Section 460.3613. Metering equipment testing requirements.

(1) The testing of any unit of metering equipment shall consist of a comparison of its accuracy with a standard of known accuracy. Units which are not properly connected or which do not meet the accuracy or other requirements of these meter and metering equipment rules at the time of testing shall be reconnected or rebuilt to meet such requirements and shall be adjusted to within the required accuracy and as close to zero error as practicable or else their use shall be discontinued.

(2)   Self-contained, single-phase meters, except for  combination   meters (meters that include demand devices or  control  devices),   shall  be  in compliance with all of the following requirements:

(a)   Be checked for accuracy at unity power factor at the point where a meter is installed, at a central testing point, or in a mobile testing laboratory within a period of from 12 months before,   to   60    days   after,   a meter is placed in service, except as

provided for in R 460.3602, and in subrule (3) of this rule, and not later than 9 months after 192 months of service for a surge-resistant meter and not later than 9 months after  96 months of service for a non-surge-resistant meter.

(b)    Notwithstanding the provisions of subdivision (a) of this subrule, upon application to the commission and upon receipt of an order granting approval, the testing of self-contained, single-phase meters in service shall be governed by a quality control plan as follows:

(i)   Meters shall be divided into homogeneous groups by manufacturers' types, except as follows:

(A)  Certain manufacturers' types shall be further subdivided into separate groups by manufacturers' serial numbers as follows:

(1)    General  Electric  type  I-30  shall  be divided  at  serial   number 20,241,829.

(2)   Westinghouse type C shall be divided at serial number 16,350,000.

(3)   Duncan type MF shall be divided at serial number 2,650,000.

(4)     Sangamo type J meters shall be divided starting with  serial  number 10,000,000.

(B)   Non-surge-resistant meters that are installed in nonurban areas shall be treated as separate groups by manufacturers' type.

(ii)    The meters in each homogeneous group shall then be further subdivided into lots of not less than 301, and not more than 10,000, meters each, except that meters of the most recent design may be combined into lots regardless of manufacturers' type, except that where the number of meters of a single type is 8,001 or more, that number of meters shall be segregated by types for the formation of lots.

(iii)   From each assembled lot, a sample of the  size specified in  table

A-2, ANSI/ASQC Z1.9, shall be drawn annually. The sample shall    be   drawn    at random.

(iv)    The meters in each sample shall be tested for accuracy pursuant to the provisions of these rules.

(v)   The test criteria for acceptance or rejection of each lot shall be based on the test at heavy load only and shall be that designated for double specification limits and an acceptable quality level (AQL) that is not higher than 2.50 (normal inspection) as shown in table B-3, ANSI/ASQC Z1.9.

(vi)     The necessary calculations shall be made pursuant to Example B-3 of ANSI/ASQC Z1.9. The upper and lower specification limits, U and L, shall be 102% and 98%, respectively.

(vii)    A lot shall be rejected if the total estimated percent defective (p) exceeds the appropriate maximum allowable percent defective (M) as determined from table B-3 as specified in paragraph (v) of this subdivision.

(viii)   All meters in a rejected lot shall be tested within a maximum period of 48 months and shall be adjusted pursuant to the provisions of R 460.3607 or shall be replaced with meters that are in  compliance with  the requirements of R 460.3607.

(ix)   During each calendar year, new  meter   samples  shall  be   drawn  as specified in this subdivision from all meters in service, with the exception that lots that have been rejected shall be excluded from  the   sampling procedure until all meters included in the rejected lots have been tested.

(x)   The utility may elect to adopt a mixed variables-attributes sampling plan as outlined in Section A9 of ANSI/ASQC Z1.9, in which case, a lot that is not  in compliance with the acceptability criteria of the variables sampling plan shall be resampled the following year using an attributes sampling plan. If the acceptability criteria of the attributes sampling plan are met, then the lot shall be considered acceptable and shall be returned to the variables sampling plan the following year. If the acceptability criteria of the attributes sampling plan are not met, then that lot shall be rejected and all meters in the lot shall be tested and adjusted or replaced within a maximum period of 36 months after the second rejection.

(xi)   The plan specified in paragraph (x) of this subdivision does not alter the rules under which customers may request special tests of meters.

(c)   Be checked for accuracy in all of the following situations:

(i)   When a meter is suspected of being inaccurate or damaged.

(ii)     When the accuracy of a meter is questioned by a customer. (See R 460.3601.)

(iii)   Before use if a meter has been inactive for more than 1 year  after having been in service.

(iv)   When a meter has been removed from service and has not been tested within the previous 48 months.

(d)    Be inspected for mechanical and electrical faults when the accuracy of the device is checked.

(e)   Have the register and the internal connections checked  before  the meter is first placed in service and when the meter is repaired.

(f)  Have the connections to the customer's circuits checked when the meter is tested on the premises or when removed for testing.

(g)    Be checked for accuracy at 50% power factor when purchased and after rebuilding.

(h)    A meter need not be tested or checked for any reason, except when a complaint is received, if the device was tested, checked, and adjusted, if necessary, within the previous 12 months.

(3)   Notwithstanding the provisions of subrules (4)(a)(ii), (5)(a) (ii) and (6)(a)(iii) of this rule, upon application to the commission and upon receipt of an order granting approval,the solid state meters described in subrules (4), (5) and (6) of this rule in service may elect  to  be  governed  by a quality control plan as follows:

(a)  ) Meters shall be divided  into  homogeneous  groups  by  manufacturers' types.

(b)   The meters in each homogeneous group shall then be further subdivided into lots of not less than 301, and not more than 10,000, meters each, except that meters of the most recent design may be combined into lots regardless of manufacturers' type, except that where the number of meters of a single type is 8,001 or more, that number of meters shall be segregated by types for the formation of lots.

(c)     From each assembled lot, a sample of the size specified in table A-2, ANSI/ASQC Z1.9, shall be drawn annually. The sample shall be drawn at random.

(d)    The meters in each sample shall be tested for accuracy pursuant to the provisions of these rules.

(e)   The test criteria for acceptance or rejection of each lot shall  be based on the test at heavy load only and shall be that designated for  double specification limits and an

acceptable quality level (AQL) that is not higher than 2.50 (normal inspection) as shown in table B-3, ANSI/ASQC Z1.9.

(f)    The necessary calculations shall be made pursuant to Example B-3 of ANSI/ASQC Z1.9. The upper and lower specification limits, U and L, shall be 102% and 98%, respectively.

(g)   A lot shall be rejected if the total estimated percent defective (p) exceeds the appropriate maximum allowable percent defective (M) as determined from table B-3 as specified in paragraph (e) of this subdivision.

(h)   All meters in a rejected lot shall be tested within a maximum period of 48 months and shall be adjusted pursuant to the provisions of R 460.3607 or shall be replaced with meters that are in compliance with the requirements of R 460.3607.

(i)   During each calendar year,  new  meter  samples   shall  be  drawn  as specified in this subdivision from all meters in service, with the exception that lots that have been rejected shall be excluded from  the   sampling procedure until all meters included in the rejected lots have been tested.

(j)    The utility may elect to adopt a mixed variables-attributes sampling plan as outlined in Section A9 of ANSI/ASQC Z1.9, in which case, a lot that is not  in compliance with the acceptability criteria of the variables sampling plan shall be resampled the following year using an attributes sampling plan. If the acceptability criteria of the attributes sampling plan are met, the lot shall be considered acceptable and shall be returned to the variables sampling plan the following year. If the acceptability criteria of the attributes sampling plan are not met, then that lot shall be rejected and all meters in the lot shall be tested and adjusted or  replaced  within  a maximum period of 36 months after the second rejection.

(k)   The plan specified in paragraph (j) of this subdivision does not alter the rules under which customers may request special tests of meters.

(4)   All single-phase meters that are not included in  the  provisions  of subrule (2) of this rule, together with associated equipment, such as demand devices, control devices and instrument transformer-rated meters, shall be in compliance with all of the following requirements:

(a)   Be checked for accuracy at unity power factor at the point where a meter is installed, at a central  testing point, or in a mobile testing laboratory as follows:

(i)   Within a period of from 12 months before, to 60 days after, a meter is placed in service, exceptions to this subrule (4)(a) of this rule are as provided for in R 460.3602 and for solid state meters.

(ii)     Not later than 9 months after 144 months  of   service  for  a surge-resistant meter and not later than 9 months after 96 months of service for a non-surge-resistant meter.

(iii)   When a meter is suspected of being inaccurate or damaged.

(iv)    When the accuracy of a meter is questioned by a customer. (See R 460.3601.)

(v)   Before use when a meter has been inactive for more than 1 year after having been in service.

(vi)   When a meter is removed from service and has not been tested within a period equal to 1/2 of the normal test schedule.

(b)    Be inspected for mechanical and electrical faults when the accuracy    of the device is checked.

(c)   Have the register and the  internal  connections  checked  before    the meter is first placed in service and when the meter is repaired.

(d)  Have the connections to the customer's circuits checked when the  meter is tested on the premises or when removed for testing.

(e)    Be checked for accuracy at 50% power factor when purchased    and      after rebuilding.

(f)  ) A meter need not be tested or checked for any reason,  except  when   a complaint is received, if the device was tested, checked and adjusted, if necessary,  within the previous 12 months.

(5)    All  self-contained,  3-phase meters  and associated equipment    shall      be in compliance with all of the following requirements:

(a)  Be tested for accuracy at unity and 50% power factor as follows:

(i)   Before being placed in service.

(ii)   Not later than 9 6 months after 120 months of service.

(iii)   When a meter is suspected of being inaccurate or damaged.

(iv)    When the accuracy of a meter is questioned    by    a    customer.    (See      R 460.3601.)

(v)   When a meter is removed from service.

(b)   Be inspected for mechanical and electrical faults when the accuracy is checked.

(c)    Have the register and internal connections checked before the meter    is first installed, when repaired and when the register is changed.

(d)   Have the  connections   to  the  customer's  circuits  and     multipliers checked when the equipment is tested for accuracy on the customer's premises.

(6)    All transformer-rated, 3-phase meters and associated    equipment     shall be in compliance with all of the following requirements:

(a)  Be checked for accuracy at unity and 50% power factor as follows:

(i)   Before being placed in service.

(ii)      On the customer's premises within 60 days after installation, unless the transformers are in compliance with the specifications  outlined   in  the  American National Standards Institute standard ANSI C-57.13, and unless the meter adjustment limits do not exceed plus or minus 1.5% at 50% power factor.

(iii)   Not later than 9 months after 72 months of service.

(iv)   When a meter is suspected of being inaccurate or damaged.

(v)   When the accuracy is questioned by a customer. (See R 460.3601.)

(vi)   When a meter is removed from service.

(b)   Be inspected for mechanical and electrical faults when the accuracy is checked.

(c)    Have the register and internal connections checked before the meter    is first placed in service and when the meter is repaired.

(d)   Have the connections to the customer's circuits  and  multipliers checked when the equipment is tested for accuracy on the premises or when removed for testing and when instrument transformers are changed.

(e)    Be checked for accuracy at 50% power factor when purchased    and      after rebuilding.

(7)   Instrument transformers shall  be  tested  in  all of the  following situations:

(a)  ) When  first  received,  unless  a  transformer   is  accompanied  by  a certified test report by the manufacturer.

(b)   When removed from service.

(c)   Upon complaint.

(d)  When there is evidence of damage.

(e)    When an approved check, such as the variable burden method in the case of current transformers that is made when the meter is tested indicates that a quantitative test is required.

(8)    Demand meters shall be    in    compliance    with    both    of    the      following requirements:

(a)  Be tested for accuracy in all of the following situations:

(i)   Before a meter is placed in service.

(ii)   When an associated meter is tested and the demand meter  is   a   block interval nonrecording type or a thermal type.

(iii)   After 2 years of service if the meter is of the recording type, but testing is not required if the meter is of the pulse-operated type and the demand reading is checked with the kilowatt-hour reading each billing cycle.

(iv)   When a meter is suspected of being inaccurate or damaged.

(v)   When the accuracy is questioned by a customer. (See R 460.3601.)

(vi)   When a meter is removed from service.

(b)  Be inspected for mechanical and electrical  faults  when  a meter  is tested in the field or in the meter shop.

History: 1983 AACS; 1995 AACS; 2008 AACS.