Custody meter proving procedure
Custody meter proving procedure
(OP)
Hi! good day :)
I am a process Eng at the Oil field production unit. For the first time, I have to work on proving the custody metering (positive displacement type).In this regard, I have read all the relevant instructions thoroughly. But an item is still not well understood.
I have question: how to make sure about the validity of the new obtained meter factor (MF) from proving?
There is 8 option for this purpose. Among them, there are two tests (Test 1 & Test 2) to be performed. In these tests, for the purpose of validating of the results, there are two sets of deviation range are determined. How these deviation range are calculated?
I highlighted in the attached text (selected pages of the meter operation manual), which is related to the proving instructions.
Also, do you also have a clear definition for these two words: "Confidence level" and "uncertainty" ? What exactly are their applications?
Thanks for your help and guidance.
I am a process Eng at the Oil field production unit. For the first time, I have to work on proving the custody metering (positive displacement type).In this regard, I have read all the relevant instructions thoroughly. But an item is still not well understood.
I have question: how to make sure about the validity of the new obtained meter factor (MF) from proving?
There is 8 option for this purpose. Among them, there are two tests (Test 1 & Test 2) to be performed. In these tests, for the purpose of validating of the results, there are two sets of deviation range are determined. How these deviation range are calculated?
I highlighted in the attached text (selected pages of the meter operation manual), which is related to the proving instructions.
Also, do you also have a clear definition for these two words: "Confidence level" and "uncertainty" ? What exactly are their applications?
Thanks for your help and guidance.





RE: Custody meter proving procedure
It's not clear what your set up is, but it sounds like you either have a portable meter prover or you have one built in on site.
either way, your prover then needs to be certified back to some sort of national measurement system on a regular basis (checking the checker)
So depending on what your prover is (PD meter, swept volume? cylinder type?) it will come with its own certification traceable back to some sort of national master meter.
The diagram on page 10 gives you an idea about what the deviation ranges are.
confidence level is the confidence that your reading is within some level of deviation based on a normal distribution curve.
Uncertainty is the calculated value of the meter reading. It is often referred to as "accuracy", but the correct terminology is uncertainty in the metering world. It provides a calculated means of adding in all the different levels of uncertainty at different values from e.g. the temperature and pressure measurements, the volumetric flow rate itself and deviation in composition, physical properties etc. Most custody transfer systems aim for an uncertainty of no more than 1 or 1.1 %.
Basically it's all best left to the metering consultant...
Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
RE: Custody meter proving procedure
I've already read the relevant API standard. But unfortunately I do not know what part of it should I focus on?
The prover setup have one built in on site and is a part of the custody metering skid.
You are right. our prover needs to be certified by third party company. And I have been selected as a representative to monitor the correctness of their work.
The schematic of the prover is attached.
Also, you are right. The diagram on page 10 shows the deviation ranges are. But their value are not determined! Because my company itself has set up this prover once, but unfortunately I can not judge the obtained results of it!If necessary, send these results to you ...
Do you know any references to to introduce me about the proving procedures and relevant items?
Thank you very much for your great help.
RE: Custody meter proving procedure
Basically it should be a check of all the components to make sure they are not passing 9 like that 4 way valve or the main line block and bleed, the detector switches are working, the sphere is still inflated properly and in good condition, things like that. Also calibrate any pressure, temperature or density transmitters.
Then may they will fit a master meter?
I don't have specialist knowledge in this area, but if you carefully read the procedures supplied and follow up the references you may learn a lot.
Getting custody transfer metering right is not easy and is quite specialised. Given the amount of money 0.5% can make, it's often well worth getting it right.
Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
RE: Custody meter proving procedure
Rather than calibrate the instruments on the prover, the third party company will determine the base volume of your prover by comparison with their own compact prover. I'm not entirely sure how this is performed in the States but in the UK we accept the new base volume on a percentage shift (from the last volume) of 0.02%, if performed on water, or 0.04% if performed on product.
As part of the calibration they will usually perform an inspection of the sphere and reseal the detector switches once complete and record the seal numbers on the certificate, this allows you to demonstrate the switches haven't been changed/disturbed and the base volume is still valid.
RE: Custody meter proving procedure
The swept volume of a prover need to be at least 1000 pulses from the meter being proved so that you can get a volume of 0.1% uncertainty.
Now the original question was "how to make sure about the validity of the new obtained meter "factor (MF) from proving?
The answer lies in the calibration certification and testing of the master meter or portable prover, traceable back to some sort of national standard for measurement.
Provers by their nature are made from very carefully controlled or even machined ID to create an internal diameter of a very defined amount which then creates a volume from point A to point B to an accuracy of 0.1% or better. Again the measurement of this diameter needs to be as accurate and to a reasonable number of decimal places, certified by the vendor.
Ultimately no commercial measurement device is 100.000% accurate with no uncertainty to give it meter speak. A total uncertainty (meter, readings, calculation, composition etc) of 1.1% of energy or whatever thing you are measuring is about as good as you get.
Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
RE: Custody meter proving procedure