Pennsylvania Code & Bulletin
COMMONWEALTH OF PENNSYLVANIA

• No statutes or acts will be found at this website.

The Pennsylvania Code website reflects the Pennsylvania Code changes effective through 53 Pa.B. 8238 (December 30, 2023).

52 Pa. Code § 59.21. Meter tests.

§ 59.21. Meter tests.

 (a)  Test schedule for other than Class A, B and C meters. Each public utility shall make and record tests of orifice, rotary displacement and turbine type service meters as follows:

   (1)  Orifice meters shall have their differential and static recording gauges tested at least once every 2 months, the diameter and condition of the orifice plate checked at least once every year, and the specific gravity of the gas determined at least once every 6 months; however, where previous or subsequent test records show that the specific gravity has not varied by an amount which would make an error in the measurement greater than is consistent with accepted engineering practice, the specific gravity of the gas may be determined at longer intervals not to exceed 1 year.

   (2)  Rotary displacement meters shall be tested and calibrated at the factory in accordance with recognized and accepted practices and shall be correct to within 1.0% when passing gas at their rated capacities. A record of the test shall be made available to and retained by the utility for the life of the meter. At least once every 10 years’ a differential-rate test shall be made and the results checked against the original test recorded at the time of installation. At least every 5 years’ the meter shall be inspected to observe the condition of the meter bearings noise, vibration, and the like, and the level and condition of the oil in the reservoirs, except that those meters installed before January 1, 1990, shall be inspected every 2 years. An observed problem shall be promptly corrected. A record of the results of these 5 year tests or 2 year tests for the pre 1990 installed meters shall be maintained by the utility for 5 years. In lieu of a differential-rate test, a test method approved by the Commission may be used.

   (3)  Turbine meters shall be tested and calibrated at the factory in accordance with recognized and accepted practices and shall be correct to within 1.0% when passing gas at their rated capacities. A record of the test shall be made available to and retained by the utility for the life of the meter. At least once every 2 years, a spin test shall be made and the results checked against the minimum spin test time specified by the manufacturer for the size meter being tested. If the spin time is not up to standard value, corrective measures shall be taken. In lieu of a spin test, a test method approved by the Commission may be used.

 (b)  Standard test schedule for Class A, Class B and Class C Meters. Unless otherwise provided by this section, each public utility shall make and record tests of Class A, Class B and Class C meters on the following schedule:

Test Period
Class(Years)
A8
B5
C2

 (c)  Extended test schedule for Class A, Class B or Class C meters. A public utility may depart from the requirements of subsection (b) for Class A, Class B or Class C meters in a testing year, and instead make and record tests using one of the test periods prescribed in paragraph (5), if the following requirements are met:

   (1)  At the end of the first immediately preceding year, not less than 98% of the meters of that class in service had been removed within whichever is the greater of 2 years plus the test period prescribed therefor in subsection (b), or the test year permitted for that class of meter for that year by paragraph (5); and, as to a meter of that class not so removed, the premises where it was located were visited and a written notice requesting an appointment for meter change was either left at the premises or posted to the mailing address of the customer as it appears in the public utility’s files. Meters removed under this paragraph shall be tested and included in the calculations under paragraph (2) unless a meter was permanently retired from service or damaged by factors other than normal age or wear such as tampering or damage beyond the control of the public utility.

   (2)  The slow meter ratios and fast meter ratios of the meter class for the second immediately preceding year and the third immediately preceding year fall below the maximum percentages prescribed in paragraph (5). Any conflict between the test periods prescribed in paragraph (5) shall be resolved by using the shortest applicable test period.

   (3)  On or before March 1 of each testing year, the public utility submits to the Commission a report showing both in absolute numbers and in percentages the facts prescribed in paragraphs (1) and (2).

   (4)  For each year in which a public utility uses the extended test schedule in this subsection, the public utility may not remove or test any meters of the same class using the statistical sampling program in subsection (d) or the variable interval program in subsection (e).

   (5)  Subject to the qualifications prescribed in paragraphs (1)—(4), a public utility may make and record tests of Class A, Class B or Class C meters on the following schedule:

Test Results from
Test Results from
Testing
Second Immediately
Third Immediately
Year’s
Preceding Year
Preceding
Permitted
Year
Test Period
(Years)
CLASS A METERS
Slow Meter
Fast Meter
Slow Meter
Fast Meter
Ratio
Ratio
Ratio
Ratio
Less
Less
Less
Less
Than (%)Than (%)Than (%)Than (%)
1010121214
88101016
668818
446620
CLASS B METERS
Slow
Fast
Slow
Fast
Meter
Meter
Meter
Meter
Ratio
Ratio
Ratio
Ratio
Less
Less
Less
Less
Than (%)Than (%)Than (%)Than (%)
101012126
8810107
66888
446610
CLASS C METERS
Slow
Fast
Slow
Fast
Meter
Meter
Meter
Meter
Ratio
Ratio
Ratio
Ratio
Less
Less
Less
Less
Than (%)Than (%)Than (%)Than (%)
8810103
66884
44665

 (d)  Statistical sampling for Class A, Class B or Class C meters. A public utility may depart from the requirements of subsection (b) for Class A, Class B or Class C meters, and instead make and record tests of Class A, Class B or Class C meters under a statistical sampling program, if the following requirements are met:

   (1)  Meters shall be divided into groups in accordance with ANSI Spec, B109.1 Part IV Sec. 4.3.2.1 or its successor. A detailed description of the composition of each group of meters, such as year set, manufacturer, case type and diaphragm material, shall be provided in the annual report to the Commission.

   (2)  Sufficient meters shall be tested annually to insure a 90% confidence level that the meter groups are performing within accuracy limits.

   (3)  For a group to remain in service, at least 80% of the meters in the sample test shall meet the accuracy limits of 98% average accuracy (2% slow) and 102% average accuracy (2% fast). If a group of meters does not meet the performance standards, corrective action shall be taken. The corrective action may consist of removing the entire group from service within 4 years or, if the group consists of one or more subgroups, implementing a selective meter removal program to improve the accuracy of the group to within acceptable limits. The selective removal program may be as follows:

     (i)   If test results indicate one or more subgroups do not meet the performance standards, the subgroup shall be identified and removed within 4 years.

     (ii)   Once identified as a group or subgroup not meeting the performance standards and during the removal process, that group or subgroup shall be removed from the sampling plan.

   (4)  The statistical sampling schedule shall be in accordance with the provisions of ANSI Spec. B109.1 Part IV Sec. 4.3.2.1 or its successor.

   (5)  A utility electing the statistical sampling program shall remain on that program for at least 4 years.

   (6)  For each year in which a public utility uses the statistical sampling approach in this subsection, the public utility may not remove or test any meters of the same class using the extended test schedules in subsection (c) or the variable interval program in subsection (e).

 (e)  Variable interval testing for Class A, Class B or Class C meters. A public utility may depart from the requirements of subsection (b) for Class A, Class B or Class C meters, and instead make and record tests of the Class A, Class B or Class C meters under a variable interval program, if the following requirements are met:

   (1)  Meters shall be divided into groups in accordance with ANSI B109.1 Part IV Sec. 4.3.2.2 or its successor. A detailed description of the composition of each group of meters, such as year set, manufacturer, case type and diaphragm material, shall be provided in the annual report to the Commission.

   (2)  The number of meters to be removed in any year will be determined from the test results of the second immediately preceding year’s incoming meters. Meters removed under this paragraph shall be tested and included in the calculations under paragraph (3) unless a meter was damaged by factors other than normal age or wear such as tampering or damage beyond the control of the utility.

   (3)  Except as provided in paragraphs (4) and (5), the ratio (r) of the number of meters in a test group to be removed to those in service in that test group shall be determined by the formula (r = .02 + .3d) where (d) is the ratio of meters which have an average accuracy of less than 98% or more than 102% as reported to the nearest 1/2%, to the total number of meters tested in the group during the second immediately preceding year.

   (4)  Meters removed in a test group in excess of the ratio (r) as described in paragraph (3) shall be credited towards the ratio (r) for a better performing test group.

   (5)  A utility may petition the Commission for an Accelerated Retirement Program (ARP) for a specific meter type that the utility may desire to purge from its system. Meters removed in an ARP in excess of the ratio (r) as described in paragraph (3) may be credited towards the ratio (r) for any other test group regardless of performance.

   (6)  A utility electing the variable interval plan shall remain on that plan for at least 4 years.

   (7)  For each year in which a public utility uses the variable interval approach in this subsection, the public utility may not remove or test any meters of the same class using the extended test schedules in subsection (c) or the statistical sampling program in subsection (d).

 (f) Meter test on request of customer. Meter tests, if requested by a customer, shall conform with all of the following:

   (1)  If a customer requests, in writing, a test of the accuracy of the meter through which gas service is supplied and the meter is not due for periodic test, the public utility shall notify the customer of the conditions under which the test will be made by the utility or by a referee. If the customer then requests the utility to proceed with the test and remits an amount equal to the scheduled cost of a referee test, the utility shall make the test promptly. If, when tested, the meter is found to be more than 2.0% fast or slow, the testing fee shall be promptly refunded to the customer.

   (2)  A customer or his representative may be present when the public utility conducts the test on the meter.

   (3)  A report giving the name of the customer requesting the test, the date of the request, the location of the premises where the meter had been installed, the type, make, size, and serial number of the meter, the date of removal, the date of the test, the result of the test and the amount of refund if the meter was found more than 2.0% fast, shall be supplied to the customer within 10 days after the completion of the test.

   (4)  The amount of the fee which may be charged by the public utility for testing meters upon the written request of a customer shall be determined by the designated rating of the manufacturer as follows. Displacement meters shall be subject to the following fee schedule:

     (i)   Meters having a rated capacity of 500 cubic feet per hour or less—$10.

     (ii)   Meters having a rated capacity of over 500 cubic feet per hour, and not more than 1,500 cubic feet per hour—$20.

     (iii)   Meters having a rated capacity of over 1,500 cubic feet per hour, orifice meters, and any meters not a displacement type—$30.

   (5)  The provisions of this section may not interfere with the practice of a public utility with reference to its regular tests of meters; except that, in the event of an application by a customer to the Commission for a referee test, the utility may not knowingly remove, interfere with or adjust the meter to be tested without the written consent of the customer and approval of the Commission.

 (g)  Installation test. Each gas meter shall be in good order and shall be correct at all test rates of flow to within 2.0% fast or slow before being installed. In the case of new meters or meters reconditioned by a manufacturer, the test results of the manufacturer can be accepted as the installation test if the utility has verified the manufacturer’s reported test results by testing a minimum of 10% or ten meters—whichever is greater—of each shipment of meters. However, in case of an emergency, a meter not meeting the requirements of this section may be installed temporarily. Each meter tested by the utility under this subsection shall also be tested for pressure as follows: every meter shall be subjected to a pressure test before being installed; the minimum test pressure shall be 1.5 times the maximum metering pressure that the meter will be exposed to during its time in service.

 (h)  Determining accuracy of small meters. For the purpose of determining compliance with subsection (e), the registration of a displacement meter shall be determined by one test at a rate of flow of approximately 1/5 of its rated capacity (check flow) and by a second test at approximately the full rated capacity of the meter. The capacity of the meter for test purposes shall be the capacity at 1/2 inch differential pressure. Prior to installation of a meter, the tests at the two rates of flow shall agree within 2.0%. In determining compliance and for the purpose of computing refunds, the check flow test may be considered as the accuracy of the meter. A utility may, at its option applicable to all of its small meters, consider the accuracy of a meter to be the algebraic mean of two errors found, one at check flow and one at capacity flow. If unusual conditions indicate that an unusual test rate of flow should be used, the test record shall show the rate used.

 (i)  Meter prover. Tests to determine the accuracy of a gas service meter used to measure gas at standard distribution pressure shall be made with a bell type meter prover unless, because of unusual capacity or construction of a meter, the method of test is considered impracticable, under which condition the test shall be made by some other method approved by the Commission.

 (j)  Testing large displacement meters. The testing of large displacement meters shall conform with the following:

   (1)  Tests on displacement meters shall be made with approved testing apparatus. The meters should be tested in their permanent locations on the premises of the customer if practicable. If critical flow provers or low pressure flow provers are used for making such tests, the accuracy should be determined at three or more rates of flow, ranging from 20% of the rated capacity at 1/2 inch differential pressure up to flow at the maximum operating rate.

   (2)  If testing with the critical flow prover, the meter shall be operated at a static pressure which approximates the average operating static pressure. If similar tests are made with a low pressure flow prover, average operating pressure should be maintained on the meter under test and the prover operated at a range of low pressures required for a prover. In installations where meters operating on pressures higher than standard service pressure are limited in their maximum operating dial rate so as to keep the differential pressure equal to or less than 2 inches water column, low pressure flow prover tests may be run with low pressure on the meter. In the flow tests by either method, the maximum dial rate shall insure that tests being run at meter differential pressures are equivalent to normal operating meter differential pressures in all cases. If it is not practicable to test rotary displacement meters with a flow prover, they shall be inspected and tested by approved methods to determine whether they conform reasonably to the original factory test data.

   (3)  An instrument or auxiliary device used in conjunction with a gas meter to correct the metered volume for pressure or temperature shall be adjusted to an accuracy level to assure that the combined percentage of error of the instrument or auxiliary device, or both, and the associated meter does not exceed plus or minus 2.0% error. This shall be verified by test prior to installation and at the time of any subsequent meter tests. In tests conducted after installation the meter and its associated auxiliary device may not be tested more than 30 days apart. Each instrument and auxiliary device shall be verified for accuracy or calibrated at least annually to verify the performance. A record of the most recent verifications shall be kept for each instrument and device.

 (k)  Testing orifice meters. Tests on orifice meters shall be made with approved testing apparatus and in their permanent locations on the premises of the customer. The accuracy of the differential pressure registration shall be determined on a rising and falling pressure throughout the entire operating pressure range of the gauge. The accuracy of the static pressure registration shall be determined at the operating pressure.

 (l)  Determining accuracy of orifice and large displacement meters. In determining the accuracy of orifice and large displacement meters, the average of the errors determined at the various rates of flow at which the meter was tested shall be taken and shall be considered as its accuracy in determining compliance with subsection (e) and for the purpose of computing refunds; however, if the rates of flow at which the meter has been registering in service may be definitely established, the weighted average error shall be determined and used.

 (m)  Test record data. Whenever a meter is tested, the original test record shall include information necessary for identifying the meter, the reason for making the test, the date and location of the test, the name of the person making the test, the reading of the meter upon removal from service, the result of the test, and data taken at the time of the test. The record shall be in complete form so as to permit the convenient checking of the methods employed and the calculations made.

 (n)  Meter records. A record shall be maintained for each meter owned or used by a public utility. It shall indicate whether acquired new or otherwise, the date of purchase, identification, and the results of the most recent test and shall contain any additional test records required to be kept by other provisions of this title relating to gas meters. This record may not be destroyed without Commission authorization.

 (o)  Capping meters. Incoming meters shall be capped when removed from service and awaiting test. Meters which have been tested or are reading for installation shall be capped and remain capped until installed.

Authority

   The provisions of this §  59.21 issued and amended under Public Utility Code,66 Pa.C.S. § §  501, 1301, 1304, 1501, 1502, 1504, 1507 and 1508.

Source

   The provisions of this §  59.21 amended through May 30, 1986, effective May 31, 1986, 16 Pa.B. 1901; amended July 3, 1997, effective August 4, 1997, 27 Pa.B. 3215; amended July 2, 1998, effective August 3, 1998, 28 Pa.B. 3050. Immediately preceding text appears at serial pages (232255) to (232263).

Cross References

   This section cited in 52 Pa. Code §  59.15 (relating to measurement of gas at higher than standard service pressure); and 52 Pa. Code §  59.22 (relating to adjustment of bills for meter error).



No part of the information on this site may be reproduced for profit or sold for profit.


This material has been drawn directly from the official Pennsylvania Code full text database. Due to the limitations of HTML or differences in display capabilities of different browsers, this version may differ slightly from the official printed version.