Direct Access Standards for Metering and Meter Data (DASMMD...Table I-1: Standards Required for...

DIRECT ACCESS STANDARDS FOR

METERING AND METER DATA

(DASMMD)IN CALIFORNIA

MARCH 1999

This Page Intentionally Left Blank

Introduction

The Direct Access Standards for Metering and Meter Data (DASMMD) are applicable toall market participants involved in Direct Access, or the provision of Direct Access-related services, in California and specify the minimum standards for safety, accuracy,and reliability of:• Meter Equipment• Meter Communications• Meter Data Management and Meter Reading − Including rules for validating, editing, and estimating meter usage data• Meter Installation, Maintenance, Testing, and Calibration

− Including classifications of meter workers

These standards were established by California Public Utilities Commission (CPUC)Decision 98-12-080 and are generally broad in scope in order to allow an openarchitecture approach to metering and meter data, expand technology choices, andprovide opportunities for all market participants on an equal basis. In instances wherethe CPUC did not adopt permanent standards in D.97-12-080, the standards establishedby D.97-12-048 and D.98-05-044 are included herein as indicated. This DASMMD isreferred to and incorporated by reference into the associated Direct Access tariffs.

Copies of the DASMMD and Rule 22/Rule 25 are available on the UDC’s websites asfollows:

UDC Website For additional informationSCE www.sce-esp.com 800-203-4634PG&E www.pge.com 415-973-1666SDG&E www.sdge.com 619-654-8211

Copies of CPUC decisions are available upon request from the CPUC by calling 415-703-1282 or on the CPUC’s website at: www.cpuc.ca.gov

For detailed information regarding ANSI standards, please consult the AmericanNational Standards Institute website at: www.ansi.org

The standards contained in the DASMMD are subject to change. As such, the UDCshould be consulted in case of doubt regarding any particular standard. Nothing in theDASMMD is intended to modify the parties’ rights and obligations under Rule 22/Rule25. The provisions of the UDC’s tariffs supersede any provisions contained in theDASMMD which may appear to be inconsistent or contrary.

ii

Table of Contents

CHAPTER A: STANDARDS FOR METER PRODUCTS

I. TABLES OF STANDARDS ...................................................................................................................... 2

I-1 STANDARDS REQUIRED FOR METER PRODUCTS USED IN DIRECT ACCESS........................................................ 2I-2. LIST OF TESTS IN ANSI C12.1 AND C12.20 STANDARDS ............................................................................... 4

II. CERTIFICATION TESTING REQUIREMENTS................................................................................... 5

II.1. GENERAL................................................................................................................................................... 5II.2. METER PRODUCT FAILURE DEFINITION ....................................................................................................... 7II.3. METER TYPE CERTIFICATION REJECTION CRITERIA ..................................................................................... 8II.4. TEST SETUP .............................................................................................................................................. 9II.5. ANSI C12.1 TESTS.................................................................................................................................. 10II.6. ANSI C12.20 TESTS................................................................................................................................ 11II.7. TEST A1 - SUNLIGHT INTERFERENCE TEST................................................................................................ 12

III. REGISTRATION AND CENTRALIZED DATABASE FOR DIRECT ACCESS COMPLIANTMETER TYPE......................................................................................................................................... 13

IV. REQUIREMENTS FOR STICKERS, SEALING AND LOCKING HARDWARE ............................. 14

V. REQUIREMENTS FOR LABELING MANUFACTURING DATE ON NEW METER PRODUCTS 14

VI. REQUIREMENTS FOR REBUILT, RETROFIT AND REPAIRED METER PRODUCTS............... 15

CHAPTER B: STANDARDS FOR METER COMMUNICATIONS

I. METER COMMUNICATION STANDARDS........................................................................................ 18

1. KYZ CONTACT OUTPUT ......................................................................................................................... 182. METER PASSWORD AUTHORIZATION ........................................................................................................ 183. CONSUMER PROTECTION ON KYZ CONTACT OUTPUT .............................................................................. 184. VISUAL METER READ .............................................................................................................................. 185. OPTICAL PORT STANDARD ....................................................................................................................... 18

CHAPTER C: STANDARDS FOR METER DATA MANAGEMENT AND METER READING

I. DEFINITION OF MDMA BUSINESS FUNCTIONS ............................................................................ 20

II. MDMA QUALIFICATION TESTING................................................................................................... 21

III. METER READING FREQUENCY........................................................................................................ 21

IV. MDMA SAFETY REQUIREMENTS..................................................................................................... 22

V. MDMA TECHNICAL/BUSINESS SUPPORT TO ESPS AND UDCS.................................................. 22

VI. MDMA PERFORMANCE STANDARDS .............................................................................................. 22

VII. MDMA PERFORMANCE EXEMPTIONS............................................................................................ 23

VIII. EDI IMPLEMENTATION...................................................................................................................... 24

IX. VALIDATING, EDITING, AND ESTIMATION................................................................................... 24

iii

CHAPTER D:STANDARDS FOR METER INSTALLATION, MAINTENANCE, TESTING, AND CALIBRATION

I. MSP METER WORKER QUALIFICATIONS ..................................................................................... 28

I.1. METER WORKER SKILL, SAFETY, AND QUALIFICATION REQUIREMENTS..................................................... 29I.1.1. CLASS 1.......................................................................................................................................... 29I.1.2. CLASS 2.......................................................................................................................................... 30I.1.3. CLASS 3.......................................................................................................................................... 31I.1.4. CLASS 4(A) ..................................................................................................................................... 32I.1.5. CLASS 4(B) ..................................................................................................................................... 33I.1.6. CLASS 5.......................................................................................................................................... 34

I.2 CERTIFICATION OF METER SERVICE PROVIDER (MSP) .............................................................. 35

II. METER INSTALLATION AND REMOVAL........................................................................................ 38

II.1. SAFETY (CUSTOMER LIFE SUPPORT, PUBLIC, ETC.) ..................................................................................... 38II.1.1. Customer Life Support..................................................................................................................... 38II.1.2. Electrical Hazards: ......................................................................................................................... 38II.1.3. Physical Hazards............................................................................................................................. 39II.1.4. Customer Premises:......................................................................................................................... 40II.1.5. Vermin ............................................................................................................................................ 40

II.2. METER SECURITY AND ACCESSIBILITY ..................................................................................................... 41II.2.1. Infraction and Evidence of Tampering/Energy Diversion ................................................................ 41

II.2.1.1. Meter Installation..................................................................................................................................... 41II.2.1.2. Meter Cover............................................................................................................................................. 41II.2.1.3. KWH Register ......................................................................................................................................... 42II.2.1.4. Meter Disk.............................................................................................................................................. 42II.2.1.5. Test Blocks .............................................................................................................................................. 42II.2.1.6. Meter Base .............................................................................................................................................. 42II.2.1.7. Meter Socket........................................................................................................................................... 42II.2.1.8. Hidden "Service Riser" Taps and Un-metered Circuits Utilizing Relay Devices ....................................... 42

II.2.2. Meter Security................................................................................................................................. 42II.2.2.1. Physical Meters and Panels ...................................................................................................................... 42II.2.2.2. Programmable Meters and Data ............................................................................................................... 42

II.3. SITE VERIFICATION ................................................................................................................................. 43II.3.1. Direct Access Metering Verification................................................................................................ 43II.3.2. Direct Access Communication Verification...................................................................................... 43II.3.3. Transformer-rated Meter Sites......................................................................................................... 43II.3.4. Pole-mounted Meter Sites................................................................................................................ 43II.3.5. Pad-mounted Meter Sites................................................................................................................. 43

II.3.5.1. Primary Metering..................................................................................................................................... 43II.3.5.2. Cabinets................................................................................................................................................... 44II.3.5.3. Cabinet Safety.......................................................................................................................................... 44II.3.5.4. Cabinet Markings..................................................................................................................................... 44

II.3.6. Grounding....................................................................................................................................... 44II.3.7. Watt Load Clock.............................................................................................................................. 44II.3.8. Customer’s Ground Fault Protection Device ................................................................................... 44

II.3.8.1. Background.............................................................................................................................................. 45II.3.8.2. Procedures ............................................................................................................................................... 45

II.3.9. PT/CT Secondary Wiring................................................................................................................. 45II.4. METER INSTALL ...................................................................................................................................... 46

II.4.1. Pre-Installation ............................................................................................................................... 46II.4.2. Self Contained Meters ..................................................................................................................... 46II.4.3. Transformer Rated Meters (CT Meter)............................................................................................. 46II.4.4. Meter Installation at New Sites........................................................................................................ 46II.4.5 Rewires and Service Upgrades ........................................................................................................ 47II.4.6. Locking Device................................................................................................................................ 47

iv

III. METER MAINTENANCE AND TESTING SCHEDULE..................................................................... 48

III.1. MAINTENANCE SCHEDULE ....................................................................................................................... 48III.2. MAINTENANCE UPON REQUEST................................................................................................................. 48III.3. STATISTICAL SAMPLING REQUIREMENTS .................................................................................................. 48III.4. CRITERIA FOR REQUIRED CORRECTIONS .................................................................................................... 48III.5. TROUBLESHOOTING AND CORRECTIVE ACTIONS ........................................................................................ 49III.6 MDMA COMMUNICATIONS ..................................................................................................................... 49III.7. DEMARCATION POINT OF METER WORK ................................................................................................... 50

IV. METER SYSTEM TESTING ................................................................................................................. 51

IV.1. METER SOCKET....................................................................................................................................... 51IV.2. METERING SYSTEM ................................................................................................................................. 51IV.3. NOTIFICATION......................................................................................................................................... 51

V. TEST STANDARDS................................................................................................................................ 52

V.1. CALIBRATION AND MAINTENANCE OF TEST STANDARDS ........................................................................... 52V.1.1. Out-of-calibration ........................................................................................................................... 52V.1.2. Basic Reference Test Standard ........................................................................................................ 52V.1.3. Portable Test Standard.................................................................................................................... 52V.1.4. Routine Accuracy Check Requirements ........................................................................................... 52

V.2. METER RE-TESTING UPON FINDING OF AN OUT-OF-CALIBRATION TEST STANDARD .................................. 52

VI. GUIDELINE FOR METER TESTING .................................................................................................. 54

MATRIX OF METER SYSTEM TESTING ............................................................................................................... 54VI.1. TASK #1: VOLTAGE TEST ........................................................................................................................ 55

VI.1.1 Secondary Distribution Voltages: .................................................................................................... 55VI.1.2 Primary Distribution and Transmission Voltages............................................................................. 56

VI.2. TASK #2: LIGHT LOAD & FULL LOAD TEST OR CUSTOMER-LOAD TEST ..................................................... 57VI.2.1 No Load or Creep Test (for mechanical & hybrid meters only) ........................................................ 57VI.2.2 Accuracy test................................................................................................................................... 57VI.2.3 Adjustments ..................................................................................................................................... 57VI.2.4 Reporting: ....................................................................................................................................... 57

VI.3 TASK #3: DEMAND TEST.......................................................................................................................... 58VI.3.1 Mechanical Demand Meters ............................................................................................................ 58VI.3.2 Solid State Meters and Electronic Registers .................................................................................... 58

VI.4 TASK #4: REGISTER VERIFICATION........................................................................................................... 59VI.4.1 Mechanical Register........................................................................................................................ 59VI.4.2 Electronic Register.......................................................................................................................... 59

VI.5 TASK #5: PHASE ANGLE TEST .................................................................................................................. 60VI.5.1 Utility phase rotation requirements ................................................................................................. 60VI.5.2 Phase shifting transformer (reactaformer) voltage........................................................................... 60VI.5.3 Power Factor Calculation: .............................................................................................................. 60VI.5.4 Current Measurement: .................................................................................................................... 60VI.5.5 Phase Angle Measurement............................................................................................................... 60VI.5.6 Phase Angle Plot: ........................................................................................................................... 60

v

VI.6 TASK #6: SEPARATE ELEMENT CHECK...................................................................................................... 61VI.6.1 Testing ............................................................................................................................................ 61VI.6.2 Correct phasing............................................................................................................................... 61VI.6.3 No forward rotation......................................................................................................................... 61VI.6.4 Cautions.......................................................................................................................................... 61

VI.7 TASK #7: BURDEN TEST .......................................................................................................................... 62VI.7.1 Procedure........................................................................................................................................ 62VI.7.2 Possible Problems ........................................................................................................................... 62

GLOSSARY OF TERMS AND ACRONYMS.................................................................................................. 63

ATTACHMENTSSTANDARDS FOR VALIDATING EDITING AND ESTIMATING MONTHLY AND INTERVAL DATA

vi


CHAPTER A

STANDARDS FOR METER PRODUCTS

Standards for Meter Products

2

I. TABLES OF STANDARDS

The following Tables provide a list of standards for meter products. Table I-1 listsstandards which shall be requirements for meter products used in Direct Access.

Table I-1: Standards Required for Meter Products Used in Direct AccessStandards Required in

CPUC D.97-12-048

Effectivedate

Comments

ANSI C12.1-1995, Code forElectricity Metering

Yes 12-17-98 To be used in conjunction with Standards forMeter Products, Section II: CertificationTesting Requirements.

ANSI C12.7-1993, Watt-hourMeter Socket

Yes 12-17-98 Applies only if a meter socket is being used.

ANSI C12.8-1981 (R1997), TestBlocks and Cabinets for Installationof Self-Contained A-Based Meters

Yes 12-17-98 Applies only if an A-base meter is used.

ANSI C12.9-1993, Test Switchesfor Transformer-rated Meters

Yes 12-17-98

ANSI C12.10-1997,Electromechanical Watt-hourMeters

Yes 12-17-98

ANSI C12.11-1987 (R1993),Instrument Transformers forRevenue Metering, 10 kV-350 kVBIL (0.6-69 kV NSV)

Yes 12-17-98

ANSI C12.13-1991, ElectronicTOU Registers for ElectricityMeters

Yes 12-17-98 Applies only if the meter has a time-of-useregister.

ANSI C12.20-1998, 0.2% & 0.5%Accuracy Class Meters (approvedbut not yet published)

Yes 3-18-2000 To be used in conjunction with Standards forMeter Products, Section II: CertificationTesting Requirements.

ANSI C37.90.1-1989 (R1994),Surge Withstand Capability (SWC)Test

Yes 12-17-98 Adds to ANSI C12.1

ANSI 57.13-1978 (1987),C57.13.1-1981 (1992), C57.13.2-1991, C57.13.3-1983 (1991),Instrument Transformers

No 12-17-98 These accuracy and safety performancestandards are used in conjunction with ANSIC12.11.


3

Table I-1: Standards Required for Meter Products Used in Direct Access (continued)

StandardsRequired in

CPUC D.97-12-048

Effectivedate

Comments

ANSI C12.18-1996, ProtocolSpecification for ANSI Type 2Optical Port

No 6-1-99 Applies only if a Type 2 Optical Port isbeing used.

Applicable FCC Regulations Yes 12-17-98 All meters and associated equipment are tomeet all applicable FCC regulations

Certification Testing Requirements Yes 4-16-99 Standards for Meter Product, Section II.Submission of meter type self-certification documents to theEnergy Division. Meter productmanufacturer to make available allbackup documentation that isrelated to the certification testingrequirements for the meter type thatis being certified.

Yes 4-16-99 Standards for Meter Product, Section III.

Stickers, sealing and lockinghardware requirements.

Yes 4-16-99 Standards for Meter Product, Section VI.

Manufacturing date to be includedon all new meter products.

Yes 4-16-99 Standards for Meter Product, Section V.

Procedures to follow whenrebuilding, retrofitting, or repairinga meter product.


For meter products which store andprovide interval meter data, themeter must be capable of providingand storing the interval meter datafor a minimum of 35 days.



4

Table I-2 provides a summary list of tests in ANSI C12.1 and C12.20 Standards, asunlight test, and ANSI C37.90.1 test. All shall be applied in conjunction with SectionII: Certification Testing Requirements. This list also shows the eight tests required to beperformed in series.

Table I-2: List of Tests in ANSI C12.1 and C12.20 StandardsLineNo.

Tests performed inseries (Sections

II.1.6.,II.5.&II.6.)

Descriptions of Certification Tests ANSI C12.1 ANSI C12.20

1 No Load Test #1 Test #12 Starting Load Test #2 Test #23 Load Performance Test #3 Test #34 Effect of Variation of Power Factor Test #4 Test #45 Effect of Variation of Voltage Test #5 Test #56 Effect of Variation of Frequency Test #6 Test #67 Equality of Current Circuits Test #7 Test #78 Internal Meter Losses Test #8 Test #89 Temperature Rise Test #9 Test #910 Effect of Register Friction Test #10 Test #1011 Effect of Internal Heating Test #11 Not applicable12 Effect of Polyphase Loading Not applicable Test #1113 Effect of Tilt Test #12 Not applicable14 Stability of Performance Test #13 Not applicable15 Independence of Elements Test #14 Not applicable16 ü Insulation Test #15 Test #1217 ü Voltage Interruptions Test #16 Test #1318 ü Effect of High Voltage Line Surges Test #17 Test #1419 Effect of External Magnetic Field Test #18 Test #1520 Effect of Variation of Ambient Temperature Test #19 Test #1621 Effect of Temporary Overloads Test #20 Test #1722 Effect of Current Surges in Ground Conductors Test #21 Test #1823 Effect of Superimposed Signals Test #22 Test #1924 Effect of Voltage Variation-secondary Time Base Test #23 Test #2025 Effect of Variation of Amb. Temp.-second. Time Base Test #24 Test #2126 ü Electrical Fast Transient/Burst Test #25 Test #2227 Effect of Radio Frequency Interference Test #26 Test #2328 Radio Frequency Conducted and Radiated Emission Test #27 Test #2429 ü Effect of Electrostatic Discharge (ESD) Test #28 Test #2530 Effect of Storage Temperature Test #29 Test #2631 ü Effect of Operating Temperature Test #30 Test #2732 ü Effect of Relative Humidity Test #31 Test #2833 Mechanical Shock Test #32 Test #2934 Transportation Drop Test #33 Test #3035 Mechanical Vibration Test #34 Test #3136 Transportation Vibration Test #35 Test #3237 Weather Simulation Test #36 Test #3338 Salt-spray Test #37 Test #3439 Rain tightness Test #38 Test #3540 Test #A1: Sunlight Interference Not yet included Not yet included41 ü Test #A2: ANSI C37.90.1, Surge Withstand Not yet included Not yet included


5

II. CERTIFICATION TESTING REQUIREMENTSThis Section describes the certification testing requirements that Meter Productsused in Direct Access metering must comply with. These tests are extracted andmodified accordingly from the Meter and Data Communication Standards(MDCS) Workshop Report, Appendix A, filed with the California Public UtilitiesCommission on July 25, 1997 and the Joint Parties Letter, Appendix B, filed withthe CPUC on November 24, 1997. This Section shall be used in conjunction withANSI C12.1 and C12.20 Standards to cover issues that are not currentlyaddressed in the ANSI C12.1 and C12.20 Standards. Some of these issues are: 1)duplication of the field electrical and environmental conditions is necessary toassure safety, 2) not all components of a meter product are required to beincluded in the meter product during certification testing, 3) reporting ofcertification tests is not based on all meter products tested, 4) no certificationrejection criteria is provided for declaration of success or failure uponcompletion of certification tests.

II.1. General

II.1.1. The tests specified shall be conducted by qualified facilities. Aqualified facility is a facility that has access to the necessaryequipment and personnel to perform the testing requirementsspecified in this document.

II.1.2. Complete performance testing is required for new meter types andfor major design changes to existing meter types. If an incrementalchange or changes are made to an existing meter type, applicabletests shall be performed to assure that Meter Products meet thecertification testing requirements as stated in this section.

II.1.3. The manufacturer shall provide a certified test report documentingthe tests and their results to the purchaser. The test report shall besigned by the appropriate manufacturer representative(s) and shallinclude appropriate charts, graphs, and data recorded duringtesting.

II.1.4. No Meter Products and metering equipment shall be installedbefore all tests, as outlined in this section, are conducted.

II.1.5. Meter Products selected for certification testing must berepresentative of production run Meter Products and must meetall applicable FCC regulations.


6

II.1.6. The following tests shall be conducted in sequence using the sameMeter Products selected as specified in II.1.5 above: Insulation,Voltage Interruptions, Effect of High Voltage Line Surges, Effect ofFast Transient/Burst, Effect of Electrostatic Discharge (ESD), Effectof Operating Temperature, Effect of Relative Humidity, and ANSIC37.90.1 (Surge Withstand). Other tests required by ANSI C12.1and C12.20 may be done either in parallel or in sequence with thesame Meter Products or a separate group of Meter Products;however, with the understanding that the same Meter Productsmust be used for all test procedures within each ANSI-numberedor FCC-numbered test.

II.1.7. All test Meter Products shall be kept as a certification proof for oneyear after the conclusion of the testing. These test Meter Productsshall be made available during this period to any purchaser forinspection, if requested.

II.1.8. Meter Products which fail during the test shall not be repaired ortested further, but can be analyzed to identify the cause of failure.

II.1.9. When the test Meter Products fail to meet these testingrequirements and after any correction is made on the new testMeter Products, all tests shall be re-started with the new test MeterProducts.

II.1.10. If requested by the purchaser, the manufacturer shall notify thepurchaser of the certification test schedule for purchaser’s testwitnessing.

II.1.11. If more than a minimum number of Meter Products arecertification tested, the test results shall be based on and reportedfor all Meter Products tested.


7

II.2. Meter Product Failure DefinitionA Meter Product shall be designated as failed if any of the followingevents occur during or after any certification test:

II.2.1. Failure of the Meter Product to perform all functions as specified ina test procedure.

II.2.2. Failure of the Meter Product to meet the fundamental technicalperformance specifications as specified by the manufacturer. Thefundamental performance must include safety, accuracy andreliability of the Meter Product, and any other functions includedin the Meter Product.

II.2.3. Signs of physical damage as a result of a test procedure.

II.2.4. The occurrence of a loss of data or other unacceptable mode ofoperation for the Meter Product as a consequence of a testprocedure.

II.2.5. Failures of either hardware, firmware or software, or acombination thereof.


8

II.3. Meter Type Certification Rejection CriteriaThe meter type certification will be rejected if any of the following eventsoccur:

II.3.1. The Meter Products fail the certification tests as specified in TableII.3.1-a below:

Table II.3.1-a: Table of failures based on Meter Products tested

# MeterProducts

Failures in different tests individually

Tested 0 1 2 3 or more

3

4 F A I L

5

6

7

P A S S

8

9 or more

Examples: The following examples explain how to apply TableII.3.1-a. Also, reference to “the series tests“ in this paragraphmeans tests required to be performed in the series manner asspecified in Section II.1.6., and reference to “the parallel tests“means testing is not required to be performed in any particularsequence (either series or parallel).

Example 1: If 3 Meter Products are selected for the series testingand one failure occurs in any test procedure, the meter typecertification will be rejected and the entire eight series tests will bestarted over from the beginning.

Example 2: If 9 Meter Products are selected for the series tests andthe first, second, and third failures occur separately in threedifferent tests or test procedures, the meter type certification willbe rejected. These failures described here mean that a failure of thefirst Meter Product during one test procedure, a failure of a secondMeter Product during another test procedure, and a failure of athird Meter Product during another test procedure different fromthe tests that the first two Meter Products have failed previously.Once such failures occur, the entire eight series tests will be startedover from the beginning.


9

However, if 3 Meter Products are selected for a parallel testperformed concurrently with the 9 Meter Products selected for theseries tests, the rejection criteria for the 3 Meter Products tested in aparallel test shall not apply to the 9 Meter Products tested in series,or vice versa. In addition, if a group of Meter Products tested in aparallel test(s) fails according to the rejection criteria, only theparticular failed test(s) needs to be repeated.

II.3.2. The failure of two or more Meter Products during the same testprocedure.

II.4. Test Setup

II.4.1. The Meter Product shall be connected to its normal operatingsupply voltage with a fully charged power failure backup systemand shall be energized throughout the duration of the testprocedures, unless otherwise stated.

II.4.2. Before testing commences, if necessary, the Meter Product shall beenergized for a reasonable period at room temperature for stressrelief.


10

II.5. ANSI C12.1 Tests All Meter Product certifications shall be performed in accordance with thecertification tests described in ANSI C12.1 (NEMA, 1995), unless notedotherwise below.

* Tests 1 through 29: no clarifications required.

* Test 30: meter covers removed during test, temperaturelimits are defined for operations underCalifornia weather conditions as+85° C = T oper-max,-20° C = T stor-min

* Test 31 through 38: no clarifications required

* Additional test A1: sunlight interference test is needed for opticalpick-up type retrofit modules (not withinscope of existing ANSI C12.1-1995 tests) and isfurther defined below.

* Additional test A2: ANSI C37.90.1 Surge Withstand Testing

The same set of selected Meter Products, as defined by unique meternumbers, will be tested with the following tests performed in series: 15,16, 17, 25, 28, 30, 31 and A2. Other tests required by ANSI C12.1 may bedone either in parallel or in sequence with the same Meter Products or aseparate group of Meter Products; however, with the understanding,however, that the same Meter Products must be used for all testprocedures within each ANSI-numbered or FCC-numbered test.

These ANSI C12.1 tests are listed and described in Table I-2 above.


11

II.6. ANSI C12.20 TestsAll Meter Product certifications shall be performed in accordance with thecertification tests described in ANSI C12.20 (NEMA, 1998) for 0.2% and0.5% accuracy class meters, unless noted otherwise below.

* Tests 1 through 26: no clarifications required.

* Test 27: meter covers removed during test, temperaturelimits are defined for operations underCalifornia weather conditions as+85° C = T oper-max,-20° C = t stor-min

* Test 28 through 35: no clarifications required

* Additional test A1: sunlight interference test is needed for opticalpick-up type retrofit modules (not withinscope of existing ANSI C12.20, NEMA-1998tests) and is further defined below.

* Additional test A2: ANSI C37.90.1 Surge Withstand Testing

The same set of selected Meter Products, as defined by unique meternumbers, will be tested with the following tests performed in series: 12,13, 14, 22, 25, 27, 28 and A2. Other tests required by ANSI C12.20 may bedone either in parallel or in sequence with the same Meter Products or aseparate group of Meter Products; however, with the understanding thatthe same Meter Products must be used for all test procedures within eachANSI-numbered or FCC-numbered test.

These ANSI C12.20 tests are listed and described in Table I-2 above.


12

II.7. Test A1 - Sunlight Interference Test

A. This test verifies the Meter Product accuracy and full functionaloperations under direct sun light.

B. The meter cover shall be removed during this test.C. The Meter Product shall be exposed to both the incandescent light

source (Lab Test) and sunlight (Outdoor Sunlight Test).

Lab Test:

D. The incandescent light source, Smith Vector #710 or equivalent,shall be used to simulate the sunlight. The incandescent light shallbe 600 watt and 3,200° K blackbody radiation as a minimum.

E. The Meter Product shall be exposed to the incandescent lightsource for a minimum of five minutes for each position of theincandescent light source.

F. The incandescent light source shall be pointed directly toward theMeter Product and positioned at a maximum direct distance of 19inches from the center of the meter rotor shaft as follows:1. Twelve positions around the meter base.2. Eight positions at a 45° angle from the meter base.3. One position at a perpendicular to the face of the meter.

G. Verify the Meter Product operations and report the direct andremote meter reads before and after each incandescent lightexposure.

Outdoor Sunlight Test:

H. The sunlight conditions shall be outdoors, clear sky, bright sunnyday, and no shades over the Meter Product.

I. The Meter Product shall be exposed to sunlight conditions for 24hours.

J. The Meter Product shall be set in a position as normally installedthe field. All Meter Products under test shall be exposed to thesunlight conditions at the same time and evenly face differentdirections starting with one Meter Product facing towards thesunrise direction.

K. Record and compare direct and remote meter reads at every hourunder the sunlight conditions.

L. To pass this test the Meter Product shall operate as specified withno observed anomalies and have an accuracy of ±0.3% on bothdirect and remote meter reads.


13

III. REGISTRATION AND CENTRALIZED DATABASE FOR DIRECT ACCESSCOMPLIANT METER TYPE

Manufacturers must state that their meters meet the Certification TestingRequirements in Section II of Chapter A. The self-certification document shall besubmitted to the Energy Division of the CPUC. During the review process, thestaff of the Energy Division shall be entitled to obtain from the meter productmanufacturer all backup documentation that is related to the certification testingrequirements for the meter that the manufacturer is certifying. If the EnergyDivision determines that the self-certification document for a particular meter isin order, the Energy Division shall post the model number of the meter type,along with the name of the meter product manufacturer, on the Commission’sweb site. The document shall be verified by an officer or authorized employeeof the manufacturer with the following statement:

DECLARATION

I, (print name and title) ___________________ hereby certify that I amempowered to act on behalf of __________________ (manufacturer’sname) and to submit this self-certification document on its behalf. Ideclare under penalty of perjury under the laws of the State ofCalifornia that the above statements are true and correct, and that ifany documents are furnished in connection with this self-certification document, that those documents are true and correctcopies.Dated ___________________, at ____________________________. (date) (place of execution)

Signature: _______________________________

If the verification is made outside of California, the verification must be made byan affidavit sworn or affirmed before a notary public.


14

IV. REQUIREMENTS FOR STICKERS, SEALING AND LOCKING HARDWAREBelow are the requirements for the stickers, sealing and locking hardware usedin Direct Access. Requirements for application of these devices are covered inChapter D: Standards for Meter Installation, Maintenance, Testing, andCalibration, Section II.1.

IV.1. Sealing and locking hardwareSealing and locking hardware shall be imprinted with company nameand/or logo and be made with material other than lead. Sealinghardware owned by the MSPs shall be orange in color and be imprintedwith its certification number.

IV.2. Life-support sealing hardwareLife-support sealing hardware used for identifying a customer premisewhich has a life support system shall be white in color and imprintedwith a red caduceus (medical symbol).

IV.3. Life-support stickerA life-support sticker used for identifying a customer premise which hasa life support system shall be imprinted with a caduceus (medicalsymbol).

IV.4. 480 V stickerA 480 V sticker used to identify a 480 Volt service panel and meter shallbe legible.

V. REQUIREMENTS FOR LABELING MANUFACTURING DATE ON NEWMETER PRODUCTSNew Meter Products shall be permanently labeled with a manufacturing date.


15

VI. REQUIREMENTS FOR REBUILT, RETROFIT AND REPAIRED METERPRODUCTS

VI.1. Rebuilt Meter ProductA Meter Product shall not be rebuilt and repackaged for resale by anyentity except its original manufacturer or a manufacturer-authorizedlicensed agent. Once a Meter Product is rebuilt and repackaged, it shallbe tested for accuracy, labeled as rebuilt and by whom, and datedaccordingly.

VI.2. Retrofit Meter ProductA Meter Product may be retrofitted with other devices or modules.Retrofitted Meter Products shall be tested in accordance with the aboveSection II.1.2 of the Certification Testing Requirements. Prior to use,retrofitted Meter Products shall be tested for accuracy, labeled asretrofitted and by whom, and dated accordingly.

VI.3. Repaired Meter ProductA Meter Product shall not be repaired for resale by any entity except itsoriginal manufacturer or a manufacturer-authorized licensed agent. Oncea Meter Product is repaired, it shall be labeled as repaired and by whom,and dated accordingly.

VI.4. Interval Data Meter ProductThe meter or the meter data system must be capable of providing andstoring required interval data for a minimum of 35 days.


CHAPTER B

STANDARDS FOR METER COMMUNICATIONS

Standards for Meter Communications

18

I Meter Communication Standards

1. KYZ Contact OutputMeter products are not required to have a contact output. If a meter product hasa contact output, it should be KYZ pulses per ANSI C12.1-1995 Standard (Codefor Electricity Metering).

2. Meter Password AuthorizationThere are three types of password authorization:1. Full read/write2. Billing read/write (other routine maintenance exclusive of programmingrevenue quantities)3. Read only

Because ESPs are responsible for safety, accuracy, and reliability of meters usedin Direct Access, they shall have the authority to issue meter passwords at theirdiscretion, but must issue read passwords to UDCs for audit purposes uponrequest. ESPs will provide meter passwords in a timely manner for UDCs toperform their scheduled functions.

3. Consumer Protection on KYZ Contact OutputFor Direct Access customers who currently have their energy managementsystems utilizing KYZ outputs, they shall be notified by their ESP if a new meterused for Direct Access will not be compatible with their energy managementsystems.

4. Visual Meter ReadAll Direct Access meters shall have a visual kWh display or a physical interfaceto enable on-site interrogation of all stored meter data. There are two reasons forrequiring a visual meter display: (1) For consumer protection: The consumer canverify that the meter read matches the bill, and (2) For on-site interrogation whenother meter communications systems fail: this would enable entities who areresponsible for billing/settlements to obtain the meter read when investigatingthe communications failure. For electromechanical meters, the dials aresufficient for this on-site interrogation. At a minimum, electronic meters musthave a physical interface to enable retrieval of all stored meter data.

If the meter has only a physical interface, the customer must be provided withthe means to be able to retrieve the stored meter data in a way that can beunderstood. In addition, appropriate passwords would need to be issued toallow authorized customers, UDCs, and MDMAs to retrieve the stored meterdata.

5. Optical Port StandardIf a Type 2 optical port is used, the port must meet ANSI C12.18.

CHAPTER C

STANDARDS FOR METER DATA MANAGEMENT

AND METER READING

Standards for Meter Data Management and Meter Reading

20

I. DEFINITION OF MDMA BUSINESS FUNCTIONS1

The role of the MDMAs are to perform the following functions:• Manage the meter reading schedule• Read and retrieve meter data• Validate, edit and estimate meter data• Calculate usage• Format data• Store data on the MDMA server• Manage data on the MDMA server• Manage data access to the MDMA server.• Meter/device management (i.e., when the meter/device was installed, what

the device type is, what the service history has been, what the serviceparameters of the meter are, etc.)

The obligations of the MDMA are (Section H.(7) of Appendix A of D.97-10-087):"MDMA services will be performed in accordance with CPUC regulations andwill be the responsibility of the party so indicated in the customer's DASR[direct access service request]. MDMA obligations include but are not limited tothe following:

(a) Meter data for DA [direct access] Customers shall be read, validated,edited, and transferred pursuant to Commission-approved standards.

(b) Regardless of whether ESP or UDC perform MDMA services both UDCand ESP shall have access to the MDMA server.

(c) The MDMA shall provide Scheduling Coordinators (or their designatedagents) reasonable and timely access to Meter Data as required to allowthe proper performance of billing, settlement, scheduling, forecasting andother functions.

(d) The MDMA is required to keep the most recent 12 months of Customerconsumption data for each DA Customer. Such data must be retained fora period of 36 months. Such data must be released on request to thecustomer or, if authorized by the customer, to any ESP or to the UDC.

(e) Within five days after installation of the meter, the MDMA must confirmthat the meter and meter reading system is working properly and that thebilling data gathered is valid.

(f) Either no more than 10 percent of the accounts will contain estimateddata, or no more than 1 percent of all the data (e.g., the 720 hourly readsper month times the number of meters) will be estimated."2

In addition, to the above, the Commission also adopted the tariff provision that"The MDMA shall read interval meters on the utility’s scheduled meter readingdate, or on such other date as may be mutually determined by the MDMA andthe utility."

1 Definition of MDMA business functions is defined in CPUC D.97-12-048, pages 28-29.2 CPUC D.97-10-087 established limits on estimated data which was later modified by D.97-12-048, page 39.


21

II. MDMA QUALIFICATION TESTING3

Per CPUC D.97-12-048 UDCs and the ESPs are to adhere to the following beforethey or any of their subcontractors are permitted to offer any MDMA services:

(1) The existing regulated utilities who perform their own meter readingand meter data management shall be allowed to perform MDMA services for theUDCs, as well as for the ESPs. All utility employees who have successfullycompleted the utility’s training programs regarding meter reading, and relatedsafety programs, shall be permitted to carry out the meter reading activitiesrequired of an MDMA. All utility employees who have successfully completedthe utility’s training programs regarding meter data management (validation,editing, etc.) and entry shall be permitted to carry out the meter datamanagement activities required of an MDMA.

(2) All non-utility entities and ESPs seeking to offer MDMA services shallbe required to submit a written request to each UDC in whose service territorythe ESP or entity seeks to offer such services. The written request shall includethe following information: name of the person or entity; business address andtelephone number; a description of the requesting party’s experience in meterreading and meter data management; and a description of what educational andtraining requirements in meter reading, meter data management, and relatedelectrical safety the MDMA will require of its employees before they are allowedto carry out the MDMA functions. The UDCs shall require the potential MDMAsto attach all pertinent training manuals and materials which describe thetraining in meter reading, safety, and meter data management that all of itsemployees have received or will undergo before the employee is allowed toperform MDMA-related activities. The request shall be verified.

(3) Upon receipt of the request, the UDC shall be required to review thewritten description and any attached materials, and to confirm in writing withthe potential MDMA whether the proposed educational and trainingrequirements are comparable with the UDC’s requirements. If the UDC statesthat the proposed MDMA’s educational and training requirements are notcomparable, the person or entity may file a formal complaint with theCommission with regard to such qualifications. If the UDC states that theproposed MDMA’s educational and training requirements are sufficient, thenthe MDMA may begin offering MDMA services so long as it meets all theMDMA-related requirements. The MSP is also free to require the MDMA tomeet other requirements that are reasonably related to the MDMA’s activities.

III. METER READING FREQUENCYTo ensure that the meter data is recorded, we will require the MDMAs to readthe meters at least once a month. Such a requirement is also consistent with thetiming of how often bills are to be rendered.

3 The MDMA qualification process is defined in CPUC D.97-12-048, pages 40-41.


22

IV. MDMA SAFETY REQUIREMENTSWe will require in the direct access tariffs that all MDMAs comply with thepertinent electrical safety provisions of Cal OSHA and the UDC’s safetyrequirements as they apply to the reading of electric meters. Prior to allowing anESP, in its role as the MDMA, or a third-party MDMA, to perform meterreading, we will require the UDCs, as discussed below, to review the safetytraining and procedures that the MDMA and its employees are to follow.With regard to the recommendation that the MDMA report meter, safety, andhazardous conditions, and that site-specific information be kept, thosesafeguards are already contained in the direct access tariffs in Sections H(3) andH(8)(e).

V. MDMA TECHNICAL/BUSINESS SUPPORT TO ESPs AND UDCsThe MDMA will provide access to technical and business assistance duringnormal business hours (8am to 5pm Pacific time, Monday through Friday, exceptholidays). At such times, staff will be available to address question and concernson data availability, corruption and adjustments, and systems technical support.

In addition, the MDMA will provide access to a support pager available 24hours a day/365 days a year to address issues of server availability. The MDMAshall respond and provide a status to all pages within 2 hours.

MDMA server availability or access issues will be dealt with as soon asreasonably possible. At the MDMA's discretion, concerns over data availability,data corruption and adjustments or non-urgent problems will be addressedduring the next business day.

VI. MDMA PERFORMANCE STANDARDSThe following MDMA performance standards shall be applied:

• The first billing cycle shall be disregarded in performance standards

• Separate estimation of ISO data to server shall be done according to VEErules in Attachment VEE (Section A for interval data and Section B formonthly data).


23

Timeliness For Validated Meter Reading Data4

The following standards shall be used to establish the time requirements forposting validated meter reading data on the MDMA server.(a) Interval Meters:

(i) 80% of all usage data must be available on the first day after thescheduled reading date of the meter.

(ii) 90% of all usage data must be available within two days of thescheduled reading date of the meter.

(iii) 99% of all usage data must be available within five days of thescheduled reading date of the meter.

(b) Non-Interval (Monthly) Data:(i) 85% of all monthly meter readings must be available by 6:00 a.m.

on the 1st working day after the scheduled meter reading date.(ii) 95% must be available by 6:00 a.m. on the 2nd working day after

the scheduled meter reading date.(iii) 99% must be available by 6:00 a.m. on the 5th working day after

the scheduled meter reading date.

VII. MDMA PERFORMANCE EXEMPTIONSIn the event of a large catastrophe (i.e., hurricanes, earthquakes, etc.) thatprevents the MDMA from reading meters, the MDMA shall estimate and postthe data. This estimated data shall be reported separately by the MDMA in theirperformance report, and not be included in any performance penalties assessedagainst the MDMA.In the event of meter failure where the meter is not accurately recording usage,the estimated data should be reported separately by the MDMA in itsperformance report, and would not be included in any performance penaltiesassessed against the MDMA, so long as the following conditions are met: (1) amanual reading has verified that the meter has failed and there is no problemwith the remote reading technology; and (2) the exemption cannot occur for anaccount more than once in a 12 month period.

4 The timeliness standards as defined in CPUC D.97-12-048, pages 31-32 and modified in D.98-12-080, pages 82-83.


24

VIII. EDI IMPLEMENTATIONThe CPUC refrained from adopting permanent EDI standards until after thereport and comments have been filed. The CPUC has directed the UDCs, ESPs,and MDMAs, to move toward using EDI to transfer meter usage information inaccordance with the schedule described below:5

We will adopt the recommendation which calls for allinterested parties to work together to create a statewideimplementation guide for the use of EDI. We will direct theEnergy Division to ensure that this result is achievedthrough the Direct Access Tariff Review Committeeestablished in D.97-10-087. We would like to implementthe use of EDI on a trial basis no later than September 1,1999, with the goal of having EDI as the only standard fortransferring meter usage data no later than February 1, 2000.With this in mind, the Direct Access Tariff ReviewCommittee needs to develop a proposed statewideimplementation guide, and to file it no later than April 2,1999. Comments to the report should also be permitted, andshould be filed within 21 days of the report’s filing. ACommission decision would then issue in June or July of1999 to address the EDI standards and implementationguidelines. Under such a schedule, market participants cangear up to move toward an EDI format, with the expectationthat a trial period will take place in the last four months of1999 and in January of 2000, and that the MEP data formatwill be discontinued on February 1, 2000.

IX. VALIDATING, EDITING, AND ESTIMATION

The proposed implementation plan for changes to the interim interval rules isdescribed in Table IX-1 below. The Optional/Required column indicates ifmarket participants will be required to make this change. The Earliest dateacceptable column indicates the earliest a market participant is allowed toimplement this change (note to UDCs - this means the VEE test would need toallow these options), and the Required by column indicates the date by whichmarket participants must implement the option (only applies to requiredoptions). MDMAs that were accepted prior to the required date must comply,but do not need to go through the acceptance process again.During the discussion, it was noted that some of the optional changes have abigger impact on some technologies than others.

5 CPUC D.98-12-080, page 86.


25

Table IX-1: Implementation Plan for the Major Changes in Interval Data RulesModification Optional/Required? Earliest date

acceptableRequired by

Spike check threshold Optional 12-17-98 n/akVARh checkthreshold

Optional 12-17-98 n/a

Use of partial days forestimation

Optional (may make abigger difference with sometechnologies than others)

12-17-98 n/a

Don’t use dayscontaining powerfailure as source forestimation

Required 12-17-98 90 days afterCommissiondecision

Allow use of accuratemeter readings scaleestimated data

Optional (may apply moreto some technologies thanothers)

12-17-98 n/a

Simplified prorationalgorithm when meterclock is off


Automated handling ofirregular usage


Handling of test modeintervals


Clarification ofselection of referencedays


High/low usage check Required 12-17-98 90 days afterCommissiondecision

kVARh check Optional 12-17-98 na

The Attachment on Validating, Editing, and Estimation (VEE) provides detailedrequirements for energy usage data VEE rules for both monthly and intervalcustomers as well as the ballot result on these VEE rules.

This Page Is Intentionally Left Blank

CHAPTER D

STANDARDS FOR METER INSTALLATION, MAINTENANCE,TESTING, AND CALIBRATION

Standards for Meter Installation, Maintenance, Testing, and Calibration

28

I. MSP METER WORKER QUALIFICATIONS

All meter workers performing meter services for Direct Access must meet therequirements as outlined in this Chapter and exercise due care for the tasks performed.Out-of-socket meter accuracy testing, meter diagnostics, and sub-metering work6

requirements are not addressed in these standards. Also, utilization of the meter’sbuilt-in diagnostics is not considered part of meter accuracy testing.

6 Sub-metering work performed in California is presently covered by California Business and Professions CodeSections #12531 (Meaning and Scope of Terms) and 12532 (Engaging in Business as Device Repairman orMaintaining Device Repair Service Unlawful without Registration).


29

I.1. Meter Worker Skill, Safety, and Qualification RequirementsThe following are the descriptions and requirements of the five meter workerclasses:

I.1.1. CLASS 1I.1.1.1. Metering Types and Voltages

This class includes single-phase, socket-based meters, 300 Vphase-to-phase maximum and does not include transformer ratedmeters. Communication wiring must be outside of energizedmeter panels.

I.1.1.2. Work to be PerformedClass 1 Meter Workers can install, remove, and replacesingle-phase, 120/240 V or 120/208 V, self-contained meters instandard socket based residential-type metering equipment.Connections of communication conductors must be outside theenergized meter panels.

I.1.1.3. Essential Technical SkillsI.1.1.3.1. Understanding of single phase electrical metering.I.1.1.3.2. Understanding of electric distribution safety procedures.I.1.1.3.3. Ability to identify energy diversion or tampering related

to this class of meter work.I.1.1.3.4. Ability to install and remove damaged and undamaged

meters.I.1.1.3.5. Understanding of the meter panel and socket layout for

the metering conditions related to this class of meterwork.

I.1.1.3.6. Ability to read meters used in this class.I.1.1.3.7. Ability to properly use tools appropriate to the work in

this class.I.1.1.3.8. Ability to connect meter communications external to the

meter panel.I.1.1.3.9. Ability to initialize meter communication modules - not

utilizing Type 2 optical ports and meter configurationsoftware.

I.1.1.4. Worker Safety and Safety EquipmentI.1.1.4.1. Job performance in accordance with employing MSP's

procedures and safety rules.I.1.1.4.2. Knowledge of hazards of electricity and ability to

perform work to avoid electrical hazards.I.1.1.4.3. Ability to comply with CAL OSHA requirements.I.1.1.4.4. On-site use of personal protective equipment.


30


This class includes all meters and skills required for Class 1.Class 2 includes up to 600 V, single-phase and poly-phase, safetysocket, and standard socket based meters, and does not includetransformer rated meters. Communication wiring may be behindthe panel, and work can be in and around energized circuits.

I.1.2.2. Work to be PerformedClass 2 Meter Workers can install, remove, and replace all metersconsistent with the above. Class 2 Meter Workers mustunderstand the operating characteristics of test-bypass facilitiesand test blocks, and may operate test-bypass facilities, but maynot install, alter, maintain, or replace wiring between the meterand test block. On panels without test-bypass facilities, single-phase and poly-phase meters will not be removed or installedwithout first disconnecting the customer load or if deemedunsafe to do so under load. Communication wiring may beinstalled inside the panel, and work can be performed in andaround energized circuits.

I.1.2.3. Essential Technical SkillsI.1.2.3.1. Cumulative including all skills for Class 1.I.1.2.3.2. Additionally, possess skills related to meter voltages and

meter forms used in Class 2.I.1.2.3.3. Ability to perform phase rotation assessments.I.1.2.3.4. Ability to operate test-bypass facilities or test blocks in a

self-contained safety socket.I.1.2.3.5. Ability to perform work required to route

communication wiring to accommodate metercommunications.

I.1.2.4. Worker Safety and Safety EquipmentI.1.2.4.1. Cumulative including all skills and safety knowledge for

Class 1.I.1.2.4.2. Electrical safety knowledge and work skills appropriate

for single-phase and three-phase metering up to 600 Vphase-to-phase, including the ability to identify andrefer to a Class 5 meter installer services above 600 Vphase-to-phase prior to performing work in the serviceequipment.


31


This class includes all meter types in Classes 1 and 2. Class 3work includes up to 600 V, A base, K base, and transformer ratedmeters with internal diagnostics. Communication wiring may bebehind the panel, and work can be in and around energizedcircuits.

I.1.3.2. Work to be PerformedIn addition to Class 1 and 2 Meter Work, Class 3 Meter Workerscan install, remove, and replace all meters consistent with theabove including transformer-rated meters with internaldiagnostics. Class 3 Meter Workers may operate test switches,but may not install, alter, maintain, or replace wiring between themeter, test switch, test block, and associated equipment.

I.1.3.3. Essential Technical SkillsI.1.3.3.1. Cumulative of Classes 1 and 2.I.1.3.3.2. Additionally, possess skills related to meter voltages and

meter forms used in Class 3.I.1.3.3.3. Ability to understand the operating characteristics of

metering transformers and how to operate test switches.I.1.3.3.4. Ability to understand, interpret, identify, and take

appropriate actions based upon built-in diagnostics ofsolid state meters.

I.1.3.4. Worker Safety and Safety EquipmentI.1.3.4.1. Cumulative of all safety skills for Classes 1 and 2.I.1.3.4.2. Ability to understand, interpret, and take appropriate

action based on built in diagnostics of solid state meters.I.1.3.4.3. Ability to work with transformer rated meters and

operate test switches and test blocks.


32

I.1.4. CLASS 4(A)I.1.4.1. Metering Types and Voltages

This class includes all meter types in Classes 1, 2, and 3.I.1.4.2. Work to be Performed

In addition to Class 1, 2, and 3 Meter Work, Class 4(A) MeterWorkers can install, remove, and replace all meters consistentwith the above. May perform in-field meter accuracy tests,calibrations, and perform all types of meter maintenance andtroubleshooting on single phase meters up to 300 V. Programsand verifies internal programs and software in solid state meters.

I.1.4.3. Essential Technical SkillsI.1.4.3.1. Cumulative of Classes 1, 2, and 3.I.1.4.3.2. Ability to perform work on metering switchboards.I.1.4.3.3. Ability to perform calibration, repair, retrofit,

troubleshooting, data collection of electric meters withinthe Class, and install, maintain, and program advancedmetering technologies, including TOU, interval data,real time pricing, remote meter communication, andload control devices.

I.1.4.4. Worker Safety and Safety EquipmentI.1.4.4.1. Cumulative of all safety skills for Classes 1, 2, and 3.I.1.4.4.2. Ability to conform processes to additional electricity

hazards and complexities associated with meteringswitchboards, testing meters, and maintaining meters inservice equipment up to 300 V.


33

I.1.5. CLASS 4(B)I.1.5.1. Metering Types and Voltages

This class includes all meter types in Classes 1, 2, 3, and Class4(A). Class 4(B) work includes all metering up to 600 V,including transformer rated meters with primary and secondaryvoltages less than 600 V. Communication wiring may be behindthe panel, and work can be in and around energized circuits.

I.1.5.2. Work to be PerformedIn addition to Class 1, 2, 3, and 4(A) Meter Work, Class 4(B)Meter Workers can install, remove, and replace all metersconsistent with the above. May perform in-field meter accuracytests, calibrations, and perform all types of meter maintenanceand troubleshooting on all meters. Programs and verifiesinternal programs and software in solid state meters.

I.1.5.3. Essential Technical SkillsI.1.5.3.1. Cumulative of Classes 1, 2, 3, and 4(A).I.1.5.3.2. Ability to perform work on metering switchboards.I.1.5.3.3. Ability to perform calibration, repair, retrofit,

troubleshooting, data collection of electric meters, andinstall, maintain, and program advanced meteringtechnologies, including TOU, interval data, real timepricing, remote meter communication, and load controldevices.

I.1.5.4. Worker Safety and Safety EquipmentI.1.5.4.1. Cumulative of all safety skills for Classes 1, 2, 3, and

4(A).I.1.5.4.2. Ability to conform processes to additional electricity

hazards and complexities associated with meteringswitchboards, testing meters, and maintaining meters inservice equipment up to 600 V.


34


This class includes all meter types in Classes 1, 2, 3, 4(A), and4(B). Class 5 meter work includes all metering above 600 V,including metering transformers, associated devices such asisolation relays and switches, and wiring between thesetransformers, associated devices, and meters. Communicationwiring may be behind the panel, and work can be in and aroundenergized circuits.

I.1.6.2. Work to be PerformedIn addition to Class 1, 2, 3, 4(A), and 4(B) meter work, Class 5Meter Workers can install, remove, and replace all metersconsistent with the above including transformer-rated meters.Replace high-voltage fuses. Perform in-field meter accuracy tests,calibrations, and perform all types of meter maintenance andtroubleshooting on all meters. Inspect wiring and instrumenttransformer ratios utilizing various apparatus as necessary.

I.1.6.3. Essential Technical SkillsI.1.6.3.1. Cumulative of Classes 1, 2, 3,4(A), and 4(B).I.1.6.3.2. Ability to identify primary metering equipment and

characteristics of service equipment rated at voltagesabove 600 V.

I.1.6.3.3. Broad knowledge and familiarity of electricaldistribution systems above 600 V and operatingcharacteristics.

I.1.6.3.4. Ability to identify UDC service voltages and electricalservice requirements.

I.1.6.3.5. Broad knowledge and familiarity of State of CaliforniaGeneral Orders.

I.1.6.4. Worker Safety and Safety EquipmentI.1.6.4.1. Cumulative of all safety skills for Classes 1, 2, 3, 4(A),

and 4(B).I.1.6.4.2. Ability to conform processes to additional electricity

hazards and complexities associated with meteringswitchboards, testing meters, and maintaining meters inservice equipment above 600 V.

I.1.6.4.3. Meet the State of California requirements for a“Qualified Electrical Worker” to perform work onmetering systems with instrument transformer atvoltages above 600 V.

I.1.6.4.4. Must have a minimum of two years of combinedtraining and experience with high-voltage (above 600 V)equipment and circuits.


35

I.2. CERTIFICATION OF METER SERVICE PROVIDER (MSP)

The following excerpt from D.97-12-048, as modified by 98-05-044 and D.98-12-080 is the

MSP certification process that all UDCs, ESPs, and MSPs must adhere to:

(1) The existing regulated utilities who perform their own electric meter installation and

removal, and meter maintenance and repair, shall be given permanent MSP certification. All utility

employees who have successfully completed the utility’s training programs regarding meter

installation and removal, meter maintenance and repair, and related electrical safety programs,

shall be permitted to install, remove, maintain and repair Direct Access meters on behalf of the

UDC acting as an MSP.

(2) All non-utility MSPs shall be required to submit a written application to the

Commission requesting “MSP Certification.” The MSP Certification will be granted to persons or

entities who possess a general electrical contractor’s license issued by the Contractors’ State

License Board. If the ESP is acting as the MSP, then the contractor’s license shall be in the name

of the ESP.7 The ESP may also subcontract the meter services to a third party, in which case the

third party would be required to have a general electrical contractor’s license.

The non-utility MSP is required to have an electrical contractor’s license because the

installation, removal or repair of an electric meter by a person other than a public utility is subject

to the Contractors’ State License Law. Generally speaking, a contractor is anyone who adds

materials to, repairs, or subtracts materials from a structure or premises. (Bus. & Prof. Code

Section 7026.) A regulated public utility is exempt from the Contractors’ State License Law when

it performs work on its own property, or when the work is undertaken in furtherance of the

distribution of electricity. (Bus. & Prof. Code Section 7042.1.) Thus, anyone else installing,

repairing, or removing an electric meter would be required to have a contractor’s license as

dictated by the current statutory provisions in the Business and Professions Code. An electrical

contractor’s license is appropriate because of the electrical voltage that is present.

The written application shall include the following information: name of the person or

entity; business address and telephone number; the name of the person or entity in which the

general electrical contractor’s license is issued; the license number and expiration date; a

7 If the ESP is a partnership, corporation, or limited liability company, the ESP shall designate a “responsiblemanaging employee” to take the license examination on the ESP’s behalf. (See Bus. & Prof. Code Section 7065.) Aresponsible managing employee shall mean an individual who is a bona fide employee of the ESP, and who isactively engaged in electrical contracting work.


36

description of the applicant’s electric meter installation, maintenance, repair and removal

experience, as well as the applicant’s training and experience regarding electrical safety; and a

description of what educational and training requirements in electrical work and electrical safety

the MSP will require of its employees before they are allowed to install, maintain, repair or

remove electric meters or metering devices. A copy of the general electrical contractor’s license

shall be attached to the application. The application shall be verified, and if verified outside

California, the verification must be made by an affidavit sworn or affirmed before a notary public.

In addition to the written application, the MSP shall either arrange for a bond in favor of

the State of California in the amount of $100,000 or provide the Energy Division with proof that

the MSP has general liability insurance that meets the specifications described below.

If a bond is used, it shall be submitted with the written application. The bond shall be for

the benefit of anyone who may be damaged as a result of the MSP’s actions in connection with

the installation, maintenance, repair, or removal of the electric meter. Should a complaint for

damages arising from the MSP’s actions be filed in civil court, and a claim is made against the

bond, a copy of the complaint shall be served by registered or certified mail upon the

Commission’s Executive Director.

The bond requirement will ensure that the MSPs adhere to all applicable provisions

governing the installation and removal of electric meters. Should an end-use customer suffer

damages as a result of the MSP’s actions, the bond will provide a source of compensation.

If the MSP elects to provide proof of liability insurance, the insurance shall meet the

following specifications: (1) the insurance policy shall be commercial general liability insurance

with coverage that, at a minimum, is the same as what is provided for in the Insurance Services

Office Commercial Liability Coverage occurrence form; (2) the policy limit shall not be less than

$1 million for each occurrence for bodily injury, property damage and personal injury, and if the

coverage is subject to a general aggregate limit, the aggregate limit shall not be less than $2

million; (3) the policy shall include, as an additional insured, the UDC in whose service territory

the MSP is operating in and the ESP for whom the MSP is performing the meter-related work;

and (4) the liability insurance policy shall include a statement that thirty days’ written notice shall

be provided to the Commission, the ESP and the UDC before the policy is canceled. Proof of

liability insurance will ensure that the MSP has sufficient coverage to cover any claims that might

be brought against the MSP for metering-related activities.


37

The application shall be submitted to the following: CPUC, MSP Certification Unit, 505

Van Ness Avenue, San Francisco, CA 94102.

(3) Upon receipt of the application for MSP certification and the bond, the staff shall

review the documents for compliance with this process. If the documents are in order, the staff

shall issue a MSP certification number to the MSP. Upon the MSP’s receipt of the MSP

certification number, the MSP may offer meter installation related services to the ESPs or to the

UDCs. By providing such services, the MSP agrees to abide by all Commission decisions,

policies, and guidelines governing the installation, maintenance, repair and removal of electric

meters. Should it be determined that the MSP is not in compliance with such requirements, the

Commission may suspend the MSP certification.

(4) After receiving its MSP certification number, each MSP shall be required to submit a

detailed work schedule to each UDC for the first 20 installations by a new MSP in that UDC’s

service area. In order to ensure that the UDCs have adequate information to make an informed

assessment as to whether or not it should attend any of the 20 installations, the MSP is required

to submit a detailed work schedule to describe the meter type that is being removed and installed,

a description of all the procedures it will follow for removing and installing the meter, and what

safety precautions will be taken during these procedures. The UDCs shall reserve the right to

attend any of these installations.

(5) The application for MSP certification and requirements can be found at the CPUC

Energy Division’s website.

Should the UDC question the ability of an MSP to work on a particular meter type, the

burden will be on the ESP to prove to the UDC that the MSP that it is using is qualified to work

on that particular meter type. If the removal or installation involves skills in a higher meter class,

the ESP may need to prove to the UDC that the MSP it is using possesses the necessary skill and

experience for removing and installing a particular kind of meter.


38

II. METER INSTALLATION AND REMOVALThis section proposes the minimum requirements for installation, maintenance,and testing for meters and metering equipment used in Direct Access. CPUCD.98-12-080 reaffirms the statement in D.97-10-087, Appendix A: Direct AccessTariff, Section H(1)(b) that:“Potential and current transformers shall be considered part of the distributionsystem and shall remain the responsibility of the UDC.”

Accordingly, and pursuant to D.97-10-087 and D.98-12-080, all instrumenttransformers, test switches, and associated wiring up to the meter socket shallremain the responsibility of the UDCs. However, per D.98-12-080, reconnectingexisting wires to a replacement of an existing meter socket, A-base socketadapter, or A-base meter may be performed by either UDCs or MSPs.

II.1. Safety (customer life support, public, etc.)For safety reasons, the following requirements must be completed prior toperforming any meter work on site and visual observations mustcontinually apply as the meter work progresses. Meter work includes,but is not limited to, meter installation, meter replacement, maintenance,programming, and testing. Visually inspect all meter sites for thefollowing conditions:

II.1.1. Customer Life SupportII.1.1.1. If a customer premise has a life support device or

equipment and a UDC requires life-support seals orstickers, they shall be installed on the meter ring or coverrespectively, to prevent avoidable service interruptionduring the process of metering work.

II.1.1.2. If such life-support or sealing hardware is found oncustomer meter covers, meter panels, test switches and/ormetering transformer panel sections, meter workers shallbe cautious in performing meter work and notinterrupting electric service to the customer premise. Ifservice will be unavoidably interrupted for any meterwork, the meter worker shall notify the customer andobtain the customer’s consent prior to performing work.

II.1.2. Electrical HazardsII.1.2.1. Various Hazardous Conditions

The hazardous condition may include exposed wiring,damaged sockets (loose or burnt wiring or jaws), auto-bypass devices (load jaws still hot when meter removed),loose or missing screws (i.e., in bypass area), missingpanels, loose or broken service insulators, service wires inbad condition/order, missing meter, improper groundingconditions, fused neutral conductor of a 2 wire or 3 wire


39

single phase service, defective service switch/disconnect,new installations which fail to conform with the UDC’selectric service requirements, etc.

II.1.2.2. 600 Volt Auto Valve (aka Lightning Arrestor or SurgeProtector)II.1.2.2.1. Auto valves are also called meter protectors or

transformer surge arrestors and may be foundon 480 V overhead meter installations, both self-contained and transformer-rated. Functional inthe field, the auto valve may be safely removedfrom the meter installation. If MSP meterworkers find that these valves cannot be safelyremoved, they must notify UDCs to schedule forremovals.

II.1.2.2.2. The auto valve can be identified as a round orsquare metal can the size of a large human fist.The auto valve is mounted through a knock outand secured with a nut on the inside of the can.There will be 3 black wires and 1 white wire.The 3 black wires are electrically connected toeach of the 3 phase wires on the service. Thewhite wire is connected to an earth ground. Theauto valve is attached to the socket side of thetest block or test switches. By de-energizing thesocket, the wires from the auto valve can beremoved. Ensure that the hole left in the serviceequipment is properly closed and sealed forweather.

II.1.2.2.3. At MSP’s discretion, an MSP may remove anddiscard the auto valve.

II.1.2.3. 480 V StickerIf the meter does not display voltage, the Meter Workershall ensure that a 480 V sticker is in place either on themeter or on the meter panel near the meter before leavingthe site.

II.1.3. Physical HazardsII.1.3.1. Tripping hazards (slippery or uneven surfaces), debris, or

materials stored in the working space, overhead hazards(stored materials or workers above meters), etc.

II.1.3.2. Environmental hazards: Chemical, caustic, hearing,biological, etc.

II.1.3.3. Inadequate or unsafe access.II.1.3.4. Meter Mounting: loose meter mounting, undue vibration,

inability to securely seal metering, unleveled meter, etc .


40

II.1.4. Customer PremisesUnsafe conditions of customer-owned stairs, railings, platforms,etc.

II.1.5. VerminWatch out for insects, snakes, and rodents when opening doors orremoving panels.


41

II.2. Meter Security and Accessibility The following requirements must be completed when performing anymeter work on site and visual observations must continually apply as themeter work progresses. Visually inspect all meter sites for the followingconditions:

II.2.1. Infraction and Evidence of Tampering/Energy Diversion

II.2.1.1. Meter InstallationII.2.1.1.1. Awareness of significant connected loads

compared to a customer's monthly energyconsumption.

II.2.1.1.2. Irregularities in the service conductor'sinsulation ("skinned" insulation, newly-tapedsections of conductors, burned or pitted serviceconductors, etc.)

II.2.1.1.3. Unauthorized connection in overhead serviceentrance, on line side of meter or meteringtransformers, or in unsealed underground pullsections or pull cans.

II.2.1.1.4. Unauthorized seals, and unsealed or improperlysealed conditions on line side raceways, testblock compartments, test switch covers, andmeter sealing rings.

II.2.1.1.5. Suspicious wiring.II.2.1.1.6. Jumpers across current leads at bottom of test

switch unsealed.II.2.1.1.7. Meter coil jumpered - Check and correct if the

neutral service wire is found connected throughthe series coil of the meter and there are groundson the load circuits that act as a jumper.

II.2.1.1.8. Lock fuse jumpered - Check for lock fuse jumperand perform a regular meter test. Jumper willbe left as found, and this condition will bereported on the test report.

II.2.1.1.9. Blown (or loosened) fuse in one leg of a 3-wireold sequence service.

II.2.1.1.10. Rate infractions in meter program/register.

II.2.1.2. Meter CoverAn unusually clean cover, small holes in the cover, burnmarks on or near the cover, visible fingerprints inside thecover, etc.


42

II.2.1.3. KWH RegisterKWH dial pointer alignment, register mesh, and registergears.

II.2.1.4. Meter DiskDisk alignment to magnets, irregularity of disk rotation,foreign objects or materials on disk/bearing, scratches orwear marks on the disk, etc.

II.2.1.5. Test BlocksDamaged wiring between test block and meter, unusualmarks, scratches or burns on test blocks, defective testblocks, etc.

II.2.1.6. Meter BaseUnusual marks and mushroomed screw slots on potentiallinks; broken meter seals; unusual wear or scratches, burnor pit marks on meter stabs/blades/terminals, etc.

II.2.1.7. Meter SocketUnusual wear or scratches, burnt or pit marks, irregularmeter socket voltages, circuit bypass jumper, etc.

II.2.1.8. Hidden "Service Riser" Taps and Un-metered CircuitsUtilizing Relay DevicesUnusual noises that could indicate a relay opening whenthe meter is removed from the socket.

II.2.2. Meter SecurityThe following are the requirements for securing physical meters,panels, programmable meters, and data

II.2.2.1. Physical Meters and PanelsMeters and meter panels shall be secured with seals andlocking devices as described in Chapter A, Standards forMeter Products, Section IV.

II.2.2.2. Programmable Meters and DataOnly programmable meters capable of supportingpasswords or other protection means to restrict access toinformation contained therein shall be used for DirectAccess, and a security password shall be applied toprevent unauthorized access to a programmable meter andunauthorized modifications of the meter data andprogram. Passwords should be controlled by followinggenerally accepted practices.


43

II.3. Site VerificationVisually inspect all meter sites for the following conditions:

II.3.1. Direct Access Metering VerificationII.3.1.1. Check for proof of Direct Access meter certification or

meter types that meet the CPUC requirements for DirectAccess.

II.3.1.2. Verify for correct customer and customer account (recordsversus meters) including billing constant, meter number,accuracy test results, customer information, address, etc.

II.3.1.3. Verify for correct meter for service characteristics (voltage,form, etc.).

II.3.2. Direct Access Communication VerificationII.3.2.1. Verify for correct phone or other communication devices

and connections if remote readII.3.2.2. Verify for successful communication handshakes and data

retrieval if remote read (This can be done by anotherperson at a remote location, and data accuracy is notverified at this time).

II.3.2.3. Verify for successful meter reading if manual read.

II.3.3. Transformer-rated Meter SitesII.3.3.1. Billing Constant: Compare between records versus sites

(CT’s/PT’s ratios when accessible).II.3.3.2. Improper Wiring: Shorted current by-pass links, reverse

current secondary wired, unmatched voltage and currentcircuits, pinched or rubbed secondary wires near panelhinges, etc.

II.3.4. Pole-mounted Meter SitesII.3.4.1. Cutouts: Open cutouts or blown triple link fuses.II.3.4.2. Grounding: Grounding electrode conductors and

connections shall not be broken at any point between theservice equipment enclosure and the ground rods or otherapproved grounding electrode.

II.3.5. Pad-mounted Meter SitesII.3.5.1. Primary Metering

II.3.5.1.1. Verify meter enclosure attached to primarymetering cabinet secured and lockable.

II.3.5.1.2. Verify meter pedestal installation mountedsecurely and cabinet lockable.


44

II.3.5.2. Cabinets: Check the cabinet is free of obvious shippingdamage, paint damage, or corrosion.

II.3.5.3. Cabinet SafetyII.3.5.3.1. Verify that there is at least one penta-head

security bolt permanently attached to highvoltage compartment door(s) and door lockinghandle.

II.3.5.3.2. Verify secondary non-polarity neutral pointsand meter enclosure/pedestal electricallyconnected to ground bus.

II.3.5.3.3. Verify all metering electrical connectionssecured and properly made.

II.3.5.4. Cabinet MarkingsII.3.5.4.1. Verify exterior warning labels properly attachedII.3.5.4.2. Verify electrical schematic and metal nameplate

mounted to the enclosure

II.3.6. GroundingSome three-phase, three-wire services have a ground connection onone of the phases. In situations where A, B, or C phase is notgrounded and there is a voltage reading at or near zero volts onany phase and reading above zero volts on any other phases, thishazard shall be reported to the UDC prior to leaving the site foremergency corrections and the customer or customer’srepresentative shall be contacted to inform them of the hazardousconditions. If it is unclear whether a phase is intentionallygrounded, the local UDC shall be contacted to clarify theconfiguration of the service. Additionally, extreme caution mustbe taken on primary (above 600 V), 480 V phase-to-phase or 277 Vphase-to-ground service.

II.3.7. Watt Load ClockCalculate and record customer's load as needed.

II.3.8. Customer’s Ground Fault Protection DeviceCheck for the existence of a ground fault protection device orequipment in customer’s electrical panel or switchboard. Thepurpose of this verification is to ensure that meter workers, whileperforming meter work on a customer’s three phase, 4-wire wyemetering installation, do not inadvertently cause an unscheduledinterruption on this service.


45

II.3.8.1. BackgroundGround-fault protection shall be provided for solidlyground wye electrical services of more than 150 V toground, but not exceeding 600 V phase-to-phase for eachservice disconnecting means rated 1000 A or more. Theproper operation of some ground-fault detection systemsrequires a single neutral-ground point which can beestablished by connecting the neutral, the equipmentgrounding conductors, and the service equipmentenclosures. Even though this protection is mandatory onall new 1000 A 277/480 V services, some customers mayhave installed it on lower voltage and amperage of a 4-wire wye service. Therefore, meter workers must followthe procedures below to avoid unscheduled serviceinterruption while performing meter work on 4-wire wyemetering installations.

II.3.8.2. ProceduresII.3.8.2.1. Visual check

Prior to performing meter work, visually inspectfor the existence of a ground fault protectiondevice in the customer’s panel, switchboard, orelectrical facilities. Note: One of the mostsensitive ground fault protection devices is theGTE-Sylvania Zinsco Model GTS-3 relay. Asilicon control rectifier (SCR) in this relay couldbe triggered by transients generated simply byopening or closing the by-pass potential switch.

II.3.8.2.2. ActionsIf a ground fault protection device is found, themeter worker must contact the customer andrequest that the customer’s representative renderthe ground-fault relay inoperative or be presentto witness the meter work performed.

II.3.9. PT/CT Secondary WiringIf accessible, secondary wiring shall be verified for correct wiringand no sign of tampering, and PTs and CTs shall be checked fornormal operating conditions. If abnormal conditions exist, theMSP shall notify the UDC serving the customer within 24 hours fortheir corrective actions.


46

II.4. Meter InstallBelow are the required procedures that a meter worker must follow wheninstalling and/or removing meters. These procedures do not necessarilyinclude all procedures of a meter installation.

II.4.1. Pre-InstallationVerify type and size of metering is appropriate for site.

II.4.2. Self Contained MetersII.4.2.1. Take closing read on existing meter.II.4.2.2. For meters with no bypass, check the load to make sure it

is safe enough to remove the meter or disconnect loadbefore replacing the meter.

II.4.2.3. For meters with a bypass, the meter socket shall bebypassed and de-energized, before replacing the meter.

II.4.2.4. Verify voltage.II.4.2.5. Verify 0 volts between line and load side of test blocks.II.4.2.6. Install new meter.II.4.2.7. Take initial meter read.

II.4.3. Transformer Rated Meters (CT Meter)II.4.3.1. Take closing read on existing meter read.II.4.3.2. Open test switch cover and verify voltages at the test

switch.II.4.3.3. Operate test switch/test block to de-energize meter

sockets.II.4.3.4. Install new meter.II.4.3.5. Check voltage before closing voltage switches on test

switch.II.4.3.6. Take initial meter read.

II.4.4. Meter Installation at New SitesInstallation of a meter(s) on a newly constructed premise where thecustomer has decided, prior to service connection, to sign up forDirect Access service must be treated differently than a changemeter order. The parties (ESP, UDC, and MSPs) shall worktogether to achieve new meter sets and service connection in atimely manner, consistent with the customer’s interest and publicsafety.

A new meter set is defined as a meter socket having a meterinstalled for the first time. This does not apply to a socket or panelreplacement due to equipment failure.


47

There are four (4) key elements which must be completed prior to aUDC authorizing a new meter set:

1) Customer’s application on file with the UDC,2) Direct Access Service Request (DASR) accepted by the UDC,3) UDC’s receipt of authority having jurisdiction’s (AHJ)

inspection clearance, and4) Service completion, or coordination for energizing service.

The UDC will issue an authorization to set a new meter and notifythe ESP if coordination is required (i.e. for instrument transformerrated metering). If coordination is required, the MSP shall use bestefforts in meeting the UDC schedule. If no coordination isrequired, the MSP shall set the new meter within two (2) workingdays of UDC notification.

Should an ESP install a new meter without authorization from theUDC, the ESP’s Service Agreement with the UDC shall be subjectto termination pending a CPUC formal investigation.

II.4.5 Rewires and Service UpgradesIf a customer rewires or upgrades their electrical service andmetering equipment and if required, the MSP may remove themeter before the work on metering equipment starts.Reinstallation of the meter is subject to the same requirements ofSection II.4.4.

Should an ESP reset and reseal an existing meter in newly replacedmetering equipment without authorization from the UDC, theESP’s Service Agreement with the UDC shall be subject totermination pending a CPUC formal investigation.

II.4.6. Locking DeviceInsure appropriate locking device is in place.


48

III. METER MAINTENANCE AND TESTING SCHEDULEThe primary purpose of the maintenance and testing program is to assure thatthe meter population owned by an entity is accurate as long as the meters are inservice. The required maintenance and testing program is a balance between thebenefits and costs of maintaining and achieving high accuracy levels throughouta meter life. This maintenance and testing program enables the entity owningthe meters to verify the accuracy of the overall meter systems and to test themeters periodically and/or on the basis of an annual statistical sampling plan.

III.1. Maintenance ScheduleElectric meters used in Direct Access shall be maintained, as a minimum,according to the following meter maintenance and testing schedule:

Table III.1-1; Minimum meter maintenance and testing scheduleMaintenance and Testing

FrequencyCustomer Maintenance and Testing Criteria

One Year Interval Customer’s annual usage of 2 million kWh or higherTwo Year Interval Customer’s annual usage between 720,000 and 2 million

kWhAnnual Statistical SamplePlan

Non residential customer’s annual usage less than720,000 kWh

Residential Meters Either a formal sampling plan performed annually ortests done upon request and removal, where applicable

Direct Current (DC) Meters Either a formal sampling plan performed annually ortests done upon request and removal, where applicable

III.2. Maintenance upon requestTesting and maintenance is required upon a reasonable request by acustomer, ESP, or UDC. Prior consultation between the parties shalldetermine the entity that would perform the test and maintenance.

III.3. Statistical Sampling RequirementsThe ANSI Z1.4 or Z1.9 Standards shall be used as the statistical samplingrequirements for testing meters. Generally, inspection level of General II(G-II) shall be used if the use of other inspection levels is not justifiable.

III.4. Criteria for required correctionsThe criteria for required corrections on the overall meter populationinclude the trigger criteria and action criteria, which are specified in TableIII.4.1 below:

Table III.4.1. Criteria for required correctionsSampling plan applied on: Trigger Criteria Action Criteria

- Overall meter population 2.5% AQL 4% AQL- A group of meter type 2.5% AQL 10% AQL


49

AQL is the Acceptable Quality Levels, which is the maximum percentageof non-conforming meters in the meter group. The non-conformingmeters are those sample meters that are tested and found outside theCPUC-required accuracy limits. The current CPUC-required accuracylimits are ±2%. The overall meter population includes all meter typesowned by an entity and is not divided into any groups of meters or metertypes.

When the trigger criteria of 2.5% AQL is not met during a scheduledmaintenance and testing, the second or more sampling plans for furthertesting and checking shall be required to monitor the accuracyperformance of the overall meter population, groups, or types. Upon theviolation of the action criteria, corrective actions shall be taken to correctthe problems. Such corrective actions may lead to removals of certaininaccurate and aging meter groups or types.

III.5. Troubleshooting and Corrective ActionsTroubleshooting shall include all applicable tasks as indicated above andshall be performed to verify the meter operation and accuracy asrequired. If the troubleshooting indicates a hazard or out-of limitaccuracy condition at a meter site, the responsible party shall takecorrective actions within 48 hours of this finding.

III.6 MDMA CommunicationsThe MSP will notify the MDMA of a problem with the meter. TheMDMA is ultimately responsible for the quality of the data from themeter and will confer with the MSP regarding potential data problemswith the meter. The MDMA will determine if the meter data should beadjusted, including the cause and correction factor.


50

III.7. Demarcation Point of Meter WorkAs part of these meter worker qualifications, a clear scope of meteringwork must be established. CPUC D.98-12-080 reaffirms D.97-10-087,Appendix A: Direct Access Tariff, Section H(1)(b) which states,

"Potential and current transformers shall be considered part of thedistribution system and shall remain the responsibility of the UDC."

Accordingly, and pursuant to D.97-10-087 and D.98-12-080, all instrumenttransformers, test switches, and associated wiring up to the meter socketshall remain the responsibility of the UDCs. However, per D.98-12-080,reconnecting existing wires to a replacement of the existing meter socket,A-base socket adapter, or A-base meter may be performed by eitherUDCs or MSPs.


51

IV. METER SYSTEM TESTINGThe following meter system testing requirements shall be applied to: 1) ensurethe accuracy of the overall metering system at a customer site within the CPUC-required limits, 2) ensure safety in meter work procedures, and 3) provide aconsistency of testing once such testing is performed for a Direct Accesscustomer.

In general, there are seven tasks involved in testing an electric meter system.The metering system, defined for use in Section IV only, is 1) the meter itself, or2) the meter and its attached equipment or module(s). These tasks are describedin Section VI, Guideline for Meter Testing, which meter workers may use as aguideline for their meter work.

The following requirements shall are applied when testing is performed onmeters, meter sockets, and metering systems:

IV.1. Meter SocketVoltage test (Task #1 in Section VI) shall be performed on meter socketsor service connections to verify correct voltages. When a phase shiftingtransformer is in use, a phase angle test (Task #5 in Section VI) shall berequired (if load is present).

IV.2. Metering SystemThe following tests are required to be performed on a metering system:

IV.2.1 Light and full load test or customer load test (Task #2 inSection VI).

IV.2.2 Demand test (Task #3 in Section VI).IV.2.3 Register verification (Task #4 in Section VI).IV.2.4 Separate element check (Task #6 in Section VI).

IV.3. NotificationMeter worker shall notify the UDCs, ESPs, and MDMAs upon finding adefective meter that affects billing.


52

V. TEST STANDARDS

V.1. Calibration and Maintenance of Test StandardsThe following are the requirements for maintenance and calibration oftest standards which are used for testing the accuracy of electric meters inthe field or in the shop:

V.1.1. Out-of-calibrationA test standard that is found to have accuracy outside the limits of±0.1% of a reference test standard.

V.1.2. Basic Reference Test StandardA basic reference test standard that is certified according to NISTrequirements and used as an accuracy reference in the laboratoryor meter shop.

V.1.3. Portable Test StandardA portable test standard that is used to certify the accuracy ofmeters in the field.

V.1.4. Routine Accuracy Check RequirementsEach test function of a test standard shall be checked andcompared with a reference standard monthly. If a test standardused in testing meters has a result of three consecutive meterstested out of the CPUC-required accuracy limits, this test standardshall be checked against the reference standard.

V.2. Meter Re-testing upon Finding of An Out-of-calibration Test StandardOnce a test standard is checked and found out-of-calibration, it shall becalibrated according to NIST requirements, and meters that were testedby this test standard since the last routine check or calibration of the teststandard shall be re-tested with an accurately-calibrated test standardaccording to the following:

V.2.1. A meter sample from the meters tested by this portable standardsince its last check or calibration shall be selected for re-testing asfollows:V.2.1.1. 3–20 meters: All meters will be re-tested for accuracy.V.2.1.2. 21–100 meters: A random sample of 20% or 20 meters,

whichever is less, will be retested for accuracy.V.2.1.3. 101–200 meters: A random sample of 10% or 20 meters,

whichever is less, will be retested for accuracy.V.2.1.4. Over 200 meters: A random sample of 10% or 50 meters,

whichever is less, will be retested for accuracy.


53

V.2.2. If at least 95% of the first sample pass the meter test for accuracy,all meters will be considered accurate.

V.2.3. If less than 95% of the first sample pass the meter test for accuracy,a second sample of the same size will be tested.

V.2.4. If a second sample is drawn, 95% of the sample meters must passthe meter test standard for accuracy.

V.2.5. If less than 95% of the second sample pass the meter test foraccuracy, all meters previously tested by the portable standard willbe re-tested and re-calibrated for accuracy.


54

VI. GUIDELINE FOR METER TESTING

Matrix of Meter System Testing

The following matrix indicates which testing task can be applied to a meter technology, and it serves as a guide in metertesting. Some of the specific tasks may not apply to newer meters:

TASKCODE

Type of Meters/Registers

1Voltage

Test

2Light & Full

Load orCustomer-Load Test

3 Demand

Test

4Register

Verification

5PhaseAngleTest

6SeparateElementCheck

7Burden

Test

PossibleOtherTypes

Í

COMMENTS

CODE

(a) Self-Contained kWh Meters X X (g), (j) Mechanical Meter (a)

(b) Transformer-Rated kWh Meters X X X X (g), (j) Mechanical Meter (b)

(c) Hybrid Meters X X X X X (i), (j) (c)

(d) Solid State Meters X X X X (g), (i),(j)

(d)

(e) Transformer-Rated kVARh Meters X X X X X (j) Mechanical Meter (e)

(f) Multi-Quadrant Meters X X �� X � X (g), (j) (f)

(g) Solid State Recorders X X (g)

(h) Mechanical Registers �� (g), (j) (h)

(i) Electronic Registers �� (g), (j) (i)

(j) Pulse Devices X (j)

(k) Self-Contained Network Meters X � � (g), (j) (k)

�� Performed if demand is present. �� Energy Consumption Investigation (ECI) is necessary. � Performed when deemed necessary or requested by customers.� The alphabetical codes in the “Possible Other Types” column refer to the “Code” column on this matrix and indicate that additional tests must be performedas required if the other types exist at the sites.


55

VI.1. Task #1: Voltage Test

The voltage test is necessary to 1) ensure safety in the meter work procedures, 2)provide a meter worker a knowledge of the correct service voltage prior to anymeter work, 3) provide a confirmation of no short-circuit or hazardousconditions in customer equipment or panel. Below is a procedure for the voltagetest:

On all services, measure the secondary voltage, with an approved volt meter,between the line phases and line to ground even if it is an ungrounded serviceand record all voltage readings on the test tag. All nominal voltages have anallowable tolerance of ±5%.

VI.1.1. Secondary Distribution Voltages Normal rating of 120 V.

Table VI.1.1-1: Nominal Voltages and allowable limits on SecondaryNominal Voltage (V) Maximum (V) Minimum (V)

120 126 114208 218 198240 252 228277 291 263480 504 456

TableVI.1.1-2: Service Voltages and allowable limits on SecondaryService Voltage (V) Measured Voltage

(V)Maximum

(V)Minimum

(V)120/240 V 3w 1ø 120 V Phase to Ground

240 V Phase to Phase126252

114228

120/208 V 3w 1ø 120 V Phase to Ground208 V Phase to Phase

126218

114198


126218

114198

120/240 V 4w 3ø 120 V Phase to Ground208 V Phase to Ground240 V Phase to Phase

126218252

114198228


291504

263456

240 V 3w 3ø 240 V Phase to Phase 252 228480 V 3w 3ø 480 V Phase to Phase 504 456


56

VI.1.2. Primary Distribution and Transmission Voltages These voltages have a secondary voltage rating of 115 V or 120 V. Thevoltage transformer primary may be connected either wye or delta.

Table VI.1.2-1: Primary and Transmission Voltages with their secondaryvoltages and ratios

Possible SystemL-L Voltage

(V)

VT PrimaryVoltage (V)

SecondaryRating

(V)

Ratio

2,400 or 4,200 2,400 120 20:14,200 or 7,200 4,200 120 35:1

7,200 or 12,000 7,200 120 60:112,000 or 20,125 12,000 120 100:1

17,200 18,000 120 150:120,125 or 34,500 20,125 115/67.08 175/300:1

60,000 34,500 115/69 300/500:169,000 40,250 115/67.08 350/600:1115,000 69,000 115/69 600/1000:1138,000 80,500 115/69 700:1230,000 138,000 115/69 1,200/2000:1500,000 287,500 115/69 2,500/4200:1


57

VI.2. Task #2: Light Load & Full Load Test or Customer-Load TestThis test is the accuracy testing to assure that the meter is accurate at variousload conditions. Typically, the accuracy of the solid state meters variesminimally in the full and light load conditions. However, the accuracy ofmechanical meters is significantly better at a full load than at a light load due tofriction and speed of disk shaft rotation. Therefore, it becomes a commonpractice to test meters for accuracy at both light and full load conditions.

Light load is 10% of test Amp rating, and full load is 100% of test Amp rating.The light and full load test is performed at 100% power factor to verify theaccuracy of the meter by comparing its test results with a standard meter ofknown traceable accuracy. Customer Load Test will consist of two or more testruns using the customer’s load. Each heavy load test run shall be at least 60seconds in duration. Heavy load is defined as any load over 10% of Test Amprating of the meter. Each light load test run shall be at least 90 seconds induration and two revolutions of the meter disk

VI.2.1. No Load or Creep Test (for mechanical & hybrid meters only)Test meters will be energized with no connected load, and the diskobserved for rotation. A minimum amount of creep is acceptable if therotation stops when an anti-creep hole reaches the meter stator.

VI.2.2. Accuracy testSingle phase and polyphase meters are to be tested with potential coilsconnected in parallel and the current coils connected in series; or an onsite test can be performed using customer load. The minimumduration of the test will be 1 disk revolution for light load and 10 diskrevolutions for full load. For solid state meters, calculated pseudorevolutions will be used. Customer Load test will be at least 60 sec induration and minimum 2 revs for heavy load test. Light load test runat least 90 sec in duration and two revolutions of the disk

VI.2.3. AdjustmentsMeters found within the limits required by the local UDC need not beadjusted.

VI.2.4. ReportingCorrection factors for light load and full load tests will be recorded asthe meter accuracy results as specified in ANSI C.12.1 (1995)

For example, for a 0.5% fast and slow meter:Correction Factor Accuracy

Fast 0.995 +0.5%Slow 1.005 -0.5%


58

VI.3. Task #3: Demand Test The demand test is performed to ensure the accuracy of the demand function ofa meter.

VI.3.1. Mechanical Demand MetersTest for the accuracy of the demand registers by checking the markedregister ratio and time interval as follows:

VI.3.1.1. Advancing Mechanism: Visually inspect the pusher arm andgears for worn out parts.

VI.3.1.2. Clutch: Check that the pusher arm returns to zero at the end ofinterval when the clutch releases it. Verify that this is a smoothoperation and that the pusher arm returns to the stop position.

VI.3.1.3. Timing Motor: Check that the timing motor is operating. If themotor is not operating, the demand reading will most likely beoff scale.

VI.3.1.4. Time Interval and Demand Test: Test the time interval toensure that it is as marked on the register name plate by timingthe motor gear and clocking an interval (i.e. 900 seconds for 15minutes), or performing a demand test. The demand test isperformed by minimally applying full load current for 125 diskrevolutions on 15 minute demand meters and for 250 diskrevolutions on 30 minute demand meters.

VI.3.1.5. Demand Mechanism: Check that the demand reset returns thepointer to zero on a reset action and that the pointer willadvance to full scale.

VI.3.1.6. Reporting: Correction factor for the demand test will berecorded as the meter accuracy result.For example for a 0.5% fast and slow meter:

Correction Factor AccuracyFast 0.995 +0.5%Slow 1.005 -0.5%

VI.3.2. Solid State Meters and Electronic RegistersBelow is a procedure for a demand test on a solid state meter or anelectronic register:

VI.3.2.1. Place the meter register in the TEST MODE.VI.3.2.2. Press the reset button to clear all test registers and start a

new demand interval.VI.3.2.3. Apply full load current for 25 disk revolutions (or

equivalent) as a minimum.VI.3.2.4. Scroll the register display to show present kW demand. If

the kW value is in pulses, calculate the actual kW value.VI.3.2.5. Calculate and record the correction factor by comparing the

meter kW value with the kW in the test standard.VI.3.2.6. Return the meter to the NORMAL MODE.


59

VI.4. Task #4: Register Verification The register verification is performed to ensure that the register parts andcomponents are working to provide and retain accurate billing data andinformation.

VI.4.1. Mechanical RegisterVisually inspect and verify that the register is correct for the watt-hourmeter, and that the register constant, gear ratio, and register ratio arecorrect for the register.

VI.4.1.1. Gear mechanism: Check that the gears are clean, and checkthe proper engagement between gears and between the firstgear and the disk shaft.

VI.4.1.2. Mechanical register: Check that the register ratio is the sameas marked. The register ratio (Rr) is determined by countingthe number of revolutions required of the first gear to causethe first dial pointer to make one complete revolution.

VI.4.2. Electronic RegisterBecause electronic registers are different for each type andmanufacturer, the manufacturer’s manual should be referred to forspecific procedures for these registers. Below is a typical procedurefor verifying electronic registers after they are programmed withproper parameters:

VI.4.2.1. Check all numerical segments and identifiers on the meterdisplay.

VI.4.2.2. Check that the meter is scrolling properly through allregisters.

VI.4.2.3. Check the function of the registers by disconnecting andthen restoring power on the meter. The meter shouldresume normal display operations and reading after thispower outage.

VI.4.2.4. Verify the programming parameters are correct.VI.4.2.5. Check the register memory and load profile data memory to

verify that data is being stored.


60

VI.5. Task #5: Phase Angle TestThe phase angle test is performed to ensure the correct wiring for a meter systemand therefore ensure the meter site accuracy because the correct wiring dependson a unique phase rotation (ABC or CBA) supplied by a UDC, customer load,and service wiring (delta vs. wye and three wires vs. four wires). The followingis an example of procedures to follow when performing a phase angle test:

VI.5.1. Utility Phase Rotation RequirementsRefer to specific UDC requirements regarding phase rotation prior toperforming this test, and check for correct wiring on a reactivemetering site and proper phase shift on each phase circuit.

VI.5.2. Phase Shifting Transformer (Reactaformer) VoltageMeasure the secondary voltage to ensure correct voltage is applied tothe potential coil in the kVarhr meter. If the phase voltage is notcorrect, check for the following possible errors: 1) incorrect wiring onthe phase shifting transformers, and 2) incorrect service voltage.

VI.5.3. Power Factor CalculationCalculate the load on the (real power) kWhr and the (reactive power)kVarhr meters and the power factor for the site. This power factor isused as a cross check for the measured phase angles.

VI.5.4. Current MeasurementMeasure each phase current and record it accordingly.

VI.5.5. Phase Angle MeasurementUse a phase angle meter to measure each phase angle and record itaccordingly.

VI.5.6. Phase Angle PlotAfter measuring all voltages, currents, and phase angles, plot a phaseangle test and cross check with the calculated power factor.


61

VI.6. Task #6: Separate Element CheckThe separate element check is performed to ensure that each element is still in agood working condition. This check is necessary because a meter has more thanone element, and it still measures energy usage even if one element is burnt ordefective.

Check individual elements for proper disk rotation or registration by performingthe following:

VI.6.1. TestingEnergize with the appropriate phase potentials and place current onindividual elements, and check for forward rotation or positiveregistration

VI.6.2. Correct phasingCheck that the proper potential coil and current coil are wiredcorrectly in the meter. If the meter is installed correctly on the socket,the meter disk will rotate forward.

VI.6.3. No forward rotationIf an element has no rotation, check for no load current and for openpotential coil(s) with a magnet or an ohmmeter.

VI.6.4. CautionsPhase angle relationships may cause reverse rotation in cases wherethere is a poor power factor, possible un-metered load, and/or shortcircuit. This will only occur on three wire three phase metered servicesites.


62

VI.7. Task #7: Burden TestBurden is the total load or impedance in the current transformer (CT) circuit dueto meter coils, leads, and other connected devices. The burden test is performedto check for the proper operating conditions of CTs. This test is necessary toverify that the output current is proportional to the CT’s nameplate ratio. Belowis an example procedure for the burden test on a CT:

VI.7.1. ProcedurePut built-in burdens of a multi-range Ammeter in series with thesecondary of the test CT to obtain the Amp readings. These Ampreadings should have the same deflection.

VI.7.2. Possible ProblemsIf a CT has a reading significantly different from the others, check forshorted turns or other problems, such as short circuited primary turns,short circuited secondary turns, high resistance connections in thesecondary circuit, short circuited secondary wiring, or grounding ofnormally ungrounded wire.

Glossary of Terms and Acronyms

63

GLOSSARY OF TERMS

Adjustment: Any change to a customer's usage data made to correct an error in theoriginal data.

Average Daily Usage (ADU): Average daily usage over a specified period of time,such as a billing period. For example, if all constants and factors have already beenapplied to the reads, the ADU could be calculated by: ADU = (current billing read -previous billing read)/(# days between billing reads).

EDI Implementation Guideline: Defines the EDI environment for using conventionswithin an industry, and provides assistance on how to implement the X12 standard.The Utility Industry Group (UIG) establishes Implementation Guidelines for the utilityindustry.

Direct Access (DA): A service option where the customer obtains its electric power andancillary services from an Energy Service Provider.

Electricity meter: A device that measures and registers the integral of an electricalquantity with respect to time.

Electronic Data Interchange (EDI): The computer-application-to-computer-applicationexchange of business information in a standard format. In the context of this report,EDI refers to use of the ANSI X12 standards.

Energy Service Provider (ESP): The party that contracts with the end-use customer toprovide commodity electric service.

End-Use Customer: A customer that takes final delivery of electric power and does notresell the power.

Estimated data: Usage or demand data that has been calculated based on standardestimation rules.

Interval data: Metered end-use data from a meter capable of recording actual energyusage for each time interval (e.g., hour, half-hour, etc.) during the billing cycle.

Irregular usage customer: Customer whose usage pattern does not follow normalusage patterns and consistently fails validation checks.

KYZ (contact output): A device coupled to a sensor or meter which producesincremental pulses with a defined value of the measured media. Also known as "Relayoutput Form C" in certain ANSI designations.


64

Local Meter Reading: The reading of meters accomplished by physically visitingindividual customer premises or meter installation sites.

Meter: A device for measuring and totaling the variable consumption of a product. Ingeneral, a meter consists of a sensor that detects and measures a flow, and anintegrating device and register that displays the total consumption in metrologicalunits.

Meter Data Management Agent (MDMA): An entity who performs a function whichentails acquiring raw end-use data, performing VEE to create validated data, providingvalidated data to specified market participants, and maintaining an archive of raw andvalidated meter data.

Meter Product: A device which measures, calculates, records and/or communicatesenergy consumption data for the purpose of determining the financial obligation for anentity consuming energy. Shall include any optional circuit boards, devices, ormodules enclosed within the meter cover.

In-inventory Meter Products: Meter products that have been purchased andstored in inventory, but not yet installed for Direct Access service in California.This does not include used, re-worked, or recycled meters.

In-service Meter Products: Meter products that have been approved and arecurrently in service for Direct Access.

Recycled Meter Products: Used meter products that are cleaned, tested, foraccuracy and good operating condition, and returned to inventory.

Retrofitted Meter Products: In-inventory or used meter products that areretrofitted with electronic modules.

Re-worked Meter Products: Used meter products that are repaired, rebuilt, orrefurbished. These do not include recycled or retrofitted meter products.

Used Meter Products: Meter products that are removed from service.

Meter Service Provider (MSP): The MSP function includes provision of the meterinstrument, installation, testing, maintenance, programming, and possibly meteringlocal area network (LAN).

Meter Type: The design and specifications for a meter product which includes allparts, components, and circuit boards, functioning as a unit. A meter type includes allcommunication technologies and any additional functions utilized by that meter typeand operated as a unit. In a situation where a meter can work with multipletechnologies or functions, but can only operate separately and individually with asingle communication technology or function, the combination of the meter and one


65

communication technology or function shall be considered as a different meter typefrom the combination of that meter and another communication technology or function.For example, meter brand "X" with a phone modem attached is a different type fromthe same meter brand "X" with a radio modem attached.

New Meter Type: Meter type that satisfies CPUC requirements described in theDASMMD. This meter type can be a new meter design or an existing meter which hasundergone a significant design change. A significant design change includes additionof a new circuit board with new meter function(s) not previously used with the metertype, or consolidation of two or more circuit boards.

Pulse overflow: Condition in which the actual usage during an interval is larger thancan be captured by the meter or recorder.

Raw data: Usage or demand data that has not gone through the validation, editing andestimation (VEE) process.

Re-framing: Changing the time frame of the metered usage data posted to the MDMAserver. This typically refers to changing the beginning or ending date/time of theusage data.

Regulations or regulatory requirements: Requirements imposed by the CPUC or otherregulatory entity; i.e., "rules & regulations."

Reliability: The probability of a product or system performing without failure aspecified function under given conditions for a specified period of time.

Remote Meter Reading: The reading of meters accomplished without physicallyvisiting individual customer premises or meter installation sites.

Specifications: Particular qualitative and quantitative attributes of a technical system orits elements, upon which the functionality of that system or its suitability for aparticular purpose depends.

Standards: Specifications established or promulgated by an official standards body,such as IEEE, ANSI, etc., for public use.

Closed Standards: Standards that are not open, i.e., that fail to meet any of thefour criteria for open standards.

De Facto Standards: Standards that are widely used in product design orreferred to by industry participants without having been sanctioned by arecognized standards body.

Direct Access Standards for Metering and Meter Data: Standards that areapproved as regulatory requirements for metering and meter data used in Direct


66

Access in California by the California Public Utilities Commission in its D.98-12-080 issued on December 17, 1998.

International Standards: Open standards as adopted by an internationalstandards body; the U.S. participates in ISO, IEC, and others.

National Standards: Open standards As adopted by a standards bodyaccredited to a national standards body; in the U.S., ANSI Standards.

Open Standards: Voluntary standards which are: (1) developed in an openforum, (2) sanctioned by an official standards body, (3) vendor-neutral, and (4)readily available to the public at a reasonable cost.

Proprietary Standards: Standards which are privately owned and for whichaccess may be unilaterally withdrawn or otherwise restricted.

Test mode: Period during which a test load is applied to a meter or recorder to verifyits accuracy.

Utility Distribution Company (UDC): The restructured descendent of an existingCPUC-regulated electric utility which provides distribution services, and is the defaultprovider of energy and revenue cycle services.

Utility Industry Group (UIG): A utility industry action group that represents membersto ASC X12. UIG develops, promotes, and establishes conventions for the use of EDIstandards, guidelines, and tools in the utility industry. Membership includes utilities,customers, suppliers, service providers, and liaisons to other organizations.

Validation check: Data check designed to identify usage or demand data that may notreflect actual usage, typically due to problems at the meter or recorder.Valid data. Usage or demand data that has gone through all required validation checksand either passed them all or has been verified.

Valid data: Usage or demand data that has gone through all required validation checksand either passed them all or has been verified.

Validated Data: Usage or demand data that has been validated, edited and estimated(VEE) in accordance with approved procedures.

VEE: Validating, Editing, and Estimating. Validation is the process of performingstandardization validation checks on usage and demand data. Estimating is theprocess of using standard estimation rules to calculate usage or demand data. Editingis the process of inserting estimated values into a validated data stream that has errors,gaps or omissions.


67

Verified data: Usage or demand data that failed at least one of the required validationchecks but was determined to represent actual usage.

Watt-hour meter: An electricity meter that measures and registers the integral, withrespect to time, of the active power of the circuit in which it is connected. The powerintegral is the energy delivered to the circuit during the integral over which theintegration extends, and the unit in which it is measured is usually the kilowatt-hour.(ANSI C12.1)


68

GLOSSARY OF ACRONYMS

ADU: Average Daily UsageANSI: American National Standards InstituteD: DecisionDA: Direct AccessDASR: Direct Access Service RequestDAWG: Direct Access Working GroupDQIWG: Data Quality and Integrity Working GroupCMEP: California Metering Exchange ProtocolCPUC: California Public Utilities CommissionCT: Current TransformerEDI: Electronic Data InterchangeESP: Energy (Electric) Service ProviderFCC: Federal Communications CommissionISO: Independent System OperatorkVARh: kilovar-hourskWh: kilowatt-hoursMDCS: Meter and Data Communication StandardsMDMA: Meter Data Management AgentMSP: Meter Service ProviderNEC: National Electric CodePG&E: Pacific Gas and Electric CompanyPSWG: Permanent Standards Working GroupPT: Potential TransformerRSIF: Retail Settlements and Information FlowsSC: Schedule CoordinatorSCE: Southern California Edison CompanySDG&E: San Diego Gas and Electric CompanyTOU: Time-of-UseUIG: Utility Industry GroupUDC: Utility Distribution CompanyV: VoltsVEE: Validating, Editing and Estimating

Attachment-VEE

STANDARDS FOR VALIDATING, EDITING,AND ESTIMATING MONTHLY AND

INTERVAL DATA


i

Table of Contents

SECTION A: CA INTERVAL DATA VEE RULES

1. INTRODUCTION............................................................................................................................................ 1

2. REQUIRED DATA VALIDATION CHECKS ............................................................................................... 1

3. INTERVAL DATA COLLECTION AND VALIDATION RULES ............................................................... 3

3.1. TIME CHECK OF METER READING DEVICE/SYSTEM ENSURES THAT THE COLLECTION DEVICE IS SYNCHRONIZED TO

THE NATIONAL TIME STANDARD BEFORE DATA COLLECTION BEGINS. ..................................................................... 33.2. COLLECT DATA............................................................................................................................................ 33.3. AS DATA IS COLLECTED................................................................................................................................ 33.4. EITHER AS DATA IS COLLECTED OR PRIOR TO PUBLISHING ON MDMA SERVER................................................ 43.5. ON THE BILLING CYCLE FOR THE METER........................................................................................................ 93.6. AFTER ALL VALIDATION CHECKS HAVE BEEN PERFORMED AND REQUIRED DATA HAS BEEN ESTIMATED,OPTIONALLY RERUN VALIDATION CHECKS TO ENSURE REASONABLENESS OF ESTIMATES........................................ 103.7. RECORD KEEPING REQUIREMENTS.............................................................................................................. 103.8. IRREGULAR USAGE CUSTOMERS ................................................................................................................. 11

4. INTERVAL DATA ESTIMATION RULES ................................................................................................. 12

4.1. IF SECTION OF DATA NEEDING ESTIMATION IS 2 HOURS OR LESS IN LENGTH, USE POINT-TO-POINT LINEAR

INTERPOLATION TO ESTIMATE THE DATA. INTERVALS CONTAINING A POWER FAILURE CANNOT BE USED AS END

POINTS FOR INTERPOLATION. ............................................................................................................................. 124.2. IF THE SECTION OF DATA NEEDING ESTIMATION IS MORE THAN 2 CONTIGUOUS HOURS, USE THE AVERAGE OF

SELECTED REFERENCE DAYS TO ESTIMATE THE DATA. ......................................................................................... 124.3. CORRECTING DATA PROBLEMS ATTRIBUTABLE TO METERING PROBLEMS.................................................... 154.4. INTERVAL IN METER DOESN’T MATCH TARIFF OR SETTLEMENT REQUIREMENTS ............................................. 15

SECTION B: CA MONTHLY DATA VEE RULES

1. INTRODUCTION.......................................................................................................................................... 17

2. REQUIRED DATA VALIDATION CHECKS ............................................................................................. 17

3. RULES FOR MONTHLY DATA VALIDATION CHECK......................................................................... 18

3.1. TIME CHECK OF METER READING DEVICE/SYSTEM..................................................................................... 193.2. TIME TOLERANCE CHECK OF METER.......................................................................................................... 193.3. HIGH/LOW USAGE ..................................................................................................................................... 203.4. HIGH/LOW DEMAND .................................................................................................................................. 263.5. TOU USAGE.............................................................................................................................................. 273.6. ZERO CONSUMPTION FOR ACTIVE METERS ................................................................................................. 283.7. USAGE FOR INACTIVE METERS ................................................................................................................... 283.8. NUMBER OF DIALS ON METER .................................................................................................................... 283.9. METER READ DEMAND DECIMAL QUANTITY DIFFERENCE ........................................................................... 293.10. METER IDENTIFICATION........................................................................................................................... 293.11. IRREGULAR USAGE MONTHLY DATA ........................................................................................................ 30

ii

4. MONTHLY DATA ESTIMATION RULES................................................................................................. 31

4.1. ESTIMATING USAGE................................................................................................................................... 314.2. ESTIMATING DEMAND ............................................................................................................................... 334.3. ESTIMATING TOU USAGE .......................................................................................................................... 344.4. ESTIMATING TOU DEMAND DATA ............................................................................................................. 36

ATTACHMENT VEE-A: INTERVAL DATA VEE TECHNICAL METHODS ........................................... 39

ATTACHMENT VEE-B: MONTHLY DATA VEE TECHNICAL METHODS........................................... 47

Standards for Validating, Editing, and Estimating Monthly and Interval DataSection A: CA Interval Data VEE Rules (Revision 2.0)

1

1. Introduction

This section defines the interval data validation, editing, and estimation techniquesrequired to participate in the California market as an MDMA.

2. Required Data Validation Checks

Data validation checks are designed to identify things that can go wrong at themeter/recorder and cause the data collected to not reflect actual usage.

These rules apply to both kWh and kVARh data, depending on the data required bythe meter’s tariff(s). If data is provided for informational purposes only (not used forbilling purposes), validation is not required. Data that has not gone through thevalidation process is raw data.

General MDMA and MSP business practices should ensure that the meter isprogrammed correctly for the required revenue data and that the MDMA system is setup to accurately maintain information such as interval size, meter constants, and whatquantity is recorded by what channel. These VEE rules do not require or describe howthe MDMA verifies that the meter is programmed correctly.

All validation checks must be run. Failure of one check does not preclude the MDMAfrom performing other validation checks.

Several words are used to describe the quality of interval data.

• Raw data - data that has not gone through the VEE process

• Valid data - data that has gone through all required validation checks and eitherpassed them all or been verified

• Verified data - data that failed at least one of the required validation checks but wasdetermined to represent actual usage

• Estimated - data that has been calculated based on standard estimation rulesbecause the raw data was not valid

The following validation checks are required for interval kWh and kVARh data. Theyare described in section 3 of this document:


2

Check PurposeTime check of meter readingdevice/system

Check for the following:• time drift of meter reading device/system

outside CPUC standardMeter identification check Check for the following:

• meter ID was reported correctly• meter has not been changed out• data is being reported for the correct meter

Time check of meter Check for the following:• time drift of meter clock outside CPUC

standard Pulse Overflow check Check for the following:

• improper scaling factor in meter• improperly sized transformer• hardware problem

Test Mode check Check for the following:• Data collected when meter was in test mode

that represents test load rather than actualusage

Sum check Check for the following in combinationmeter/recorder installations:• Crossed channels between meter and recorder• pulse relay problems Check for the following for all installations:• invalid PT and CT ratios,• invalid meter constants

Spike check Check for the following for all installations:• transmission error• spike resulting from meter test. Note that a spike can also occur after an outage -in this case the data is valid, but may or may notbe used for peak billing depending on the tariffand company policy.

kVARh check (for kWh dataonly if corresponding kVARhdata available)

Check for the following:• kWh channels are correctly mapped to kVARh

channel• meter is operating correctly


3

Check Purpose High/Low Usage check Check for the following in all installations:

• dropped phases• inaccurate meter constants• energy diversion• fast/slow meters Also check for the following in combinationmeter/recorder installations: erratic pulse input to recorder

3. Interval data collection and validation rules

If interval data is read more often than required for billing, checks need to beperformed at different times in the process. Some must be done as the data is readfrom the meter; some can be done anytime between when the data is collected from themeter and the end of the cycle, and others have to be done on a billing period basis atthe end of the billing cycle. They are broken out that way in this description.

3.1. Time check of meter reading device/system ensures that the collection deviceis synchronized to the national time standard before data collection begins.

3.2. Collect data

3.3. As data is collected

3.3.1. Check meter identification – verify that the meter’s identification matcheswhat is expected.

3.3.2. Perform Time Tolerance check on meter and data.

The time tolerance check is performed to minimize and correct meterclock drift and to minimize and correct the data problems associated withmeter clock drift.

How to do Time Tolerance check on meter

To perform a time tolerance check on the meter, compare meter time todata collection device time. Note that depending on the communicationtechnology used, network latency must be taken into account.

Pass/Fail Criteria• If meter time is within 3 minutes of time standard, the meter passes

the time tolerance check. (Note that if the meter time is within the3-minute tolerance, the meter time can optionally be corrected.)


4

• If meter time is off by more than 3 minutes, the meter time must becorrected. If the meter fails the time tolerance check for threeconsecutive months, the meter must be physicallyinspected/tested.

How to do Time Tolerance check on data

To perform a time tolerance check on the data, compare the number ofintervals retrieved from the meter to the number of intervals expectedgiven the elapsed time.

Pass/Fail Criteria• If the actual number of intervals is equal to the expected number,

the data passes the time tolerance check.

• If the actual number of intervals differs from the expected number,the data fails the time tolerance check. The data to be correctedincludes all intervals from the last time the meter time wasdetermined to be good (i.e., within the 3-minute tolerance) andwhen it was discovered that the meter time was off by more than 3minutes and the meter time was reset.

If data fails the Time Tolerance Check…1) If the meter time was off by less than or equal to 75 minutes,

prorate the data using one of the algorithms in Attachment VEE-A,Section 1.1.

2) If the meter time was off by more than 75 minutes, the data mustbe estimated.

3.4. Either as data is collected or prior to publishing on MDMA Server

3.4.1. Perform Pulse Overflow Check

Inspect each interval for this condition. If a pulse overflow occurs, themeter requires physical meter test/maintenance. Intervals with pulseoverflows must be estimated.

3.4.2. Perform Test Mode Check

If the MDMA determines the meter was in test mode, the MDMA mustensure that the customer is not billed for the test load. If no actualcustomer usage data is available for the time in which the meter was intest mode, zero usage is reported for that period; this data is valid. If themeter is inadvertently left in test mode, the data must be estimated ifactual usage data is not available from the meter .


5

3.4.3. Perform Sum Check

The sum check is performed to ensure that the difference between theenergy use recorded in the intervals and the energy use recorded in themeter over the same time period is within an acceptable range. Thischeck may be done on either consumption or pulse data, provided thedata scaling is consistent throughout the period.

How to do the Sum Check1) Calculate the energy use recorded in the intervals by summing the

intervals between the start and stop meter readings.

2) Calculate the energy use recorded by the meter by taking thedifference between the start and stop readings accounting forpossible rollover between start and stop readings. For example, ifthe start reading was 99968 and the stop reading was 00294, andthe meter reading rolls over at 99999, the difference would be 326.

3) Compare the energy use recorded in the intervals to the energy userecorded by the meter. Note that the values must be in the sameunits for the comparison.

Pass/Fail Criteria• If difference is <= two meter multipliers, the data passes the sum

check. (meter multiplier = CTR x VTR, where CTR is currenttransformer ratio and VTR is voltage transformer ratio)

• If difference is > two meter multipliers, the data fails the sumcheck.

If data fails the Sum Check1) Several optional steps may be taken to resolve the sum check

failure.

(a) Reread the meter and redo the sum check from original startmeter reading to new stop meter reading.

(b) Redo the sum check, taking into account the differencesbetween the time of the start read and the start of the firstinterval, and the time of the stop read and the end of the lastinterval. See Attachment VEE-A, Section 1.2 for moreinformation.

(c) Redo the sum check, taking into account missing orincomplete intervals. See Attachment VEE-A, Section 1.2 formore information.


6

(d) Additional checks may be performed, based on thetechnology used, to verify that the interval data is anaccurate representation of usage as measured by the meterreadings.

2) If sum check is not resolved, perform manual inspection of data.

(a) Verify meter and pulse multipliers. If a multiplier wasincorrect, redo the sum check using the correct multipliers.

(b) Check for a meter change between the start and stop meterreadings. If the meter was changed, redo the sum check foreach meter independently.

(c) Manually inspect data. If the data seems reasonable, it canbe considered verified.

(d) If the data does not seem reasonable, perform physicalmeter test/inspection. If meter tests OK, the data can beconsidered verified. If a problem is found with the meter,the data must be estimated. (Note: if the problem existedprior to this billing period, previously posted data must beadjusted and re-posted.) If unable to visit site and performmeter test prior to posting the data, the data must beestimated.

(e) If interval data is available but meter readings are notavailable, manually inspect the data. Data that seemsreasonable (compared with historical data) can beconsidered verified. Any data that does not seemreasonable must be estimated.

3) If the sum check failure cannot be resolved, the data must beestimated.

3.4.4. Perform Spike Check

The spike check is performed to identify intervals with suspiciously highusage relative to the surrounding intervals. This check may be done oneither consumption or pulse data, provided the data scaling is consistentthroughout the period.

How to do the Spike Check1) For each 24-hour period, identify the highest and third highest

peaks. (Normally the 24-hour period is from midnight tomidnight. If the data is at the beginning of the span and doesn’t


7

start at midnight, use enough data from the next day of data to get24 hours of data. If the data is at the end of the span and doesn’tstop at midnight, use enough data from the next to last day of datato get 24 hours.)

2) If the highest peak is less than or equal to the spike check thresholdof 10 pulses, skip the spike check. (A spike check threshold is usedto eliminate false spikes for meters with very low usage.)

3) If the highest peak is greater than the spike check threshold of 10pulses, subtract the third highest peak from the highest peak anddivide by the third highest peak.

Pass/Fail Criteria• If ((highest peak - third highest peak)/third highest peak) <= 1.8,

the interval passes the spike check.

• If ((highest peak - third highest peak)/third highest peak) > 1.8, theinterval fails the spike check.

If data fails the Spike Check1) Optionally reread the meter. If you get different value from reread,

redo spike check.

2) If value is the same on reread or you cannot reread the meter,perform manual inspection of data.

(a) Look for similar patterns on similar days. If a similarpattern is found and this seems reasonable, the data can beconsidered verified.

(b) Optionally check with customer for unusual conditions atthe time of the spike. If a legitimate reason for spike isfound, the data can be considered verified.

3) If no similar pattern or legitimate reason for spike is found, theinterval with the spike must be estimated.

4) If there is a regular pattern of failing this check, the customer maybe an irregular usage customer. See section on Irregular UsageCustomers for additional information.

3.4.5. If interval kVARh data is available, perform kVARh Check

The kVARh check is performed to identify intervals where reactive load(kVARh) is present and active load (kWh) is not, indicating a suspicioususage pattern and possible meter malfunction. This check is onlyrequired when both kWh and kVARh are used for billing. If kVARh datais available but not used for billing, the check is optional. This check may


8

be done on either consumption or pulse data, provided the data scaling isconsistent throughout the period.

How to do the kVARh Check1) If multiple kWh channels map to a single kVARh channel, or

multiple kVARh channels map to single kWh channel, theappropriate channels must be totaled prior to this check.

2) If there are any kWh intervals with zero consumption, check thecorresponding kVARh interval.

Pass/Fail Criteria• If the corresponding kVARh interval is also zero or less than or

equal to the kVARh check threshold of 4 pulses, the kWh datapasses the kVARh check. (A kVARh check threshold is used toeliminate false errors for meters with very low usage.)

• If the corresponding kVARh interval is greater than the kVARhcheck threshold of 4 pulses, the kWh interval fails the kVARhcheck.

If data fails the kVARh Check1) Several optional steps may be taken to resolve the kVARh failure.

(a) Investigate to determine if this data represents actualcustomer usage, in which case the data can be consideredverified.

(b) If multiple kWh channels map to a single kVARh channel,investigate to determine if the problem can be directlytraced to specific kWh channels. If this is the case, only datafor those channels must be estimated. If the problem is notattributable to specific channels, all kWh channels need to beestimated.

2) If no legitimate reason for the kVARh failure is found, the intervalswith failures must be estimated.



9

3.5. On the billing cycle for the meter

3.5.1. High/Low Usage Check

This test must be performed on the data that has passed or been verifiedfor previous checks, with no estimated values included. This identifiesmetered usage that is suspiciously high or low relative to historical usage.It may optionally also be performed on all data (valid and estimated) toprovide a reasonableness check on the estimates derived using thestandard estimation techniques.

This check must be done on consumption data, not pulses.

How to do the High/Low Usage Check1) If last year’s data is available, calculate average daily usage for

same billing month last year; use summed VEE or historical billinginterval data if available, if not use VEE or historical billing usage(i.e., difference between register readings).

2) If last year’s data is not available, calculate average daily usage forthe previous billing month; use summed VEE or historical billinginterval data if available, if not use VEE or historical billing usage(i.e., difference between register readings).

3) If last year’s data and last month’s data are not available, skip thehigh/low usage check.

4) Calculate average daily usage for this billing month using eithersummed VEE data (if check includes estimated data) or sum of allintervals not requiring estimation (if check does not includeestimated data). If not all intervals are included in the sum,prorate the sum accordingly.

Pass/Fail Criteria• If |(historical daily average - this month’s daily average)| <= 0.5*

historical daily average, the data passes the high/low usage check.

• If |(historical daily average - this month’s daily average)| > 0.5*historical daily average, all data in the billing month fails thehigh/low usage check.

If data fails the High/Low Usage Check1) Perform manual inspection of data.

(a) Look at recent history for the meter. If monthly usage hasbeen on a trend in the appropriate direction and this seemsreasonable, the data can be verified.


10

(b) Optionally check with customer for changed usage patterns.If changed usage patterns match change in data, the data canbe verified.

(c) Check to see if some of the data looks reasonable; reasonablelooking data can be verified. For example, if a meter failssometime during the month, the data at the beginning of themonth may be OK, while the data after the meter failuremay be obviously bad.

(d) If the data does not seem reasonable, perform physicalmeter test/inspection. If meter tests OK, the data can beverified. If a problem is found with the meter, the data mustbe estimated. (Note: if the problem existed prior to thisbilling period, previously posted data must be adjusted andre-posted.) If unable to visit site and perform meter testprior to posting the data, the data must be estimated.

2) If the data is investigated and found to be accurate, the data isverified.

3) If the data fails high/low usage check, suspect data must beestimated.


3.6. After all validation checks have been performed and required data has beenestimated, optionally rerun validation checks to ensure reasonableness ofestimates.

3.7. Record Keeping Requirements

If data failed one or more validation checks, the specific checks that the datafailed must be recorded on an interval level, and:1) If the data was manually verified, that information must be recorded on

an interval level. Verified data is valid.

2) If the validation failure(s) were not resolved through accepted methods,the data must be estimated.

For each interval that is estimated, the MDMA must record the estimationalgorithm used. Interval data estimation algorithms include:• less than 2 hours (4.1)

• greater than 2 hours (4.2) – not scaled based on usage

• greater than 2 hours (4.2) – scaled based on usage


11

• time-drifted intervals prorated (Attachment VEE-A, Section 1.1)

• time-drifted intervals prorated (Attachment VEE-A, Section 1.2)

• intervals adjusted (4.3)

• intervals manually estimated (4.2.8)

• intervals estimated due to meter interval programmed incorrectly (4.4)

• load profile template used (4.2.7)

3.8. Irregular Usage Customers

An irregular usage customer is one whose usage pattern does not follow normalusage patterns and consistently fails the spike check, kVARh check, or high/lowusage check. An MDMA can identify a customer as an irregular usage customerif:

1) the customer data fails the standard validation check for three consecutivemonths and the MDMA verifies that the data represents the actualcustomer usage, OR

2) MDMA is notified by the customer’s ESP or previous MDMA of theirregular usage pattern.

The data used to identify an irregular usage customer could be data collected bythe MDMA, or historical data provided by the previous ESP or MDMA. AnMDMA may modify the spike check and/or high/low usage check, and skip thekVARh check if an irregular usage customer consistently fails the check. TheMDMA must notify both the customer’s ESP and UDC of the customer’sirregular usage status and what modified checks will be performed.

The goal of the modified checks is to automate the manual procedures theMDMA would perform to verify that this is the customer’s normal usagepattern. An MDMA may use a variation of the spike check or high/low usagecheck based on the actual usage pattern. Note that the MDMA may not skip thespike check or high/low usage check. If the data passes the modified check, thedata is valid and does not need to be marked as verified.

• Examples of modifications for the spike check include modifying thespike check value (180%) or the pulse threshold value (10 pulses).

• Examples of modifications for the high/low usage check includechanging the percentages (+/- 50%), using the year’s average instead ofone billing period’s average, or comparing to the minimum andmaximum values for the past year.


12

For some customers, irregular usage patterns are symptomatic of the businessand will always be present, such as co-generation customers. For othercustomers, irregular usage patterns may be a temporary condition, such as whena factory adds a second shift and fails the high/low usage check for the first 12months. The MDMA must determine whether a customer is a permanent ortemporary irregular usage customer. Temporary irregular usage customersmust be reviewed annually to determine if they are still irregular usagecustomers or should be returned to the normal checks.

4. Interval Data Estimation Rules

Estimate intervals needing estimation using the following estimation rules

4.1. If section of data needing estimation is 2 hours or less in length, use point-to-point linear interpolation to estimate the data. Intervals containing a powerfailure cannot be used as end points for interpolation.

How to apply Point-to-Point Linear Interpolation1) If the section occurs in the middle of the data, the “first point” is the last

valid interval before the section, and the “second point” is the first validinterval after the section.

2) If the section occurs at the beginning of the span, use the last interval fromthe historical data as the first point if the historical data is available andvalid. Otherwise, use the second point (the first valid interval after thesection) as the first point – this will cause the load to be estimated as a flatload.

3) If the section occurs at the end of the span, use the first point (the lastvalid interval before the section) as the second point – this will cause theload to be estimated as a flat load.

4.2. If the section of data needing estimation is more than 2 contiguous hours, usethe average of selected reference days to estimate the data.

Rules and definitions for selecting reference days for estimation:• “Same weekdays” are defined as the same day of week as the day that

needs estimation. In the case of holidays, “same weekdays” are holidays.

• “Like days” are defined as the same daytype (i.e., weekday, weekend, orholiday) as the day that needs estimation.

• A standard list of holidays will be used, regardless of the tariff or serviceterritory of the meter. The “legal” definitions of the holiday is used; if theholiday falls on a Sunday, the “legal” holiday is the following Monday.Otherwise the “legal” holiday is the same date as the actual holiday. Theholidays used are the following:


13

• New Years Day

• Presidents Day

• Memorial Day

• Independence Day

• Labor Day

• Veterans Day

• Thanksgiving Day

• Christmas Day

• Only “valid” intervals can be used for estimation. Valid intervals aredefined as those that have passed all validation checks or have beenverified. Estimated intervals cannot be used for estimation.

• Data from days with a power failure cannot be used for estimation.Power failures can cause irregular usage patterns, resulting in data that isnot typical for the customer.

• Valid intervals from “partial” days can optionally be used for estimation.“Partial” days are defined as those containing estimated data or those onwhich data collection began at some time other than midnight.

• Historical data up to 90 days prior to the day needing estimation andfrom the current billing period may be used for estimation.

• Reference days are chosen to be the closest chronologically to the dataneeding estimation, regardless of seasonal crossover. This may includedays after the day requiring estimation. If two potential reference daysare equidistant from the day requiring estimation, use the earlier day first.For example, if June 2, 1998 needed estimation and the billing period wasfrom June 1 to June 30, the reference days used would be May 19, May 26,and June 9, provided they contained valid data.

4.2.1. Develop a daily profile

1) Find the three “same weekdays” with valid data closest in time tothe day with the data needing estimation based on the rules listedin the previous section. If the day with data needing estimation isa holiday, the “same weekdays” are holidays, not the same day ofweek. If not enough historical holidays exist in the current billingperiod or previous 90-day period, use Sundays. Calculate theaverage daily profile using the three selected days.


14

2) If only two same weekdays are available from the current billingperiod and 90 days of historical data, calculate the average dailyprofile using the two selected days.

3) If only one same weekday is available from the current billingperiod and 90 days of historical data, use it as the daily profile.

4) If no same weekdays are available in the current billing period and90 days of historical data, look for the three “like” days that areclosest chronologically to the day with intervals needingestimation. For example, if the intervals needing estimation wereon Tuesday, use Monday, Wednesday, and Thursday. Only useweekdays with weekdays; only use weekends with weekends; onlyuse holidays or Sundays with holidays. Calculate the averagedaily profile using the three selected days.

5) If only two like days are available from the current billing periodand 90 days of historical data, calculate the average daily profileusing the two selected days.

6) If only one like day is available from the current billing period and90 days of historical data, use it as the daily profile.

4.2.2. Use the daily profile to estimate the required data

1) Estimate the data needing estimation by applying the appropriateintervals from the average daily profile to fill the missing intervals.

2) If start and stop meter readings are available and known to begood, they may optionally be used to scale the estimated intervaldata. See Attachment VEE-A, Section 1.3 for more information.

4.2.3. If there are no similar days or like days, use the load profile for thecustomer’s class to estimate the data. Use this month’s usage, if available,to scale the load profile. If this month’s usage is not available, use lastmonth’s usage or last year’s usage, whichever is determined to be morereasonable, to scale the load profile. Refer to the UDCs’ load profiledocumentation for more information on applying load profiles. Note thatthe load profiles are for hourly data. If 15 minute data is required,assume a flat load throughout the hour (i.e., each 15-minute intervalwould have 1/4th the hourly usage).

4.2.4. If there is no historical data that can be used, the data must be estimatedmanually and the process and assumptions documented.


15

4.3. Correcting Data Problems Attributable to Metering Problems

If on investigation the cause of the data problem is determined to be a problemwith the meter or meter installation and the data can be corrected by scaling theintervals and meter readings, the MDMA will be notified of:

1) The time period requiring correction.

2) The scaling factor to be applied to each interval in that period.

Examples of these situations include a meter running slow, a meter running fast,one or two phases dropped, etc.

When the data is posted, it is marked as estimated if it had not been previouslyposted, and marked as adjusted if it had previously been posted

4.4. Interval in meter doesn’t match tariff or settlement requirements

If the meter programming and the MDMA requirements are inconsistent, thedata is calculated as follows:

1) The meter is programmed to collect data at a smaller interval thanrequired by its tariffs, and the meter's interval evenly divides into theinterval required by the tariff – for example, the meter was programmedto collect 5-minute data, and the tariff requires 15-minute data. Sum the 5-minute intervals into 15-minute intervals on even 15-minute boundaries.If the data passed all the other validation checks, it is valid and does notneed to be marked as estimated or verified.

2) The meter is programmed to collect data at a larger interval than requiredby the meter’s tariff. For example, the meter is programmed to collect 60-minute intervals, but the tariff require 15-minute intervals. Prorate thedata by assuming an even load distribution during the interval. In thisexample, the usage in the 60-minute interval would be divided by 4 toestimate the usage in a 15-minute interval. The data is marked asestimated. The meter must be reprogrammed to the correct interval.

3) The meter is programmed to collect data at a smaller interval thanrequired by its tariff, but the meter’s interval doesn't evenly divide intothe interval required by the tariff. For example, the meter is programmedto collect 10-minute intervals, and the tariff requires 15-minute intervals.The data would be estimated and marked as estimated. To estimate data,all collected intervals that are contained within the required reportinginterval are included in the appropriate reporting interval. Collectedintervals that cross the boundaries of required reporting intervals areincluded proportionally in both reporting intervals. In this example, ifthere were three 10-minute intervals containing 10 kWh, 20 kWh, and 30


16

kWh, the corresponding estimated 15-minute intervals would contain 20kWh (10 + 0.5*20) and 40 kWh (0.5*20 + 30). This is similar to theprorating technique discussed in Attachment VEE-A Section 1.1. Themeter must be reprogrammed to the correct interval.

Standards for Validating, Editing, and Estimating Monthly and Interval DataSection B: CA Monthly Data VEE Rules (Revision 1.1)

17

1. Introduction

This document defines the data validation, editing, and estimation techniques requiredto participate in the California market as an MDMA for monthly data. Monthly dataincludes consumption, demand, and Time-of-Use (TOU) consumption and demand..

2. Required data validation checks

Data validation checks are designed to identify things that can go wrong at themeter/recorder and cause the data collected not to reflect actual usage.The following checks are required for monthly data validation for kWh and kW data.Similar checks would apply to kVARh and kVAR data if those values are required.

Check Purpose

Time check of meter readingdevice/system (applies todevices/systems collecting TOUdata only)

Check for the following:

• time drift of meter reading device/systemoutside CPUC standard

Time check of meter (applies tometers collecting TOU data only)


• time drift of meter clock outside CPUCstandard

High/low usage check Check for the following:

• misread

• fast/slow meter

• broken meter

• incorrect multipliers

• energy diversion

• dropped phases

High/low demand check (appliesto demand readings only)


• misread

• fast/slow meter

• broken meter



• dropped phases


18

Check Purpose

Time-of-use check (applies toTOU data only)


• misread

• fast/slow meter

• broken meter



• dropped phases

Zero consumption for activemeters



• meter doesn’t register

Usage for inactive meters Check for the following:unauthorized usage at a site for which there is nocustomer with financial responsibility

Number of dials on meter Check for the following:

• wrong meter

• misread

Dial decimal quantity Check for the following:

• wrong meter

• misread

Meter identification Check for the following:

• that the meter ID was reported correctly

• the meter has not been changed out

• the data is being reported for the correct meter

3. Rules for Monthly Data Validation Check

Some monthly data validation checks must be done at the time of meter reading, andother checks could be done anytime after the meter is read until the data is posted. Forexample, the meter identification check must be performed at the time of the meterreading, while the high/low usage check can be performed in a handheld system as themeter is read, or back in a host system. All checks are not applicable to all types ofdata. The following table summarizes which checks must be done for each type ofdata, and provides a recommended sequence.


19

Check Must bedone attime ofmeter read?

Consumption Demand TOUConsumption

TOUDemand

Time in meter Yes n/a n/a 3 3High/Lowusage

No 3 n/a 4 n/a

High/Lowdemand

No n/a 3 n/a 4

TOU usage No n/a n/a 6 n/aZeroconsumptionfor activemeters

No 4 n/a 5 n/a

Number ofdials

Yes 2 n/a 2 n/a

Number ofdemanddecimalplaces

Yes n/a 2 n/a 2

Meter ID Yes 1 1 1 1

Most of the checks and estimation algorithms are based on historical data for the samecustomer and the same site. In areas with wide fluctuations in weather, this may notprovide the best data for residential customers, as residential usage patterns vary muchmore with changes in weather than larger customers. A separate set of High/Lowusage validation check and estimation rules are provided based on day-before usage ofsimilar customers in the same geographic area.

3.1. Time Check Of Meter Reading Device/System

This check only applies for meter reading devices and systems collecting Time-of-Usedata. Time check of meter reading device/system ensures that the collection device issynchronized to a national time standard before data collection begins

3.2. Time Tolerance Check Of Meter

The time tolerance check is only required if the meter is collecting Time-of-Use (TOU)data. It verifies that the meter’s time is correct, and that TOU data represents theappropriate time periods. Note that depending on the communication technologyused, network latency must be taken into account.

3.2.1. If time in meter is within +/- 3 minutes of the time standard, the data haspassed Time Tolerance check. Note that if the meter is within +/- 3minutes of the standard, the time in the meter can optionally be corrected.


20

3.2.2. If time in meter is off > 3 minutes but <= 55 minutes, the data passes theTime Tolerance check. The data does not need to be estimated, but theMDMA must record the fact that the meter’s time was off by this amountin case there is a later question about the data. The meter time mayoptionally be reset.

3.2.3. If time in meter is off > 55 minutes, the data fails the time tolerance checkand must be estimated. The time in the meter must be reset. If the meterfails the time tolerance check after being reset for three consecutivemonths, the meter must be physically inspected/tested.

3.3. High/Low Usage

The High/Low Usage check validates cumulative consumption (kWh). Twomethods are provided - one based on historical data, and one based on previousday data from similar customers. An MDMA may implement either check,depending on weather characteristics and density of meter population served bythe MDMA. The second check requires a minimum density of meter populationto be statistically accurate; this still needs to be determined.

3.3.1. Method based on historical data

3.3.1.1. Calculate the average daily usage (ADU) for the present billingperiod.

For example, if all constants and factors have already beenapplied to the reads, the ADU could be calculated by:

ADU = (current billing read - previous billing read)/(# daysbetween billing reads)

If the previous billing read were on June 1, and the presentbilling read is on June 30, there would be 29 days betweenbilling reads.

3.3.1.2. Calculate the historical ADU

3.3.1.2.1. If there is not at least one month (minimum 27 days)of historical billing data available for the samecustomer and site, this check is not performed.

3.3.1.2.2. If data for the same customer and site is available,calculate the ADU for the same billing period lastyear. Use this as the historical billing ADU. One wayto determine which billing period last year is “thesame” is to choose the mid-point of this year’s billing


21

period, and find the billing period last year thatincluded the same date. For example, if the billingperiod this year was from April 13 to May 13, themid-point would be April 28. If the billing periodhas an even number of days, use the day after themiddle as the mid-point. For example, if the billingperiod was from June 1 to June 30, the mid-pointwould be June 16. Another way to determine whichbilling period is “the same” is to find the billingperiod last year with its read date in the samecalendar month as the read date of the data beingvalidated.

3.3.1.2.3. If there is no data from a year ago but there is data forthe last billing period (minimum 27 days), calculatethe ADU for the last billing period and use this as thehistorical ADU.

3.3.1.3. Compare the present billing period ADU with the historicalbilling period ADU. If the present billing period ADU isbetween 40% and 200% (inclusive) of the historical ADU, thedata passes this check. (Note that some systems may convertADU to a billing period usage to perform the check.) Optionaltrend factors that take into account peer group usage based ondemographics, climactic areas, and customer class may beapplied to the ADU to refine the High/Low comparison check.Sample trend factor calculations are to be provided at a laterdate.

3.3.1.4. If the present billing period ADU is not within 40% to 200%(inclusive) of the historical billing period ADU, the data failsthis check. Optionally re-read the meter.

1. If the reread is essentially at the same time as the originalread and a different value is obtained, assume the firstreading was a misread and perform the check again withthe new reading. If the same reading was obtained,assume the meter reading is correct; the data failed theHigh/Low Usage check but is verified. OR

2. If the reread is not at the same time as the original read, re-compute the average daily usage using the new reading.If the new ADU is within +/- 20% of the previous ADU,the data fails the High/Low usage check but is verified. If


22

the new ADU is not within +/- 20% of the previous ADU,the data fails the High/Low usage check.

3.3.1.5. Data that fails the High/Low usage check and has not beenverified may be investigated and manually verified ifjustification is found; otherwise the data must be estimated.

3.3.2. Method based on previous day usage of similar customers

Note that this method requires a certain density of customer data forresidential customers in the same geographic area, where weatherpatterns are typically consistent throughout the geographic area.

3.3.2.1. The following steps are performed at the end of each meterreading cycle day for each geographical area in order tovalidate and estimate usage the following day:

3.3.2.1.1. At the end of the reading day, for each good meterread (open account, billed, between 27-33 days &ADU =< 100), perform the following calculations todetermine an ADU for the billing period:

1. Calculate ADU (= KWH/days in billing period)2. Add ADU to Sum of Current ADU3. Calculate ADU squared4. Add ADU squared to Sum of Current ADU

squared5. Add 1 to total meters6. Calculate last month’s ADU7. Calculate current ADU times last month’s ADU8. Calculate last month’s ADU squared

3.3.2.1.2. Determine which range of usage (high, medium orlow) the current ADU should be grouped with bycomparing current ADU to yesterdays ADU Low andHigh Range Factors (Reference 3.3.2.1.3 for ADU lowand high range factor calculation methodology)

1. If the current ADU is less than yesterday’s ADUlow range factor:• Add current ADU to Sum of current low ADU• Add last month’s ADU to Sum of last month’s

low ADU


23

• Add current ADU times last month’s ADU toSum of current low ADU times last month’sADU

• Add current ADU squared to Sum of currentlow ADU squared

• Add last month’s ADU squared to Sum of lastmonth’s low ADU squared

• Add 1 to total low meters

2. If the current ADU is not less than the ADU lowrange factor from yesterday and is less than theADU high range factor, add the figures to themedium range following same format is in3.3.2.1.2. step 1.

3. Otherwise, add the current ADU to the ADU high

range following the same format in 3.3.2.1.2.step 1.

3.3.2.1.3. Calculate an aggregated ADU for current data foreach geographic area

3.3.2.1.3.1. Sum together the ADU values for eachgeographic area

3.3.2.1.3.2. Calculate the mean for the total ADU(=Sum of Current ADU / total meters)

3.3.2.1.3.3. Calculate the standard deviation for thetotal

3.3.2.1.3.4. Calculate the current ADU low and highrange factors:

• ADU Low Range Factor = mean -.43 StandardDeviation. If ADU Low range factor is less thanthe total current mean * .5, the ADU low rangefactor becomes the mean half.

• ADU High Range Factor = mean + .43 StandardDeviation

• NOTE: By determining the low & high factors,the Medium Range = (mean - .43 StandardDeviation) to (mean + .43 Standard Deviation)


24

3.3.2.1.4. For each of the three ranges determined above (low,medium, and high), calculate a percent of change ofmonthly usage for each geographic area.

3.3.2.1.4.1. After each meter’s current billing period’s(ADU) is grouped in 3 Ranges (Low,Medium, and High) as specified in3.3.2.1.2, the following data are summedup by ADU range and area:

• Number of customers• Sum of all last month’s ADU• Sum of all current month’s ADU• Sum of {each last month’s ADU times

current month’s ADU}• Sum of {all last month’s ADU squared}

i.e., Square all ADU, then sum them.• Sum of {all current month’s ADU

squared}

From the data above modified ADUmean factors and standard deviationfactors are determined for each range asfollows:

Modified Mean Factor:Sum of {last month ADU times currentmonth ADU} divided by the sum of {alllast month’s ADU squared}

Modified Standard Deviation Factor:Step 1: (Sum of {all current month’s ADU

squared} minus (Mean squared times sumof {all last month’s ADU squared})divided by (Total Meters minus 1).

Step 2: Take square root of Step 1

3.3.2.1.5. Calculate high and low range factors.

Calculate high and low range factors (HRF and LRF)for each of the 3 usage ranges within a geographicarea. The mean is used to calculate estimated reads,and the high and low range factors are used in thisvalidation check. 2.8 and 3.5 are used in the belowexample to represent the range deviation factor and


25

will allow for an appropriate meter read error rate.This factor can be changed to control the error rate.

High Range Factor Formula:HRF = 1 + {(2.8X Modified Standard Deviation XNumber of meters) / (Sum of Current month’s ADU)}

Low Range Factor Formula:LRF = 1 - {(3.5 X Modified Standard Deviation XNumber of meters) / (Sum of Current month’s ADU)}

3.3.2.2. As each meter is read, perform the following using the valuescalculated from the previous meter reading days’ data.

3.3.2.2.1. Determine the usage from the preceding billingmonth and the preceding billing reading for thecustomer and site.

3.3.2.2.2. Calculate low limit for this month’s usage bymultiplying the preceding month’s usage by the lowrange factor determined above.

3.3.2.2.3. Calculate high limit for this month’s usage bymultiplying the preceding month’s usage by the highrange factor determined above.

3.3.2.2.4. If the current usage is between the low and high limitcalculated in the previous two steps, the data passesthe High/Low check.

The following is a representation of how the High and LowRange Factors are used to validate meter usage:

Sample High/Low Usage Check:

Customer’s previous usage = 400 kWh

High Range Value: 400kwh X 1.115 HRF* = 446Low Range Value: 400kwh X .885 LRF* = 354

Usage values falling between 354 and 446 are accepted. Usage values outside this range fail thecheck.

* HRF/LRF = High and Low Range Factors, see description above.


26

3.3.2.2.5. If the current usage is outside the low and high limit,the data fails the High/Low check. Optionally re-read the meter.

1. If the reread is essentially at the same time as theoriginal read and a different value is obtained,assume the first reading was a misread andperform the check again with the new reading.If the same reading was obtained, assume themeter reading is correct; the data failed theHigh/Low Usage check but is verified. OR

2. If the reread is not at the same time as theoriginal read, re-compute the average dailyusage using the new reading. If the new ADU iswithin +/- 20% of the previous ADU, the datafails the High/Low usage check but is verified.If the new ADU is not within +/- 20% of theprevious ADU, the data fails the High/Lowusage check.

3.3.2.2.6. Data that fails the High/Low usage check and has notbeen verified may be investigated and manuallyverified if justification is found; otherwise the datamust be estimated.

3.4. High/Low Demand

The High/Low Demand Check compares the demand against historical data as areasonableness check.

3.4.1. Determine the peak demand for this billing period.

3.4.2. Determine the historical peak demand.

3.4.2.1. If there is not at least one month (at least 27 days) of historicalbilling demand data available, skip this check.

3.4.2.2. If demand data for this customer and site is available for thesame billing period last year, use that as the historical peakdemand. (Refer to section 3.3.1.2.2 to determine same billingperiod last year.)

3.4.2.3. If demand data is not available for the same billing period lastyear but there is demand data for the last billing period, use thepeak demand from the preceding billing month as the historicalpeak demand.


27

3.4.3. Compare the present peak demand with the historical peak demand. Ifthe present peak demand is between 40% and 200% of the historical peakdemand, the data passes this check.

3.4.4. If the present peak demand is not within 40% to 200% of the historicalpeak demand, the data fails the High/low demand check. Optionally re-read the meter if the same data is still available in the meter. Forexample, the same data would still be available if demand reset has notyet been performed or meter stores preceding billing period data.

1. If the reread is essentially at the same time as the original readand a different value is obtained, assume the first reading was amisread and perform the check again with the new reading. Ifthe same reading was obtained, assume the meter reading iscorrect; the data failed the High/Low Demand check but isverified. OR

2. If the reread is not at the same time as the original read and thenew demand value is within +/- 20% of the previous demandvalue, the data fails the High/Low demand check but is verified.If the re-read results in the same value, the data fails theHigh/Low demand check but is verified.

3.4.5. Data that fails the High/Low Demand check and has not been verifiedmay be investigated and manually verified if justification is found.Otherwise the data must be estimated.

3.5. TOU Usage

The TOU usage check compares the sum of the kWh meter readings for allperiods against the current season total kWh meter reading. Note that this checkmust be done in whatever units are read from the meter. For example, if themeter provides kWh, the kWh values must be summed and compared. If themeter provides pulses, the pulse values must be summed and compared.

3.5.1. For the current billing period, calculate the total kWh by summing all theperiods, including all seasons.

3.5.2. Compare the calculated total kWh with the current total kWh read fromthe meter. If they are within +/- the number of periods (active orinactive) summed together, the data passes the check. If they are not, thedata fails the TOU Usage check and must be estimated. (Note: someTOU rates may include more periods in one season than another, causing“inactive” periods. For example, a summer season may have threeperiods, and a winter season only two. The period that appears onlyduring the summer is “inactive” during the winter season.)


28

For example, assume there were two periods, peak and off-peak, and aseason change occurred during the month.

Period Previous Season Current SeasonPeak 50 150Off-Peak 100 200

Determine the valid range by calculating the sum of the periods +/- thenumber of periods, or (50 + 100 + 150 + 200) +/- 2. If the current totalkWh read from the meter was between 488 and 502 inclusive, the datawould pass the check. If the current total kWh read from the meter wasless than 488 or greater than 502, the data would fail the check.

If the meter is programmed to provide readings for all but one of theperiods, this test is modified to verify the sum of the periods withreadings is <= the total kWh.

3.6. Zero Consumption for Active Meters

The Zero Consumption checks for zero usage during the billing month.

3.6.1. If the meter is an active meter (i.e., is associated with a customer who hasfinancial responsibility), calculate the usage for the present billing month.

3.6.2. If the usage is greater than 0, the data passes the zero consumption check.

3.6.3. If the usage is 0, the data failed the zero consumption check. Optionallyverify the meter reading by re-reading the meter and/or testing themeter. If the reread is the same and the usage is still 0, the data failed theZero Consumption check but is verified. If a new, different meter readingis obtained, run all the checks again using the new data

3.6.4. Data that fails the zero consumption check may be manually investigatedand verified if justification is found (for example, a building or equipmentthat is only used seasonally). If the data is not validated or verified, itmust be estimated.

3.7. Usage for Inactive Meters

An inactive meter is one for which there is no customer with financialresponsibility. This does not apply to non-UDC MDMAs and is not a requiredcheck.

3.8. Number of Dials on Meter

This check applies to cumulative consumption only. It checks that the number of“dials” (digits) reported in the read is consistent with the number of dials (or


29

digits) on the meter display. This check is performed for both remote and localreads if supported by the meter reading technology. If the meter readingtechnology doesn’t support this check, it is not performed.

3.8.1. Determine the number of digits in the meter reading.

3.8.2. If the number of digits in the meter reading is consistent with the numberof digits/dials on the meter, the data passes this check.

3.8.3. If the number of digits is not consistent, re-read the meter to verify thatthe correct meter is being read and that it has the correct number ofdigits/dials. If the re-read provides the same values, the meter readingfailed the Number of Dials on Meter check but is verified. The situationmust be investigated and records must be corrected. If the re-readproduces different values, perform the check again with the new values.

3.9. Meter Read Demand Decimal Quantity Difference

The Meter Read Demand Decimal Quantity Difference check verifies that thenumber of demand decimal places displayed on the meter is correct. Note thischeck is only performed for on-site meter reads, and is not performed for remotemeter reads.

3.9.1. When the meter is read on-site, the meter reader compares the number ofdecimal places displayed by the meter with the number of decimal placesexpected. If they are the same, the reading passes the Meter Read DialDecimal Quantity check.

3.9.2. If they are not the same, re-read the meter to verify that the correct meteris being read and that it has the correct number of decimal places. If there-read provides the same values, the meter reading failed the Meter ReadDial Decimal Quantity Difference check but is verified. The situationmust be investigated and records must be corrected. The meter may needto be re-programmed. If the re-read produces different values, performthe check again with the new values.

3.10. Meter Identification

There are two types of Meter Identification checks depending on how the meteris read - Internal Meter Identification check and External Meter Identificationcheck. The following table summarizes when each check is required:


30

Meter Reading Method Perform External ID Check? Perform Internal ID Check?Remote No YesOptical Port Yes YesManual Yes No

1. If the meter is read remotely or via its optical port, the Internal MeterIdentification check is performed. This compares the meter’s internalidentification (often its serial number) with the identification expected by themeter reading system. If they match, the data passes this check. If they don’tmatch, the MDMA must investigate why the meter is different than indicatedby their records and resolve the inconsistency

2. If the meter is red locally (including via an optical port), the External MeterIdentification Check is performed. This compares the Meter ID on the meternameplate with the Meter ID expected by the meter reading system If theymatch, the data passes this check. If they don’t match, the MDMA mustinvestigate why the meter is different than indicated by their records andresolve the inconsistency.

3.11. Irregular Usage Monthly Data

An irregular usage customer is one whose usage pattern at a specific locationdoes not follow normal usage patterns and consistently fails the High/Lowusage or zero consumption for active meters checks. A customer may bedetermined to be an irregular usage customer by an MDMA if the MDMAverifies that:

1. the customer data fails the standard validation check for three consecutivemonths, and that the data represents the actual customer usage, OR

2. the MDMA is notified by the customer’s previous MDMA.

The data used to determine a customer is an irregular usage customer could bedata collected by the MDMA, or historical data provided by the previous ESP orMDMA. An ESP may notify the MDMA that a customer is a potential irregularusage customer based on conversations with the customer, triggering aninspection of the data. If a customer is determined to be an irregular usagecustomer, the MDMA may optionally omit the check the customer normallyfails. For example, if an irregular usage customer typically fails the High/Lowusage check but not the zero consumption check, the zero consumption checkmust still be performed but the high/low usage check may be omitted. TheMDMA must notify both the customer’s ESP and UDC of the customer’sirregular usage status and what checks will not be performed.

For some customers, irregular usage patterns are symptomatic of the business


31

and will always be present, such as agricultural or seasonal customers. For othercustomers, irregular usage patterns may be a temporary condition, such as whena factory adds a second shift and fails the high/low usage check for the first 12months. The MDMA must determine whether a customer is a permanent ortemporary irregular usage customers. Temporary irregular usage customersmust be reviewed at least annually to determine if they are still irregular usagecustomers or should be returned to the normal checks.

4. Monthly Data Estimation Rules

Note that the MDMA must record the estimation algorithm used for each data elementthat is estimated. The MDMA must retain this information for the same periodrequired for raw and validated data (3 years). Monthly data estimation algorithmsinclude:• Estimation based on previous year’s data• Estimation based on preceding billing period’s data (>= 27 days)• Estimation based on similar customers• Estimated demand based on average load• Other estimation method (MDMA must document when this is used)

4.1. Estimating Usage

Two methods to estimate usage are provided. They are similar to the twomethods of performing the High/Low Usage check. The first is based onhistorical usage for the same customer and site; the second is based on historicalusage for the same customer and site combined with a factor based on presentusage of customers of the same class and same geographic area. The number ofdecimal places included in ADU calculations must be sufficient so thatsignificant rounding errors do not occur. The recommended value is 2 decimalplaces. Final estimated usage is truncated to an integer.

4.1.1. Method 1 - Based on Historical Usage

4.1.1.1. Calculate ADU to be applied

4.1.1.1.1. If billing data is available from the same customerand same site for the same billing period last yearand it is not estimated, calculate the ADU for thesame billing period last year and use this value as theADU. Refer to section 3.3.1.2.2 to determine the samebilling period last year. Optional trend factors thattake into account peer group usage based ondemographics, climactic areas, and customer classmay be applied to the ADU to provide a more


32

accurate estimation. Sample trend factor calculationsare to be provided at a later date.

In this case, the estimation algorithm is estimationbased on previous year’s data.

4.1.1.1.2. If there is no data from the previous year but there isa full preceding billing month (at least 27 days)calculate the ADU for the preceding billing monthand use this value for the ADU. In this case, theestimation algorithm is estimation based onpreceding billing period’s data.

4.1.1.1.3. If neither of the previous two options are available,data must be estimated based on any available data,such as similar customers, load profiles, averageusage for the customer class, meter reads since lastbilling read, other historical data, etc. In this case, theestimation algorithm is other estimation algorithm.The MDMA must document how the data isestimated.

4.1.1.2. Calculate the number of days since the last good meter readingwithin the current billing cycle to the end of this billing period.If the meter is read monthly, this would typically be lastmonth’s billing meter reading. If the meter is read morefrequently, this could be more recent than last month’s billingreading.

4.1.1.3. Multiply the ADU (including any constants or factors) by thenumber of days since the last good reading. If necessary,divide this value by a meter constant or other factor to convertit to the same units reported in the meter reading. Truncate thevalue to an integer, and add the truncated value to the last goodreading to obtain an estimated reading. This is the estimatedmeter reading. Mark the reading as being estimated using theappropriate algorithm.

4.1.2. Method 2 - Based on historical usage and similar customers

4.1.2.1. For the residential meter population (i.e., same geographic areaand customer class), utilize the following determinants asdetermined in 3.3.2:

• ADU Low Range Factor (3.3.2.1.3.4)• ADU High Range Factor (3.3.2.1.3.4)


33

• Low Range Modified Mean Factor (3.3.2.1.4.1)• Medium Range Modified Mean Factor (3.3.2.1.4.1)• High Range Modified Mean Factor (3.3.2.1.4.1)

4.1.2.2. Calculate the ADU from last month’s billing period for thatcustomer.

4.1.2.3. Calculate the modified ADU for a specific meter by multiplyinglast month’s ADU (from step 4.1.2.2) by yesterday’s mediumrange modified mean factor (above) for that geographical area.

4.1.2.4. Determine if the modified ADU is in yesterday’s low, medium,or high range.

• If the modified ADU is less than the ADU low range factor,yesterday’s low range modified mean factor is used to calculateestimated ADU in the succeeding steps.

• If the modified ADU is equal to or greater than ADU low rangefactor but less than the ADU high range factor, yesterday’smedium range modified mean factor is used to calculateestimated ADU in the succeeding steps.

• If the modified ADU is greater than or equal to the ADU highrange factor, yesterday’s high range modified mean factor isused to calculate estimated ADU in the succeeding steps.

4.1.2.5. Multiply the prior ADU by the modified mean factordetermined in 4.1.2.4. This becomes the new estimated ADU.

4.1.2.6. Continue with steps 4.1.1.2, and 4.1.1.3 using the estimatedADU calculated in the preceding step.

4.2. Estimating Demand

4.2.1. If demand data is available from the same customer and same site for thesame billing period last year and it is not estimated, use that demand asthe estimated demand. Refer to section 3.3.1.2.2 to determine the samebilling period last year. In this case, the estimation algorithm is estimationbased on previous year’s data.

4.2.2. If there is no demand data from the previous year but there is demanddata from a full preceding billing month (at least 27 days), use thepreceding month’s demand as the estimated demand. In this case, theestimation algorithm is estimation based on preceding billing period’sdata


34

4.2.3. If neither of the above two options are available, calculate the averagedemand for the billing period. This is done by dividing the actual orestimated usage by the number of hours in the billing period. Use thisvalue as the estimated demand. For example, if the billing period is 30days, divide the usage for the billing period by 720 (the number of hoursin 30 days)

4.3. Estimating TOU Usage

For missing TOU usage data, each period must be estimated separately, usinghistorical data from the same TOU period (defined by time frames) and seasonas the data requiring estimation. Optional trend factors that take into accountpeer group usage based on demographics, climactic areas, and customer classmay be applied to the ADU to provide a more accurate estimation. Sampletrend factor calculations are to be provided at a later date. The number ofdecimal places included in ADU calculations must be sufficient so thatsignificant rounding errors do not occur. The recommended value is 2 decimalplaces. Final estimated usage is truncated to an integer.

4.3.1. For each period requiring estimation, the following steps are performed.Note that if there is a season change during the time period requiringestimation, each season needs to be done separately. If season crossoveroccurs in the month requiring estimation, reference data could be selectedfrom the last month with crossover between the same seasons, or the lastfull month of the season. There are two cases to consider. For example:

1. If season changes occur on October 1 and May 1, and the billingmonth April 15 to May 15 (including the season crossover) requiresestimation, reference data for the winter period may be chosen fromthe billing period that contained the October 1 crossover, or from thepreceding billing month. Reference data for summer could bechosen from the billing period that contained the October 1crossover, or from the last full summer month. If the referencemonth selected does not contain the same seasons as the monthrequiring estimation, an appropriate month containing the correctseasons should be selected.

2. If season changes occur on October 1 and May 1, and the billingmonth May 15 to June 15 requires estimation, reference data may bechosen from the billing period that contained the last full month ofsummer data, or from the summer portion of the preceding billingmonth.


35

4.3.1.1. Calculate ADU to be applied

4.3.1.1.1. If billing data is available from the same customer,same site, and same TOU period for the same billingperiod last year, and last year’s data is not estimated,calculate each period’s ADU for the same billingperiod last year and use the values as each period’sADU. Refer to section 3.3.1.2.2 to determine the samebilling period last year. For example, if the billingperiod this year was from April 13 to May 13, themid-point would be April 28. If a season changeoccurred during the month, use data from theappropriate season as reference data. Optionally, usedata from the month before or after the same billingperiod last year to get at least one week’s worth ofdata for the season. In this case, the estimationalgorithm is estimation based on previous year’s data.

4.3.1.1.2. If there is no data from the previous year but there isat least one month’s data (minimum 27 days)available from the preceding billing month and it isnot estimated, calculate each period’s ADU for thepreceding billing month and use the values for eachperiod’s ADU. If a season change occurred duringthe month, use data from the appropriate season asreference data. Optionally, use data from the monthbefore or after the same billing period last year to getat least one week’s worth of data for the season. Inthis case, the estimation algorithm is estimation basedon preceding billing period’s data.

4.3.1.1.3. If there is less than one week’s historical dataavailable, each period’s data must be estimated byother methods based on any available data, such assimilar customers, load profiles, average usage for thecustomer class, meter reads since last billing read,other historical data, etc. In this case, the estimationalgorithm must be documented. The estimationalgorithm is other estimation algorithm.

4.3.1.2. Calculate the number of days since the last good meter readingwithin the current billing cycle to the end of this billing periodfor each period requiring estimation. If the meter is readmonthly, this would typically be last month’s billing reading. If


36

the meter is read more frequently, this could be more recentthan last month’s billing reading.

4.3.1.3. For each period, multiply the ADU for that period by thenumber of days requiring estimation. This is the estimatedusage by period. Sum the periods to derive the total estimatedusage for the billing period. Mark the reading as beingestimated using the appropriate algorithm.

4.4. Estimating TOU Demand Data

For missing TOU demand data, estimate each period required for billingseparately. Note that if there is a season change during the time periodrequiring estimation, each season needs to be done separately.

4.4.1. For each period requiring estimation, the following steps are performed.Note that if there is a season change during the time period requiringestimation, each season needs to be done separately. If season crossoveroccurs in the month requiring estimation, reference data could be selectedfrom the last month with crossover between the same seasons, or the lastfull month of the season. There are two cases to consider. For example:

1. If season changes occur on October 1 and May 1, and the billingmonth April 15 to May 15 (including the season crossover) requiresestimation, reference data for the winter period may be chosen fromthe billing period that contained the October 1 crossover, or from thepreceding billing month. Reference data for summer could bechosen from the billing period that contained the October 1crossover, or from the last full summer month. If the referencemonth selected does not contain the same seasons as the monthrequiring estimation, an appropriate month containing the correctseasons should be selected.

2. If season changes occur on October 1 and May 1, and the billingmonth May 15 to June 15 requires estimation, reference data may bechosen from the billing period that contained the last full month ofsummer data, or from the summer portion of the preceding billingmonth.

4.4.1.1. If demand data is available from the same customer and samesite for the same billing period last year and it is not estimated,use that demand as the estimated demand. Refer to section3.3.1.2.2 to determine the same billing period last year. In thiscase, the estimation algorithm is estimation based on previousyear’s data.


37

4.4.1.2. If there is no demand data from the previous year but there isdemand data from a full preceding billing month (at least 27days), use the preceding month’s demand as the estimateddemand. In this case, the estimation algorithm is estimationbased on preceding billing period’s data.

4.4.1.3. If neither of the above two options are available, calculate theaverage demand for the billing period. This is done bydividing the actual or estimated usage by the number of hoursin the billing period. Use this value as the estimated demand.For example, if the billing period is 30 days, divide the usagefor the billing period by 720 (the number of hours in 30 days).


Standards for Validating, Editing, and Estimating Monthly and Interval Data

Attachment VEE-A: Interval Data VEE Technical Methods (Revision 2.0)

39

This attachment provides technical information for Section A (CA Interval Data VEERules) of Attachment VEE.

1.1. Prorating Time Drifted Data

Two options are provided for correcting the data when the actual number ofintervals retrieved from the meter does not equal the expected number ofintervals based upon the elapsed time.

1.1.1. Option 1 for Prorating Time Drifted Data

This section describes how to normalize interval data when the clock inthe meter does not agree with the clock in the computer reading themeter. This phenomenon is called Clock Drift. Clock drift can be both anegative or positive value, depending upon whether the real time (at thecomputer) is greater than [negative drift] or less than [positive drift] thanthe clock in the meter. This is illustrated below.

For each of the illustrations shown above, the actual interval ofmeasurement is different. We assume that the meter and the computersystems are synchronized at some time, T0 {for example, the last meterread} and that the meter is now being read at a read time, Tr. For eachcase above, assume that the meter reading system reads at the same time.The elapsed time is given by

Elapsed Time = ∆∆Te = Tr - T0

t∆

0 1 2 3 4 5 6 7 8 9 . . .

t∆

10 2 3 4 5 6 7 8 9 10 11 12 13 . . .

NormalTime

ShortInterval

t∆

1 2 3 . . .0 4 5 6Time

1Long

Interval



40

The elapsed time for the shorter interval case is given by

Short Interval Elapsed Time = ∆∆Tes = Tms - T0

where Tms is the time that the meter clock gives when the meter is read.The elapsed time for the longer interval case is given by

Longer Interval Elapsed Time = ∆∆Tel = Tms - T0

where Tml is the time that the meter clock gives when the meter is read.Note that Tms > T0 , i.e. the meter clock is running faster therefore clockingmore intervals and more elapsed time. And that Tml < T0, i.e., the meterclock is running slower, hence fewer intervals and a shorter elapsed time.

In each case, the internal clock in the meter is registering that the intervallength is the length that is specified for the meter, namely ∆t or 15minutes for CA interval meters. However, since the clock is runningfaster for the short interval case and slower for the long interval case, wemust adjust the values such that the correct usage is obtained for eachmeter. The total drift time for the clock can be calculated for each case asfollows:

Total Drift Time For Short Intervals TDs = ∆∆Te - ∆∆Tes= Tr - Tms

and

Total Drift Time For Long Intervals TDl = ∆∆Te - ∆∆Tel= Tr - Tml

Note that TDs would be negative had we not taken the absolute value.

The actual number of intervals for each case can be calculated using theelapsed time measured by the meter and dividing it by the preset meterinterval, ∆t or 15 minutes. Or when truncated to an integer:

Expected Number of Intervals, N = ∆∆Te / ∆∆t

Number of Short Intervals Ns = ∆∆Tes / ∆∆t

Number of Long Intervals Nl = ∆∆Tel / ∆∆t



41

Two cases are considered.

Case # 1 is when N = Ns = Nl, i.e. the clock drift is small enough that noadditional interval is generated. For this situation, we could elect to donothing since for a thirty day reading schedule for 15 minute intervaldata, there will be about 2880 intervals and a time drift error of less than0.04% in each interval. The actual drift time for each interval, ignoringany fractional intervals at the end of the read period can be calculated as

∆∆DTs = TDs / N

and for the long interval case,

∆∆DTl = TDl / N

This result could be used in combination with the actual interval width toincrease or decrease the usage in the particular interval.

Case #2 involves a situation when N ≠ Ns or N ≠ Nl and hence there are afewer or greater number of intervals that expected. Here we suggest thata procedure be adopted that distributes the measured interval values intointervals that have the correct length, namely ∆t. Thus we increase ordecrease the interval length to correspond to actual interval length thatshould have been in the meter based on the drift, ∆t + ∆DTl or ∆t - ∆DTs.Thus the interval distribution that is shown in the above figure wouldnow have proper interval lengths in minutes. The procedure fordistribution this data is fairly straightforward. We simply insert the real15-minute interval grid on the actual interval grid taken from the meterand adjusted by the above technique. Next we divide each of theobtained usage values at the corresponding 15-minute grid points.



42

This figure shows how to allocate each of the respective values. Forexample in the first corrected interval, the usage would equal the sum ofthe usage in the incorrect time interval case plus the fraction to the left ofthe corresponding interval in the correct zone. For example, if the driftper interval was found to be 4 minutes for short interval case, then thetime grid would be 0, 11, 22, 33, 44, 55, 66, 77, etc. The amount taken fromthe second short interval would be the usage in interval 2 multiplied by4/11 or 34%. For the case of the third correct interval, there would becomponents from interval 3, 4, and 5 of the incorrect interval usage. Fromthe third interval, it would be 3/11 of the usage, from the fourth intervalit would be 100% of the usage, and from the fifth interval it would be1/11 of the usage. This scheme should be easy to implement. The onecaution is that additional manipulation of the usage data will be requiredif the interval values are stored as kW for the interval, rather than kWh.But this should not be a problem.

1.1.2. Option 2 for Prorating Time Drifted Data

The objective of this algorithm is to create the expected number ofintervals while preserving the usage recorded in the actual number ofintervals.

When the Actual number > Expected Number…

1) Truncate the extra intervals.

2) The total usage will decrease by the amount recorded in the truncatedintervals. In order to preserve the recorded usage, the usage in theremaining intervals will need to be scaled up as described below.

ShortInterval

t∆

10 2 3 4 5 6 7 8 9 10 11 12 13 . . .

Time

t∆

0 1 2 3 4 5 6 7 8 9 . . .

NormalTime



43

When the Actual number < Expected Number…

1) Interpolate using the last good interval to create the expected numberof intervals.

2) The total usage will increase by the amount in the interpolatedintervals. In order to preserve the recorded usage, the usage in eachinterval will need to be scaled down as described below.

To scale the usage in the truncated or interpolated intervals…

1) Calculate the recorded usage.

Recorded Usage = total usage for the period. It can be derived from thesum of the usage in the “actual” intervals or from good meter readings.

2) Calculate the pre-scaled usage.

Pre-scaled Usage = The total usage in the intervals after they have beentruncated or interpolated to the expected number of intervals.

3) Calculate the scaling factor.

Scaling Factor = (Recorded Usage/Pre-scaled usage).

4) Multiply the usage in each interval by the scaling factor to createcorrected usage. The sum of the usage in the scaled intervals shouldnow be equal to the recorded usage.

5) Flag all intervals for the period as estimated.

1.2. Sum Check Failure Troubleshooting Techniques

The objective of the sum check is to compare the energy use recorded by themeter to the energy use recorded by the pulse recorder over the same timeperiod. Due to data collection methods, often the period represented by themeter reads does not correspond exactly to the period represented by theinterval data. For example, the period of data collection may span from 5/1/9801:12 AM to 6/1/98 01:22 AM, with the meter readings corresponding exactly tothis time period. With 15-minute interval data, the interval data for this sameperiod of data collection would begin at 5/1/98 01:00 AM and end at 6/1/9801:15 AM. The difference of 12 minutes from the start meter reading and 7minutes from the end meter reading could be the source of error in the failure ofthe sum check.



44

1.2.1. Account for Start and End Time Differences

The following technique enables the MDMA to resolve sum check failuresby taking into account time differences between the meter readings andthe interval data.

Redo the sum check, taking into account the differences in time betweenthe time of the start read and the start of the first interval, and the time ofthe stop read and the end of the last interval:

1) Calculate a prorated start meter reading to be used in this check by doingthe following:

(a) Calculate the percentage of an interval that has elapsed between thestart time of the first interval and the time of the start meter reading.For example, if the meter was read at 3:30 PM, the first interval in anhourly interval data stream would start at 3:00 PM. The percentage oftime elapsed is (30 min./60 min.) = 50%.

(b) Multiply the usage from the first interval by the percentage from theprevious step. For example, if the usage in the first interval is 240kWh, the percentage usage is (240*0.50) = 120 kWh.

(c) Determine how many meter increments are represented by thepercentage usage from the previous step. For a meter multiplier ofone, the usage is equal to the number of meter increments, so 120 kWhis equal to 120 meter increments. For a meter multiplier of 80, 120kWh is equal to 1 meter increment (i.e., 120 divided by 80 androunded down to the nearest integer).

(d) Calculate a prorated start meter reading by subtracting the number ofmeter increments from the previous step from the actual start meterreading. For example, if the start meter reading is 55555, and thenumber of meter increments is equal to 120, the prorated start meterreading would be (55555-120) = 55435.

2) Calculate an allowable margin of error to be used in this check by doingthe following:

(a) Calculate the percentage of an interval that has elapsed between theend time of the last interval and the time of the stop meter reading.For example, if the meter was read at 11:15 AM, the last interval in anhourly interval data stream would start at 11:00 AM. The percentageof time elapsed is (15 min./60 min.) = 25%.



45

(b) Multiply the usage from the last interval by the percentage from theprevious step. For example, if the usage in the last interval is 120kWh, the percentage usage is (120*0.25) = 30 kWh.

(c) Determine how many meter increments are represented by thepercentage usage from the previous step. For a meter multiplier ofone, the usage is equal to the number of meter increments, so 30 kWhis equal to 30 meter increments. For a meter multiplier of 80, 30 kWhwould result in .375 meter increments.

(d) Calculate the allowable margin of error by adding 2 to the valuecalculated in the previous step.

(e) Redo the sum check using the prorated start and original stop meterreadings and the allowable margin of error instead of the twomultipliers.

1.2.2. Account for Missing or Incomplete Intervals

With some metering and data collection technologies, it is possible for themeter or cumulative usage register to reflect accurate usage even whenthe interval data is missing or incomplete. The following techniqueenables the MDMA to resolve the sum check failure for those intervalsthat were successfully collected.

If some intervals are missing or incomplete, redo the sum check afterscaling the difference between the adjusted start read and the stop readby the percentage of good intervals:

1) Count the number of good intervals in the data stream.

2) Calculate the percentage of good intervals by dividing the count fromthe previous step by the number of intervals elapsed between starttime and stop time.

3) Multiply the percentage by the difference between the start readingand the stop reading. (Note that you may use the actual start and stopreadings or the prorated start and stop readings from 1.2.1 in thisstep.)

4) Compare the new difference with the sum of the usage in the goodintervals. Note that the values must be in the same units for thecomparison.



46

5) If the difference is <= allowable margin from 1.2.1, the good intervalspass the sum check. The missing or incomplete intervals need to beestimated.

1.3. Scaling estimated data using good meter readings

If start and stop meter readings are available and are known to be good, theymay optionally be used to scale the estimated interval data as follows:

1) Determine the total usage for the time period based upon the meter readings.

Total Usage = ((Stop Reading-Start Reading)*Meter Multiplier)

2) Sum together the valid intervals.

3) Subtract the sum of the valid intervals from the total usage to determine thetotal estimated usage.

Total Estimated Usage = Total Usage – Sum of Valid Intervals

4) Sum together the previously estimated intervals.

5) Calculate the scaling factor by dividing the total estimated usage by the sumof the estimated intervals.

Scaling Factor = Total Estimated Usage/Sum of Estimated Intervals

6) Multiply each estimated interval by the scaling factor.


Attachment VEE-B: Monthly Data VEE Technical Methods

47

This attachment provides an example of validating and estimating usage based onprevious day usage of similar customers. The text is a duplicate of the correspondingsections (Sections 3.3.2 and 4.1) in CA Monthly VEE Rules. The numbers to the left ofthe text correspond to the numbers on the spreadsheet at the end of the attachment.

3.3.2. Method based on previous day usage of similar customers

Note that this method requires a certain density of customer data forresidential customers in the same geographic area, where weatherpatterns are typically consistent throughout the geographic area.

3.3.2.1. The following steps are performed at the end of each meterreading cycle day for each geographical area in order tovalidate and estimate usage the following day:

3.3.2.1.1. At the end of the reading day, for each good meterread (open account, billed, between 27-33 days &ADU =< 100), perform the following calculations todetermine an ADU for the billing period:

1. Calculate ADU (= KWH/days in billing period)2. Add ADU to Sum of Current ADU3. Calculate ADU squared4. Add ADU squared to Sum of Current ADUsquared5. Add 1 to total meters6. Calculate last month’s ADU7. Calculate current ADU times last month’s ADU8. Calculate last month’s ADU squared

3.3.2.1.2. Determine which range of usage (high, medium orlow) the current ADU should be grouped with bycomparing current ADU to yesterdays ADU Low andHigh Range Factors (Reference 3.3.2.1.3 for ADU lowand high range factor calculation methodology)

1. If the current ADU is less than yesterday’s ADUlow range factor:

• Add current ADU to Sum of current low ADU• Add last month’s ADU to Sum of last month’s low

ADU• Add current ADU times last month’s ADU to Sum

of current low ADU times last month’s ADU• Add current ADU squared to Sum of current low

1

2

34

5

6

Represents the numericalassociation on the attachedspreadsheet calculations.Used for mapping only.

#



48

ADU squared• Add last month’s ADU squared to Sum of last

month’s low ADU squared• Add 1 to total low meters 2. If the current ADU is not less than the ADU low

range factor from yesterday and is less than theADU high range factor, add the figures to themedium range following same format is in3.3.2.1.2. step 1.

3. Otherwise, add the current ADU to the ADU high

range following the same format in 3.3.2.1.2.step 1.

3.3.2.1.3. Calculate an aggregated ADU for current data foreach geographic area

3.3.2.1.3.1. Sum together the ADU values for eachgeographic area

3.3.2.1.3.2. Calculate the mean for the total ADU(=Sum of Current ADU / total meters)

3.3.2.1.3.3. Calculate the standard deviation for thetotal

3.3.2.1.3.4. Calculate the current ADU low and highrange factors:

• ADU Low Range Factor = mean -.43 StandardDeviation. If ADU Low range factor is less thanthe total current mean * .5, the ADU low rangefactor becomes the mean half.

• ADU High Range Factor = mean + .43 StandardDeviation

• NOTE: By determining the low & high factors,the Medium Range = (mean - .43 StandardDeviation) to (mean + .43 Standard Deviation)

3.3.2.1.4. For each of the three ranges determined above (low,medium, and high), calculate a percent of change ofmonthly usage for each geographic area.

3.3.2.1.4.1. After each meter’s current billing period’s(ADU) is grouped in 3 Ranges (Low,

8

9

7

10

11

12



49

Medium, and High) as specified in3.3.2.1.2, the following data are summedup by ADU range and area:

• Number of customers• Sum of all last month’s ADU• Sum of all current month’s ADU• Sum of {each last month’s ADU times

current month’s ADU}• Sum of {all last month’s ADU

squared} i.e., Square all ADU, thensum them.

• Sum of {all current month’s ADUsquared}

From the data above modified ADUmean factors and standard deviationfactors are determined for each range asfollows:

Modified Mean Factor:Sum of {last month ADU times currentmonth ADU} divided by the sum of {alllast month’s ADU squared}

Modified Standard Deviation Factor:Step 1: (Sum of {all current month’s ADU

squared} minus (Mean squared times sumof {all last month’s ADU squared})divided by (Total Meters minus 1).

Step 2: Take square root of Step 1

3.3.2.1.5. Calculate high and low range factors.

Calculate high and low range factors (HRF and LRF)for each of the 3 usage ranges within a geographicarea. The mean is used to calculate estimated reads,and the high and low range factors are used in thisvalidation check. 2.8 and 3.5 are used in the belowexample to represent the range deviation factor andwill allow for an appropriate meter read error rate.This factor can be changed to control the error rate.

High Range Factor Formula:HRF = 1 + {(2.8X Modified Standard Deviation XNumber of meters) / (Sum of Current month’s ADU)}

20

21

1314

15

17

16

18

19



50

Low Range Factor Formula:LRF = 1 - {(3.5 X Modified Standard Deviation XNumber of meters) / (Sum of Current month’s ADU)}

3.3.2.2. As each meter is read, perform the following using the valuescalculated from the previous meter reading days’ data.

3.3.2.2.1. Determine the usage from the preceding billingmonth and the preceding billing reading for thecustomer and site.

3.3.2.2.2. Calculate low limit for this month’s usage bymultiplying the preceding month’s usage by the lowrange factor determined above.

3.3.2.2.3. Calculate high limit for this month’s usage bymultiplying the preceding month’s usage by the highrange factor determined above.

3.3.2.2.4. If the current usage is between the low and high limitcalculated in the previous two steps, the data passesthe High/Low check.

4.1. Estimating Usage

4.1.2. Method 2 - Based on historical usage and similar customers

4.1.2.1. For the residential meter population (i.e., same geographic areaand customer class), utilize the following determinants asdetermined in 3.3.2:

• ADU Low Range Factor (3.3.2.1.3.4)• ADU High Range Factor (3.3.2.1.3.4)• Low Range Modified Mean Factor (3.3.2.1.4.1)• Medium Range Modified Mean Factor (3.3.2.1.4.1)• High Range Modified Mean Factor (3.3.2.1.4.1)

4.1.2.2. Calculate the ADU from last month’s billing period for thatcustomer.

4.1.2.3. Calculate the modified ADU for a specific meter by multiplyinglast month’s ADU (from step 4.1.2.2) by yesterday’s mediumrange modified mean factor (above) for that geographical area.

23

24

22

25

26

27

28



51

4.1.2.4. Determine if the modified ADU is in yesterday’s low, medium,or high range.

• If the modified ADU is less than the ADU low range factor,yesterday’s low range modified mean factor is used tocalculate estimated ADU in the succeeding steps.

• If the modified ADU is equal to or greater than ADU lowrange factor but less than the ADU high range factor,yesterday’s medium range modified mean factor is used tocalculate estimated ADU in the succeeding steps.

• If the modified ADU is greater than or equal to the ADUhigh range factor, yesterday’s high range modified meanfactor is used to calculate estimated ADU in the succeedingsteps.

4.1.2.5. Multiply the prior ADU by the modified mean factordetermined in 4.1.2.4. This becomes the new estimated ADU.

4.1.2.6. Continue with steps 4.1.1.2, and 4.1.1.3 using the estimatedADU calculated in the preceding step.

30

29


Attachment C-VEE-B: Monthly Data VEE Technical Methods

Spreadsheet Calculations:

52

# of Meters 19# of Days in Month 30

Meter Current Current Last Month Last Month Meter Current ADU * Current ADU Last Month RangeNumber Usage ADU Usage ADU Reading Last Month ADU Squared ADU Squared Grouping

1 125 4.17 135 4.50 3731 18.75 17.36 20.25 Low2 275 9.17 250 8.33 7475 76.39 84.03 69.44 High3 114 3.80 142 4.73 3697 17.99 14.44 22.40 Low4 178 5.93 165 5.50 4888 32.63 35.20 30.25 Medium5 254 8.47 260 8.67 7358 73.38 71.68 75.11 High6 299 9.97 189 6.30 6835 62.79 99.33 39.69 High7 178 5.93 190 6.33 5278 37.58 35.20 40.11 Medium8 158 5.27 136 4.53 4176 23.88 27.74 20.55 Medium9 165 5.50 140 4.67 4329 25.67 30.25 21.78 Medium10 235 7.83 255 8.50 7033 66.58 61.36 72.25 High11 218 7.27 235 7.83 6500 56.92 52.80 61.36 High12 110 3.67 119 3.97 3286 14.54 13.44 15.73 Low13 98 3.27 105 3.50 2912 11.43 10.67 12.25 Low14 85 2.83 129 4.30 3117 12.18 8.03 18.49 Low15 169 5.63 110 3.67 3913 20.66 31.73 13.44 Medium16 200 6.67 168 5.60 5221 37.33 44.44 31.36 High17 147 4.90 178 5.93 4688 29.07 24.01 35.20 Low18 165 5.50 201 6.70 5281 36.85 30.25 44.89 Medium19 180 6.00 135 4.50 4446 27 36.00 20.25 Medium

Sum 111.77 108.07 681.63 727.99 664.82Mean 5.88Std Dev 1.98 Ranges Low HighADU Low Range Factor 5.03 Low 0 5.03ADU High Range Factor 6.73 Medium 5.03 6.73

High 6.73 or higher

Meters Curr ADU Last mo ADU Curr * Last Mo ADU Curr Mo sqrd Last Mo ADU SqdLow 6 22.63 26.93 103.97 87.95 124.33Medium 7 39.77 35.90 204.26 226.38 191.27High 6 49.37 45.23 373.40 413.66 349.22

Modified Mean

Mod Std Deviation

Low Range Factor

High Range Factor

Low 0.84 0.45 0.58 1.33Medium 1.07 1.17 0.28 1.58High 1.07 1.70 0.28 1.58

Example validation routine for meter read the following day using the determinants calculated above:

Customer's Last Month's Usage = 200 kWh ADU = 6.67 Customer's usage falls in Medium Range

High Range Value: 200 kWh X 1.58 = 316Low Range Value: 200 kWh X 0.28 = 56

Usage falling between 56 and 316 is acceptable. Usage outside of this range fails the high/low usage check.

1 23 4 5 6

Assume yesterday's ADU low range factor was 4.95 and the ADU high range factor was 6.4.

789

1011 12

1315 1814 16 17

19 20 2122

Example estimation routine for meter read the following day using the determinants calculated above:

Customer's Previous Usage = 200 kWh ADU = 6.67

Modified ADU = 6.67 X 1.07 (Medium Range Modified Mean Factor) = 7.14 Modified ADU

Determine where 7.14 is grouped according to the ADU low and high range factors (5.03 & 6.73 respectively) 7.14 > 6.73, therefore, the high range modified mean factor of 1.07 is used.

ADU Estimated usage = last month's ADU (6.67) X 1.07 = 7.14, rounded to 7Total usage = ADU Estimated usage X number of days in the billing period. 30

2928

26

24

27

25

23

Date post:	13-Nov-2020
Category:	Documents
Upload:	others
View:	3 times
Download:	0 times

Direct Access Standards for Metering and Meter Data (DASMMD...Table I-1: Standards Required for...

Documents