Nordic Innovation Centre Telephone +47 47 61 44 00 ISSN 0283-7234Stensberggata 25 Fax +47 22 56 55 65 NO-0170 OSLO [email protected]Norway www.nordicinnovation.net INTERCOMP A RISON AND METHOD FOR UNCERTAINTY CALCULATION OF WATER A ND HEAT FL OW Kaj Bryder TR610Ap p ro ve d2004- 11
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.
Intercomparison and method for uncertainty calculationof water and heat flow
Abstract:
A relatively high utilisation of district heating and limited resources of drinking water
makes metering of water and heat of great importance in both the Nordic and the
Baltic countries. This has further resulted in a heavy focus on testing and calibration
of the meters as well as on the uncertainty of the meter installation.
Combined with European wishes for establishing an accredited and uniform basis fortype testing and verification of e.g. water and heat meters this underlines the
importance of exchange of know how, intercomparison and setting up of common
methods for uncertainty calculations in the field of water and heat flow measurement.
The objective of this Nordtest-project “Intercomparison and method for uncertainty
calculation of water and heat flow” is to improve metering of water and district
heating in the Nordic and Baltic area by carrying out an intercomparison, and
preparation of a method for uncertainty calculations for calibration and for meter
installations by exchange of know-how at two work shops. This report is part of the
exchange of know how and is based on work shops held on 20 March 2002 at DTI in
Aarhus and 17–18 February 2004 at LNMC in Riga as well as it comprises the background and the results of the intercomparisons and the background of the method
of uncertainty calculations. The method of uncertainty calculations is carried out as a
separate document set up according to normal Nordtest specifications.
Technical Group:
Building and Construction (HVAC/VVS)
ISSN:
0283-7234
Language:
English
Pages:
51 p. + 3 annexes
Key Words:
Intercomparison, uncertainty, water and heat flow, test method
2.1.1 The importance of intercomparisons according to EA...................................... 17
2.1.2 Basic types of intercomparisons ........................................................................ 17
2.1.3 Comparing of results of intercomparisons......................................................... 18
2.2 DTI participations and experiences.............................................................................. 18
2.2.1 DTI participation in intercomparisons............................................................... 18
2.2.2 The first international comparison..................................................................... 20
2.2.3 Danish comparisons and follow-up ................................................................... 24
2.3 LEI participations and experiences .............................................................................. 272.4 LNMC participations and experiences......................................................................... 28
2.5 SP participations and experiences................................................................................ 29
2.5.1 SP participation in comparisons ........................................................................ 29
2.5.2 Examples from various intercomparisons ......................................................... 30
3. The Nordic-Baltic intercomparison ................................................................................. 45
3.1 Setting up the intercomparison .................................................................................... 45
3.1.1 Reference meter packages and the calibrations ................................................. 45
3.1.2 Plan and schedule .............................................................................................. 46
The objective of this Nordtest-project “Intercomparison and method for uncertaintycalculation of water and heat flow” is to improve metering of water and district heating in the
Nordic and Baltic area by carrying out an intercomparison, and preparation of a method for
uncertainty calculations for calibration and for meter installations by exchange of knowhow at
two work shops. This report is part of the exchange of know how and is based on work shops
held on 20 March 2002 at DTI in Aarhus and 17–18 February 2004 at LNMC in Riga as well
as it comprises the background and the results of the intercomparisons and the background of
the method of uncertianty calculations. The method of uncertainty calculations is carried out
as a separate document set up according to normal Nordtest specifications.
Participants of the project
The project has been intitiated and carried out as a co-operation between Latvian National
Metrology Centre, Lithuanian Energy Institute (LEI), Swedish National Testing and Research
Institute and Danish Technological Institute, which has been project manager.
Detailed information about the participants:
- Latvian National Metrology Centre (LNMC),
K. Valdemara iela 157, Riga, LV 1013- Janis Jansons (water meters)
This Nordtest-project “Intercomparison and method for uncertainty calculation of water and
heat flow” comprises:
• Legal requirements for water and heat meter calibration and testing in the Nordic and the
Baltic countries (Denmark, Sweden, Latvia and Lithuania)
• Presentation of various calibration and testing facilities
• Presentation and discussion of existing intercomparisons in which the laboratories have
participated
• Basis of and results from the Nordic_Baltic intercomparison
• Basis of the method of uncertainty calculations.
The project has been carried out in co-operation between: Latvian National Metrology Centre,Lithuanian Energy Institute (LEI), Swedish National Testing and Research Institute and
Danish Technological Institute, which has been manager of the project and also coordinator
for this report.
The results of the project are:
- An intercomparison which can be used for future improvements of the calibration and
testing laboratories
- A method for uncertainty calculations for calibration of water and heat meters and based
on partly existing standards and manual about uncertainty calculations, partly on know
how and experiences from each of the laboratories
- A general exchange of know how and experiences between the laboratories about water
and heat meter calibration and testing
- The laboratories from the Baltic countries have gained better know how about water and
heat meter calibration and testing in the old EU countries.
conductivity for measurements with magnetic inductive meters. Therefore, EN 1434 stipulates
conductivity greater than 200 µS/cm, which is used as criteria.
When measuring heat, the flow measurement is supplemented with the measurement of the
temperature difference between a temperature sensor placed in the district heating flow pipeand a sensor placed in the return pipe. Calibrating heat meters normally takes place through
temperature simulation in two temperature-regulated baths.
Figure 1.2-3: Test meters mounted in the large test rig. The diverter can be
seen in the background at the top of the photo.
Figure1.2-4: Test meters mounted in the medium test rig.
1.4.1 LNMC facilities for testing of cold water meters up to 18 m3 /h
The LNMC flow facilities for cold water meters are illustrated in Figure 1.4-1. The test rig is
manufactured by Karl Adolf Zenner, Germany in 1998 and according to standard ISO
4064/1-83. The calibration of meters in the range DN 15 – DN 40 is based on a 300 kg
weighing tank or a reference meter.
Figure 1.4-1: Diagram of LNMC flow calibration facilities for cold water meters.
Components The test bench is fitted with the following components:
1. Stand with tubes and outlet to fix the water meters.2. Flow rate control system with flow rate meters (4 pieces).3. Pumps (3 pieces).4. Measuring tank for the flow provision.5. Scales (300 kg with resolution e = 2 gr) with container (280 l).6. Tanks for water keeping with emptying device (2 pieces).7. Pipes, shutting-off devices, armature, ball valves.
8. Termometer, manometers.
The specifications of the facilities from Table 1.4-1.
Table 2.2-1: DTI participation in national and international comparisons since 1991.
EU-BCR = European intercomparison co-financed by the BCR programme of EUEU-SMT = European intercomparison co-financed by the SMT programme of EUUK/DTI = Department of Trade & Industry, UKUN = International intercomparison co-financed by UNDanish Ring = Intercomparison carried out by “Intercomparison scheme for volumetric and heat meter
laboratories in Denmark”“References” refer to projects desribed in The Flow Laboratory for Water and Heat Measurement – Laboratory,activities and references, Danish Technological Institute, June 2001, /DTI-Flow/
In 1990 DTI became pilot laboratory of a BCR-project11: “Intercomparison of flow meter
calibration facilities for the measurement of flow at various temperatures”. The project
comprised flow measurements at 3 flow levels:
- “Small package” up to 1500 l/h
- “Medium package” up to 16.000 l/h
- “Large package” up to 180.000 l/h
All packages were further tested at water temperatures of 20, 50 and 70°C.
The project was co-financed by BCR with app. DKK 2m and after continual expansions the
project comprised a total of 12 laboratories (e.g. . PTB in Germany, NEL in Great Britain,
CETIAT in France and IMBC in Italy) in 11 different countries (Denmark, Sweden, Finland,Great Britain, Germany, the Netherlands, Belgium, France, Italy, Switzerland and Spain).
The calibration comprised 3 meter packages: “large”, “medium” and “small” with pipe
dimensions of 100 mm, 25 mm and 15 mm, respectively, which were circulated as illustrated
in Figure 2.2-1.
Figure 2.2-1: Various packages and rings from the BCR
intercomparison carried out from 1990 to 1994.
1 Bureau Commutaire Reference (BCR), which was the title of a special EU programme covering metrology.
From the very beginning the intercomparison scheme was organised with a steering group
with representatives from Danish utility plants and meter manufacturers, respectively, with
their own laboratories as well as from Danish Technological Institute. In addition to Danish
Technological Institute, the steering group consists of representatives from Danish utility
plants and from the Association of Manufacturers of Energy and Flow Meters in Denmark. Atthe beginning, DANAK participated as observer, but as the intercomparisons recently have
become part of DANAK’s requirements, DANAK now participates as conclusive member.
The first intercomparison took place from 1994 - 1995 and a total of 13 Danish meter
laboratories participated with a total of 17 calibration benches. The intercomparison was
based on the circulation of a meter package consisting of two meters of different make
mounted in series and covering a volume flow from 125 l/h to 2000 l/h and a heat flow of app.
90 kW.
Figure 2.2-6: Intercomparisons initiated by Denmark – organised
in three Danish, and later in one international ring.
After the procurement of the meter package a number of tests were carried out on the chosen
reference meters, i.a. with regard to repeatability (showing the degree of correlation between
repeated measurements when the individual measurements are carried out under similar
conditions) and their reproducibility (showing the degree of correlation between measurement
results when the individual measurements are carried out under varying conditions). DTI
participated in the intercomparison as equivalent laboratory and as coordinating laboratory
and due to the many participants the intercomparison was organised in three rings as shown in
Figure 2.2-6.
At the end of 1995 contacts were taken to SP (Sveriges Provnings- och Forskningsinstitut),
Sweden, and to PTB (Physikalisch Technische Bundesanstalt, Berlin), in Germany, in order to
ensure international continuity. Both laboratories accepted the invitation to participate and in
that way the scope of intercomparisons was broadened to comprise a total of 15 laboratories
The background and the result of the DTI-PTB-SP intercomparison is desribed in detail in the
report Intercomparison of Flow Metering between DTI, PTB and SP, Final report, Danish
Technological Institute, September 1998 /DTI-1998/
As a natural follow-up - and also at the request of DANAK - it was at a later point decided torepeat the intercomparison with the meter package up to a volume flow of 2000 l/h and a heat
flow rate of app. 90 kW (“medium” package). In addition, the package was supplemented by
a “large” meter package exclusively covering a volume flow of up to 16,000 l/h (16 m3/h).
Today the “Intercomparison scheme for volumetric and heat meter laboratories in Denmark”
comprises the following packages.
- “Small” package comprising a coriolis mass flow meter and a magnetic inductive meter
of up to 125 l/h
- “Medium” package comprising two magnetic inductive meters of different make
covering a flow range of 125 l/h (2 m3/h) and heat measurements
- “Large” package comprising two magnetic inductive meters of different make covering a
flow range of 2000 - 16000 l/h.
The participants have i.a. comprised the laboratories stated in Table 2.2-2.
Table 2.2-2: The participants in the intercomparisons carried out under the “Intercomparison
scheme for volumetric and heat meter laboratories in Denmark”.
Documentation of the intercomparison follows from:
• Intercalibration flow meter – Instructions for first round robin test, Flow rate 60 l/h -
2000 l/h, Nordtest Project No. 1610-02, Danish Technological Institute, 4 March 2003,
/NT-Instruct/
• EA-2/03 Interlaboratory Comparison, European co-operation of Accreditation (former
EAL), March 1996, /NT-Results/
3.1.2 Plan and schedule
According to the schedule originally laid out in it was decided to start up the intercomparisons
at the end of 2002, but it was also decided that there should be time for improving facilities
and staff both before and when carrying out the intercomparisons. Such circumstances e.g.
running-in of a test facility at LNMC in Latvia, and problems in fitting the tests into
scheduling of commercial tests and calibrations resulted, however, in several delays of theintercomparison so that it was not finishing until February 2004. The resulting timetable
follows from Table 3.1-2.
Table 3.1-2: Timetable of the intercomparison.
3.2 Intercomparison instruction
Based on former instructions and the discussions at the meetings an informative instruction
for the laboratories was carried out. The instruction follows from Annex 1.
3.3 Intercomparison results
A complete overview of the results of the intercomparison follows from Tables 3.3-1 to 3.3-5
and Figure 3.3-1 and 3.3-2. The tables also includes a calculation of the En-value that gives
an estimate of the test result compared with the accredited uncertainties of the laboratory (the
average from all laboratories are used as reference values). The complete test results follow
The reference standard consists of 2 magnetic inductive volume flow meters/volume meters, 2calculation units and 2 sets of flow and return temperature sensors.
The volume flow meters are mounted in series and thus constitute a complete unit that must
not be dismantled by any of the participants. The straight pipe length upstream of the first
volume flow meter is 200mm or approx. 12 X d i, as the pipe dimension is 2" ISO (DN
16mm). The pipe length is the same between the meters, and the straight length downstream
after the last meter is approximately 10 X di.
The inlets and outlets to/from the individual meters are conical, as the internal diameters of the
meters are less than those of the connected pipe diameters.
The units of calculation and the volume flow meters secondary device are assembled on an
inclined plate in a frame stand, which has 2 interior pipe carriers on which the volume flow
meters are mounted during transportation. All electrical connections are switched on,
including the temperature sensors of the type Pt 500.
The principal dimensions of the package appear from figures 1 and 2 as well as from section
2.2 and the manufacturers= specifications of the measurement equipment appear from the
sections 2.3 - 2.5.
In order to ensure that the volume flow meter can function satisfactorily, the conductivity of
the water in the measuring bench is required (manufacturer requirement) to be equal to or
Connect the meters to the power supply and ensure that they are switched on for at least an
hour before finally starting the calibrations (see APPENDIX B).
3.5 Calibration procedure
1 Inspect and install the flow meter package as indicated under item 3.2. Do not
dismantle the package!
2 Carry out the electrical connections as indicated under item 3.3.
3 Fill the measuring length with water of 20°C and vent as indicated under item 3.4.1.
4 Apply voltage to the meters and the joint connector as indicated under item 3.4.2
5 Flush the meter package at volume flow no. 3 (2000 l/h).
This is to ensure that the electrodes are without any disturbing coating.
Then reduce the volume flow to volume flow no. 2 (125 l/h) and perform three
repeated measurements at this volume flow (at stable flow conditions) and a water
temperature (Tv) of 20°C 3°C.
6 Calculate the error indications of the meters and report the results to the DTI flow
laboratory, John Frederiksen or Jens B. Vestergaard, who will decide whether the
package has changed during transport or not.
7 When signal has been received indicating that the meter package is considered free of
defects, start by exercising the meter package for half an hour at volume flow no.3:
2000 l/h (Tv = 20°C 3°C).
Then reduce the volume flow to volume flow no. 2. Then carry out 5 repeated
calibrations of the meter package at this volume flow.
Then increase the water flow to flow no. 3: 2000 l/h
When stable conditions have been ensured, calibrate the package 5 consecutive times
at this water flow.
Then increase the water flow to flow no. 1: 60 l/h and
when stable conditions have been ensured, carry out 5 repeated calibrations at this
volume flow.
End calibrations by repeating the calibrations at volume flow no. 2: 125 l/h, and repeat
5 times. This repetition of flow no. 2 is to be carried out to see whether changes may
have occurred in the meter package meanwhile as well as to show something about the
reproducibility of the laboratory.
If the DISPLAY is used to read the meter, use the standing start/stop method and do
not read the display and the residual value until 10 secs. after each stop before as wellas after the measurement to allow the calculation unit time to complete its internal
The Nordic Innovation Centre initiates and financesactivities that enhance innovation collaboration anddevelop and maintain a smoothly functioning market in
Return address:
Nordic Innovation Centre,
Stensberggata 25
NO-0170 Oslo, Norway
NORDTESTNORDTEST is a Nordic Innovation Centrebrand offering competence and experticein the field of harmonizing of norms andmethods,a large Nordic net-work of experts,more than 650 recommended Nordic testingmethods and 550 published technical reports.