Number Of Proving Runs And Repeatability - ABB CoriolisMaster FCB100 Product Manual

FCB100, FCB400 COMMISSIONING AND PROVING | PG/CORIOLIS/PROVING-EN REV. A

6 Number of proving runs and repeatability

Change from one to two columns
'Validity of the average meter factor to a random uncertainty of
0.00027', is the API requirement for proving.
The higher the number of runs in a proving, the lower the
repeatability requirements.
The API MPMS 4.8 requirement is shown below:
Runs at proving repeatability to meet ±0.027 % uncertainty of meter
factor
Number of runs
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Change from one to two columns
Repeatability Uncertainty
0.02 % ±0.027 %
0.03 % ±0.027 %
0.05 % ±0.027 %
0.06 % ±0.027 %
0.08 % ±0.027 %
0.09 % ±0.027 %
0.10 % ±0.027 %
0.12 % ±0.027 %
0.13 % ±0.027 %
0.14 % ±0.027 %
0.15 % ±0.027 %
0.16 % ±0.027 %
0.17 % ±0.027 %
0.18 % ±0.027 %
0.19 % ±0.027 %
0.20 % ±0.027 %
0.21 % ±0.027 %
0.22 % ±0.027 %
Sets of runs can also be averaged for the repeatability
calculation.
One technique is the averaging of three sets of five proving runs.
An example of this technique is as follows:
Proving run
Sample 1 40009
Sample 2 40007
Sample 3 40005
Sample 4 40002
Sample 5 39998
Repeatability 0.028 %
Average 40004.2
Repeatability all 0.01 %
Averaging allows for more flow fluctuations during proving, the
more data that is averaged, typically the better the repeatability.
No 1 No 2
39996
40002
40007
40004
39998
0.028 %
40001.4
0.01 %
11
No 3
39995
40001
40008
39998
40004
0.033 %
40001.2
0.01 %