Michigan Administrative Code
Department - Licensing and Regulatory Affairs
Public Service Commission
Technical Standards For Gas Service
Part 5 - METERS METERING EQUIPMENT INSPECTIONS AND TESTS
Section R. 460.2351a - Statistical quality sampling program for diaphragm-type meters

Universal Citation: MI Admin Code R. 460.2351a

Current through Vol. 25-4, March 15, 2025

Rule 51a.

(1) A utility shall comply with the provisions of R 460.2351, except that a utility that receives approval from the commission may adopt the requirements of this rule for statistical sampling and quality control of in-service diaphragm meters. Statistical sampling and quality control must be supervised by an individual trained in statistical sampling techniques.

(2) A utility may use any of the following statistical quality control programs for meter testing, as adopted by reference in R 460.2344:

(a) ANSI B109.1-2019, "Diaphragm-Type Gas Displacement Meters, Under 500 Cubic Feet per Hour Capacity."

(b) ANSI B109.2-2000 (R2008), "Diaphragm-Type Gas Displacement Meters, 500 Cubic Feet per Hour Capacity and Over."

(c) ANSI/ASQ Z1.9-2003 (R2018), "Sampling Procedures and Tables for Inspection by Variables for Percent Nonconforming."

(d) ANSI/ASQ Z1.4-2003 (R2018), "Sampling Procedures and Tables for Inspection by Attributes."

(3) A utility may use an alternative statistical quality sampling program if approved by the commission. An application to use an alternative program must include all of the following information:

(a) A description of the sampling program that must include all of the following:
(i) The type or types of meters subject to the sampling plan.

(ii) The frequency of testing.

(iii) The procedures to be used for the sampling.

(iv) The meter test method to be used.

(v) The accuracy of the testing and of the sampling plan.

(b) An explanation of the reason or reasons for the requested sampling plan.

(c) An analysis that demonstrates that, with respect to assuring the accuracy of the meters tested, the requested sampling program is at least as effective as the standards listed in subrule (2) of this rule.

(4) Meters for quality control sampling must be separated into homogenous groups by year set and may be further separated by manufacturer, capacity rating, model, case type, diaphragm material, year manufactured, or other distinguishing characteristics. When 1 or more groups established are believed to be too small for practical quality control sampling, they may be combined with another group of similar operating characteristics to establish a larger sampling base. Combined groups must have sample size and acceptance-rejection numbers based on the combined total of meters. Samples must be drawn by a random method that ensures each meter in the group has an equal chance of being selected.

(5) All meter groups, or combined meter groups, must be subject to acceptance or rejection on the basis of the statistical results unless it becomes obvious that the rejected meters are predominantly from 1 identifiable subgroup which may be shown by test data to have been affected by location, age, or other common factors. If this result should occur, the identifiable subgroup may be separated and the remaining meters treated as a new combined group with appropriate sample size and acceptance-rejection numbers.

(6) A meter removed from a customers premises and tested as part of any business practice not related to the statistical quality control program must be included only in the programs sample if the meter is randomly selected according to subrule (4) of this rule.

(7) Not later than March 1 of each year, utilities shall file a report of the meters that have been tested during the preceding calendar year. The report must include separate sections addressing results for meters tested as part of the statistical quality control program and meters tested as part of routine meter removals or exchanges. The report must detail both of the following:

(a) All of the following meter characteristics:
(i) Set year.

(ii) Type of case.

(iii) Manufacturer.

(iv) Type of diaphragm.

(v) Revenue classification, either commercial and industrial or residential.

(b) The number of meters in each meter class tested and found within the norm and within each 1% variance from norm between 94% accuracy and 106% accuracy. Meters that are slower than 94% and faster than 106% must each be grouped separately. Meters that are determined to be nonregistering must be reported to either have been repaired, tested, and returned to the field, or retired.

Disclaimer: These regulations may not be the most recent version. Michigan may have more current or accurate information. We make no warranties or guarantees about the accuracy, completeness, or adequacy of the information contained on this site or the information linked to on the state site. Please check official sources.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.