and are aiming at

starting research projects comparable to COMTES700. Europe is currently in the position of a techno-

logical leadership.

Also in Europe, 700°C power plant technology offers potential for CO2-avoidance by modernizing the

coal-fired power plant fleet. In 2011, a study came to the conclusion that increasing renewables installa-

tion won’t be sufficient to realize the 2020 CO2-reduction targets of the European Union. Replacement

of less efficient by advanced thermal power plants with highest efficiencies has to contribute to the

reduction targets [7], which will be the case even beyond 2020.

If the European Union sticks to the aim to realise carbon capture & storage, 700°C power plant technol-

ogy remains an indispensable prerequisite for a sustainable implementation of this technique. Efficiencylosses of 7-12% in consequence of carbon capture & storage require highest efficient power plants to

minimize inevitable energy losses.

In some European countries where national incentive schemes - in addition to the European Union

Emission Trading System - have spurred rapid growth of renewables installed capacity, balancing of

fluctuating energy supply from variable sources like wind and solar is a very big challenge. If other

options like grid optimisation, storage facilities or energy demand management are limited, thermal

power plants will face the need for flexible operation (part load operation, frequent start-up/shut down).

Depending on the specific national energy mix coal-fired power plants might be used as back-up ca-

pacity. In particular, components made from Ni-based alloys could be also applied in USC power plantsat reduced wall thickness according to improved material properties. Investigations on this option have

been commenced [8]. But the same rapid growth of renewables, which requires back-up capacity, might

kick advanced coal-fired power plants off the market, if its utilisation is reduced beyond profitability.

Disregarding of the fact that cost reductions are to be expected as a result of the technically focussed

follow-up test programmes, the commercial viability of the 700°C technology is depending on electrici-

ty prices, CO2-emission prices, fuel costs/availability and political priorities resulting in different regu-

6 Design of a new generation of 12% chromium steels, VGB Research project 348,

http://www.vgb.org/en/research_project348.html

7 Calculation of CO2-avoidance potential by modernizing coal-fired power plant in the EU until 2020, VGBResearch project 307, http://www.vgb.org/en/research_project307.html

8 Roland Jeschke, et al., Flexibility through highly efficient technology, VGB PowerTech, 05/2012, p. 66.

160

8/13/2019 KINA25921ENN_002.pdf


latory frameworks. These conditions differ significantly subject to the reference framework. The Asian

view point might remain different from Europe, and national priorities may differ in Europe as well.

Last but not least coal is competing with the advantages of renewable energy which on the other hand

include risks related to the fact that these are still evolving technologies.

The technology of a coal-fired power plant with live steam temperatures of 700°C provides an im-

portant option for sustainable energy generation with regard to climate protection and preservation ofresources. Its final marketability depends on the development of rapidly changing global energy mar-

kets.

161

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


(16) Christian Folke, Symposium “Material Innovativ”, 29 March 2006, Universität Bayreuth.

(17) Stephen L. Goodstine, et al., Developments in Materials Technology for Ultrasupercritical Steam

Power Plants, Clearwater Coal Conference, 21 – 25 May 2006, Clearwater, Florida/USA.

(18) Ingo Telöken, et al., Integration and First Operating Experience with a Component Test Facility

Installed into a 700 MW Coal-fired Power Station, PowerGen Europe, 30 May – 1 June 2006, Cologne.

(19) Andreas Helmrich, et al., Materials development for advanced steam boilers, Conference on Mate-

rials for Advanced Power Engineering, 18 – 20 September 2006, Liège/Belgium.

(20) Helmut Tschaffon, The European Way to 700 °C Coal-fired Power Plant, 8th Liège Conference on

Materials for Advanced Power Engineering, 18 – 20 September 2006, Liège/ Belgium.

(21) Stephan Wegerich, COMTES700 – Auf dem Weg zum Kohlekraftwerk mit 50 % Wirkungsgrad,

38. Kraftwerkstechnisches Kolloquium, Dresden/Germany, 24 – 25 October 2006.

(22) Gerhard Weissinger, et al., European Development Program for the 700 °C Power Plant, The Sec-

ond Annual Conference of the Ultra-supercritical Thermal Power Technology Collaboration Network,

Quingdao/China, October 2006.

(23) Thomas Gräb, Schweißtechnische Verarbeitung dickwandiger Komponenten aus Nickelbasiswerk-

stoff für 700°C-Kraftwerke, 2. FDBR-Werkstofftagung, Düsseldorf/Germany, 22 November 2006.

(24) Ingo Telöken, COMTES700 – Auf dem Weg zum Kohlekraftwerk mit 50 % Wirkungsgrad, Ruhr-

Universität Bochum, 5 February 2007.

(25) Lutz Werner, Regelventile in der Kraftwerkstechnik, FDBR-Workshop “Kraftwerke sind mehr als

Turbinen und Kessel”, 28 Februar y 2007, Gelsenkirchen.

(26) Günther Kasparek et al., Projekt COMTES700 – Neue Maßstäbe bei den Isolierdicken. ISOLIER-

TECHNIK Jg. 33, Heft 2/2007, S. 22 – 29.(27) Kai Hesel, Planung einer Versuchsstrecke für Hochtemperatur-Werkstoffe im Kraftwerk Scholven,

22. Rohrleitungstechnische Tagung des FDBR, 6 – 7 March 2007, Gelsenkirchen.

(28) Ralf-Udo Husemann et al., Status of Development of Materials for 700 °C Technology in Coal-

fired Power Stations, 6th NIMS-MPA-IfW Workshop, Tsukuba/Japan, 14 March 2007.

(29) Christian Folke et al., Operating experience with COMTES700 – on track towards the 50plus pow-

er plant, CCT2007, 15 – 17 May 2007, Sardinia/Italy.

(30) Christian G. Stolzenberger, The European Roadmap for the 700 °C USC Power Plant, 7th Interna-

tional Charles Parsons Turbine Conference, Glasgow/UK, 11 – 13 September 2007.

(31) Jørgen Bugge, et al., Development of PF-fired High Efficiency Power Plants (AD700), 20th World

Energy Congress "Energy Future in an Interdependent World", Rome/Italy, 11 – 15 November 2007.

(32) Christian G. Stolzenberger et al., Aktueller Stand und Perspektiven der 700 °C-Technologie, 39.

Kraftwerkstechnisches Kolloquium 2007, 11 – 12 October 2007, Dresden/Germany.

(33) Christian Folke et al., Von COMTES700 zu 50plus – Entwicklungsschritte auf dem Weg zum

Kohlekraftwerk mit 50 % Wirkungsgrad, 39. Kraftwerkstechnisches Kolloquium 2007, 11 – 12 October

2007, Dresden/Germany.

(34) Gregor Gierschner, COMTES700 – On the Track Towards the 50plus Power Plant, New Build

Europe 2008, Düsseldorf, 4 – 5 March 2008.

163

8/13/2019 KINA25921ENN_002.pdf


(35) Dr. M. Bader, H. Tschaffon, COMTES700 – Betriebserfahrungen an dickwandigen Alloy617-

Komponenten bei 700°C, VGB Konferenz "Kraftwerke im Wettbewerb 2009", Prague, 29. und 30 April

2009

(36) Reinhard Klemm, Auf dem Weg zur 700 °C-Technologie, MESSTEC & Automation, Heft 5/2009

(37) Jürgen Pick, Nickel-Basiswerkstoffe für 700°C-Kraftwerke – Erste Erfahrungen, VDMA Technik

Forum Industriearmaturen, Frankfurt 21 January 2010

(38) C. Ullrich, Dr. M. Bader, Dr. O. Wachter, G. Gierschner, First experience with the repair of service

exposed alloy 617 (700°C), VGB Conference "Maintenance in Power Plants 2010", Bremen, 24 and 25

February 2010

(39) Christian Stolzenberger, COMTES700: a multinational EU R&D project for 700°C technology

development, CSM Workshop “Materials and Technologies for Energy Efficiency”, Rome, 31 March

2010

(40) C. Ullrich, G. Gierschner, C. Stolzenberger, H. Tschaffon, „Erfahrungen beim Testbetrieb von

Komponenten für das 700°C Kraftwerk“, 10.

VDI-Fachkonferenz, Kassel, 7 and 8 June 2011

(41) Gregor Gierschner, Christian Ullrich, Helmut Tschaffon, E.ON New Build & Technology GmbH,

From COMTES700 to COMTES+ – Component Tests for a Flexible 700°C Power Plant, MPA Semi-

nar, 6 – 7 October 2011

164

8/13/2019 KINA25921ENN_002.pdf


3 LIST OF TABLES AND FIGURES

Table 1: Availability of technical solutions for critical components .................................................. 18

Table 2: Calculated data of the CTF at full load of the boiler ............................................................ 22

Table 3: Pressure drop of the CTF ...................................................................................................... 22

Table 4: Materials for headers, pipes and tubes and the applied codes............................................... 26

Table 5: Geometrical data of the evaporator ....................................................................................... 27

Table 6: Data from stress calculation .................................................................................................. 27

Table 7: Thermal design data of the evaporator .................................................................................. 28

Table 8: Calculated steam and wall temperatures of the evaporator ................................................... 28

Table 9: Geometrical data of the riffled tubes..................................................................................... 30

Table 10: Geometrical data of the superheater...................................................................................... 31

Table 11: Data for stress calculation ..................................................................................................... 31

Table 12: Thermal design data of the superheater ................................................................................ 33

Table 13: Calculated flows of the CTF ................................................................................................. 34

Table 14: Design data final injection .................................................................................................... 34

Table 15: Design data HP-bypass injection .......................................................................................... 34

Table 16: Course of Commissioning..................................................................................................... 37

Table 17: Dimensions and suppliers of the evaporator materials ......................................................... 41

Table 18: Dimensions and suppliers of steam piping materials ............................................................ 42

Table 19: Dimensions and suppliers of the superheater materials ........................................................ 43

Table 20: Dimensions and suppliers of filler materials......................................................................... 43

Table 21: Chemical composition of novel materials applied for the CTF ............................................ 45

Table 22: Grain size no G according to EN ISO 643 (G-7 ÷ G-1) and ASTM E 112 (G0 ÷ G13) ....... 51

Table 23: NDT conducted on CTF parts ............................................................................................... 55

Table 24: Erection and start-up of the major components .................................................................... 87

Table 25: List of laboratories. ............................................................................................................. 118

Table 26: Summary of results for evaporator tubes ............................................................................ 119 Table 27: Summary of results for evaporator tubes similar welds ...................................................... 120

Table 28: Summary of results for evaporator dissimilar welds .......................................................... 121

Table 29: Summary of results for SH similar/dissimilar welds .......................................................... 123

Table 30: The investigated components .............................................................................................. 124

Table 31: Mechanical properties of various samples. RT = room temperature, HT = 700°C ............ 125

Table 32: Mechanical properties of pipe sample SA1 and weld SAW1. ............................................ 126

Table 33: Investigated dissimilar welds .............................................................................................. 127

Table 34: Test programme with six different electrode and TIG orbital welds .................................. 135

Table 35: Positions and failure types with reference to Figure 130 and 131. ..................................... 141

Table 36: Fatigue and creep lifetime consumption based on EN 12952-4:2011 ................................ 143

Table 37: Total consumption of lifetime based on EN 12952-4:2011 ................................................ 144

165

8/13/2019 KINA25921ENN_002.pdf


Figure 1: Construction of the test heat exchange surfaces and location in the boiler ........................ 20

Figure 2: Isometric drawing of the CTF ............................................................................................. 21

Figure 3: P&I diagram of the CTF ..................................................................................................... 25

Figure 4: Outline of the evaporator panel........................................................................................... 27

Figure 5: Calculated wall temperatures for T24 ................................................................................. 29

Figure 6: Calculated wall temperatures for HCM12 and Alloy 617B ................................................ 29

Figure 7: Geometry of riffled tubes .................................................................................................... 30

Figure 8: Arrangement of the superheater surface in the host boiler ................................................. 32

Figure 9: Outline of attemperator arrangement .................................................................................. 34

Figure 10: Pressure drop of nozzles vs. flow capacity of attemperator ................................................ 35

Figure 11: Details of spray attemperator 3 ........................................................................................... 35

Figure 12: Flash tank ............................................................................................................................ 36

Figure 13: 3.1A certificate of HCM12 evaporator tube ....................................................................... 41

Figure 14: PMA of Alloy 617B ............................................................................................................ 42

Figure 15: 3.1B certificate of Alloy 617B welding rod ........................................................................ 44

Figure 16: Interim certificate of the Notified Body (RWTÜV) ........................................................... 46

Figure 17: Cover sheet of WPQR HCM12/Alloy 617B BW ............................................................... 47

Figure 18: Final machining of inner surface of Alloy 617B ................................................................ 48

Figure 19: Alloy 617B. with grain size number G= -3 ......................................................................... 48

Figure 20: Forging of Alloy 617B (OD219.1 x 50mm) ....................................................................... 49

Figure 21: TIG fusion welding without filler metal ............................................................................. 49

Figure 22: TIG welded base material and after PT on the right side .................................................... 50

Figure 23: TIG/SMAW welding of Alloy 617 tube in fixed position .................................................. 50

Figure 24: New heat treatment method for Alloy 617 components ..................................................... 51

Figure 25: Indication of DMV 617 SH-Tube ....................................................................................... 52

Figure 26: Microsection of DMV 617 SH-Tube .................................................................................. 52

Figure 27: Superheater outlet header .................................................................................................... 52

Figure 28: Spray Attemperator ............................................................................................................. 53 Figure 29: Evaporator outlet header (P92) ........................................................................................... 53

Figure 30: Manufactured panel with T24/HCM12/Alloy 617B. .......................................................... 54

Figure 31: Indication found after PT .................................................................................................... 54

Figure 32: PT testing of Alloy 617B V-weld (OD219.1 x WT50mm) and while welding ......................with TIG/SMAW ................................................................................................................ 56

Figure 33: Cross sectional drawing of gate valves GH 251 and DN125 (KKS no. 26NA76S004) ..... 60

Figure 34: Cross sectional drawing of the globe valves VHA510 and DN50 ..........................................(KKS no. 26NE22S521) ..................................................................................................... 61

Figure 35: Cross sectional drawing of safety valve SOH (KKS no. 26NA77S094) ............................ 62

Figure 36: Cross sectional drawing of swing check valve KRH400 and DN80 ......................................

(KKS no. 26NA76S508) ..................................................................................................... 63

Figure 37: Cross sectional drawing of check valve VR500 and DN50 (KKS no. 26NE22S522) ........ 63

Figure 38: Left: Cross sections of start-up valve (KKS no. 26NA76S002); Rright: control valve ..........(KKS no. 26NA76S006) ..................................................................................................... 64

Figure 39: HP-bypass valve (KKS no. 26NA77S003) ......................................................................... 64

Figure 40: Overview of casing neck weld of gate valve. Red arrows show cracks after dye .................. penetrant testing (KKS no. 26NA76S004-01) .................................................................... 66

Figure 41: Adhesion damage of seat ring (KKS no. 26NA76S004-01) ............................................... 66

Figure 42: Seat of the HP-bypass after 22,000 hours of operation ....................................................... 67

Figure 43: Dissmanteld HP-bypass valve body for detailed investigation ........................................... 68

Figure 44: Cross section of the HP- bypass valve (seam ‘A’) with cracks (arrows) and the ....................appropriate microscopic view on a crack propagated from a thread root ........................... 68

Figure 45: Hardness gradient in valve cross section ............................................................................ 69

166

8/13/2019 KINA25921ENN_002.pdf


Figure 46: Outer temperature profile of a thermocouple ...................................................................... 74

Figure 47: Pressure measurement equipment as used for the CTF. First shut-off valves .......................(see WP3 Valves), impulse pipes and transmitter shut-off valve made of Alloy 617B ...... 75

Figure 48: Design drawing of pressure gauge stop valve tailor-made for the CTF ............................. 76

Figure 49: Pressure gauge stop valve built in for long-term design testing under high temperature .......conditions (insulation dismantled). The white insulation is high temperature insulation .......

microtherm, the other is a standard insulation material. ..................................................... 76

Figure 50: Flow measuring venture nozzle design as chosen for the CTF ........................................... 77

Figure 51: External view of the installation position of the venture flow nozzle................................. 77

Figure 52: Readings during commissioning of the closed loop control ............................................... 78

Figure 53: Evaporator and superheater temperture control monitor screen shot.................................. 78

Figure 54: Temperature differentials vs. the temperature band ............................................................ 79

Figure 55: Control system architecture ................................................................................................ 80

Figure 56: Old Scholven F control room .............................................................................................. 82

Figure 57: COMTES700 area of the control room ............................................................................... 82

Figure 58: The specific insulation factor of insulation materials at different temperatures ................. 83

Figure 59: The companies involved in the erection phase of the CTF ................................................. 86

Figure 60: Scaffolding in front of the evaporator inside the boiler ...................................................... 88

Figure 61: Left: superheater tubes inside the boiler. Right: marks on superheater .............................. 88

Figure 62: Left: superheater outlet header before dismantling. Right: end of 700°C steam pipe ............at the end of the first dismantling phase ............................................................................. 88

Figure 63: Left: Determination of cutting points. Right: Pipe parts with identification marks after .......severing ............................................................................................................................... 89

Figure 64: Deposits in a pre-heating pipe detected during the dismantling ......................................... 90

Figure 65: Crack in spray injection nozzle after HP-bypass valve detected during the dismantling ... 90

Figure 66: Left: Superheater tubes in shelves. Right: Evaporator ........................................................ 90

Figure 67: Left: Stored pipes. Right: Stored valves ............................................................................. 91

Figure 68: Hours of operation .............................................................................................................. 92 Figure 69: Number of starts .................................................................................................................. 93

Figure 70: Temperature distribution in the superheater ....................................................................... 93

Figure 71: Main activities and non-availabilities ................................................................................. 94

Figure 72: CTF parts with failures ....................................................................................................... 95

Figure 73: Failure in the superheater .................................................................................................... 96

Figure 74: Intergranular crack at tube 1 ............................................................................................... 96

Figure 75: Left: Inner surface. Right: Outer surface ............................................................................ 97

Figure 76: Location of spray attemperator 2 in the CTF ...................................................................... 98

Figure 77: Spray attemperator 2 ........................................................................................................... 98

Figure 78: Indications at stub welds of the injection nozzle assembly ................................................ 99

Figure 79: Injection nozzle after grinding ............................................................................................ 99

Figure 80: Cracks at the stubs of spray injection nozzle assemblies and extraction of boat-shaped ........samples.............................................................................................................................. 100

Figure 81: Spray attemperator 2 support construction with bracket .................................................. 101

Figure 82: Indications at bracket of spray attemperator 2 .................................................................. 101

Figure 83: Results of the UT of spray attempeartor at bracket .......................................................... 102

Figure 84: Condition of protective pipe, seen after cutting off crack-affected pipe section .............. 102

Figure 85: Results of the visual inspection, crack of protective pipe in the area of injection ..................nozzle assembly ................................................................................................................ 103

Figure 86: Different cuttings during the repair works ........................................................................ 103

Figure 87: Damage affected area before and after removal of insulation .......................................... 104

Figure 88: Investigation of the crack .................................................................................................. 105

Figure 89: Location of spray attemperator 3 ...................................................................................... 106

167

8/13/2019 KINA25921ENN_002.pdf


Figure 90: Spray attemperator 3 ......................................................................................................... 106

Figure 91: Defective weld at counter bearing of spray injection nozzle assembly ............................ 107

Figure 92: Sample for laboratory investigation .................................................................................. 107

Figure 93: Metallographic investigation ............................................................................................ 108

Figure 94: Crack of spray attemperator 3 inlet weld .......................................................................... 108

Figure 95: Deflection of turbine valve ............................................................................................... 109

Figure 96: Laboratory investigation of crack at spray attemperator 3 inlet weld ............................... 110

Figure 97: Old protective pipe. Left: Pipe broken in 5 segments. Right: deformation at .......................injection nozzle assembly part of the attemperator ........................................................... 111

Figure 98: crack at circumferential weld and stub weld of the Alloy 617B main pipe ...................... 112

Figure 99: Further indications at stubs ............................................................................................... 113

Figure 100: Indications at stubs after the heat treatment ...................................................................... 114

Figure 101 Indications at circumferential welds after heat treatment ................................................. 114

Figure 102: Cutting of indications ........................................................................................................ 115

Figure 103: Indications at repair weld .................................................................................................. 116

Figure 104: Left: LOM image of SP2. Right: BSE-SEM image of BP1 (arrows: see text). ................ 124

Figure 105: Steam side oxide. Left: BSE-SEM image of H3. Middle: LOM image of SPB2 after ...........etching. Right: LOM image of BP1 .................................................................................. 125

Figure 106: Shallow pits developed on the steam side of DE1 ............................................................ 126

Figure 107: Left: View from inside header H3 with bore holes after penetrant testing. ............................Right: Hot crack observed in weld SHHW1 near outer surface ....................................... 126

Figure 108: Hardness measurements performed across dissimilar welds between P92 and ......................Alloy 617B. r = root layer, t = top layer ........................................................................... 127

Figure 109: Hardness measurements performed across the dissimilar weld between 10CrMo9-10 ..........and Alloy 617B (DW8). r = root layer, t = top layer ........................................................ 128

Figure 110: Left: The HAZ of P92. Right: Alloy 617B (right) in dissimilar weld, DW7 .................... 128

Figure 111: Left (DW3): SEM images and EDS line scan across the interface between weld metal ........

and P92. Right (DW8): Weld metal and 10CrMo9-10. The contents of Ni, Cr, Fe, and ........Co at end positions in wt% ............................................................................................... 129

Figure 112: Left (DW4): Steam side oxide at the transition between, P92 and weld metal. .....................Right (DW8): 10CrMo9-10 and weld metal ..................................................................... 129

Figure 113: Crack that occurred during operation of the CTF. Left: Typical crack path close to .............a stub. Right: Associated microstructure .......................................................................... 130

Figure 114: Indications that occurred directly after welding of service exposed Alloy 617B in ..............he area of welds: a) crack in the weld, b) PT-indications during non-destructive testing, .....c) crack in the heat affected zone ...................................................................................... 131

Figure 115: Left: Indications before operation. Right: Indications after 500 hours of operation......... 132

Figure 116: General test set-up of bending test .................................................................................... 132

Figure 117: Result of the 3-point bending test: a) service exposed sample, b) load-time diagram ............for sample out of a), c) service exposed + 980°C for 3h, d) load time diagram for the ..........sample out of c) ................................................................................................................. 133

Figure 118: Microstructure of service exposed Alloy 617B after different heat treatments ......................(22,000h at 700°C): a) service exposed, b) service exposed + 980°C for 3h, c) service ........exposed + 1160°C for 1h .................................................................................................. 134

Figure 119: Medium size of carbides in service exposed and heat treated Alloy 617B ....................... 134

Figure 120: Test pipe with heating mats and insulation ....................................................................... 135

Figure 121: Arrangement of 18 thermocouples for the 1160°C heat treatment test ............................. 136

Figure 122: Heating of the test pipe ..................................................................................................... 136

Figure 123: Measured temperatures during the heat treatment test...................................................... 136

Figure 124: Rapid cooling down process after the heat treatment with water ..................................... 137

Figure 125: Measured temperatures during cooling in the heat treatment test .................................... 138

168

8/13/2019 KINA25921ENN_002.pdf


Figure 126: Comparison between electrode welds and TIG orbital welds ........................................... 138

Figure 127: Metallographic investigation of the six test welds ............................................................ 139

Figure 128: Test set-up for mechanised ultrasonic investigations ....................................................... 140

Figure 129: Cross section of qualification component with reflector positions .........................................(notches and drilled holes) ................................................................................................ 140

Figure 130: Comparison of mechanised ultrasonic testing (left) and micro-section (right) ................. 141

Figure 131: Microsections with higher magnification ......................................................................... 141

Figure 132: Consumed lifetime in Alloy 617B superheater tubing ...................................................... 142

Figure 133: CAE geometry of spray attemperator ............................................................................... 144

Figure 134: Analysed part positions of spray attemperator 2 ............................................................... 145

Figure 135: Fixed-point thermocouple principle .................................................................................. 146

169

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


P91 Ferritic-martensitic steel: EN 10216-2; VdTÜV 552/2 X10CrMoVNb9-1, W. Nr.1.4903

P92 Ferritic-martensitic steel: EN 10216-2; VdTÜV 511 X10CrMoVNb9-2, W. Nr.1.4901;

PGIM Power Generation Information Manager

PMA Particular material appraisalProvia Commissioning tool for data storage and visualisation based on Excel prWHT Pre weld heat treatment

PT Penetrant testing

PWHT Post weld heat treatment

Rear side The section of the tube periphery away from the direction of the burners

Rohr2 Pipe stress analysis programme for static and dynamic analysis of complex pip-ing and skeletal structures

RT Radiographic testing

Sanicro 25 Austenitic steel: VdTÜV 555 Sanicro® 25

SAW / SMAW Submerged Arc welding / Shielded metal arc welding

Similar weld Weld between two tubes of similar material quality e.g. HCM12 to HCM12

Steam side oxidation Oxide on the inner periphery of the tubesT24 Low alloy ferritic steel: EN 10216-2 7CrMoVTiB10-10, Steel number 1.7378;

VdTÜV 533 7CrMoVTiB10-10, W.Nr. 1.7378; ASTM A-213 T24

TIG Tungsten inert gas welding or Gas tungsten arc welding

TiO2 Titanium dioxide

TRD German standard of technical rules for steam boiler

TXP Modern DCS system produced by Siemens for modern Power plants with inte-

grated BPS system

Type N Type N (Nicrosil – Nisil) (Nickel-Chromium-Silicon/Nickel-Silicon) thermocouples are suitable for use at high temperatures, exceeding 1.200 °C, due to theirstability and ability to resist high temperature oxidation.

Type S Type S thermocouples use platinum or a platinum – rhodium alloy for each con-ductor. These are among the most stable thermocouples, but have lower sensitivi-ty than other types and are usually used only for high temperature measurementsdue to their high cost and low sensitivity.

UT Ultrasonic testing

WM Weld metal

WPQR / WPS Welding procedure qualification record / welding procedure specification

ZeO3 Zirconium trioxideγ'- phase Precipitates in Ni-base alloys consisting of Ni, Al and possibly Ti

η - phase Precipitates in Ni-base alloys consisting of Ni and Ti

171

8/13/2019 KINA25921ENN_002.pdf


8/13/2019 KINA25921ENN_002.pdf


European Commission EUR 25921 — Component test facility for a 700 °C power plant (Comtes700) Luxembourg: Publications Office of the European Union 2013 — 171 pp. — 21 × 29.7 cm

ISBN 978-92-79-29379-5doi:10.2777/98172

8/13/2019 KINA25921ENN_002.pdf


K I -N A -2 5 9 2 1 -E N -N

The objective of Comtes700 was a test of components for the 700 °C power

plant technology. A test facility of steam generator components up to a wall

thickness of 50 mm was erected in an existing power plant. A partial steam flow

from the host plant was heated up to 700 °C and operated for 22 400 hours.

Compared to ferritic components, machining of nickel-based components lasted

a minimum of four times longer. Maximum grain sizes in semi-finished products

were achieved by a modified heat treatment. After adjustment of testing

technology, overlaps of alloy 617B tubes were avoided.

Evaporator materials functioned satisfactorily during operation. Updatedmaterial properties and reduced mechanical loads restricted the reliability

statement. Superheater materials only failed once in a dissimilar weld. Some

superheater materials may not reach a lifetime of 200 000 hours due to

fireside corrosion. Thick-walled components from 30 to 50 mm in wall thickness

displayed a susceptibility to stress relaxation cracking in welds. Additional

stresses were caused by two-phase flow and insufficient design of spray

attemperator. A solution for the workshop and repair welds was to conduct

KINA25921ENN_002.pdf

Documents