V-17—Temperature correction of provers

Category: Volume
Issue date:
Effective date:
Revision number: 4
Supersedes: V-17 (rev. 3)

Table of contents

1.0 Scope

This bulletin applies to narrow neck local standards (provers) as well as volumetric measures that are used to calibrate or certify volumetric measuring devices or other provers.

Exemption phase out

Previous versions of this bulletin exempted provers with a nominal capacity of less than 50 L. That is, prover shell volume changes due to temperature were simply ignored for these smaller provers as these changes were usually relatively small.

Measurement Canada (MC) decided that these calculations must be made for smaller provers as well. While it is true that the changes in the volume of open-neck provers are small when the temperature is moderate, when proving at very high or very low temperatures, the shell correction can account for a significant portion of the applicable limit of error for the meter being tested. Since April 1, 2020, prover shell corrections must be made for all open neck provers regardless of the nominal capacity. These calculations must therefore be made for 20 L provers or test measures.

2.0 Purpose

The purpose of this bulletin is to define the policy relating to the correction of a prover or measure's indicated volume to account for volume changes due to expansion or contraction of the material of which it is constructed when the actual temperature is different than the reference temperature.

3.0 References

  • Measurement Canada's Guidelines for the Calibration and Certification of Volumetric Standards
  • Field Inspection Manual—Volumetric Measuring Devices, Part 4, Section 23: Specialized Test Procedure–Prover Shell Temperature Corrections (STP-23)

4.0 Background

Provers are calibrated to a specific volume at a reference temperature, which is usually 15 °C for narrow neck provers used for petroleum and chemical products. If a prover is used at any temperature other than the reference temperature, a correction is required to address the expansion or contraction of the material used to construct the prover.

It is important to understand that a prover's volume changes with temperature and this change is different for various construction materials. For this reason, the prover's true volume must be calculated based on the operating temperature. It also means that when comparing provers constructed of different materials, a correction must be applied to both provers, even if the temperature of each is the same. For example, stainless steel has a cubical coefficient of thermal expansion (CCTE) of 0.0000518 per °C, which is about one and a half times that of mild steel at 0.0000335 per °C. Therefore, the volume of a stainless steel prover will change more significantly than that of a mild steel prover as the operating temperature moves away from the reference temperature. This correction factor is referred to as the correction for the temperature of the steel of the prover (Ctsp). In this expression, steel is a general term used to refer to the material used to construct the prover. Once determined, this correction factor is multiplied by the indicated prover volume (IVp) in order to establish the corrected volume of the prover (CVp).

Note: We do not apply a correction for the meter material; only for the prover material.

Measurement Canada considers it essential to correct the volume of provers according to shell temperature when testing meters.

While the correction may be made using a formula and the appropriate coefficient for the prover, in some cases it may be simpler to refer to the specific correction table in Appendix A of STP-23.

5.0 Policy

Measurement Canada inspectors and service providers authorized by Measurement Canada to inspect devices must correct the volume of all volumetric provers and measures to the appropriate reference temperature (e.g. 15 °C) when conducting meter inspections or examinations using in-service or acceptance limits of error.

Note: This is a change from previous versions of the policy that exempted provers of less than 50 L.

Corrections must always be based upon the actual CCTE applicable to the prover shell material. If this information is not available on the prover data plate or accompanying documentation, please contact MC for guidance in determining an appropriate coefficient.

Corrections can be made to the nearest 0.1 °C. As the tables presented in the Appendix A of STP-23 show corrections to the nearest 1.0 °C, linear interpolation may be used to determine the correction when the temperature is not a whole number. This additional step is not required if using the formula.

6.0 Cubical coefficient of thermal expansion for some common materials

To calculate the change in prover volume due to the expansion or contraction of the prover shell, you must know the CCTE per °C for the material used to construct the prover. This data is available from the manufacturer and is usually stamped on the prover's data plate or indicated on the prover's certificate of designation. If the information is not readily available, please contact Measurement Canada.

Note: In some cases, the CCTE may be identified with Greek letters, typically alpha (α) or gamma (γ). Table 1 contains some commonly accepted CCTE values.

Table 1: Typical cubical coefficient of thermal expansion for some common materials
Material Cubical coefficient of thermal expansion / °C
Mild steel 0.000 033 5 or 3.35 x 10-5
Stainless steel type 304 (18/8) 0.000 051 8 or 5.18 x 10-5
Stainless steel type 4.77 0,000 0477 or 4.77 x 10-5
Stainless steel type 316 0.000 045 4 or 4.54 x 10-5
Stainless steel type 17-4 0.000 032 4 or 3.24 x 10-5
Borosilicate glassware (Pyrex®, Bomex®, etc.) 0.000 010 0 or 1.00 x 10-5
Notes:
  1. The base or reference temperature that was used to calibrate the prover (usually 15 °C) and the prover shell temperature at the time of using (reading) the prover must be known before performing the calculations.
  2. Provers made of borosilicate glass, such as Pyrex® and Bomex®, is usually calibrated at a reference temperature of 20 °C.
  3. Sanitary provers for use with milk are usually calibrated at a reference temperature of 4.4 °C and are exclusively built of food-grade stainless steels.

The prover shell temperature is usually determined using a thermometer mounted in a thermowell in the prover body. Although the temperature of the liquid in the vessel is being sensed, once stabilized, it's deemed to be representative of the prover shell temperature. Test measures and smaller volume provers that do not incorporate a test thermowell may use the average temperature determined during a test run, at the device under test's thermowell, to establish the prover shell temperature. Glassware temperatures are to be determined by way of a submersible temperature probe inserted into the fluid contained within the measure at the time of reading the volume.

7.0 Temperature determination when using 20 L test measures

Shell corrections for 20 L test measures may be made using the average temperature determined by a certified thermometer installed in the device under test's thermowell during a test run. This average temperature must be obtained during every test run in order to properly apply shell corrections.

For devices that do not incorporate automatic temperature compensation and where test thermowells are not present, there are several methods of determining the temperature. Measurement Canada recommends the following ones:

  • A temperature reading taken within a permanently affixed thermowell; this permanent solution will require that the modified standard be recertified ahead of use.
  • A temperature reading obtained from a hand-held infrared thermometry device (suggested resolution: 0.1°C). (MC's Legal Metrology and Laboratory Services does not calibrate this type of device.)

Other means of temperature determination will be considered on a case-by-case basis by MC's volumetric technical authority. The primary considerations at the time of this determination will be testing safety and result accuracy.

Authorized service providers may request a review of their proposed alternate method of temperature determination via their MC point of contact.

8.0 Revisions

The purpose of this revision 4 was to:

  • add a section to clarify the acceptable means of temperature determination;
  • refer requests for decisions pertaining to alternate means of temperature determination to the volumetric technical authority;
  • remove and integrate the procedural related inclusions (procedure, examples and correction factor tables) to the actual test procedure (STP-23) as this bulletin is a policy , not a procedure;
  • add guidance on when and how to determine an authorized equivalent shell temperature for 20 L test measures that do not incorporate a test thermowell;
  • add the CCTE for stainless steel to Table 1;
  • make various editorial and formatting changes.

The purpose of revision 3 was to:

  • make the cubical coefficient of thermal expansion calculation applicable to all prover shells, including those of less than 50 L;
  • add an example of a calculation for a 20 L test measure.

The purpose of revision 2 was to:

  • add common prover sizes;
  • expand tables to cover temperatures from -30 °C to 40 °C;
  • make separate tables for mild steel and stainless steel type 304 (18/8) provers;
  • add a column with volume correction factors;
  • make editorial and format changes;
  • add a section dealing with selecting and calculating appropriate corrections.

The purpose of revision 1 was to:

  • revise the bulletin format;
  • make minor editorial changes;
  • remove the requirement for inspectors to annotate the inspection certificate to show that the prover was corrected for temperature;
  • define the physical size and use condition when volumetric standards require volume correction for the temperature of the shell and include a column for 1500 L stainless steel prover corrections.
 Report a problem
Date modified: