Top Banner
Coversheet Aramco Overseas Company Vendor Document P.O. No.: 4502941884 Project : Berri Ethane & LNG Recovery Plant Doc Code : T.B.A. Description: Cryogenic Test on Pneumatic Actuated Vanessa Series 30.000 Valves Item / Tag: F20-HOV-4021 F20-HOV-4024 Revision: 0 Date: 06 July 2010 Tyco Ref: 3023769 Tyco Doc. No. : 23769 040 (KVS-10/1497-CR)
17
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: 23769_040 (KVS-10-1497-CR)

Coversheet

Aramco Overseas Company Vendor Document

P.O. No.: 4502941884

Project : Berri Ethane & LNG Recovery Plant

Doc Code : T.B.A.

Description: Cryogenic Test on Pneumatic Actuated Vanessa Series 30.000 Valves

Item / Tag: F20-HOV-4021 F20-HOV-4024

Revision: 0

Date: 06 July 2010

Tyco Ref: 3023769

Tyco Doc. No. : 23769 – 040 (KVS-10/1497-CR)

Page 2: 23769_040 (KVS-10-1497-CR)

Tipo: Type: TEST PROCEDURE

Emesso da: Titolo:Issued by: ENGINEERING

Numero:

Title:

CRYOGENIC TEST ON PNEUMATIC ACTUATED VANESSA SERIES 30,000 VALVES FOR BERRI ETHANE RECOVERY PLANT, SAUDI ARAMCO P.O. N° 4502941884

Number: KVS-10/1497-CR Page: 1 of 16

SOSTITUISCE DOC. N° REV. DATA DESCRIZIONE

REPLACES DOC. N° REV. DATE DESCRIPTION

Rev.: Data: Descrizione: Preparato: Controllato: Rev.: Date: Description: Prepared: Checked:

00 05/07/2010 Issued for Comments and/or Approval A. NEGRI (Eng. Dept.) 

D. VIDONI (H&S Dept.)

Approvato: Approved:

L. VERTOVA (Eng. Mgr.)

P. BOTTI (Operations Mgr.)

F. CASTAGNETTI (Q.A. Mgr.)

CRYOGENIC TEST ON PNEUMATIC ACTUATED VANESSA SERIES 30,000

VALVES FOR BERRI ETHANE RECOVERY PLANT, SAUDI ARAMCO P.O. N°

4502941884

Page 3: 23769_040 (KVS-10-1497-CR)

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

KVS-10/1497-CR 00 05/07/2010 Issued for Comments and/or Approval 2

VANESSA TEST PROCEDURE

KVS-10/1497-CR

Table of Contents

1 SCOPE 3

2 APPLICABILITY 3

3 REFERENCE DOCUMENTS 4

4 RESPONSIBILITY AND SAFETY 4

5 EXECUTIVE MODALITIES 5

5.1 TEST INSTRUMENTATION 5

5.2 TEST PARAMETERS 6

5.3 TEST RIG PREPARATION 7

5.4 INITIAL PROVING AT AMBIENT TEMPERATURE 7

5.5 COOLING DOWN 9

5.6 CRYOGENIC TEMPERATURE SEAT TESTING 9

5.7 CRYOGENIC TEMPERATURE EMISSIONS TESTING 10

5.8 WARM UP AND FINAL INSPECTION 11

5.9 ACCEPTANCE CRITERIA 11

5.10 CERTIFICATE AND REPORTS 12

6 ARCHIVE 12

7 ATTACHMENTS 12

Page 4: 23769_040 (KVS-10-1497-CR)

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

KVS-10/1497-CR 00 05/07/2010 Issued for Comments and/or Approval 3

1 SCOPE This procedure is intended to give more detail about the cryogenic testing steps mentioned in the

applicable VANESSA Quality Control Plan.

The scope of such testing is basically to prove that the VANESSA valves in object are intrinsically easily

operable and tight, through the seat and towards atmosphere, at both ambient and cryogenic (-196°C)

temperatures, as per the applicable specifications.

Valves seat leakage will be measured via a mass flow-meter (or bubble cylinder) while the valve

bodies are immersed in liquid nitrogen.

The external leakage (i.e. the so-called “fugitive emissions”)1 will be detected using a mass

spectrometer probe in a “sniffing method” : valve body and bonnet shell, packing area and bottom flange will

be investigated, at both cryogenic and ambient temperatures.

Valve operability (by means of the relevant pneumatic actuator) will be also checked throughout the

test.

2 APPLICABILITY This procedure defines the main requirements and guidelines regarding the cryogenic testing

(including seat, external leakage and operability checks) of the VANESSA Series 30,000 valves to be supplied,

through TYCO VALVES & CONTROLS DISTRIBUTION NL (Breda), to SAUDI ARAMCO for the BERRI ETHANE RECOVERY

PLANT.2

This procedure refers to VANESSA ack. ner VA10/1497 (TYCO V&C’S NL P.O. ner 5049697; ARAMCO

P.O. ner 4502941884) and aims to satisfy the relevant Customer’s requirements.

The units to be tested shall be the two (2) double flanged, ASTM A 351 CF8M body material, cryogenic

configuration VANESSA  Series  30,000 valves listed in attached APPENDIX C, with the relevant  BIFFI  ALGAS pneumatic actuators (one valve per each item).

1 A fugitive emission is defined as any chemical or mixture of chemicals which represents an unanticipated and/or spurious leak (see also definitions in ISO 15848-1 standard and Shell MESC SPE 77/312 specification).

2 See http://www.saudiaramco.com/index.htm .

Page 5: 23769_040 (KVS-10-1497-CR)

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

KVS-10/1497-CR 00 05/07/2010 Issued for Comments and/or Approval 4

3 REFERENCE DOCUMENTS

The reference document to be followed for anything not mentioned in the present procedure shall be

Appendix A of BS 6364 standard (September 1998).

For general information and comparison, the following standards and specifications may also be consulted:

EN 1779 (August 1999)

ISO 15848-1, Appendix B (January 2006);

ISO 15848-2 (August 2006);

Shell MESC SPE 77/200 (August 2008);

Shell MESC SPE 77/306 (February 2005);

Shell MESC SPE 77/312 (February 2005);

Shell MESC SPE 77/312 (June 2007);

VANESSA SCF 010 NDE (May 2007);

VANESSA SDF 066 (February 2003).

4 SAFETY AND RESPONSIBILITY All activities shall be in accordance with the HSE Rules (conforming to OHSAS 18001 standard) of

VANESSA Company and with the national laws in force about safety.

The special testing operations shall be executed within a dedicated area, properly confined and

signalled to prevent the entry of unauthorized Personnel.

The specialized Personnel of the VANESSA Assembly & Testing Department will be fully responsible for

the execution of the test according to the criteria of the present procedure.

Page 6: 23769_040 (KVS-10-1497-CR)

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

KVS-10/1497-CR 00 05/07/2010 Issued for Comments and/or Approval 5

Each Operator shall have received a task-oriented safety training and be aware of the operation

specific risks, as well as of the related preventive and protective measures (including technical solutions,

procedures and personal protection equipment).

In particular, the VANESSA safety procedures usually applicable in case of special testing may be are

the following:

- PRO.SPP.02 about high pressure gas tests and admittance to the relevant “bunker” areas;

- IOS.09 about safe admittance to confined spaces.

The leak testing Operators shall be qualified according to EN 473 and/or ASNT TC-1A standards (at least “Level 1”) for helium leak detection method (LT and/or MSLT).

The VANESSA Quality Assurance Department has the complete authority in the critical test decisions

(e.g.: in case of non-compliance or failure) and will verify the results of the tests; VANESSA Engineering Department may also be consulted, if necessary.

5 EXECUTIVE MODALITIES

5.1 TEST INSTRUMENTATION

The test will be carried out by using the standard and special equipment periodically verified

according to the VANESSA Quality Management System and to internal calibration procedures. All measuring

instruments shall be supported by calibration certificates, within their period of validity3 at the date of testing,

which shall be submitted for review and approval, if required by witnessing party prior to testing.

3 The validity period of each instrument’s calibration is stated within the VANESSA Quality System depending on the type of instrument, the usage frequency, the instrument Manufacturer’s indications and the operative experience. Any further calibrations will be performed only on demand and at the Customer’s expense.

Page 7: 23769_040 (KVS-10-1497-CR)

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

KVS-10/1497-CR 00 05/07/2010 Issued for Comments and/or Approval 6

The main elements of test equipment (see APPENDIX A) are:

- two flanged caps for each valve, with appropriate inlet and outlet piping for test medium and connections for the thermocouples, in order to measure internal valve temperature;

- at least five (5) low temperature (type T) thermocouples;

- the relevant thermocouple signal reading and recording system;

- a controlled helium gas supply;

- helium-calibrated gas mass flow-meters to detect and measure seat leakage;

- the cryogenic tank and the controlled liquid nitrogen supply;

- adequate pressure gauges.

A mass spectrometer (a VARIAN 979C model, having a sensitiveness level of 2*10-9 mbar*l/s) with

sniffer probe and the relevant (capillary type) standard leak,4 for fugitive emissions detection, completes the

test instrumentation (see APPENDIX B).

5.2 TEST PARAMETERS

Ambient temperature TAMB shall be between +10° and +38°C (+50° and +100°F).

Minimum test temperature TMIN shall be referred to the valve internals and comprised within a

tolerance range of ±5°C (±9°F) around the value of -196°C (-321°F).

Being the valves pneumatic actuated, the maximum test pressure pMAX shall be equal to the design

differential pressure (as per the applicable valve data-sheet) the actuator has been selected for.

In the specific case of interest: pMAX = 41.2 barg for both valves. According to the most common specifications and practices, the pMAX value shall be reached in three

(3) to five (5) equal pressure increments, so that the first step pressure value p1 shall be a fraction of pMAX: in

the specific case of interest, it is advisable to apply four (4) increments, of about 10.3 bar each.

According to ISO 15848-2 standard, clause 5.3, and to safety principles, the test pressure pFE during

fugitive emissions detection shall be equal to 6.0 barg.

4 Mass spectrometer is calibrated before each test by using a well defined and calibrated leakage and compensating for background reading, as per VANESSA internal procedure SCF 010 NDE.

Page 8: 23769_040 (KVS-10-1497-CR)

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

KVS-10/1497-CR 00 05/07/2010 Issued for Comments and/or Approval 7

5.3 TEST RIG PREPARATION The valve to be tested shall have passed the standard shell and seat tightness tests at ambient

temperature.5

Valve and related accessories shall be clean, dry and not painted. The test equipment shall be

installed as shown in APPENDIX A, with valve stem in vertical position. The valve shall be tested with the

relevant pneumatic actuator.

Each cap will be equipped with a fitting for inlet and outlet of test medium. Furthermore, connections

are required for the thermocouples to measure the valve internal temperature. In total, five (5) thermocouples

shall be used to monitor the temperature of valve and coolant. Following temperatures shall be measured

throughout the cryogenic temperature test:

- valve body external surface;

- valve inside, shaft side;

- valve inside, disc side;

- bonnet outside (stuffing box area);

- coolant.

The leakage test medium and gas for purging shall be commercially (i.e. more than 97%) pure helium.

The coolant shall be liquid nitrogen.

5.4 INITIAL PROVING AT AMBIENT TEMPERATURE

An initial proving test6 shall be performed at ambient temperature, in order to ensure that the valve

and the system are in a suitable condition for the test to proceed.

The valve disc shall be set in half-open position and the system shall be pressurized with helium gas

up to pFE = 6.0 barg (see paragraph 5.2 above).

The leak detection system shall be calibrated following VANESSA internal procedure.7

5 According to the relevant Quality Control Plan.

6 See BS 6364, Appendix A, clause A.3.1.2.

7 See Note 5 above.

Page 9: 23769_040 (KVS-10-1497-CR)

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

KVS-10/1497-CR 00 05/07/2010 Issued for Comments and/or Approval 8

The valve body shall be checked for porosity and leakage, while the following areas shall be inspected

with the mass spectrometer’s sniffing probe:

- the stem seal and stuffing box (packing area);

- the bottom flange;

- the body to bonnet flange;

- each valve end to cap junction and test rig piping (check only).

Any leakage found in the system (test piping, joints, needle valves, caps flanges, etc.) shall be repaired

and re-checked before proceeding.

If leakage exceeding the maximum acceptable (see paragraph 5.9 below) is detected from the valve,

the pressure shall be released, the relevant bolts shall be re-tightened (up to the proper torque value, as

specified in VANESSA VMS 027 standard assembly procedure) and the test shall be repeated.

Once ensured that the system and the valve are tight enough towards atmosphere, the initial valve seat

leakage and operability shall be checked too.

The valve shall be closed with the relevant torque and gradually pressurized on the shaft side with

nitrogen or helium gas up to pMAX = 41.2 barg (see paragraph 5.2 above); any pressure shall be released from

the downstream (disc) side, then the seat leakage shall be measured for at least 5 min (after pressure

stabilization).

The maximum acceptable shall be the same as for the cryogenic temperature testing (see paragraph

5.9 below).

On completion of the seat leakage detection, the valve shall be opened against pressure and cycled

closed and open again five (5) times, to check for ease of operability by means of the pneumatic actuator.8

At the end, the valve shall be depressurized, opened and lowered into the tank for cooling down.

8 Being a quarter-turn, triple offset, torque seated valve, the VANESSA Series 30,000 has a rubbing free design throughout the whole 90 degrees rotation, during which the torque requirements are very low: the only applicable resistant torque features are the End to Close and the Start to Open torques, which control the seating and unseating operations of the valve.

Page 10: 23769_040 (KVS-10-1497-CR)

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

KVS-10/1497-CR 00 05/07/2010 Issued for Comments and/or Approval 9

5.5 COOLING DOWN Cooling down of the test valve shall be carried out in an apposite insulated tank, that shall be filled

with liquid nitrogen up to the top of the valve body neck flange (see APPENDIX A).

Throughout the cooling down operation, the valve shall be kept in open position and a purge of about

0.5 barg helium gas shall be maintained, to prevent the freezing of moisture inside the valve.

Temperatures shall be recorded (at least every 10 min) throughout the cooling down period.

On reaching of the specified TMIN temperature (as defined in paragraph 5.2 above), helium gas

purging shall be stopped. The valve shall be kept for at least 15 min in open position to let all the temperatures

stabilize.

5.6 CRYOGENIC TEMPERATURE SEAT TESTING

When thermocouples readings show that valve temperatures have stabilized within the specified

range, cryogenic testing may begin.

The valve shall be closed with the relevant torque and gradually pressurized on shaft side up to the

relevant first step p1 = 10.3 barg pressure value; pressure and temperature shall be stabilized for at least 10

min, then the leakage shall be measured over an observation period of at least 5 min.

The disc side of the valve shall be then depressurised and the pressure on shaft side shall be gradually

increased (without discharging), in order to apply the following pressure increment (i.e. up to p2 = 20.6 barg);

the minimum stabilization and observation period durations shall be the same as the first step (10 min and 5

min, respectively).

This shall be repeated in four (4) equal pressure increments up to p4 = pMAX = 41.2 barg to be reached

in the last step.

On completion of the last seat leakage detection step, the disc side of the valve shall be depressurized

while the relevant pMAX value shall be maintained on the shaft side; then, at least 1 min after pressure

stabilization, the valve shall be cycled open and closed five (5) times, in order to check for ease of operability.

Page 11: 23769_040 (KVS-10-1497-CR)

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

KVS-10/1497-CR 00 05/07/2010 Issued for Comments and/or Approval 10

After that, the valve shall be cycled to the closed position and pressurized with helium on the disc

side, up to the relevant first step p1 pressure value, in order to repeat, in the opposite direction, the same seat

leakage test in five (5) steps described above.

On completion of the last seat leakage detection step, the valve shall be depressurized and opened.

5.7 CRYOGENIC TEMPERATURE EMISSIONS TESTING

The disc shall be set in half-open position and the valve body shall be pressurized up to pFE = 6.0 barg

(see paragraph 5.2 above).

The leak detection system shall be calibrated following VANESSA internal procedure9.

The valve shall be lifted from the liquid nitrogen and laid on the floor.

Valve body shall be immediately checked for porosity and leakage, while the following areas shall be

inspected with the mass spectrometer’s sniffing probe:

- the stem seal and stuffing box (packing area);

- the bottom flange;

- the body / bonnet flange;

- each body flange / cap junction and test rig tubing (check only).10

9 See Note 5 above.

10 As long as it does not affect the detection of fugitive emissions from the valve (packing, bottom and bonnet flange), the leakage from the test rig is out of the scope of this test. See also para. 5.4 above.

Page 12: 23769_040 (KVS-10-1497-CR)

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

KVS-10/1497-CR 00 05/07/2010 Issued for Comments and/or Approval 11

5.8 WARM UP AND FINAL INSPECTION

After successful completion of all tests, the valve and caps assembly shall be definitively removed

from the tank and allowed to warm up spontaneously to ambient temperature (with the disc in the full open

position, if possible).

Once the metal temperature allows safe handling, the caps shall be removed, in order to visually

inspect the valve and the relevant pneumatic actuator (without disassembling) and to check them for any

signs of unanticipated wear or damage.

5.9 ACCEPTANCE CRITERIA

Regarding the seat leakage, the maximum allowable shall be 100 mm3/s per mm of valve DN = 150 ml/min per in of valve ND. The resulting maximum seat leak rates are listed, for each valve to be tested, in

attached APPENDIX C.

Regarding the external leakage, the maximum acceptable helium leak rate shall be 1*10-4 mbar*l/s

of actual helium leak rate.

Regarding the valve operability, by means of the relevant pneumatic actuator (with the proper

settings) it shall be always possible to easily: close the valve and reach the required tightness, open the valve

against pressure and perform both strokes in a regular manner, without any resistances, steps or delays.

Page 13: 23769_040 (KVS-10-1497-CR)

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

KVS-10/1497-CR 00 05/07/2010 Issued for Comments and/or Approval 12

5.10 CERTIFICATE AND REPORTS All results of the tests shall be properly recorded and certified by VANESSA Quality Assurance

Department. The relevant reports and certificates shall be signed by the witnessing Test Inspector (whenever

applicable).

6 ARCHIVE This document shall be sent to the Customer for approval and/or comments. It shall be archived by

the Issuer and, once it has been approved, by the VANESSA Quality Assurance Department (together with the

relevant test reports and certificates). Everybody in charge of using this document shall verify to hold the

latest valid revision of the document itself.

7 ATTACHMENTS

Appendix A: CRYOGENIC TEST EQUIPMENT… … page 13

Appendix B: EXTERNAL LEAKAGE TEST EQUIPMENT… … page 14

Appendix C: TEST ITEMS SUMMARY TABLE… … page 15

Appendix D: INSTRUCTIONS IN ITALIAN LANGUAGE… … page 16

Page 14: 23769_040 (KVS-10-1497-CR)

APPENDIX A: CRYOGENIC TEST EQUIPMENT

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

Issued for Comments and/or Approval 13 KVS-10/1497-CR 00 05/07/2010

Page 15: 23769_040 (KVS-10-1497-CR)

APPENDIX B: EXTERNAL LEAKAGE TEST EQUIPMENT

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

Issued for Comments and/or Approval 14 KVS-10/1497-CR 00 05/07/2010

Page 16: 23769_040 (KVS-10-1497-CR)

APPENDIX C: TEST ITEMS SUMMARY TABLE

Numero: Rev.: Data: Descrizione: Pagina:

Number: Rev.: Date: Description: Page:

Issued for Comments and/or Approval 15 KVS-10/1497-CR 00 05/07/2010

Page 17: 23769_040 (KVS-10-1497-CR)

Numero: Rev.: Data: Descrizione: Pagina: Number: Rev.: Date: Description: Page:

KVS-10/1497-CR 00 05/07/2010 Issued for Comments and/or Approval 16

APPENDIX D: INSTRUCTIONS IN ITALIAN LANGUAGE

TIPO DI PROVA: Criogenica + Sniffer.

PROCEDURA Vanessa N°:  KVS-10/1497-CR

Paragrafo di 

riferimento:

PRINCIPALI PROCED. DI SICUREZZA:  PRO.SPP.02; IOS.09. 4

APPLICABILITA’:  14” & 18” / cl.300 in CF8M 2; APP.C

SEQUENZA OPERATIVA: 

‐ a)  montaggio caps (con termocoppie tipo “T”) su valvola asciutta e controllo delle coppie di serraggio di tutti i bulloni (in particolare quelli del packing, come da VMS 027);

‐ b)  posizionamento della idonea vasca nella buca del bunker;

‐ c)  collegamento dell’assieme valvola + caps al circuito di prova;

5.3

‐ d)  prova sniffer preliminare della valvola e del circuito di prova a TAMB con elio a pFE = 6.0 barg, (se si trovano perdite: scaricare, controllare e ripristinare le coppie di serraggio laddove necessario, riparare eventuali giunti difettosi, etc…; poi ripetere);

‐ e)  rilascio della pressione ed inserimento della valvola nella vasca posizionata dentro al bunker;

‐ f)  prova preliminare a TAMB di tenuta della sede contro pMAX = 41.2 barg sul lato stelo;

‐ g)  disinsediamento contro pMAX = 41.2 barg sul lato stelo, completa apertura e chiusura (prova funzionale in pressione) per 5 volte a TAMB;

5.4

‐ h)  rilascio della pressione con valvola semi-aperta; raffreddamento forzato a -196°C con azoto liquido e flussaggio di elio dentro la valvola (registrando le temperature almeno ogni 10 min);

5.5

‐ i)  prova di tenuta della sede a -196°C contro p1 = 10.3 barg sul lato stelo;

‐ l)  prova di tenuta della sede a -196°C contro p2 = 20.6 barg sul lato stelo;

‐ m)  prova di tenuta della sede a -196°C contro p3 = 30.9 barg sul lato stelo;

‐ n)  prova di tenuta della sede a -196°C contro p4 = pMAX = 41.2 barg sul lato stelo;

‐ o)  disinsediamento contro pMAX = 41.2 barg sul lato stelo, completa apertura e chiusura (prova funzionale in pressione) per 5 volte a -196°C;

‐ p)  ripetizione dei cinque steps di prova della sede da i) ad o) con pressione dell’elio sul lato disco;

‐ q)  rilascio della pressione con valvola semi-aperta;

5.6

‐ r)  prova ad elio a pFE = 6.0 barg a -196°C e rilevazione fughe con sniffer attorno a packing, flangia di fondo e della colonnina per almeno 15 min, dopo aver estratto molto lentamente e con massima attenzione la valvola dalla vasca;

‐ s)  rilascio della pressione con valvola semi-aperta;

5.7

‐ t)  rimozione dal circuito e posizionamento della valvola in area delimitata (opportunamente segnata con apposita cartellonistica) per ritorno spontaneo a TAMB;

‐ u)  smontaggio caps ed ispezione visiva finale di valvola + attuatore pneumatico. 5.8

ACCETTABILITA’:  - 150 ml/min per in di ND (vedi APPENDIX C); - emissioni: 1*10-4 mbar*l/s di elio max.; - operabilità senza intoppi tramite attuatore pneumatico.

5.9

CERTIFICAZIONE:      a cura dell’Ufficio Q.A. VANESSA. 5.10