PUBLIC UTILITIES COMMISSION OF THE STATE OF CALIFORNIA
CA-12
ENERGY DIVISION RESOLUTION G-3299
December 21, 2000
Resolution G-3299. Sempra Energy, on behalf of San Diego Gas & Electric Company (SDG&E), proposes to modify its deviation from General Order 58-A, "Standards for Gas Service" and its Gas Meter Performance Control Program. Approved.
By Letter dated May 8, 2000, from Sempra Energy Company.
__________________________________________________________
SUMMARY
This resolution approves Sempra's request by Letter dated May 8, 2000, to change: (1) the way it currently deviates from General Order (G.O.) 58-A, Section 13c (Periodic and Other required Tests of Gas Meters); and (2) its Gas Meter Performance Control Program (Program) for testing and controlling the performance of its 175 - 1000 cubic feet per hour (cfh) capacity diaphragm gas meters as authorized by the Commission by Resolution (Res.) G-3061, adopted June 23, 1993. Sempra is proposing two changes to its Program: a change in the gas meter Control Group Sampling start date and a change to one Probability Decision Curve.
We approve Sempra's request on the basis that they have provided adequate justification to warrant a deviation from G.O. 58-A, Section 13c and its Program. The change to the gas meter Control Group Sampling start date will save the removal and testing of over 2,000 gas meters annually. This will result in an estimated annual capital, operation, and maintenance cost savings of $60,938. The change to one Probability decision curve will result in estimated annual capital, operation, and maintenance cost savings of $129,938. Some of the annual cost savings should be passed on through SDG&E's PBR mechanism. SDG&E will split these annual cost savings between shareholders and ratepayers.
BACKGROUND
G.O. 58-A, Section 13, Periodic and Other Required Tests of Gas Meters, Paragraph a., provides that no gas meter installed after April, 1989, shall be allowed to remain in service more than ten years from the date when last tested without being retested.
G.O. 58-A, Section 13, Paragraph c., provides that "Under certain conditions utilities may be authorized to deviate from Section 13.a. and use a statistical meter control program based on meter performance as demonstrated by sample testing in lieu of periodic testing of each meter. Applications to deviate shall be based on accepted principles of statistical sampling."
By Res. G-1426 of December 10, 1968, the Commission approved SDG&E's initial Program and also approved statistical sampling of small domestic gas diaphragm meters rated up to 300 cfh in accordance with SDG&E's statistical sampling program. By Res. G-2907 of March 14, 1990, the Commission approved expansion of the Program to include diaphragm-type gas meters with a capacity up to and including 500 cfh.
Based on the statistical analysis from Program data, many more customer meters were registering below actual volumes than above actual volumes; and it appeared that meters tended to run slower and register lower as they age. By Res. G-2927 of December 6, 1990, the Commission authorized a Program modification to set customer meters at zero percent below actual check flow volumes, rather than at 0.25% below actual check flow volumes, and with customer meters allowed to register no more than 0.3% above actual under full volume flows.
By Res. G-3061 of June 23, 1993, the Commission authorized SDG&E to expand its existing Program to include all diaphragm meters rated from 50 to 1,000 cfh. Instead of testing all meters, SDG&E tests statistically significant samples of all groups of gas meters annually to determine whether the meters should be replaced.
Sempra filed its letter dated May 8, 2000, to change: (1) the way it currently deviates from General Order (GO) 58A, Section 13c (Periodic and Other required Tests of Gas Meters); and (2) its Gas Meter Performance Control Program for testing and controlling the performance of its 175 - 1000 cfh capacity diaphragm gas meters as authorized by the CPUC in Res. G-3061.
Sempra is proposing two changes to SDG&E's Gas Meter Performance Control Program: (1) a change in the gas meter Control Group Sampling start date; and (2) a change to one Probability Decision Curve.
Sempra has performed a study of the meter control groups (1986 - 1998) for the first 10 years of their life and found that all maintained accuracy within the performance standards of SDG&E's program. Sempra is proposing to change the present practice of removing and sampling gas meters starting in year one of the control group life. Instead, it wants to start sampling gas meters in year 10 of the control group life and continue sampling each year throughout the life of the control group. In their proposal letter dated May 8, 2000, SDG&E states, "We believe this is consistent with the intent of GO 58-A 13.a. which indicates, "No gas meter hereafter shall be allowed to remain in service more than ten (10) years from the time when last tested in the matter herein provided...". This change to the testing parameters will save the removal and testing of over 2,000 gas meters annually. In 1999 this change would have saved the removal and testing of approximately 2,016 gas meters."
Sempra is proposing to change the language in the Gas Meter Performance Control Program to read, "A minimum sample of 50 meters will be obtained for each control group each year, starting in the 10th year of the control groups life."
Sempra is also proposing to change the Probability Decision Curve for "Percent of meters between Plus and Minus 2%." Presently, the requirements indicate that any meter group should be scheduled for removal as soon as sampling indicates that: (1) more than 20% of the meters in the group are registering outside the range of 2% fast to 2% slow; or (2) more than 10% of the meters in the group are registering more than 2% fast. SDG&E proposes to change the first condition to allow up to 25% of the meter control group to register outside the range of 2% fast to 2% slow. The standard for percent of fast meters will remain at its current level.
The current Probability Decision Curves show control groups of meters to perform accurately when: (1) 80% or more of the meters tested are within +/- 2% of zero error; and (2) 90% of the meters tested are not more than 2% fast.
The proposed Probability Decision Curves will show control groups of meters to perform accurately when: (1) 75% or more of the meters tested are within +/- 2% of zero error; and (2) 90% of the meters tested are not more than 2% fast.
Sempra's proposed change is one of many procedures being aligned for consistency between SDG&E and the Southern California Gas Company (SoCalGas) under the Sempra Parent Company. In Res. G-2928 adopted on November 9, 1990, SoCalGas was authorized to change its Gas Meter Performance Program and meter proof setting standards. SoCal Gas was allowed up to 25% of the meters in a group to register outside the range of 2% fast to 2% slow. Also, Res. G-3298 allowed SoCalGas to use optimal proof settings based upon the statistical performance of each type of meter.
Sempra believes the proposed program change would effectively extend the average service life of SDG&E meter groups by 4 - 5 years. As in the present standard, the proposed standard accepts no greater ratio of fast meters. It allows only a slightly greater number of customers to be served through slow meters.
NOTICE
Notice of Sempra's proposed modifications was made by publication in the Commission Daily Calendar on September 1, 2000.
PROTESTS
No protests were filed.
DISCUSSION
Energy Division has reviewed Sempra's Letter, G.O. 58-A, and Resolutions G-1426, G-2927, G-3061, and G-2928. In Sempra's Letter, Sempra attached Exhibits A and B. Exhibit A is entitled "Gas Meter Performance Study." In this exhibit, SDG&E performed a study of the meter control groups (1986 - 1998) for the first 10 years of their life and found all groups maintained accuracy within the standards of SDG&E's program. In response to an Energy Division data request, SDG&E stated that there are 62 control groups of meters for this type of gas meter included in SDG&E's Meter Performance Control Program. There were 29 control groups for the 1986 - 1998 study, which means 47% of the total number of control groups were tested in the study. It is the Energy Division's opinion that this is a reasonable approach.
G.O. 58-A, Section 13c addresses removal and testing of gas meter sets for inaccurate measurement only. This is not a safety issue and therefore will not adversely affect the level of safety in the gas system. Gas safety issues are addressed under G.O. 58A Section 11 Service and Meter Installations and G.O. 112E. Thus, there should be no reduction in safety from changing this Program. We find the proposal to change the gas meter Control Group Sampling start date reasonable and believe it should be approved.
Exhibit B of Sempra's Letter is entitled "SDG&E's Gas Meter Performance Control Program." Sempra is proposing to change the language in the Program to read, "A minimum sample of 50 meters will be obtained for each control group each year, starting in the 10th year of the control group life." In response to an Energy Division data request, SDG&E has stated that the samples will be selected at random with a minimum of 50 meters per control group. In Exhibit A of Sempra's Letter, SDG&E performed a study of the meter control groups (1986 - 1998) for the first 10 years of their life and found all maintained accuracy within the standards of SDG&E's program. We believe the random sampling and results of this study of the meter control groups are reasonable and approve Sempra's proposed language.
SDG&E is also proposing to change the Probability Decision Curve for, "Percent of meters between Plus and Minus 2%." SDG&E proposes to change the first condition to allow up to 25% of the meter control group to register outside the range of 2% fast to 2% slow. The standard for percent of fast meters will remain at its current level.
In Res. G-3298 adopted on November 9, 1990, SoCalGas was authorized by the Commission to change its Gas Meter Performance Program and meter proof setting standards. SoCal Gas was allowed up to 25% of the meters in a group to register outside the range of 2% fast to 2% slow. Sempra's Letter states, "SoCal Gas has used this procedure for 10 years by extending the gas meters life by approximately 4 - 5 years without reducing the accuracy of the control groups meters. This proposed program change would effectively extend the average service life of SDG&E meter groups by 4 - 5 years."
In an Energy Division data request, SDG&E was asked to explain why this would occur. SDG&E explained that by changing the probability decision curves for +/- 2% of zero error, from 80% to 75%, removal of a meter control group is deferred until a higher percentage (5%) do not meet the +/- 2% criteria. The 4 - 5 year extending of meter life was based on SoCalGas' meter performance program experience. Sempra's proposed change to one Probability decision curve is reasonable and should be approved.
The meters utilized by SDG&E have the same specifications as those used by SoCalGas. Hence, we believe the SDG&E meters will perform in a manner comparable to the SoCalGas meters.
The Letter states, "This change to the testing parameters will save the removal and testing of over 2000 gas meters annually. In 1999 this change would have saved the removal and testing of approximately 2016 gas meters." In response to an Energy Division data request, SDG&E stated that the proposed changes in the Letter would result in annual cost savings from two sources: the change in the Probability Decision Curve, and by not having to sample meters until their tenth year of service life. Changing the Probability Decision Curve from the existing 80% testing level to the proposed 75% testing level is expected to yield $129,938 in annual capital, operation, and maintenance cost savings because SDG&E would be able to reduce the statistical sample size of the testing. In addition, not having to sample meters until their tenth year of service life is expected to yield $60,938 in annual capital, operation, and maintenance cost savings. In summary, there will be a total of $190,876 in estimated annual cost savings that SDG&E will realize by the approval of this deviation. Some of the annual cost savings may be passed on through SDG&E's PBR mechanism. SDG&E will split these annual cost savings between shareholders and ratepayers.
Energy Division has reviewed SDG&E's proposed changes for its deviation from G.O. 58-A and concludes that by adopting the changes the quality of metering will not decrease. Since these proposed changes concern meter accuracy and not safety issues, they would not adversely affect the level of safety in the gas system. SDG&E will continue to perform normal operation and maintenance procedures as scheduled to ensure safe meter operation. These include routine leak surveys for gas leaks, monitoring for atmospheric corrosion, as well as the routine visual inspection done by the meter readers monthly as they read the meters. The quality of metering provided to the customer should not decrease under the proposed modifications. The effect on customers should be positive given that the reduction in number of meter transactions decrease the opportunity for error and also decrease customer inconvenience due to meter change-outs. Finally, some of the savings due to reduced meter testing will be passed on to ratepayers.
COMMENTS
This is an uncontested matter in which the resolution grants the modifications requested. The only party, SDG&E, has stipulated to waive the 30-day waiting period required by P.U. Code Section 311(g)(1), and opportunity to file comments on the draft resolution. Accordingly, this matter will be placed on the Commission's agenda directly for prompt action.
FINDINGS
1. By Res. G-1426 adopted on December 10, 1968, the Commission authorized a deviation from G.O. 58-A, allowing sampling of certain meters instead of testing all meters after ten years.
2. By Res. G-2927 adopted on December 6, 1990, the Commission authorized a deviation to include larger meters than authorized in 1968.
3. By Res. G-3061 adopted on June 23, 1993, the Commission authorized SDG&E to expand its existing Program to include all diaphragm meters rated from 50 to 1,000 cfh.
4. By Res. G-2928, adopted on November 9, 1990, SoCalGas was authorized to change its Gas Meter Performance Program and meter proof setting standards. SoCal Gas was allowed up to 25% of the meters in a group to register outside the range of 2% fast to 2% slow. Also, Res. G-3298 allowed SoCalGas to use optimal proof settings based upon the statistical performance of each type of meter.
5. Sempra is proposing two changes to its Gas Meter Performance Control Program. The first change is a change in the gas meter Control Group Sampling start date. The second change is a change to one Probability Decision Curve.
6. These proposed modifications will not adversely affect the level of safety in the gas system.
7. The two requested changes by Sempra will not decrease the quality of metering and will benefit both the utility and the customer and should be approved.
8. The change to the gas meter Control Group Sampling start date will save the removal and testing of over 2,000 gas meters annually. This will result in estimated annual capital, maintenance, and operation cost savings of $60,938.
9. The change to one Probability Decision Curve will result in estimated annual capital, maintenance, and operation cost savings of $129,938.
10. Some of the annual cost savings may be passed on through SDG&E's PBR mechanism. SDG&E will split these annual cost savings between shareholders and ratepayers.
THEREFORE IT IS ORDERED THAT:
1. Sempra Energy's request, on behalf of SDG&E, to propose to modify its deviation from General Order 58-A, "Standards for Gas Service" and its Gas Meter Performance Control Program is approved.
2. Some of the annual cost savings shall be passed on through SDG&E's PBR mechanism.
This Resolution is effective today.
I certify that the foregoing resolution was duly introduced, passed and adopted at a conference of the Public Utilities Commission of the State of California held on December 21, 2000; the following Commissioners voting favorably thereon:
_____________________
WESLEY M. FRANKLIN
Executive Director
