20191227Gas Standards Manual 2020.pdf

Avista Corp.
l4l I East Mission P.O. Box 3727
Spokane, Washington 99220-0500
Telephone [phone]
Toll Free 800-727-9170
Dec,ernber 26,2019
Diane Hanian, Secretary
Idaho Public Utilities Commission
1 l33l W. Chinden Blvd., Bldg. 8, Ste. 201-A
Boise, Idaho 83717
Re: AW-G-I8-07
Dear Ms. Hanian:
Sincerely,
lt{Lind,wGenai,*{
Linda Gervais
Senior Manager, Regulatory Policy
Avista Utilities
509-495-4975
I inda. gervais(r4avistacorp.com
Enclosure
=C]'c-i mno C)-rm?4ri I
t$@
.ii'iststa
Enclosed for electronic filing with the Commission is Avista Corporation's, dba Avista
Utilities ("Avista" or "the Company"), 2020 Gas Standards Manual and Standard Operating
Procedures for its Gas Meter Measurement Performance Program ("SOP"). This filing is being made
in compliance with the Commission's Order No. 34178 in Case No. AVU-G-18-07, and with the
Company's associated natural gas tariff LP.U.C. No 27, Schedule 170, which states that Avista is to
provide electronic copies ofthese documents to the Commission upon request or whenever substantive
changes are made.
Due to the voluminous nature of tie Gas Standards Manual, the first 26 pages of the document
provide a summary of the changes made for 2020.
Ifyou have any questions regarding this filing, please contact Jaime Majure at [phone]
or iaime.maiure(D.avistacorp.com.
AVISTA UTILITIES
GAS METER
MEASUREMENT
PERFORMANCE PROGRAM
(PMC PROGRAM)
GAS METER SHOP - STANDARD OPERATING PROCEDURE
Title: Gas Meter Measurement Performance Program Date: 12-10-19
Jurisdiction: Washington, ldaho, Oregon Manager: Jeff Webb PaBe 1 of 13
CONTE,NTS
I. CONTENTS2. SCOPE3. GENERAL
A. COMPLIANCEb. HISTORY
4. GAS METER MEASUREMENT PROGRAM REQUIREMENTS
5. APPENDX -COMPLIANCE STANDARDSa. Avista ID Gas Tariff Rule No. 27, Section 25, effective November 1,2018
b. Avista OR Gas Tariff Rule No. 18, effective dates April 1, 2008 thru January I , 2012c. OAR 860-023 @ublic Utility Commission)
d. Avista WA Gas TariffRule No. 170, Section 20, effective November 30, 201 1
e. WAC 480-90 "Gas companies - operations", Section 333, 338, 343, &348
GAS METER SHOP - STANDARD OPERATING PROCEDURE
Title: Gas Meter Measurement Performance Program Date:12-10-19
Jurisdiction: Washington, ldaho, Oregon Manager: Jeff Webb PaBe 2 of 13
Deviations or changes to this SOP (Standard Operating Procedure) shall be approved by the Gas
Measurement Manager. Material changes require concurrence with the governing commission. Non-
material changes shall be submitted to the commissions in conjunction with annual meter performance
statistics.
GAS METER SHOP - STANDARD OPERATING PROCEDURE
Title: Gas Meter Measurement Performance Program Date: 12-10-19
Jurisdiction: Washington, ldaho, Oregon Manager: Jeff Webb Page 3 of 13
SCOPE
This procedure covers the methodology, testing requirements, and annual reporting guidelines for
Avista's gas meter measurement performance testing program (PMC Program) for new and in-service
meters.
GENERAL
COMPLIANCE
Idaho: AvistalD Gas Tariff Rule No.27, Section 25, effective November 1,2018, has been approved
and prescribes the minimum inspection and testing requirements.
Oregon: Gas meler testing requirements for Avista are promulgated by Oregon Administrative Rules,
Chapter 860, Division 023 "Service Standards", Section 0015 (Testing Gas and Electric Meters). Avista
Gas Tariff Rule No. 18, Section 20, effective September 1,2012, has been approved and is in compliance
with the requirements of the applicable OAR and prescribes the minimum inspection and testing
requirements. It is the intent that this program is in accordance with the aforementioned Oregon
Administrative Rules.
l;l/ashington: Gas meter testing requirements for Avista are promulgated by Washingxon Administrative
Code, Chapter480-90 "Gas companies - operations", Section 333 (lnitial accuracy ofmeters), Section
338 (Metering tolerance), Section 343 (Statement of meter test procedures), and Section 348 (Frequency
of periodic meter tests. Avista Gas Tariff Rule No. 170, Section 20, effective November 30, 201 l, has
been approved and is in compliance with the requirements of WAC 480-90 and prescribes the minimum
inspection and testing requirements. It is the intent that this program is in accordance with WAC 480-90.
HISTORY
Avista's current random meter measurement performance program is in accordance with ANSI Zl.9
(lnspection by Variables). Random sampling and testing is completed for all domestic meters 1000 CFH
and smaller. The program was revised and upgraded in 2009. The prior program was in accordance with
inspection standard ANSI Zl .4 (hspection by Attributes). The change in inspection standard and
programmatic changes were completed in order to enhance the quality and accuracy ofthe measurement
performance program.
In 201I the gas meter testing program was enhanced as follows:
Random sampling progam - Redefined a meter population. Previously a meter population was
defined as meters ofthe same model, installed in the same State jurisdiction, and purchased in the
same year. The revised definition for a meter population now includes meters of the same model,
size, and manufactured in the same year. The new definition removes the criteria to test meters by
state. A statistical analysis of meter testing results for all three (3) states revealed that the mean
accuracy and standard deviation of accuracy about the mean are similar for all three states and
that any difference is not statistically significant, indicating that state test results can be
combined. Random sampling gas meter testing requirements for 2012 were developed using the
revised meter population defi nition.
Installed turbine Meters: Changes were made to expand turbine testing requirements to include
both single rotor and dual rotor auto adjust meters.
Installed rotary meters: Changes were made to provide clarification regarding necessary test
requirements for rotary meters.
GAS METER SHOP - STANDARD OPERATING PROCEDURE
Title: Gas Meter Measurement Performance Program Date:12-10-19
Jurisdiction: Washington, ldaho, Oregon Manager: leff Webb PaBe 4 of 13
In 2019 a new subpart titled "Meter Data Managemenl Calibration" was added to section VI Installed
Meter Testing Program, part c. This subpart defines an altemative means to address meter families that
are experiencing a consistent drift in mean accuracy identified by the sample meter testing program.
Because these families are experiencing a consistent drift in accuracy, the meter accuracy can be adjusted
back into company specification by changing the Installation Constant value in Avista's Meter Data
Management system rather than replacing the meter. Meter families experiencing this drift in accuracy
will continue to be tested under tightened inspection and their Installation Constant reevaluated every
year.
GAS METER SHOP - STANDARD OPERATING PROCEDURE
Title: Gas Meter Measurement Performance Program Date: 12-10-19
Jurisdlction: Washington, ldaho, Oregon Manager: Jeff Webb Page 5 of 13
GAS METER MEASUREMENT PROGRAM REQUIREMENTS
I. Tests of Customer Meters
Avista shall maintain a quality control program to veriff performance accuracy ofnew
meters and installed meters.
b. Testine of Small Capacitv Meters. All meters with capacities up to and including one
thousand ( I,000) cubic feet per hour (clh) at %" differential pressure that have been in
service ten ( l0) or more years as established by either the last install date or manufacture
date shall be tested tkough annual random sampling using sample size methods in
accordance with ANSVASQ-Z1.9 or other generally accepted statistical method that
conforms to the requirements ofthe applicable s€ctions ofANSI B.l09.l and 8109.2
measuring in-service performance of diaphragm type gas meters.
Testing of Lareer Capacitv Meters. Al[ meters with capacities greater than one thousand
(1,000) cfh that have been in service ten (10) years as established by either last install or
manufacture date shall be replaced or field tested based on a periodic schedule.
Diaphragm type meters shall be tested in accordance with applicable sections ofANSI
Bl09.l andBl09.2. Rotary type meters shall be tested in accordance with applicable
sections of ANSI B 109.3.
II. Customer Meter Accuracy
Accuracv of Meters
i. Diaphragm Meters:
L A new gas meter installed for the use of any customer shall not be more
than one (l) percent slow or fast.
2. Every meter removed from service and opened for repair shall have the
meter tolerance adjusted to 100% plus or minus 0.57o, with no greater
spread than 0.7% between the check and open prior to being reinstalled.
3. Meters not opened for repairs may be reinstalled without adjustment if
found to be not more than two (2) percent in error slow or fast, per
Section III (Customer Meter Test Loads).
ii. Rotary Meters
1. A new gas meter installed for the use of any customer shall not be more
than one (1) percent slow or fast.
2. Meters inspected by differential pressure testing shall confirm that the
meter is performing within 150% ofthe manufacturer's specification for
differential pressure at the operating pressure.
3. Every meter tested by proving shall be adjusted to be not more than two
(2) percent slow or fast before being reinstalled.
iii. Turbine Meters
a
c
a
GAs METER SHOP - STANOARD OPERATING PROCEDURE
Title: Gas Meter Measurement Performance Program Date:12-10-19
Jurisdiction: Washington, ldaho, Oregon Manager: JeffWebb Page 6 of 13
I . A new gas meter installed for the use of any customer shall not be more
than one (l) percent slow or fast.
2. Meters inspected by spin testing shall confirm that the meter is
performing within the manufacture's limits.
3. Every meter tested by proving shall be adjusted to be not more than two
(2) percent slow or fast before being reinstalled.
b. Removal of Defective Meters from Service. No meter that is mechanically defective shall
be placed in service or allowed to remain in service after the defect has been discovered.
When any gas meter is not connected and i.n service, the inlet and outlet will be capped to
prevent the drying out of diaphragms and contamination ofthe meter with dirt and debris.
The exception shall be when stored indoors during the testing and/or repair process.
III. Customer Meter Test Loads
a. Testing of Meters, Test Equipment. All tests to daermine the accuracy of registrations of
gas service meters shall be made with a suitable meter prover or testing equipment as
detailed in this standard.
b. Gas Flows During Testing. The flow rate to be used in testing meters, except orifice ty?e
meters and high volume meters with capacity in excess oftest equipment capability, will
be at twenty (20) percent (Check) and one hundred (100) percent (Open) oftheir
capacity. For the purpose of determining the accuracy ofthese meters, the average ofthe
two results shall be used.
c. High Volume Meters. Turbine and rotary meters that have rated capacities in excess of
the mechanical limits of the proving equipment shall be tested at a minimum of two (2)
points of flow with each at or above a minimum of l5% ofthe rated capacity ofthe
meter. For purposes ofthe test the two results shall be averaged and shall conform to the
standards set forth in Section II (Customer Meter Accuracy).
IV. Meter Test Procedures and Test Records
a. Reporting
i. After December 3l of each year, meter test results will be summarized, analyzed
and forwarded to all the three State Utility Commissions. Retention and filing of
records will be in accordance with the appropriate Cornmission requirements.
b. Test Records
After December 3l of each year, meter test results will be summarized and
analyzed. Armual results ofall meter accuracy tests will be tabulated and
retained in accordance with the appropriate Commission as follows:
l. Idaho: Avista ID Gas Tariff Rule No. 27, Section25
2. Oregon: OAR 860-023-0015(4)
3. Washinglon: WAC 480-90-353
Avista shall keep records oftests ofthe accuracy of each of its meters, which
may be superseded by a later test, but shall be maintained for not less than two
These records shall VC:
l1
GAS METER SHOP - STANDARD OPERATING PROCEDURE
Date: 12-10-19Title: Gas Meter Measurement Performance Program
Manager: Jeff Webb Page 7 of 13Jurisdiction: Washington, ldaho, Oregon
I . sufficient information to identiff the meter,
2. the reason for the test,
3. the date ofthe test and reading ofthe meter,
4. an identification of the person making the test, and
5 . accuracy, as found and as left, together with enough of the data taken at
the time ofthe test to permit the convenient checking ofthe methods
employed and the calculations.
Technical Standard. Technical performance requirements for diaphragm natural gas
meters shall be per the current versions of ANSIBl09.1 and ANSI B109.2. Technical
performance requirements for rotary type natual gas meters shall be per the current
versions of ANSI B109.3.
d. Uniquely Defective Test An individual meter test resuh within a random sample of
more than l0% error shall be declared a uniquely defective test and disregarded. A
substitute test will be made with meter selected fiom within the same test family. Subject
meter shall be adjusted or replaced.
V. New Meters
a. New meters shall be factory tested and certified to mget accuracy criteria specified by
Section II (Customer Meter Accuracy).
i. Manufacturers shall provide results ofnew meter testing that conforms to
generally accepted practice for meter accuracy and performance. Diaphragm
meters shall meet the standards set forth in ANSI B 109.1 and ANSI B.109.2.
Rotary meters shall meet the standards set forth in ANSI B109.3.
b. Inspection and acceptance testing by Avista prior to installation of new meters.
i. Each meter shipment will be inspected for physical damage. Meters found to be
damaged or in damaged packaging will be tested, repaired and/or calibrated or
returned to the manufacturer as described herein. All costs for tests, retum
shipping and,/or calibration to meters described in this section shall be bome by
the manufactur€r.
ii. Normal acceptance testing described herein, not associated with physical damage
found on arrival ofthe shipment, will be performed for all meters not factory
tested prior to shipment. Expanded testing beyond normal test quantities for
shipments found to be non-conforming through acceptance testing will be paid
for by the manufacturer or the shipment retumed to the manufacturer per
negotiations between Avista and the manufacturer.
c. New diaphragm t)?e meters, less than or equal to 1000 CFH:
i. The methodology for the random sampling of new meters is derived fiom
ANSyASQ 21.9 or other generally accepted statistical method that conforms to
the requirements ofthe applicable sections ofANSI 8.109.1 and B109.2.
I . The lot size to determ ine random sample quantity shall be the size of the
shipment.
2. Acceptable Quality Limit (AQL) for analysis of new meters will equal
1.5.
GAS METER SHOP - STANDARD OPERATING PROCEDURE
Title: Gas Meter Measurement Performance Program Date: 12-10-19
Jurisdiction: Washington, lda ho, Oregon Manager: leff Webb Page 8 of 13
C
ii. Sample size: is based on Normal inspection levels
1 . Normal inspection is the default inspection level.
2. Sample size for meter types with five (5) test histories are eligible for
application of switching rules in general accordance with the guidelines
contained within the Standard as modified below.
3. Switching Rules:
a. Normal to tightened: Switching nrles for transition from normal
to tightened inspection shall be applied if two (2) out offive (5)
consecutive lots have been rejected on original inspection.
Testing beyond the sample size for normal inspection level will
be paid for by the manufacturer.
b. Tightened to Normal: Switching nrles for transition from
tightened to normal inspection may be applied when testing has
been at the tightened level and five (5) consecutive batches have
been acceptable on original inspection.
c. Normal to Reduced: Switching rules for transition from normal
to reduced impection (inspection level II to inspection level II!
may be applied if:
i. Preceding five (5) lots have been on normal inspection
and none have been rejected, AND
ii. Meter model has been in steady production without
major design modifications (as determined by the
Company).
d. Reduced to Normal: Switching rules for transition fiom reduced
to normal (inspection level III to inspection level II) shall be
applied if:
i. A batch is rejected under reduced inspection, OR
ii. Meter model has not been in steady production or a
major design modification has occurred (as determined
by the Company).
4. Random sampling per the Standard shall be discontinued if five (5)
consecutive lots are not accepted. Acceptance procedures as detailed
herein shall not be resumed until corrective action has been taken.
d. New diaphragm meters, greater than 1000 CFH.
i. New large capacity diaphragm meters shall be tested against metering tolerance
of Section II.
ii . Meters found to be outside of tolerance shall be adj usted per section IL If the
meter cannot be adjusted within these standards, it shall be returned to the
manufacturer.
e. New rotary meters.
i. Rotary meters shall meet the accuracy requirements as specihed in Section II and
appropriate requirements of ANSI B 109.3.
f. New turbine meters.
GAs METER SHOP - STANDARD OPERATING PROCEDURE
Title: Gas Meter Measurement Performance Program Date: 12-10-19
Jurisdiction: Washington, ldaho, Oregon Manager: Jeff Webb Page 9 of 13
New turbine meters shall meet the accuracy requirements as specified in Section
II and shall be spin tested at installation to verifu they will operate within the
accuracy limits as specified by the manufacture.
VL lnstalled meter testing program
a. General: Meters shall be inspected and tested against metering tolerance prescribed in
Section II. Meters found to be outside the tolerances shall be immediately adjusted or
replaced.
b. Aee: Testing requirements are applicable to meters that are 10 years old or older. The
first year of required testing is calculated as Manufacturing year + 9.
Domestic meters. diaohrasm tvbe I000 CFH and smaller randoml v samnled: A random
sample of domestic meters shall be selected, tested against tolerances prescribed by
Section II, and analysis conducted using the Standard. The annual random sampling
program shall begin during the tenth ( 1Oft) year after meter manufacture date. The
methodology for sample sizes and analysis for the meter testing progam is derived from
ANSVASQ 21.9 (hereafter may be referred to as the Standard) or other generally
accepted statistical method that conforms to the requirements ofthe applicable sections of
ANSI 8.109.1 and B 109.2 measuring in-service performance of gas meters.
i . The lot size to determine random sample quantity shall be the size of the meter
population. A meter population should include randomly sampled meters in
approximate proportion to the number of meters installed in each operating area.
ii. A meter population is defined as meters ofthe same model, size, and
manufactured year.
1. Meter population is synonymous with the term lot as used in the
Standard.
2. Major design changes to a meter model within a single year shall be a
new population for sampling
iii. All of the meters within a population will be identified by a "flag" in Avista's
Service Suite mobile application at the beginning of a calendar year. Meters
shall be chosen at random and in sufficient quantities to meet the glidelines for
sampling as detailed in the standard. Once the required number of meters in each
family is removed for testing the "flag" will be removed in Service Suite
indicating that no more meters in that family are required for testing. Meter
populations of 50 or fewer units are subject to modified sample size. Sample
quantity shall be set to avoid repeat testing of a meter more than once every ten
years.
iv. Inspection lrvels, Sample Size and Test result analysis:
l. Genera Analysis ofpopulation sample results shall conform to the
guidelines contained in the Standard. Acceptable Quality Limit (AQL)
for analysis will equal 10.0.
2. Normal inspection is the default level ofinspection.
3. Inspection levels for meter types with frve (5) year test histories are
eligible for application of switching rules.
4. Switching rules:
GAS METER SHOP - STANDARD OPERATING PROCEDURE
Title: Gas Meter Measurement Performance Program Date: 12-10-19
Jurisdiction: Washington, ldaho, Oregon Manager: Jeff Webb Page 10 of 13
a. Normal to tightened: Switching rules for transition from normal
to tightened inspection shall be applied if two (2) lots out offive
(5) consecutive lots have been rejected.
b. Tightened to Normal: Switching rules for transition from
tightened to normal inspection may be applied when testing has
been at the tightened level and five (5) consecutive batches have
been acceptable.
c. Normal to Reduced: Switching rules for transition from normal
to reduced inspection may be applied if:
i. Preceding ten (10) lots have been on normal inspection
and none have been rejected.
d. Reduced to Normal: Switching rules for transition fiom reduced
to normal shall be applied if:
i. A batch is rejected.
5. !'ailure ofmeter population. A meter population shall be declared
defective (failed) when:
a. Three (3) consecutive yearly inspections for a population under
tightened inspection are not accepted based on AQL of 10.0 for
overall performance (double specification limit) 4IID76A
b. Two (2) consecutive yearly inspections for a population under
tightened inspection are not accepted based on AQL of 10.0 for
overall performance (single specification limit) which measures
meters rurming fast, also referred to as Failed Fast.
These failed meter families shall either be removed from service as
described by subpart 6 or have their lnstallation Constant updated in the
Meter Data Management system as described in subpat 7.
6. Removal of failed meter population: Removal of a meter population of
less than 2500 units may take place within one ( I ) year. Removal of
meter population of more than 2500 units may take place over two (2)
years. Removal of meter population of more than 5000 units may take
place over three (3) years. Removal of meter population of more than
7500 units may take place over four (4) years.
7. Meter Data Manaeement calibration: If a statistical analysis shows that a
meter family is experiencing a consistent drift in mean accuracy, the
meter inaccuracy may be corrected by adjusting the entire family's
lnstallation Constant value in the Meter Data Management system rather
than removing the meters from service. Meter families experiarcing this
drift in accuracy will continue to be tested under tightened inspection and
their lnstallation Constant reevaluated every year.
d. Installed diaphragm meters, greater than 1000 CFH, periodic test.
i. 1001 CFH through 3000 CFH: Inspected and proved every ten (10) years or
sooner.
ii. Larger than 3000 CFH: Inspected and proved every five (5) years or sooner.
iii. All meters shall be tesled against metering tolerance of Section II.
GAS METER SHOP - STANDARD OPERATING PROCEDURE
Title: Gas Meter Measurement Performance Program Date: 12-10-19
Jurisdiction: Washington, ldaho, Oregon Manager: Jeff Webb Page 11 of 13
iv. Meters found to be outside of tolerance shall be adjusted to within the tolerances
as specified in Section II.
e. Installed rotary meters, periodic test.
i. Testing ofinstalled rotary meters will be done using either a portable transfer
prover or differential testing device. A differential test shall confirm a minimum
oftwo (2) points. Results ofa differential test or prover test shall show that the
meter is performing within the accuracy as detail in Section II.
ii. Meters shall be inspected and tested every five (5) years or sooner.
iii. Meters found to be outside of tolerance shall be adjusted to within the tolerances
as specified in Section II or removed from service.
f. Installed turbine meters, periodic test.
i. lnstalled single rotor turbine meters shall be inspected and spin tested annually
and determined to be within the manufactures acceptability limits. Turbine
meters failing the spin test shall be removed from the held and repaired prior to
any subsequent installation.
ii. Installed auto adjust meters shall be inspected annually and determined to be
operating within the manufactures acceptability limits. Meter Delta A (AA), the
difference between the main and sensing rotor pulses, that exceed the
manufactures recommended operating parameters shall be repaired or replaced.
iii. AII turbine meters will be tested with a transfer proven or at an approved offsite
facility at a maximum interval of every ten (10) years. Tested meters shall meet
or exceed the meter accuracy requirements of Section II.
Vll. Meter Test Equipment and Application
a. Meter test equipment
i. Transfer Prover. The accuracy of the testing equipment is ascertained through:
L Monthly, in-house self-testing procedures.
2. Sending of Standard Meter Module to the manufacturer for periodic
calibration. The period between factory calibrations shall not exceed
five (5) years.
ii. Sonic Nozzle Prover and Bel[ Prover.
I . The accuracy of the testing equipment is ascertained through an
automatic test diagnostic. The diagnostic is completed each time the
prover is powered on.
2. The test equipment shall be factory calibrated every two (2) yeals.
b. Meters shall be tested on either: Sonic Nozzle, Bell Prover, Transfer Prover, or at an
approved offsite facility.
GAs METER SHOP - STANDARD OPERATING PROCEDURE
Title: Gas Meter Measurement Performance Program Date: 12-10-19
Jurisdiction: Washington, ldaho, Oregon Manager: Jeff webb Page 12 of 13
APPENDIX _ COMPLIANCE STANDARDS
. Avista ID Gas Tariff Rule No. 27, Section 25e Avista OR Gas TariffRule No. l8. OAR 860-023 (Public Utility Commission). Avista WA Gas Tariff Rule No. 170, Section 20. WAC 480-90 "Gas companies - operations", Section 333, 338,343, & 348
GAS METER SHOP - STANDARD OPERATING PROCEDURE
Title: Gas Meter Measurement Performance Pro8ram Date:12-10-19
Jurisdiction: WashinBton, ldaho, Oregon Manager: J eff Webb Page 13 of 13