Energy Conservation Program: Test Procedures for Central Air Conditioners and Heat Pumps

DEPARTMENT OF ENERGY

10 CFR Parts 429 and 430

[Docket No. EERE-2016-BT-TP-0029]

RIN 1904-AD71

Energy Conservation Program: Test Procedures for Central Air Conditioners and Heat Pumps

AGENCY:

Office of Energy Efficiency and Renewable Energy, Department of Energy.

ACTION:

Final rule.

SUMMARY:

On August 24, 2016, the U.S. Department of Energy (DOE) published a supplemental notice of proposed rulemaking (SNOPR) to amend the test procedure for central air conditioners and heat pumps. That SNOPR serves as the basis for this final rule. This final rule amends the test procedure and specific certification, compliance, and enforcement provisions related to this product. In this final rule, DOE makes two sets of amendments to the test procedure: Amendments to appendix M that would be required as the basis for making efficiency representations starting 180 days after final rule publication and a new appendix M1 that would be the basis for making efficiency representations as of the compliance date for any amended energy conservation standards. The new appendix M1 establishes new efficiency metrics SEER2, EER2, and HSPF2 that are based on the current efficiency metrics for cooling and heating performance, but generally have different numerical values than the current metrics. Broadly speaking, the amendments address off-mode test procedures, test set-up and fan delays, external static pressure conditions for testing, represented values for CAC/HP that are distributed in commerce with multiple refrigerants, the methodology for testing and calculating heating performance, and testing of variable-speed systems.

DATES:

The effective date of this rule is February 6, 2017. The final rule changes of appendix M will be mandatory for representations of efficiency starting July 5, 2017. Representations using appendix M1 will be mandatory starting January 1, 2023. The incorporation by reference of certain publications listed in Appendix M1 is approved by the Director of the Federal Register on February 6, 2017 February 6, 2017. The incorporation by reference of certain publications listed in Appendix M was approved by the Director of the Federal Register as of July 8, 2016.

ADDRESSES:

The docket, which includes

Federal Register

notices, public meeting attendee lists and transcripts, comments, and other supporting documents/materials, is available for review at

regulations.gov

. All documents in the docket are listed in the

regulations.gov

index. However, some documents listed in the index, such as those containing information that is exempt from public disclosure, may not be publicly available.

The docket Web page can be found at

https://www.regulations.gov/docket?D=EERE-2016-BT-TP-0029

. The docket Web page will contain simple instruction on how to access all documents, including public comments, in the docket.

FOR FURTHER INFORMATION CONTACT:

Ashley Armstrong, U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, Building Technologies Program, EE-2J, 1000 Independence Avenue SW., Washington, DC 20585-0121. Telephone: (202) 586-6590. Email:

Ashley.Armstrong@ee.doe.gov

.

Johanna Jochum, U.S. Department of Energy, Office of the General Counsel, GC-33, 1000 Independence Avenue SW., Washington, DC, 20585-0121. Telephone: (202) 287-6307. Email:

Johanna.Jochum@hq.doe.gov

.

For further information on how to review public comments and the docket contact the Appliance and Equipment Standards Program staff at (202) 586-6636 or by email:

CACHeatPump2016TP0029@ee.doe.gov

.

SUPPLEMENTARY INFORMATION:

This final rule incorporates by reference into part 430 specific sections, figures, and tables in the following industry standards:

(1) ANSI/AHRI 210/240-2008 with Addenda 1 and 2, (“AHRI 210/240-2008”): 2008 Standard for Performance Rating of Unitary Air-Conditioning & Air-Source Heat Pump Equipment, ANSI approved October 27, 2011;

(2) ANSI/AHRI 1230-2010 with Addendum 2, (“AHRI 1230-2010”): 2010 Standard for Performance Rating of Variable Refrigerant Flow (VRF) Multi-Split Air-Conditioning and Heat Pump Equipment, ANSI approved August 2, 2010.

Copies of AHRI 210/240-2008 and AHRI 1230-2010 can be obtained from the Air-Conditioning, Heating, and Refrigeration Institute, 2111 Wilson Boulevard, Suite 500, Arlington, VA 22201, USA, 703-524-8800, or by going to

http://www.ahrinet.org/site/686/Standards/HVACR-Industry-Standards/Search-Standards.

(3) ANSI/ASHRAE 23.1-2010, (“ASHRAE 23.1-2010”): Methods of Testing for Rating the Performance of Positive Displacement Refrigerant Compressors and Condensing Units that Operate at Subcritical Temperatures of the Refrigerant, ANSI approved January 28, 2010;

(4) ANSI/ASHRAE Standard 37-2009, (“ANSI/ASHRAE 37-2009”), Methods of Testing for Rating Electrically Driven Unitary Air-Conditioning and Heat Pump Equipment, ANSI approved June 25, 2009;

(5) ANSI/ASHRAE 41.1-2013, (“ANSI/ASHRAE 41.1-2013”): Standard Method for Temperature Measurement, ANSI approved January 30, 2013;

(6) ANSI/ASHRAE 41.6-2014, (“ASHRAE 41.6-2014”): Standard Method for Humidity Measurement, ANSI approved July 3, 2014;

(7) ANSI/ASHRAE 41.9-2011, (“ASHRAE 41.9-2011”): Standard Methods for Volatile-Refrigerant Mass Flow Measurements Using Calorimeters, ANSI approved February 3, 2011;

(8) ANSI/ASHRAE 116-2010, (“ASHRAE 116-2010”): Methods of Testing for Rating Seasonal Efficiency of Unitary Air Conditioners and Heat Pumps, ANSI approved February 24, 2010;

(9) ANSI/ASHRAE 41.2-1987 (Reaffirmed 1992), (“ASHRAE 41.2-1987 (RA 1992)”): “Standard Methods for Laboratory Airflow Measurement”, ANSI approved April 20, 1992.

Copies of ASHRAE 23.1-2010, ANSI/ASHRAE 37-2009, ANSI/ASHRAE 41.1-2013, ASHRAE 41.6-2014, ASHRAE 41.9-2011, ASHRAE 116-2010, and ASHRAE 41.2-1987 (RA 1992) can be purchased from ASHRAE's Web site at

https://www.ashrae.org/resources--publications

.

(10) ANSI/AMCA 210-2007, ANSI/ASHRAE 51-2007, (“AMCA 210-2007”) Laboratory Methods of Testing Fans for Certified Aerodynamic Performance Rating, ANSI approved August 17, 2007.

Copies of AMCA 210-2007 can be purchased from AMCA's Web site at

http://www.amca.org/store/index.php

.

For a further discussion of these standards, see section IV.M.

Table of Contents

I. Authority and Background

A. Authority

B. Background

II. Synopsis of the Final Rule

III. Discussion

A. Testing, Rating, and Compliance of Basic Models of Central Air Conditioners and Heat Pumps

1. Representation Accommodation

2. Highest Sales Volume Requirement

3. Determination of Represented Values for Multi-Split, Multi-Circuit, and Multi-Head Mini-Split Systems

4. Service Coil Definition

5. Efficiency Representations of Split-Systems for Multiple Refrigerants

6. Representation Limitations for Independent Coil Manufacturers

7. Reporting of Low-Capacity Lockout for Air Conditioners and Heat Pumps With Two-Capacity Compressors

8. Represented Values of Cooling Capacity

9. New Efficiency Metrics

B. Amendments to Appendix M Testing To Determine Compliance With the Current Energy Conservation Standards

1. Measurement of Off Mode Power Consumption: Time Delay for Units With Self-Regulating Crankcase Heaters

2. Refrigerant Pressure Measurement Instructions for Cooling and Heating Heat Pumps

3. Revised EER and COP Interpolation Method for Units Equipped With Variable-Speed Compressors

4. Outdoor Air Enthalpy Method Test Requirements

5. Certification of Fan Delay for Coil-Only Units

6. Normalized Gross Indoor Fin Surface Area Requirements for Split Systems

7. Modification to the Test Procedure for Variable-Speed Heat Pumps

8. Clarification of the Requirements of Break-In Periods Prior to Testing

9. Modification to the Part Load Testing Requirement of VRF Multi-Split Systems

10. Modification to the Test Unit Installation Requirement of Cased Coil Insulation and Sealing

11. Correction for the Calculation of the Low-Temperature Cut-Out Factor for Single-Speed Compressor Systems

12. Clarification of the Refrigerant Liquid Line Insulation

C. Amendments to Appendix M1

1. Minimum External Static Pressure Requirements

2. Default Fan Power for Rating Coil-Only Units

3. Revised Heating Load Line Equation

4. Revised Heating Mode Test Procedure for Units Equipped With Variable-Speed Compressors

D. Effective Dates and Representations

1. Effective Dates

2. Comment Period Length

3. Representations From Appendix M1 Before Compliance Date

E. Comments Regarding the June 2016 Final Rule

1. Determination of Represented Values for Single-Split Systems

2. Alternative Efficiency Determination Methods

3. NGIFS Limit for Outdoor Unit With No Match

4. Definitions

5. Inlet Plenum Setup

6. Off-Mode Power Consumption

IV. Procedural Issues and Regulatory Review

A. Review Under Executive Order 12866

B. Review Under the Regulatory Flexibility Act

C. Review Under the Paperwork Reduction Act of 1995

D. Review Under the National Environmental Policy Act of 1969

E. Review Under Executive Order 13132

F. Review Under Executive Order 12988

G. Review Under the Unfunded Mandates Reform Act of 1995

H. Review Under the Treasury and General Government Appropriations Act, 1999

I. Review Under Executive Order 12630

J. Review Under Treasury and General Government Appropriations Act, 2001

K. Review Under Executive Order 13211

L. Review Under Section 32 of the Federal Energy Administration Act of 1974

M. Description of Materials Incorporated by Reference

N. Congressional Notification

V. Approval of the Office of the Secretary

I. Authority and Background

A. Authority

Title III, Part B

1

of the Energy Policy and Conservation Act of 1975 (“EPCA” or “the Act”), Public Law 94-163 (42 U.S.C. 6291-6309, as codified) sets forth a variety of provisions designed to improve energy efficiency and established the Energy Conservation Program for Consumer Products Other Than Automobiles.

2

These products include central air conditioners and central air conditioning heat pumps,

3

(single-phase

4

with rated cooling capacities less than 65,000 British thermal units per hour (Btu/h)), which are the focus of this Final Rule. (42 U.S.C. 6291(1)-(2), (21) and 6292(a)(3))

1

For editorial reasons, Part B was codified as Part A in the U.S. Code.

2

All references to EPCA in this document refer to the statute as amended through the Energy Efficiency Improvement Act of 2015, Public Law 114-11 (Apr. 30, 2015).

3

This rulemaking uses the term “CAC/HP” to refer specifically to central air conditioners (which include heat pumps) as defined by EPCA. 42 U.S.C. 6291(21.)

4

Where this rulemaking uses the term “CAC/HP”, they are in reference specifically to central air conditioners and heat pumps as defined by EPCA.

Under EPCA, DOE's energy conservation program generally consists of four parts: (1) Testing; (2) labeling; (3) Federal energy conservation standards; and (4) certification, compliance, and enforcement. The testing requirements consist of test procedures that manufacturers of covered products must use as the basis of: (1) Certifying to DOE that their products comply with applicable energy conservation standards adopted pursuant to EPCA, and (2) making other representations about the efficiency of those products. (42 U.S.C. 6293(c); 42 U.S.C. 6295(s)) Similarly, DOE must use these test procedures to determine whether covered products comply with any relevant standards promulgated under EPCA. (42 U.S.C. 6295(s))

EPCA sets forth criteria and procedures DOE must follow when prescribing or amending test procedures for covered products. (42 U.S.C. 6293(b)(3)) EPCA provides, in relevant part, that any test procedures prescribed or amended under this section shall be reasonably designed to produce test results which measure the energy efficiency, energy use, or estimated annual operating cost of a covered product during a representative average use cycle or period of use, and shall not be unduly burdensome to conduct.

Id.

In addition, if DOE determines that a test procedure amendment is warranted, it must publish proposed test procedures and offer the public an opportunity to present oral and written comments on them. (42 U.S.C. 6293(b)(2)) Finally, in any rulemaking to amend a test procedure, DOE must determine to what extent, if any, the amended test procedure would alter the measured energy efficiency of any covered product as determined under the existing test procedure. (42 U.S.C. 6293(e)(1))

The Energy Independence and Security Act of 2007 (EISA 2007), Public Law 110-140, amended EPCA to require that, at least once every 7 years, DOE must review test procedures for all covered products and either amend the test procedures (if the Secretary determines that amended test procedures would more accurately or fully comply with the requirements of 42 U.S.C. 6293(b)(3)) or publish a notice in the

Federal Register

of any determination not to amend a test procedure. (42 U.S.C. 6293(b)(1)(A))

DOE's existing test procedures for CAC/HP adopted pursuant to these provisions appear under Title 10 of the Code of Federal Regulations (CFR) part 430, subpart B, appendix M (“Uniform Test Method for Measuring the Energy Consumption of Central Air Conditioners and Heat Pumps”). These procedures establish the currently permitted means for determining energy efficiency and annual energy consumption for CAC/HP. The procedures established in the new appendix M1 include new efficiency metrics to represent cooling and heating performance whose values will be altered as compared to the current metrics. The new metrics include seasonal energy efficiency ratio 2 (SEER2), energy efficiency ratio 2 (EER2), and heating seasonal performance factor 2 (HSPF2). Use of the test procedures of appendix M1 will become mandatory to demonstrate compliance on the compliance date of revised energy conservation standards.

Section 310 of EISA 2007 established that the Department's test procedures for all covered products must account for standby mode and off mode energy consumption. (42 U.S.C. 6295(gg)(2)(A)) For CAC/HP, standby mode is

incorporated into the SEER and HSPF metrics, while off mode power consumption is separately regulated. This final rule includes changes relevant to the determination of both SEER and HSPF (including standby mode) and off mode power consumption.

B. Background

DOE initiated a round of test procedure revisions for CAC/HP by publishing a notice of proposed rulemaking in the

Federal Register

on June 2, 2010 (June 2010 NOPR; 75 FR 31223). Subsequently, DOE published several supplemental notices of proposed rulemaking (SNOPRs) on April 1, 2011 (April 2011 SNOPR; 76 FR 18105), on October 24, 2011 (October 2011 SNOPR: 76 FR 65616), and on November 9, 2015 (November 2015 SNOPR; 80 FR 69277) in response to comments received and to address additional needs for test procedure revisions. The June 2010 NOPR and the subsequent SNOPRs addressed a broad range of test procedure issues. On June 8, 2016, DOE published a test procedure final rule (June 2016 final rule) that finalized test procedure amendments associated with many but not all of these issues. 81 FR 36991.

On November 5, 2014, DOE published a request for information for energy conservation standards (ECS) for CAC/HP (November 2014 ECS RFI). 79 FR 65603. In response, several stakeholders provided comments suggesting that DOE amend the current test procedure. The November 2015 SNOPR addressed those test procedure-related comments, but, as mentioned in this preamble, not all of the related issues were resolved in the June 2016 final rule.

On July 14, 2015, DOE published a notice of intent to form a Working Group to negotiate a NOPR for energy conservation standards for CAC/HP and requested nominations from parties interested in serving as members of the Working Group. 80 FR 40938. The Working Group, which ultimately consisted of 15 members in addition to one member from Appliance Standards and Rulemaking Federal Advisory Committee (ASRAC) and one DOE representative, identified a number of issues related to testing and certification. The term sheet summarizing the Working Group recommendations included several recommendations associated with test procedures. (CAC ECS: ASRAC Term Sheet, No. 76)

5

5

This final rule addresses proposals and comments from two rulemakings: (1) Stakeholder comments and proposals regarding the CAC test procedure (CAC TP: Docket No. EERE-2009-BT-TP-0004); and (2) stakeholder comments and proposals regarding the CAC energy conservation standard from the Working Group (CAC ECS: Docket No. EERE-2014-BT-STD-0048). Comments received through documents located in the test procedure docket are identified by “CAC TP” preceding the comment citation. Comments received through documents located in the energy conservation standard docket (EERE-2014-BT-STD-0048) are identified by “CAC ECS” preceding the comment citation. Further, comments specifically received during the CAC/HP ECS Working Group meetings are identified by “CAC ECS: ASRAC Public Meeting” preceding the comment citation.

On August 24, 2016 DOE published a SNOPR (August 2016 SNOPR) proposing several amendments to the test procedure and to certification, compliance, and enforcement provisions, including a proposal to establish a new appendix M1 to be used for testing under any new energy conservation standard. 81 FR 58164. That SNOPR addressed issues not resolved by the June 2016 final rule and also proposed test procedure amendments to implement several of the items summarized in the ASRAC Working Group Term Sheet.

II. Synopsis of the Final Rule

In this final rule, DOE revises the certification requirements and test procedure for CAC/HP based on public comment on various published materials and the ASRAC negotiation process discussed in section I.B. This final rule establishes two sets of test procedure changes: One set of changes to appendix M (effective 30 days after publication of a final rule and required for testing and determining compliance with current energy conservation standards); and another set of changes to create a new appendix M1 that would be used for testing to demonstrate compliance with any amended energy conservation standards (agreed compliance date of January 1, 2023, by the Working Group in the CAC rulemaking negotiations (CAC ECS: ASRAC Term Sheet, No. 76)). With the exceptions discussed in sections III.B.3 and III.B.7, the changes to appendix M do not alter measured efficiency. However, the new appendix M1 establishes new efficiency metrics for cooling and heating performance, SEER2, EER2, and HSPF2.

In this final rule, DOE makes the following changes to certification requirements:

(1) Codifying the CAC/HP ECS Working Group's recommendation regarding delayed implementation of testing to demonstrate compliance with amended energy conservation standards;

(2) Relaxing the requirement that a split system's tested combination be a high sales volume combination;

(3) Revising requirements for certification of multi-split systems in light of the adoption of multiple categories of duct pressure drop that the indoor units can provide;

(4) Making explicit certain provisions of the service coil definition;

(5) Revising the certification of separate individual combinations within the same basic model for each refrigerant that can be used in a model of split system outdoor unit and certification of details regarding the indoor units with which unmatched outdoor units are tested;

(6) Revising representation limitations for independent coil manufacturers;

(7) Revising the certification of low-capacity lockout for air conditioner and heat pumps with two capacity compressors;

(8) Revising the requirements for represented values of cooling and heating capacity; and

(9) Adding new efficiency metrics SEER2, EER2, and HSPF2 to reflect the changes in the test procedure that result in significant change in the efficiency metric values.

DOE implements the following changes to appendix M:

(1) Requiring a limit on the internal volume of lines and devices connected to measure pressure at refrigerant circuit;

(2) Revising the method to calculate EER and coefficient of performance (COP) for variable-speed units for calculating performance at intermediate compressor speeds;

(3) Requiring a 30-minute test without the outside-air apparatus connected (a “free outdoor air” test) to be the official test as part of all cooling and heating mode tests which use the outdoor air enthalpy method as the secondary measurement;

(4) Relaxing the requirement for secondary capacity checks, requiring instead use of a secondary capacity measurement that agrees with the primary capacity measurement to within 6 percent only for the cooling full load test and, for heat pumps, for the heating full load test;

(5) Revising the certification of the indoor fan off delay used for coil-only tests;

(6) Modifying the test procedure for variable-speed heat pumps; and

(7) Modifying the part load testing requirement of VRF multi-split systems and test unit installation requirement of cased coil insulation and sealing.

DOE adopts the following provisions for new appendix M1:

(1) New higher external static pressure requirements for all units, including unique minimum external static pressure requirements for mobile home systems, ceiling-mount and wall-mount systems, low- and mid-static multi-split systems, space-constrained systems, and small-duct, high-velocity systems;

(2) A unique default fan power for rating mobile home coil-only units and new default fan power for all other coil-only units;

(3) Revisions to the heating load line equation in the calculation of the heating mode efficiency metric, HSPF2;

(4) Amendments to the test procedures for variable-speed heat pumps that change speed at lower ambient temperatures and add a 5 °F heating mode test option for calculating full-speed performance below 17 °F; and

(5) Establishment of a 4-hour or 8-hour delay time before the power measurement for units that require the crankcase heating system to reach thermal equilibrium after setting test conditions.

The test procedure amendments to appendix M for subpart B to 10 CFR part 430 established in this final rule pertaining to the efficiency of CAC/HP will be effective 30 days after publication in the

Federal Register

(referred to as the “effective date”). Pursuant to EPCA, manufacturers of covered products are required to use the applicable test procedure as the basis for determining that their products comply with the applicable energy conservation standards. (42 U.S.C. 6295(s)) 180 days after publication of a final rule, any representations made with respect to the energy use or efficiency of CAC/HPs are required to be made in accordance with the results of testing pursuant to the amended test procedures. (42 U.S.C. 6293(c)(2))

The test procedures established in this final rule for appendix M1 to subpart B of 10 CFR part 430 pertaining to the efficiency of CAC/HP are effective 30 days after publication in the

Federal Register

. The appendix M1 procedures will be required as the basis for determining that CAC/HP comply with any amended energy conservation standards (if adopted in the concurrent CAC/HP energy conservation standards rulemaking) and for representing efficiency as of the compliance date for those amended energy conservation standards.

DOE revises the test procedure and requirements for certification, compliance, and enforcement in this final rule effective on February 6, 2017. The amended test procedure of appendix M is mandatory for representations of efficiency as of July 5, 2017. The new test procedure of appendix M1 is mandatory for representations of efficiency as of January 1, 2023.

III. Discussion

This section discusses the revisions to the certification requirements and test procedure that DOE adopts in this final rule.

A. Testing, Rating, and Compliance of Basic Models of Central Air Conditioners and Heat Pumps

1. Representation Accommodation

In the August 2016 SNOPR, DOE proposed to implement the following recommendations from the CAC/HP ECS Working Group regarding representations for split systems in 10 CFR 429.16 and 429.70:

○ DOE will implement the following accommodation for representative values of split system air conditioners and heat pumps based on the M1 methodology:

○ By January 1, 2023, manufacturers of single-split systems must validate an AEDM that is representative of the amended M1 test procedure by:

Testing a single-unit sample for 20-percent of the basic models certified.

The predicted performance as simulated by the AEDM must be within 5 percent of the performance resulting from the test of each of the models.

Although DOE will not require that a full complement of testing be completed by January 1, 2023, manufacturers are responsible for ensuring their representations are appropriate and that the models being distributed in commerce meet the applicable standards (without a 5% tolerance).

○ By January 1, 2023, manufacturers must either determine representative values for each combination of single-split-system CAC/HP based on the M1 test procedures using a validated AEDM or through testing and the applicable sampling plan.

○ By January 1, 2023, manufacturers of multi-split, multi-circuit, or multi-head mini-split systems must determine representative values for each basic model through testing and the applicable sampling plan.

○ By July 1, 2024, each model of condensing unit of split system CAC/HP must have at least 1 combination whose rating is based on testing using the M1 test procedure and the applicable sampling plan. 81 FR at 58167 (Aug. 24, 2016)

Lennox and AHRI commented that they supported DOE's proposal, although AHRI noted it supported DOE's proposal with certain exceptions. (Lennox, No. 25 at p. 2; AHRI, No. 27 at p. 1) While AHRI did not note the exceptions, DOE assumes these may be related to their comments regarding test requirements for two-stage air conditioners (Id at p. 2), effective dates for appendix M in the June 2016 Final Rule and this final rule (Id at p. 8), and AEDM options for multi-split systems (Id at p. 20). These issues are discussed separately in III.D and III.E. As these exceptions are tangential to the original proposal, DOE has adopted the accommodations as proposed.

2. Highest Sales Volume Requirement

In the August 2016 SNOPR, based on recommendations by the CAC/HP ECS Working Group, DOE proposed removing the requirement for single-split-system air conditioners that the individual combination required for testing be the highest sales volume combination (HSVC). Specifically, DOE proposed that for every basic model, a manufacturer must test the model of outdoor unit with a model of indoor unit.

6

81 FR at 58202 (Aug. 24, 2016)

6

As adopted in the June 2016 Final Rule, for single-split-system air conditioners with single-stage or two-stage compressors, the model of indoor unit must be coil-only.

ACEEE, NRDC, ASAP, and NEEA supported DOE's proposal to adopt the CAC/HP ECS Working Group recommendations regarding removing the HSVC, as described in the SNOPR. (ACEEE, NRDC, and ASAP, No. 33 at p. 8; NEEA, No. 35 at p. 1) DOE received no other comment on this issue. Therefore, DOE adopts this proposal in this final rule. DOE notes that some stakeholders commented on related items that were finalized in the June 2016 Final Rule. These are discussed in section III.E.1.

3. Determination of Represented Values for Multi-Split, Multi-Circuit, and Multi-Head Mini-Split Systems

In the August 2016 SNOPR, DOE proposed that multi-split, multi-head mini-split, and multi-circuit systems could be tested and rated with five kinds of indoor units: Non-ducted, low-static ducted, mid-static ducted, conventional ducted, or small-duct, high velocity (SDHV). DOE proposed that when determining represented values (including certifying compliance with amended energy conservation standards), at a minimum, a manufacturer must test and rate a “tested combination” composed entirely of non-ducted units. Under the proposed rule, if a manufacturer were to offer the model of outdoor unit with

models of low-static, mid-static, and/or conventional ducted indoor units, the manufacturer would be required, at a minimum, also to test and rate a second “tested combination” with the highest static variety of indoor unit offered. The manufacturer would also be allowed to choose to test and rate additional “tested combinations” composed of the lower static varieties. In each case, the manufacturer would test with the appropriate external static pressure. DOE did not propose use of AEDMs for these systems. 81 FR at 58169 (Aug. 24, 2016)

DOE also proposed to maintain its requirement from the June 2016 final rule that, if a manufacturer also sells a model of outdoor unit with SDHV indoor units, the manufacturer must test and rate the SDHV system (

i.e.,

test a combination with indoor units that all have SDHV pressure capability). DOE also proposed to continue to allow mix-match ratings across any two of the five varieties by taking a straight average of the ratings of the individual varieties, and to allow ratings of individual combinations through testing. 81 FR at 58169 (Aug. 24, 2016)

NEEA commented that they supported DOE's proposals regarding certification of multi-split, multi-circuit, and multi-head mini-split systems. (NEEA, No. 35 at p. 1-2) Lennox and Nortek commented that they supported DOE's proposals regarding tested combinations for multi-split, multi-head mini-split, and multi-circuit systems. (Lennox, No. 25 at p. 3-4; Nortek, No. 22 at p. 3) AHRI commented that they supported DOE's proposals regarding tested combinations for multi-split and multi-circuit systems. (AHRI, No. 27 at p. 2)

AHRI and Mitsubishi commented that they were concerned with DOE's proposal to add low-static and mid-static testing requirements to appendix M. They commented that the “low-static” and “mid-static” terminology and the associated testing requirements were negotiated for appendix M1, and implementing this requirement before the effective date of the 2023 standard would not be in alignment with the Working Group's recommendation. (AHRI, No. 27 at p. 2-3; Mitsubishi, No. 29 at p. 2)

DOE notes that it intended the low-static and mid-static requirements to apply to appendix M1 only. In the August 2016 SNOPR, 10 CFR 429.16(a)(1) and (b)(2)(i) included tables regarding determining represented values and minimum testing requirements. In both of these tables, DOE only discussed the static variety in regards to testing in accordance with M1 or making representations on and after January 1, 2023. In addition, the definitions for the static varieties are only found in appendix M1. However, DOE acknowledges that 10 CFR 429.16(c)(3) may have included unclear language on this topic. DOE has modified this language in this final rule.

AHRI and Mitsubishi commented that multi-head mini-split systems do not belong in the requirements for multi-split and multi-circuit systems because they operate as 1-to-1 combinations, and it is not possible to turn off one indoor unit for testing. In addition, they stated that these systems do not have multiple-ducted and non-ducted combinations. AHRI and Mitsubishi requested that DOE remove multi-head mini-split systems from non-applicable testing requirements and other sections and instead include multi-head mini-split in the same line as “Single-Split-System” in the table in 10 CFR 429.16(b)(2). (AHRI, No. 27 at p. 2; Mitsubishi, No. 29 at p. 1-2; Mitsubishi, Public Meeting Transcript, No. 20 at p. 113-114)

In response, DOE notes that, though the August 2016 SNOPR proposed additional requirements regarding tested combinations, the certification and testing requirements for multi-head mini-split systems became associated with the testing requirements for multi-split and multi-circuit systems in the June 2016 final rule, and were not proposed in the August 2016 SNOPR. The only related change proposed in the August 2016 SNOPR pertains to requirements for different static varieties. Furthermore, although multi-head mini-split systems are grouped with multi-split and multi-circuit systems in the certification requirements, appendix M and M1 do not require this equipment to turn off any indoor units during testing. In addition, DOE does not believe, based on the information provided by AHRI and Mitsubishi, that the proposed language in 10 CFR 429.16 presents a problem for multi-head mini-split systems. The certification and testing requirements allow only non-ducted representations if that is all that is sold, or representations of only one kind of ducted combination, if that is all that is sold. The fact that multi-head mini-split systems are sold in few combinations should not preclude manufacturers from meeting these requirements. For these reasons, DOE is not removing multi-head mini-splits from its grouping with multi-split and multi-circuit systems in 10 CFR 429.16.

DOE received no other comment on the proposals in the August 2016 SNOPR for determining represented values for multi-split, multi-circuit, and multi-head mini-split systems and DOE adopts all of the proposed requirements in this final rule. DOE also notes that in the August 2016 SNOPR, DOE omitted mention in 10 CFR 429.16(a)(1) that non-SDHV multi-split, multi-circuit, and multi-head mini-split systems may also include space-constrained units, so DOE has clarified that in this final rule.

4. Service Coil Definition

In the June 2016 final rule, to distinguish newly installed cased and uncased coils from replacement cased and uncased coils, DOE added a definition for service coils and explicitly excluded them from indoor units in the indoor unit definition.

In the August 2016 SNOPR, DOE proposed to modify the adopted definition of service coil to more explicitly define what “labeled accordingly” meant. Specifically, DOE proposed that a manufacturer must designate a service coil as “for indoor coil replacement only” on the nameplate and in manufacturer product and technical literature. In addition, DOE proposed that the model number for any service coil must include some mechanism (

e.g.,

an additional letter or number) for differentiating a service coil from a coil intended for an indoor unit. 81 FR at 58169-58170 (Aug. 24, 2016)

AHRI, Nortek, and Ingersoll Rand commented that they support DOE's proposal. (AHRI, No. 27 at p. 3, Nortek, No. 22 at p. 3, Ingersoll Rand, No. 38 at p. 2) DOE received no other comments on this issue. Therefore, DOE is adopting this proposal in this final rule.

5. Efficiency Representations of Split-Systems for Multiple Refrigerants

DOE made numerous proposals in the August 2016 SNOPR regarding efficiency representations for multiple refrigerants, and they elicited voluminous and multi-faceted responses. The proposals themselves can be divided into three broad categories, including (1) representations for multiple refrigerants, (2) certification report requirements for outdoor units with no match, and (3) clarifying what outdoor units must have no-match efficiency representations. By far most of the responses addressed the third category—discussion thereof has been divided up into the following sub-topics: DOE authority, altering the measured efficiency, specific no-match criteria, and normalized gross indoor fin surface (NGIFS) (addressed in sections III.A.5.c through III.A.5.f).

a. Representations for Multiple Refrigerants

In the August 2016 SNOPR, to address instances in which the manufacturer indicates that more than one refrigerant is acceptable for use in a unit, DOE proposed that a split-system air conditioner or heat pump, including an outdoor unit with no match, must be certified as a separate individual combination for every acceptable refrigerant. Specifically, each individual combination would be certified under the same basic model. DOE's existing requirements for basic models would continue to apply; therefore, if an individual combination or an outdoor unit with no match fails to meet DOE's energy conservation standards using any refrigerant indicated by the manufacturer to be acceptable, then the entire basic model would fail. DOE also proposed that manufacturers must certify the refrigerants for every individual combination that is distributed in commerce. For models where the manufacturer only indicates one acceptable refrigerant, this proposal would simply entail certifying to DOE the refrigerant for which the model is designed. Finally, DOE proposed that any outdoor unit model that has certain characteristics (

e.g.,

if it is distributed in commerce without a specific refrigerant), a manufacturer must determine the represented value as an outdoor unit with no match. For some outdoor units, the proposal called for representations both as an outdoor unit with no match and as part of a combination, both as part of the same basic model. 81 FR at 58170 (Aug. 24, 2016).

The August 2016 SNOPR proposed that a refrigerant's acceptability for use in an outdoor unit would be based on its being covered under the unit's warranty, either explicitly or based on refrigerant characteristics.

Id.

at 58201.

AHRI, Nortek, Ingersoll Rand, and Carrier/UTC supported DOE's proposal that manufacturers should be required to certify efficiency ratings for all refrigerants that they have designed their equipment to use. (AHRI, No. 27 at p. 3; Nortek, No. 22 at p. 3; Ingersoll Rand, No. 38 at p. 2; Carrier/UTC, No. 36 at p. 3) AHRI, Nortek, and JCI suggested that DOE revise the requirement so that, if a manufacturer approves an air conditioner or heat pump for multiple refrigerants by listing them on the nameplate, such a product is subject to DOE certification and enforcement requirements for each approved refrigerant. AHRI, Nortek, and JCI commented that manufacturers should have the option to rate all compatible refrigerants as one basic model with the same efficiency rating, or to list different efficiencies for different refrigerants as separate basic models. AHRI, Nortek, and JCI contend that the determination of different efficiency ratings for different refrigerants should be allowed based on testing, or the appropriate use of AEDMs. (AHRI, No. 27 at p. 6; Nortek, No. 22 at p. 6; JCI, No. 24 at p. 9) Ingersoll Rand commented similarly. (Ingersoll Rand, No. 38 at p. 2)

ACEEE, NRDC, and ASAP commented that they support the proposed requirement to assign separate model numbers to systems designed for more than one refrigerant. (ACEEE, NRDC, and ASAP, No. 33 at p. 4; Lennox, No. 25 at p. 5)

Goodman commented that they agreed with DOE's proposal in principle, but were concerned that clarification regarding the refrigerants that are approved for use in a product may not always be clear, and that a refrigerant may be used in the field if information about approved refrigerants is weak or not readily identifiable. Goodman proposed regulatory text to address this issue, emphasizing reliance on a product's nameplate to indicate which refrigerants are approved. Specifically, the suggestion was that any refrigerant listed on the unit nameplate of any portion of the basic model be considered to be approved. Further, Goodman's suggestion also includes as “approved for use” those non-zero ozone-depleting refrigerants with similar thermophysical properties to a refrigerant listed on the nameplate, (Goodman, No. 39, p. 2-3)

In response to these comments DOE has revised the requirements so that indication of which refrigerants require certification of performance is based on the unit nameplate that is required by safety standards (

e.g.,

UL 1995) to list all approved refrigerants (see newly designated paragraph (a)(3) of section 10 CFR 429.16).

DOE does not understand Goodman's reference to “any portion of the basic model”. If an individual combination of a basic model includes an indoor unit whose nameplate lists a refrigerant that is not listed on the outdoor unit's nameplate, such listing on the indoor unit's nameplate would not make the refrigerant approved for use in the outdoor unit. The refrigerant would therefore not be approved for use with that individual combination and presumably would not be required for certification with the basic model. Hence, if listing on the unit's nameplate is a sufficiently strong indication of which refrigerants are approved for use, it is not clear that any refrigerant listed on the indoor unit's nameplate but not on the outdoor unit's nameplate should be considered approved for use with the outdoor unit. Consequently, DOE has not included the “any portion of the basic model” language in its requirements. DOE has not adopted this language due to manufacturers' representations that the refrigerant listings on the nameplate are respected sufficiently that installers would not use a refrigerant in a system if it is not listed on the outdoor unit's nameplate.

DOE also is not convinced that the “approved refrigerants” need to include any non-zero ozone depletion potential refrigerant that has similar thermophysical properties to a refrigerant approved for use on the unit nameplate. DOE is only aware of HCFC-22 as a non-zero ozone depletion refrigerant that is used for split system air conditioners—no such alternatives are approved in the EPA SNAP list for residential and light commercial air conditioning and heat pumps.

7

HCFC-22 and refrigerants with properties similar to HCFC-22, whether non-zero ozone depletion or not, are addressed separately in the no-match requirements (see section III.A.5.e).

7

https://www.epa.gov/snap/acceptable-substitutes-residential-and-light-commercial-air-conditioning-and-heat-pumps.

Additionally, in the August 2016 SNOPR, DOE did not intend to require testing of each refrigerant. In this final rule, DOE is clarifying the requirement to allow the manufacturer to test the unit with one refrigerant and to use an AEDM for other refrigerants. This clarification appears in paragraph (a)(3) of § 429.16, but DOE has also modified paragraph (c)(2) of this section to emphasize this clarification for outdoor units with no match. Additionally, in this final rule, DOE is adding a provision in paragraph (a)(3) of § 429.16 to allow grouping of refrigerants in reporting provided that the representative values represent the least efficient refrigerant. In response to ACEEE, NRDC, and ASAP, DOE does not believe the additional reporting burden of requiring that each refrigerant have its own model number and efficiency representation is justified if the rating represents the least efficient refrigerant. In response to AHRI and Nortek, DOE is requiring that all of the refrigerants for the given model of outdoor unit be part of the same basic model. This is consistent with the basic model definition adopted in the June 2016 final rule, which groups all combinations with a given model of

outdoor unit into the same basic model. 81 FR at 37053 (June 8, 2016).

b. Certification Report Requirements for Outdoor Units With no Match

DOE proposed to require reporting of additional non-public information for the indoor unit that is tested with an outdoor unit with no match. This would include the indoor coil face area, depth in the direction of airflow, fin density (fins per inch), fin material, fin style (

e.g.,

wavy or louvered), tube diameter, tube material, and numbers of tubes high and deep. These additional requirements would apply to outdoor units with no match, whether or not the outdoor unit was also certified as part of an individual combination. 81 FR at 58172 (Aug. 24, 2016).

Unico, Goodman, ACEEE, NRDC, and ASAP supported DOE in requiring that specific indoor coil descriptions be specified for outdoor units with no match. (Unico, Inc., No. 30 at p. 2; Goodman, No. 39 at p. 5; ACEEE, NRDC, and ASAP, No. 33 at p. 4)

AHRI generally did not support DOE's proposals for outdoor units with no match, but noted that the following fin styles are available as options in the AHRI Directory: Flat corrugated, high performance, lanced, louvered, andN/A. (AHRI, No. 27 at p. 7) Rheem commented that the proposed list of indoor unit details are insufficient as a measure of indoor coil performance. Rheem opposed reporting of additional non-public information for the indoor unit that is tested with an outdoor unit with no match. (Rheem, No. 37 at p. 2) Nortek similarly commented that DOE's attempt to have manufacturers describe a fin style and tube diameter is obsolete and that with the varying materials and technologies in the market, the burden of characterizing fins as “lanced, flat, corrugated”, etc. is of no value. (Nortek, No. 22 at p. 7)

In response to the comments from AHRI, DOE will include options noted by AHRI for fin style in the certification template. In response to the comments from Rheem and Nortek, DOE notes that the reporting of information on the indoor unit is necessary for DOE's assessment and enforcement testing. DOE notes that, although Rheem indicated that the listed information is insufficient, they provided no recommendations regarding alternative ways that DOE can verify performance claimed for outdoor units with no match. Therefore, DOE adopts this requirement in this final rule.

c. DOE Authority

Per DOE's regulations in Appendix M established in the June 2016 final rule, the model of outdoor unit must be tested with an indoor unit meeting specified criteria. 81 FR at 37051 (June 8, 2016). 81 FR at 58171 (Aug. 24, 2016). Under the certification requirements proposed in the August 2016 SNOPR, DOE expanded the scope of outdoor units that would be required to be tested as outdoor units with no match. The specific criteria proposed to require such a rating are discussed in greater detail in section III.A.5.e, but they include having no designated refrigerant, a warranty that specifies refrigerant properties similar to those of HCFC-22 to define refrigerant acceptability (rather than or in addition to specific refrigerants), shipping without refrigerant or with a charge that requires addition of more than a pound of charge during setup, and shipping with any amount of R-407C. As proposed, any such unit would need to be certified as an outdoor unit with no match.

Multiple stakeholders commented on various aspects of DOE's authority to establish such requirements.

AHRI and Nortek commented that DOE has authority over manufacturers, but that DOE cannot expand that authority to make the manufacturer selling a legal product liable for the conduct of a distributor, contractor or individual consumer. They emphasized that an objective standard that could be the basis of DOE's certification and enforcement requirements will capture the conduct through which the manufacturer is distributing in commerce and marketing the equipment. (AHRI, No. 27 at p. 4; Nortek, No. 22 at p. 3-4)

DOE agrees that DOE has authority over manufacturers but notes that EPCA defines manufacture as “to manufacture, produce, assemble, or import.” (42 U.S.C. 6291(10))

AHRI and Nortek commented that the test requirements for outdoor units with no match represent design requirements and that DOE does not have authority to impose design requirements for central air conditioners. They noted that EPCA clearly states for some products that a standard may be a design requirement or a performance standard, but not both, and that EPCA does not even give DOE the option of considering design requirements for central air conditioners. AHRI and Nortek commented that when the use of a component with specific design requirements is mandated by the test procedure, it is in fact a design requirement for the product, since that test procedure must be used to determine the product's efficiency. (AHRI, No. 27 at p. 4-5; Nortek, No. 22 at p. 4)

In response, DOE does not agree that the test procedure imposes a design requirement as DOE does not impose any design restrictions on the outdoor unit. However, DOE must establish test procedures that are reasonably designed to measure energy efficiency during a representative average use cycle as determined by DOE (42 U.S.C. 6293 (b)(3)), which is why the indoor unit characteristics are specified. This requirement is analogous to the requirement to use higher external static pressure (ESP) when testing an SDHV system. DOE also notes that its delineation of outdoor units with no match is for units that are predominantly used to replace failed HCFC-22 outdoor units. As such, DOE has developed a straightforward approach to defining the characteristics of an indoor unit which is representative of such applications in order to allow the test procedure for these units to be representative of field installation. The extension of this concept to additional categories of outdoor units with no match (other than those designed for HCFC-22) does not invalidate this premise. For example, DOE has no evidence that outdoor units designed for use with R-407C are installed to a significant extent with new indoor units. Further discussion regarding the specific criteria to identify outdoor units with no match is in section III.A.5.e.

AHRI and Nortek commented that DOE's proposal for outdoor units with no match would be an expansion into technical and policy issues that are outside of DOE's authority under EPCA, were not within Congress' intent in granting DOE authority over energy efficiency standards, and are the jurisdiction of the EPA. They assert that the proposed approach would effectively ban the sale of otherwise legal products by requiring the very restrictive no match testing. (AHRI, No. 27 at p. 5; Nortek, No. 22 at p. 4-5) Similarly, JCI commented that DOE's R-407C proposal effectively bans the use of R-407C in split-system CACs and HPs by proposing to burden R-407C units with more stringent testing requirements than units designed for use with any other EPA-SNAP approved refrigerant, requiring testing with an inefficient indoor unit, and thus requiring outdoor unit efficiency that is either technically impossible or economically inviable to meet. JCI commented that this refrigerant-specific test procedure requirement constitutes back-door regulation of R-407C by DOE even though R-407C is already subject to

direct regulation by EPA under the Clean Air Act, and EPA has permitted the use of R-407C in split system CAC/HPs. In proposing to manipulate the CAC/HP test procedure in a way that would eliminate the use of R-407C in split-system CAC/HPs, JCI stated that DOE is acting beyond its legal authority under EPCA. (JCI, No. 24 at p. 3-4)

Ingersoll Rand agrees with AHRI's position that these proposed requirements exceed DOE's statutory authority. (Ingersoll Rand, No. 38 at p. 3)

On the other hand, ACEEE, NRDC, and ASAP commented that DOE regulates energy efficiency and has a legal obligation to ensure that manufacturers comply with its standards. According to ACEEE, NRDC, and ASAP, the August 2016 test procedure SNOPR does precisely that by ensuring that units intended as replacement units have to meet the same rules regardless of the refrigerant they are designed to use. ACEEE, NRDC, and ASAP commented that in the SNOPR, DOE clearly set out to close a loophole in its own regulations that, if left unaddressed, would result in the sale of units that do not meet existing standards, resulting in higher energy consumption. ACEEE, NRDC, and ASAP commented that closing that loophole is the purpose of DOE's “no-match” requirements for certifying these units. ACEEE, NRDC, and ASAP further commented that DOE is not banning the sale of R-407C units and that selling outdoor unit replacements using R-407C is and will continue to be perfectly legal—in fact, manufacturers may produce and sell outdoor units with no match using any refrigerant they want, including R-22 and R-407C. They commented that these units will need to meet the efficiency of DOE's existing minimum standards, rather than skate by with a certified value not achieved in the real world. They expressed the view that DOE's SNOPR effectively addresses the efficiency performance of products on the market today. (ACEEE, NRDC, and ASAP, No. 33 at p. 11) ACEEE, NRDC, and ASAP also indicated that some products, including the R-407C products introduced to the market in 2016, can only meet the existing standards by pairing the outdoor unit with an oversized indoor unit, even though the units are sold as replacements for outdoor units in which the existing indoor unit is not replaced. They further stated that other combinations in which the outdoor and indoor units are mismatched are unlikely to be sold in these combinations in any significant quantity. (ACEEE, NRDC, and ASAP, No. 33 at p. 4) Lennox also commented that “a manufacturer” rated an outdoor unit for R-407C by matching the outdoor unit with an unusually large indoor coil and sold it with one pound of refrigerant charge as a replacement for HCFC-22 units. (Lennox, No. 25 at p. 4)

Contrary to the comments of AHRI, JCI, Nortek, and Ingersoll Rand, EPCA requires DOE to establish appropriate test procedures with which to measure product efficiency for a representative average use cycle. (42 U.S.C. 6293(b)(3)) DOE's proposals regarding outdoor units with no match are based on efficiency considerations and supported by DOE's authority granted by EPCA to regulate product efficiency and to establish appropriate test procedures with which to measure product efficiency. JCI commented that when consumers are offered the option to use R-407C, as opposed to HCFC-22, they take advantage of it, citing that sales of R-407C are rising proportionately with JCI's sales of R-407C units, and pointing out that they are giving customers the opportunity to avoid HCFC-22 refrigerant without entirely replacing their CAC/HP systems. (JCI, No. 24 at p. 7) These statements support DOE's expectation that the sales of these R-407C units are primarily, if not entirely, for no-match installations in which the indoor unit is not replaced. Although JCI claims that DOE cannot extend its arguments made for HCFC-22 outdoor units (

i.e.,

that they are clearly no-match installations because there is no valid EPA-approved combination that includes an HCFC-22 outdoor unit (JCI, No. 24 at p. 5)), DOE asserts that the possibility that there are or could be a few valid R-407C combinations sold does not in itself make sales of combinations (rather than no-match sales) the representative efficiency value for R-407C.

JCI also claimed that DOE has no authority to regulate outdoor units with no match because they are not a central air conditioner or a heat pump as defined by EPCA. (JCI, No. 24 at p. 4) DOE notes that in the June 2016 Final Rule, DOE reasonably interpreted the statutory definition to specify the following: “A central air conditioner or central air conditioning heat pump may consist of: a single-package unit; an outdoor unit and one or more indoor units; an indoor unit only; or an outdoor unit with no match. In the case of an indoor unit only or an outdoor unit with no match, the unit must be tested and rated as a system (combination of both an indoor and an outdoor unit).” 81 FR at 37056 (June 8, 2016). In that rule, DOE noted that this interpretation did not change the scope of DOE's product coverage and is in line with the current certification requirements for CAC/HP. 81 FR at 36999.

d. Altering the Measured Efficiency

In the August 2016 public meeting, JCI commented that they offer a matched combination with R-407C, and that the tested combination is available in the AHRI database. JCI noted that the product has been available since spring 2016, and it is too early to say that there is no tested combination of this product. JCI also questioned how long after introduction of an outdoor unit product an assessment can be made whether there is or is not a highest sales volume combination. (JCI, Public Meeting Transcript, No. 20 at pp. 124-132) In written comments, JCI cited EPCA requirements that when amending test procedures, DOE must consider to what extent the amendments alter the measured efficiency of covered products, and then amend the applicable energy conservation standards if a determination is made that the test procedure amendment alters the measurement. (42 U.S.C. 6293(e)(1-2)) JCI commented that DOE has not done this for its amendments associated with no-match R-407C products. JCI explained that the no-match proposals would force manufacturers to re-test previously certified compliant products using a new testing standard that is technically impossible to meet, which would render the previously-compliant R-407C systems non-compliant. (JCI, No. 24 at p. 6)

This test procedure provides a mechanism of assessing the performance of no-match products, such as those that use R-407C, which can then be used to provide a reasonable level of assurance that all field-match combinations of the new, unmatched outdoor units will achieve the established efficiency levels. The current test procedure requires that single-stage split system air conditioners be tested using the highest sales volume tested combination. 10 CFR 429.16. It is DOE's understanding that condensing units utilizing R407C typically do not have a highest sales volume indoor unit that satisfy the requirements of the test procedure and thus, could not be tested under the current regulatory regime. Further, if the condensing units were to have a highest sales volume indoor unit for testing, DOE believes the results of such testing would overstate the performance of R407C systems as installed. DOE believes this is the case because R407C systems typically get installed with existing indoor units, which are not properly sized, in order

to achieve the system efficiency that would result from a new matched pair system. Thus, DOE believes that manufacturers of R407C condensing units should have sought a waiver for the current test procedure requirements pursuant to the procedures at 10 CFR 430.27. EPCA requires DOE to adopt test procedures that are reasonably designed to produce test results which measure energy efficiency of a covered product during a representative average use cycle or period of use. (42 U.S.C. 6293(b)(3)) To meet this requirement for outdoor units with no match, DOE is now adopting an alternative approach similar to the proposal with modification for testing and determining represented values for no-match R407C products based on stakeholder comments. DOE notes that under the approach adopted in this final rule, the testing method for no-match systems does not consider HSVC. In this rulemaking, the only proposal regarding HSVC was to remove the requirement for single-split system air conditioners, which DOE adopts as discussed in section III.A.2. The application of HSVC to current applicable regulations is not within the scope of this rulemaking. Therefore, DOE will not address its application in this rule.

JCI also questioned whether DOE performed any analysis on how the new requirements for units with R-407C refrigerant impact consumers. (JCI, Public Meeting Transcript, No. 20 at pp. 137-139)

In response, DOE does not evaluate impacts on consumers for test procedure amendments. The test procedure amendments are developed to provide efficiency representations for representative average use cycles. (42 U.S.C. 6293(a)(3)) As discussed in section III.A.5.d, DOE developed the test approach for outdoor units with no match on this basis. Thus, the energy conservation standard rulemaking's consideration of consumer impacts accounts for the impacts that might be associated with specific test procedure changes.

e. Specific No-Match Criteria

DOE proposed in the August 2016 SNOPR that manufacturers must determine efficiency representations for outdoor units as outdoor units with no match if they meet any of the following criteria: Having no designated refrigerant, a warranty that specifies refrigerant properties similar to those of HCFC-22 to define refrigerant acceptability (rather than or in addition to specific refrigerants), shipping without refrigerant or with a charge that requires addition of more than a pound of charge during setup, and shipping with any amount of R-407C. 81 FR at 58170-58172 (Aug. 24, 2016).

JCI and Goodman commented that there are other refrigerants, including MO-99 and NU-22, that are used as replacements for HCFC-22. JCI questioned why those refrigerants were not specifically called out in the proposed test procedure as R-407C was, while Goodman indicated that the proposal would do nothing to address these other HCFC-22 replacement refrigerants. (JCI, Public Meeting Transcript, No. 20 at p. 140; Goodman, No. 39 at p. 3)

JCI also stated that they have competitors that have published guidelines around the application of R-410A units into existing indoor applications, and questioned why those units would not have to be held to the same test approach for outdoor units with no match.

In response, it has always been the case that some outdoor units are installed as replacements for failed outdoor units. However, in most cases an outdoor unit model would also be sold in substantial numbers as a combination with indoor units. This is in contrast to R-407C units, which are predominantly sold in scenarios in which the outdoor unit is replaced, and the indoor unit is not replaced. Hence the test procedure is representative of an average use cycle for R-410A units without requiring that it be tested as a unit with no match.

JCI also commented that the benefits of R-407C will increase over time if products designed for this refrigerant based on “additional valid matches” are allowed to be sold, but that the proposed requirements would significantly limit any such possibility. JCI asserted that it can create a larger market for complete R-407C systems and that DOE should not limit the potential for such innovation. (JCI, No. 24 at p. 7)

ACEEE, NRDC, and ASAP and Lennox supported the proposed requirement that an outdoor unit distributed without a designated refrigerant must be tested and certified as an outdoor unit with no match. (ACEEE, NRDC, and ASAP, No. 33 at p. 4; Lennox, No. 25 at p. 5)

AHRI and Nortek commented that DOE's categorization of dry-ship units is overly-broad and does not necessarily equate to outdoor units with no match. AHRI and Nortek commented that units with long line sets require more than one pound of charge to be added in the field. AHRI and Nortek contended that it is also very realistic that manufacturers will not be able to ship units with mildly flammable refrigerants factory charged which will require adding refrigerants in the field during installation. (AHRI, No. 27 at p. 6; Nortek, No. 22 at p. 6) JCI, Ingersoll Rand, Goodman, Carrier/UTC also disagreed with DOE's proposal for similar reasons. Ingersoll Rand, Goodman, and Carrier/UTC gave examples of situations in which the entire charge required for a system could not be contained within the outdoor unit by itself as shipped from the factory, and would require more than a pound of refrigerant to be added, including for MicroChannel Heat Exchangers and long line sets. (JCI, No. 24 at p. 7-8; Ingersoll Rand, No. 38 at p. 2; Goodman, No. 39 at p. 3-4; Carrier/UTC, No. 36 at p. 3; JCI and Ingersoll Rand, Public Meeting Transcript, No. 20 at pp. 140-141) Goodman further commented that the regulatory text should restrict the one pound rule to laboratory tests and suggested regulatory text to address this issue as well as the small diameter tubing issue. (Goodman, No. 39 at p. 3-4) Lennox supported the intent of DOE's proposal but found it to be too restrictive because of the existence of products in which the internal volume of the product does not allow it to be fully charged from the factory. (Lennox, No. 25 at p. 5) Goodman, Lennox, and JCI were particularly concerned with potential unintended consequences and potentially impeding innovation as the industry moves toward lower global warming potential (GWP) refrigerants, in which cases the manufacturer may choose to ship split-system units designed for use with A2L refrigerants without the refrigerant factory-installed. (Goodman, No. 39 at p. 4) Lennox commented that the safety requirements and codes and standards required for a transition to A2L

8

refrigerants are not developed and that there is a high probability that some form of mitigation to ensure product safety will be required, for example, requiring that such units be dry-shipped,

i.e.

with a dry nitrogen charge rather than with refrigerant. Lennox commented that DOE should maintain a path that allows dry-shipping products (DOE understands this to mean not requiring no-match testing for these products) to ensure the most efficient transition to low-GWP products with the least

negative consumer impacts. (Lennox, No. 25 at p. 5)

8

A2L is a safety classification for refrigerants that have low toxicity and lower flammability. See

https://www.epa.gov/snap/refrigerant-safety

. Most refrigerants in current use (

e.g.

R-410A) have an A1 classification, indicating both low toxicity and no flame propagation.

First Co. objected to the requirement to test an outdoor unit as a no-match outdoor unit if more than a pound of refrigerant would have to be added during set up. First Co. commented that the proposals are based on a single charge value when there are multiple charge values for different coils. First Co. requested DOE drop this requirement entirely. (EERE-2016-BT-TP-0029, No. 21 at p. 5)

In response to these comments DOE has revised the criteria for outdoor units with no match. Specifically, manufacturers must determine efficiency representations, and certify such representations, for outdoor units as an outdoor unit with no match if:

• The outdoor unit is approved for use with, determined by listing on the outdoor unit nameplate, HCFC-22 or refrigerants with similar thermophysical properties, as specified in § 429.16(a)(3) (the discussion below addresses similarity);

• There are no designations of approved refrigerants on the outdoor unit nameplate; or.

• The outdoor unit is shipped requiring more than two pounds of charge when tested according to the test procedure (

e.g.,

with 25 feet of interconnecting lines), unless (a) an A2L refrigerant is listed as approved on the nameplate, or (b) the factory charge listed on the nameplate is 70 percent or more of the outdoor unit's internal refrigerant circuit volume times the density for 95 °F refrigerant liquid.

DOE agrees with JCI and Goodman that outdoor units approved for use with refrigerants similar to HCFC-22 (other than R-407C) are likely to be intended for no-match use in the field. Hence, DOE is changing the criteria so that approval for use of any such refrigerant similar to HCFC-22 would make the outdoor unit subject to the no-match requirements. DOE does not find it likely that a large market for complete systems based on R-407C or other refrigerants similar to HCFC-22 would likely emerge in the near future given the initial trends associated with introduction of R-407C products, as discussed section III.A.5.c. As suggested by ACEEE, NRDC, and ASAP (ACEEE, NRDC, and ASAP, No. 33 at p. 3), R-410A is nearly universally used as the refrigerant that has replaced HCFC-22 in CAC/HP systems. Other refrigerants approved by the EPA in its SNAP listing for acceptable substitutes in residential and light commercial air conditioning and heat pumps

9

are rarely used in new split systems. DOE considered the approved refrigerants in the SNAP list and refrigerants understood to be suitable for use in HCFC-22 systems (“Refrigerants for R-22 Retrofits”, No. 46 at p. 1) and developed an HCFC-22 similarity criterion that would apply for these likely replacement options. DOE determined that the HCFC-22 replacement refrigerants would be selected and no other refrigerant that is likely to be approved for use in new split systems would be selected if the saturation pressure associated with 95 °F refrigerant temperature is within 18 percent of the pressure for HCFC-22. Hence, DOE adopts this as a criterion for no-match status of an outdoor unit. DOE recognizes that there may be A2L refrigerants that would themselves have similar pressures that in future may be approved on EPA's SNAP list for these products. To ensure that transition from global warming refrigerants is not restricted, DOE acknowledges that some revisions to these requirements may need to be developed as manufactures start to adopt such refrigerants in new split systems. DOE will consider such testing and certification revisions and propose options in a future rulemaking.

9

https://www.epa.gov/snap/acceptable-substitutes-residential-and-light-commercial-air-conditioning-and-heat-pumps.

DOE is also revising the no-match criteria regarding dry shipping and required refrigerant addition as indicated above in response to manufacturer comments and additional research. First, DOE recognizes that where an installation requires long line sets, that a higher quantity of refrigerant may have to be added. DOE agrees with Goodman's suggestion to base this limit on a standardized scenario, specifically the addition of charge in a DOE test, for which 25 feet of refrigerant lines are specified. Second, DOE has adopted the exception associated with small-volume outdoor coils (factory charge 70 percent or more than the coil internal volume times refrigerant density) suggested by Goodman. However, DOE reviewed its own available test data for CAC/HP systems and determined that, for tests in which the added charge quantities were clearly recorded, a large percentage of tests required addition of 1 pound or more of refrigerant. Review of the data showed that nearly all of the tests could be conducted with the addition of less than 2 pounds of refrigerant. Hence, DOE is revising the charge addition requirement accordingly. First Company's comments addressed differences in indoor coil volumes, but did not provide specific information regarding the potential differences in charge that could be associated with different coil sizes—the additional pound doubles the allowed charge addition for a unit before requiring a no-match test and, based on DOE test experience, is sufficient to address nearly all tested systems. Because these systems were charged without consideration of this new requirement and would likely have required less charge addition if pre-charged with the limit in mind, and also considering that at least one manufacturer (Goodman) agreed with the one-pound limit on the basis of additional clarifications that DOE has adopted (the low-coil-volume exclusion and clarification that the limit applies for ratings testing), DOE believes that the finalized criteria are sufficiently flexible to avoid requiring no-match testing for any outdoor units that should not be tested this way.

DOE also acknowledges the issues associated with A2L refrigerants and small-volume heat exchanger technologies. DOE agrees with Goodman's suggestions for providing exceptions to the no-match requirements in these cases and has adopted the suggestions in this final rule.

f. NGIFS

In the July 2016 final rule, DOE set requirements for the indoor units that are used in tests of outdoor units with no match. 81 FR at 37065 (June 8, 2016). The August SNOPR proposed extension of this requirement to additional types of outdoor units with no match. 81 FR at 58170 (Aug. 24, 2016).

AHRI and Nortek commented that it will not always be the case that outdoor units with no match are a result of the phase-out of R-22 refrigerant and that in the future there will be a transition between nonflammable and mildly flammable refrigerants. They further suggested that when higher GWP refrigerants, such as R-410A are phased out, there will likely be a period of time when R-410A condensing units will be sold as outdoor units with no match, and that they will likely be shipped dry. AHRI and Nortek commented that while a NGIFS no higher than 1.0 sq.in./Btu/hr may be representative of R-22 units circa 2006, NGIFS of 1.0 makes no sense for R-410A, resulting in energy measurements that are not representative of the unit in the field. (AHRI, No. 27 at p. 5-6; Nortek, No. 22 at p. 5) Ingersoll Rand commented similarly. (Ingersoll Rand, No. 38 atp. 2) Ingersoll Rand further commented that the NGIFS definition is only appropriate for

3/8

″ tube coils and cannot be used for coils with smaller

diameter tubes or with microchannel heat exchangers. Ingersoll Rand commented that NGIFS does not account for fin design or tube pattern which affects heat transfer, and its adoption will create the potential for testing loopholes in the future. Ingersoll Rand commented that it would be better to set a limit on coil cabinet volume based on coils sold in the 5 years prior to the elimination of a refrigerant. (Ingersoll Rand, No. 38 at p. 2)

DOE acknowledges that the old indoor units that are matched with no-match outdoor units in field installations will not always be old HCFC-22 indoor units. DOE will consider adjustments to the no-match requirements consistent with available information in a future rulemaking. However, DOE does not necessarily agree that a phaseout of high GWP refrigerants will by itself mean a step change of the existing population of indoor units to characteristics typical of more recent R-410A systems. Consideration will have to be given to whether the NGIFS value is allowed to rise to reflect representative field conditions or whether there are alternative approaches that would be more effective in addressing issues associated with installation of no-match outdoor units.

In response to Ingersoll Rand's comment regarding applicability of NGIFS, DOE responds that the vast majority of indoor units that are field-matched with no-match outdoor units have

3/8

-in OD tubing. Further, DOE selected the NGIFS value based on the assumption that manufacturers would use enhanced fin surfaces (

e.g.,

lanced, louvered, wavy) for such tests. DOE also notes that such surfaces were in general use during the time period before phaseout of HCFC-22 for new systems. (See,

e.g.,

page 1-11 of the 1997 technical support document for room air conditioners, which indicates that such surfaces were in use for central air conditioners at the time,

https://www1.eere.energy.gov/buildings/appliance_standards/pdfs/tsdracv2.pdf

.)

6. Representation Limitations for Independent Coil Manufacturers

In the June 2016 final rule, DOE adopted language in 10 CFR 429.16 specifying that a basic model may only be certified as compliant with a regional standard if all individual combinations within that basic model meet the regional standard for which that basic model would be certified and that an ICM cannot certify a basic model containing a representative value that is more efficient than any combination certified by an OUM containing the same outdoor unit. 81 FR at 37050 (June 8, 2016).

Based on letters submitted by several stakeholders (Docket No. EERE-2016-BT-TP-0029-0006, -0005, and -0003), in the August 2016 SNOPR, DOE proposed to remove the sentence: “An ICM cannot certify a basic model containing a representative value that is more efficient than any combination certified by an OUM containing the same outdoor unit.” and replace it with the following language in 10 CFR 429.16(a)(4)(i): An ICM cannot certify an individual combination with a rating that is compliant with a regional standard if the individual combination includes a model of outdoor unit that the OUM has certified with a rating that is not compliant with a regional standard. Conversely, an ICM cannot certify an individual combination with a rating that is not compliant with a regional standard if the individual combination includes a model of outdoor unit that an OUM has certified with a rating that is compliant with a regional standard. 81 FR at 58172 (Aug. 24, 2016)

AHRI, Nortek, Unico, First Co., ADP, ACEEE, NRDC, and ASAP, Ingersoll Rand, Rheem, Carrier, Lennox, and JCI supported DOE's proposal. (AHRI, No. 27 at p. 7; Nortek, No. 22 at p. 7; Unico, Inc., No. 30 at p. 2; First Co, No. 21 at p. 3; ADP, No. 23 at p. 3; ACEEE, NRDC, and ASAP, No. 33 at p. 5; Ingersoll Rand, No. 38 at p. 3; Rheem, No. 37 at p. 2; Carrier/UTC, No. 36 at p. 4; Lennox, No. 25 at p. 11; JCI, No. 24 at p. 9; ADP, Public Meeting Transcript, No. 20 at p. 143) Therefore, in this final rule, DOE is adopting this language as proposed.

7. Reporting of Low-Capacity Lockout for Air Conditioners and Heat Pumps With Two-Capacity Compressors

In the August 2016 SNOPR, DOE proposed to require that the lock-out temperatures for both cooling and heating modes for CAC/HPs with two-capacity compressors be provided in the certification report. 81 FR 58163, 58172 (Aug. 24, 2016).

NEEA commented that they strongly support the proposed reporting requirement. (NEEA, No. 35 at p. 2) AHRI, Nortek, Ingersoll Rand, JCI, and Carrier/UTC commented that low-capacity lockout for air conditioners and heat pumps with two-capacity compressors is considered intellectual property, and that they are concerned about the possibility of reverse engineering products if this information is publicly reported. (AHRI, Public Meeting Transcript, No. 20 at p. 101; AHRI, No. 27 at p. 7; Nortek, No. 22 at p. 8; Ingersoll Rand, No. 38 at p. 3; JCI, No. 24 at p. 17-18; Carrier/UTC, No. 36 at p. 3)

In the existing requirements and the requirements proposed in the August 2016 SNOPR, DOE lists product-specific items that needs to be included in certification reports in 10 CFR 429.16(e), with subsection (2) listing public items, and subsection (4) listing additional items that would not be posted to DOE's public certification database. DOE notes that it included the proposal to require reporting the outdoor temperature(s) at which the unit locks out low capacity operation (where applicable) in proposed § 429.16(e)(4) of the August 2016 SNOPR. Because, under the proposal, the item would not be posted to DOE's public certification database, DOE is maintaining this requirement in this final rule.

8. Represented Values of Cooling Capacity

In the August 2016 SNOPR, DOE proposed to revise the regulatory text in three locations (10 CFR 429.16(b)(3), 10 CFR 429.16(d), 10 CFR 429.70(e)(5)(iv)) to allow a one-sided tolerance on cooling and heating capacity that allows underrating of any amount, but only overrating up to 5 percent (

i.e.,

the certified capacity must be no greater than 105 percent of the mean measured capacity or the output of the AEDM), as intended in the June 2016 final rule. As adopted in the June 2016 final rule, DOE would still use the mean of the measured capacities in its enforcement provisions.

AHRI, Mitsubishi, Rheem, Carrier, JCI, Nortek, Ingersoll Rand, ADP, Lennox, and Goodman opposed DOE's proposal for tolerance on cooling capacity. They commented that the same rules that apply to efficiency should be applied to capacity, where manufacturers should be permitted to rate cooling and heating capacity only as high as the tested value or AEDM output. (AHRI, No. 27 at p. 7; Mitsubishi, No. 29 at p. 2; Rheem, No. 37 at p. 2; Carrier/UTC, No. 36 at p. 4; JCI, No. 24 at p. 9; Nortek, No. 22 at p. 8; Ingersoll Rand, No. 38 at p. 3; ADP, No. 23 at p. 3-4; Lennox, No. 25 at p. 6; Goodman, No. 39 at p. 12; Carrier/UTC and Lennox, Public Meeting Transcript, No. 20 at p. 145) Additionally, Carrier commented that de-rating capacity would result in a consumer getting more capacity than expected but that overrating capacity as suggested in this proposal would result in a loss to the consumer. In addition, the double sided tolerance would statistically result in much higher risk for manufacturers. (Carrier/UTC, No. 36

at p. 4; Carrier/UTC, Public Meeting Transcript, No. 20 at p. 144)

ACEEE, NRDC, ASAP supported the use of one-sided tolerance tests where possible, stating that there may be legitimate business reasons to label and sell units that are more efficient than their certified values and that consumers can only be pleased if a product does better than claimed. (ACEEE, NRDC, and ASAP, No. 33 at p. 5)

Unico commented that they strongly support one-sided tolerance for capacity, without which a manufacturer cannot rate conservatively. Unico stated that it recognizes that, for some product classes other than small-duct high-velocity, there is a very small chance that a manufacturer could conservatively rate a system with the express intent to avoid testing with a slightly higher external static pressure. Unico believes the advantage that this provides is insignificant. (Unico, Inc., No. 30 at p. 2)

NEEA commented that they do not necessarily support the proposal, stating that they were not able to ascertain if DOE's one-sided tolerance for capacity reporting would result in a system being rated with a lower building load as a result of reporting an overly conservative value, and thus an overrated cooling and/or heating performance. (NEEA, No. 35 at p. 2)

First Co. agreed with DOE's proposal to allow one sided tolerance on represented values of cooling and heating capacity, but commented that the proposed language in § 429.70(e)(5)(iv) does not accurately reflect DOE's intention. First Co. believes that in the first sentence after the words “. . . by more than 5 percent” the text should read “or tests worse than its certified cooling capacity by more than 5 percent.” (First Co, No. 21 at p. 3)

DOE understands that overrating capacity could result in a loss to the consumer and could put the manufacturer at risk. In response to the comments received, in this final rule DOE is revising the tolerance on cooling capacity to be similar to the tolerance on efficiency, where the cooling capacity should be less than or equal to the lower of: (1) The mean of the sample and (2) the lower 90 percent confidence limit of the true mean divided by 0.95; or less than or equal to the AEDM output. DOE agrees with Unico that conservatively rating to gain some advantage is not a significant risk. In response to NEEA, DOE notes that the building loads, calculated by sections 4.1 and 4.2 of both appendix M and appendix M1 of the August 2016 SNOPR, use the tested heating and cooling capacities, not the rated capacities. Therefore, there is no concern of overrating cooling or heating performance.

In response to First Co.'s comments, DOE notes that the August 2016 SNOPR, § 429.70(e)(5)(iv), regarding AEDM verification testing, inadvertently stated that DOE would notify a manufacturer that a unit fails to meet its certified rating if the tested cooling capacity is greater than 105 percent of its certified cooling capacity. In this final rule, the section has been revised to indicate DOE will notify a manufacturer that a unit fails to meet its certified rating if the tested cooling capacity is lower than its certified cooling capacity. This is consistent with DOE's revisions to its tolerance on cooling capacity.

9. New Efficiency Metrics

During the August 2016 Public Meeting, EEI, PG&E, Goodman, Rheem, and Unico recommended renaming the efficiency metrics whose values will be altered as compared to the current metrics, which includes HSPF, SEER, and EER. The purpose of this would be to help avoid confusion in the marketplace and to allow more relevant utility incentive programs. (EEI, PG&E, Goodman, Rheem, and Unico, Public Meeting Transcript, No. 20 at pp. 85-91)

Additionally, EEI submitted a written comment suggesting that a new efficiency acronym be used under the revised test procedure in order to avoid market confusion and to ensure that consumers are aware that significant changes have been made in how heat pumps are tested and rated. EEI suggested the use of several specific acronyms. (EEI, No. 34, page 6) The California IOUs similarly commented that the proposed changes to appendix M1 efficiency ratings are so substantial that they should be given new descriptors. The California IOUs stated that value changes will cause confusion in the marketplace unless they are re-labeled as “EER2,” “SEER2,” and “HSPF2,” or with other labels determined by DOE to be appropriate. (California IOUs, No. 32 at p. 5)

In response to the comments, in this final rule, DOE is creating new efficiency metrics to represent cooling and heating performance whose values will be altered as compared to the current metrics. The new metrics include seasonal energy efficiency ratio 2 (SEER2), which will replace seasonal energy efficiency ratio (SEER); energy efficiency ratio 2 (EER2), which will replace energy efficiency ratio (EER); and heating seasonal performance factor 2 (HSPF2), which will replace heating seasonal performance factor (HSPF). These labels are consistent with those used in the CAC/HP ECS Working Group Term Sheet. New efficiency metrics SEER2, EER2, and HSPF2 reflect the changes in the test procedure in appendix M1 that result in change in the measured efficiency values. The definitions for these metrics are identical to those for the original metrics except that they are determined in accordance with appendix M1 instead of in accordance with appendix M.

B. Amendments to Appendix M Testing To Determine Compliance With the Current Energy Conservation Standards

Under EPCA, any test procedure that DOE prescribes or amends shall be reasonably designed to produce test results which measure energy efficiency and energy use of a covered product during a representative average use cycle or period of use. (42 U.S.C. 6293(b)(3)) In the August 2016 SNOPR, DOE proposed several revisions to appendix M to subpart B of 10 CFR part 430 to improve the test representativeness and repeatability. 81 FR 58164 (Aug. 24, 2016) In addition, DOE held a public meeting at DOE headquarters in Washington, DC, on August 26, 2016 (Public Meeting Transcript, Docket No. EERE-2016-BT-TP-0029-0020). Based on the comments DOE received from the August 2016 Public Meeting and from the August 2016 SNOPR comment period, DOE is modifying its approach and adopting revisions to its procedures in Appendix M, which is independent of Appendix M1.

1. Measurement of Off Mode Power Consumption: Time Delay for Units With Self-Regulating Crankcase Heaters

In the August 2016 SNOPR, DOE proposed revisions to the off-mode test procedure imposing time delays to allow self-regulating crankcase heaters to approach equilibrium before making measurements. DOE proposed a 4-hour time delay for units without compressor sound blankets and an 8-hour time delay for units with compressor sound blankets. 81 FR at 58173 (Aug. 24, 2016)

In the SNOPR public meeting, JCI commented that adding four or eight hour time delays is a substantial testing burden and requested that DOE consider developing an approach to predict the final values without much extra test time. They reiterated this request in written comments and suggested that a time-based correlation developed by manufacturers could be built into the AEDM for the off-mode metric. (JCI, Public Meeting Transcript, No. 20 at p. 31; JCI, No. 24 at p. 10)

AHRI and Nortek commented that they generally support establishing delay time but were concerned that manufacturers would have to retest all units again within 180 days of the publication of the final rule so soon after initiating off-mode testing after the June 2016 final rule first established the off-mode test procedures. AHRI asserted that this revision represents a significant and unnecessary testing burden. AHRI suggested that DOE should either allow the off-mode rating to be based on appendix M modifications finalized in the June 2016 Final Rule (DOE assumes this is a request to clarify that products tested within 180 days of the June 8 final rule need not be retested again using the time delays) or move this revision to appendix M1 (AHRI, No. 27 at p. 8; Nortek, No. 22 at pp. 8-9). Carrier commented that the estimated time to implement this change would be at least six additional months (Carrier, No. 36 at p. 5). Rheem disagreed with the implementation time frame because this change will double the testing time and supported moving the change to appendix M1 (Rheem, No. 37 at p. 2). Ingersoll Rand commented that completing all the required testing would extend beyond the effective date (Ingersoll Rand, No. 38 at p. 3).

ACEEE, NRDC, and ASAP commented that DOE's approach to the thermal response delay issue for self-regulating crankcase heaters seems reasonable and responsive, but also sub-optimal considering that the measured self-regulating heater's power at the end of the specified delay times could be higher or lower with compressors having more or less thermal mass. ACEEE, NRDC, and ASAP recommended that DOE allow manufacturers to select alternative delay times if shorter or longer delays are required for specific models. (ACEEE, NRDC, and ASAP, No. 33 at p. 6).

Lennox, the CA IOUs and NEEA supported DOE's proposal. (Lennox, No. 25 at p. 11; CA IOU, No. 32 at p. 4; NEEA, No. 35 at p. 2)

DOE agrees that this additional delay time requirement could change the off-mode power measurement for some tested combinations that manufacturers may have already tested using the test procedure of the June 2016 Final Rule. DOE does not intend to introduce unnecessary test burden due to the close timing between the June 2016 Final Rule and this final rule. Therefore, DOE has decided to remove this requirement from appendix M and adopt it only in appendix M1. As for JCI's suggestion to develop a time-based correlation to allow prediction of the final measurement based on the trend in the measurement over a limited time period, DOE does not have sufficient test data to be confident that such an approach would provide a predictable result. In fact, depending on the equation used to fit the curve created by the first few data points, the details of the particular compressor design, and the history of testing just prior to conducting an off-mode test, DOE is concerned that a wide range of results might be obtained for any given unit, including a prediction of infinite wattage. DOE understands JCI's concern and agrees that such an approach could be considered in the future with more analysis and testing to validate an approach. Hence, DOE will not adopt a shortened test using curve fitting to predict ultimate off-mode power input. Regarding JCI's mention of an AEDM for off-mode, DOE does not regulate what analytic evaluation can be used in an AEDM—there is nothing in the AEDM requirements that would prevent a manufacturer from adopting an AEDM that uses the results of a shortened test as its input, as long as the requirements in 10 CFR 429.16 and 429.70 are satisfied. Thus, this notice does not adopt a shortened test procedure using curve fitting and prediction to determine off-cycle power input for systems with self-regulating crankcase heaters.

DOE received no comment suggesting different time delays than those proposed by DOE. Hence, DOE has adopted in appendix M1 the proposed time delays for measurement of off-mode power for units with self-regulating crankcase heaters or heater systems in which the crankcase heater control is affected by the heater's heat.

In addition, DOE notes that the August 2016 SNOPR inadvertently included in the regulatory text a certification requirement for the duration of the crankcase heater time delay for the shoulder season and heating season, if such time delay is employed. DOE does not actually require this information and has not adopted this requirement in the final rule.

2. Refrigerant Pressure Measurement Instructions for Cooling and Heating Heat Pumps

In the August 2016 SNOPR, DOE proposed limiting the internal volume of the pressure measurement system (

i.e.

the pressure gauge or transducer and the capillary tube and tube fittings connecting the transducer to the refrigerant lines) at pressure measurement locations that may switch from liquid to vapor state when changing operating modes and for all locations for systems undergoing cyclic tests for cooling/heating heat pumps. Specifically, DOE proposed the limit to be 0.25 cubic inch per 12,000 Btu/h. DOE also proposed the default internal volumes to be assigned to pressure transducers and gauges of 0.1 and 0.2 cubic inches, respectively, if transducer or gauge datasheets do not provide their internal volume. 81 FR at 58174 (Aug. 24, 2016)

During the 2016 August Public Meeting, Carrier commented that manufacturers typically test with up to six pressure transducers and the proposed limit would prohibit the level of testing during manufacturers' development stage and limit the number of pressure transducers to two. Carrier requested a reconsideration of the tolerance. (Carrier, Public Meeting Transcript, No. 20 at pp. 70-75)

AHRI requested clarification of “locations where the refrigerant state changes from liquid to vapor for different parts of the test.” AHRI commented that it is standard industry practice to place pressure taps with capillary tubes at six locations and advised that one of its members reported that, in their test chambers, the average internal volume of each pressure line is 0.91 cubic inches. Hence, AHRI asserts that DOE's proposed limit is too tight, such that the allowed number of pressure transducers would be zero for a unit that has a capacity less than 3 tons, and only one for larger-capacity units. In addition, AHRI commented that, for a cyclic test, the refrigerant state change occurs so quickly during transient startup that the effects (if any) will be within the tolerance of the measuring equipment. According to AHRI, for steady-state tests of units with the cooling mode restrictor located in the outdoor unit, there are at most two locations where the refrigerant state changes from liquid to two-phase between heating and cooling. AHRI's comment provided a table showing the refrigerant states at the six typical measurement locations for a cooling/heating heat pump having two expansion devices (one each in the indoor and outdoor units) for four test scenarios: Cooling steady-state, cooling transient start-up, heating steady-state, and heating transient start-up. The comment provided a similar table showing the refrigerant states for a heat pump with a single expansion device in the outdoor unit. In these tables, the transient startup scenario entries were all “two-phase”. In addition, the only differences in refrigerant state between steady-state heating and steady-state cooling were highlighted in the single-expansion-device table for the liquid

service valve and indoor coil inlet locations. AHRI commented that the refrigerant weight difference (

e.g.,

associated with transfer of refrigerant in and out of the pressure lines) is extremely small (particularly considering standard charging conditions in the field), and would have a negligible effect on the system performance. AHRI requested that DOE eliminate restrictions on pressure transducer internal volume or increase them significantly in order to ensure proper system analysis. (AHRI, No. 27 at pp. 8-11) JCI, Carrier, Ingersoll Rand and Goodman concurred with AHRI's comment. (JCI, No. 24 at p. 10-12; Carrier/UTC, No. 36 at p. 5-6; Ingersoll Rand, No. 38 at p. 3; Goodman, No. 39 at p. 9) Ingersoll Rand further requested that there be clarification that this requirement would apply only to assessment and enforcement testing, not for developmental testing. (Ingersoll Rand, No. 38 at p. 3)

Lennox commented that this proposal is not practical or in alignment with current practice for either manufacturer or audit testing, and requested DOE remove or extensively revise this requirement to align with current practices. (Lennox, No. 25 at p. 12) Rheem disagreed with DOE's proposal, and commented that the amount of refrigerant trapped in pressure measuring devices can be adequately accounted for through proper refrigerant charging instructions. (Rheem, No. 37 at p. 3) Unico agreed there should be volume limits but did not have a comment on the value. Unico commented that most systems have a high tolerance for charging while some systems, particularly systems with microchannel coils, have a very low tolerance. (Unico, No. 30 at p. 3) ACEEE, NRDC, and ASAP appreciated DOE's interest but stated that it could not judge whether the proposed volumetric limits are the right ones. (ACEEE, NRDC, and ASAP, No. 33 at p. 6) The CA IOUs agreed with DOE's proposal (CA IOU, No. 32 at p. 4)

DOE has considered all of the comments received and is making revisions based on those comments. First, DOE agrees that the transient startup phase of a cyclic test may be sufficiently short that any transfer of refrigerant in or out of the pressure lines at this time could have very little impact on measured cyclic performance. The scenario for cyclic test performance enhancement at the end of the on cycle discussed in the August 2016 SNOPR could still occur (see 81 FR at 58174 (Aug. 24, 2016)), but there is no data available to demonstrate that this effect is significant.

DOE notes that the tables provided in the AHRI comment showing refrigerant states at different refrigerant circuit locations represent states in the refrigerant lines and not in the pressure measurement systems, which could be different. For example, while the refrigerant state is always vapor at the discharge location during steady-state operation, the pressure measurement system is at a lower temperature than the saturation temperature associated with the prevailing pressure level. Hence, the vapor in the pressure line will condense. The condensed liquid may flow out of the capillary line back into the system, but this is unlikely if the pressure measurement system is lower than the measurement location. Also, it is somewhat unclear whether surface tension inside a small-diameter capillary tube would impede the flow of condensed liquid back into the system, or whether the vapor flowing into the system to replace the liquid would hold up the liquid's return flow. DOE considered the potential states within the pressure measurement systems rather than at the measurement locations when evaluating the potential for refrigerant transfer between steady-state operating modes. DOE made some reasonable assumptions for this assessment, making liberal assumptions where there is some doubt about what will occur—specifically, DOE did not assume that for the above scenario that liquid return flow to the system would be impeded. DOE's assessment of likely refrigerant states for a single-expansion-valve heat pump is summarized in Table III-1. The table adds a seventh potential refrigerant circuit location, between the outdoor coil and the expansion valve, which DOE expects that some manufacturers may monitor during developmental testing to determine subcooling achieved during cooling mode operation.

Table III-1—Refrigerant States in Pressure Measurement Systems for a Single-Expansion-Valve Heat Pump

Operating mode

Pressure measurement system above or below tap location

Steady-state cooling

Above

Below

Steady-state heating

Above

Below

1. Compressor Discharge

Vapor

Liquid **

Vapor

Liquid **.

2. Between Outdoor Coil and Expansion Valve

Liquid

Liquid

Vapor *

Two-phase.

3. Liquid Service Valve

Vapor *

Two-phase

Liquid

Liquid.

4. Indoor Coil Inlet

Vapor *

Two-phase

Liquid

Liquid.

5. Indoor Coil Outlet

Vapor

Vapor

Vapor

Liquid **.

6. Common Suction Port (

i.e.

vapor service valve)

Vapor

Vapor

Vapor

Liquid **.

7. Compressor Suction

Vapor

Vapor

Vapor

Vapor.

* Any liquid that enters the pressure measurement system will evaporate because the system is at a warmer temperature than the saturation temperature associated with the pressure.

** Liquid will condense in the pressure measurement system because the system is at a cooler temperature than the saturation temperature associated with the pressure, and will not drain back into the refrigeration circuit.

DOE notes that the liquid that might transfer out of one pressure measurement system as the operating mode switches from cooling to heating may transfer into another pressure measurement system and therefore not affect total charge operating within the refrigerant circuit. Also, because of the large density difference between liquid and vapor, DOE believes that the charge in the pressure measurement system would be negligible if the refrigerant within it is two-phase or vapor. Hence, the likely transfer of refrigerant out of the refrigeration circuit as the system switches from cooling to heating would be equal to the liquid density (calculated for 100 °F bubble point conditions) multiplied by the volume differential obtained by adding the volumes of the downward-run pressure measurement systems at locations 5 and 6 (as designated in Table III-1) to the volumes of any pressure measurement systems at locations 3 and 4 and subtracting the volume of any pressure measurement system at location 2. For

a system with two expansion valves, the transferred refrigerant would represent only the volumes of downward-run pressure measurement systems at locations 5 and 6.

DOE realizes the refrigerant transfer could be mitigated by complex phenomena occurring within the pressure measurement systems, some of which, for example surface tension, are mentioned above. Another mitigating phenomenon would be the filling of the pressure measurement system with compressor oil, which would displace any refrigerant that might transfer into it. Hence, DOE is relaxing the requirement proposed in the August 2016 SNOPR in new section 2.2.g (see 81 FR at 58207 (Aug. 24, 2016)) that the volume differentials listed above represent no more than 0.5 percent of refrigerant charge. DOE is instead adopting a requirement in section 2.2.g that the volume differential represent no more than 2 percent of the charge listed on the outdoor unit nameplate. Basing the limit on the outdoor unit nameplate charge will provide more flexibility for pressure measurement systems for those heat pumps that have more charge and would hence be less sensitive to this issue. However, due to the uncertainty regarding the actual potential behavior regarding refrigerant transfer, DOE also is imposing a pressure measurement system volume limit of 1 cu. in. for location 2 for single-expansion-device heat pumps, in order to prevent a test laboratory from using a very large volume for this location to offset the volumes of locations 3, 4, 5, and 6.

For a two-expansion-device heat pump with pressure measurement systems at locations 5 and 6 above the pressure tap locations, this approach imposes no volume limits. Also, for single-expansion-valve heat pumps with pressure measurement systems at locations 5 and 6 above the pressure tap locations and the volume at locations 2 offsetting the volumes at locations 3 and 4, there will also be no volume limit, other than the 1 cu. in. limit at location 2. DOE believes that these revisions to the proposal will allow manufacturers to make pressure measurements at the locations typically used for development and ratings testing while also providing some assurance that unforeseen impacts associated with refrigerant transfer between operating modes will be mitigated. However, DOE notes that the test procedure is for determining the performance of the product for the purpose of efficiency representations, not for development testing. DOE does not require pressure measurements installed at all 7 locations indicated in Table III-1. If manufacturers require use of pressure lines for development testing that exceed the volume requirements, they have the option of using isolation valves to isolate the tap locations not needed for ratings tests as the test transitions from development to determination of ratings for purposes of certifying compliance with applicable standards. Another option is to use pressure transducers that are more resistant to the temperature changes that occur in the test chamber. In any case, DOE may consider revisions to the requirements in the future if testing shows that they can be revised further to both improve test repeatability and allow more flexibility in making pressure measurements.

3. Revised EER and COP Interpolation Method for Units Equipped With Variable-Speed Compressors

In the August 2016 SNOPR, DOE proposed to require use of bin-by-bin interpolations for all variable-speed units (including variable-speed multi-split and multi-head mini-split systems), to calculate performance when operating at an intermediate compressor speed to match the building cooling or heating load. This method consists of using interpolation of EER or COP for each temperature bin based on the estimates of capacity and power input for the specific bin temperature. (EER is equal to cooling capacity divided by power input, while COP is proportional to heating capacity divided by power input.) 81 FR at 58175 (Aug. 24, 2016)

Nortek, JCI, Mitsubishi, Carrier, Rheem, Ingersoll Rand and AHRI expressed support for DOE's proposal but stated concerns that it would impact ratings and would, as a result, be more appropriate for inclusion in appendix M1 as opposed to Appendix M. (Nortek, No. 22 at p. 9; JCI, No. 24 at p. 12; Mitsubishi, No. 29 at p. 2; Carrier, No. 36 at p. 6; Rheem, No. 37 at p. 3; Ingersoll Rand, No. 38 at p. 4; AHRI, No. 27 at p. 11) AHRI also commented that its members were in the process of collecting data on the impact this proposed change would have on ratings and committed to providing additional information to the Department within 30 days of the close of the comment period. (AHRI, No. 27 at p. 11) DOE notes that the additional data were not provided. Goodman also requested DOE implement this change as part of appendix M1. (Goodman, No. 39 at p. 6) Unico recommended that this proposal be moved to appendix M1, and if it remains as an appendix M change, DOE should allow that the higher rating of both methods be used, but only if the bin-by-bin method results in a failure. (Unico, No. 30 at p. 3-4) Lennox, CA IOU, ACEEE, NRDC, and ASAP, and NEEA all supported DOE's proposal. (Lennox, No. 25 at p. 12; CA IOU, No. 32 at p. 4; ACEEE, NRDC, and ASAP, No. 33 at p. 6; NEEA, No. 35 at p. 2)

Central air conditioning heat pumps include single-speed, two-speed, and variable-speed products, all within the same product class that when tested in accordance with the DOE test procedure will have different measured efficiencies. Pursuant to 42 U.S.C. 6293(e), DOE is required to determine to what extent, if any, the proposed test procedure would alter the measured efficiency of the covered product. DOE proposed changes to the interpolation method for variable speed units only. For single-speed and two-speed products there would be no change in measured efficiency because they would not be impacted by this change in test procedure. However, variable-speed products would be impacted by this change in test procedure, so the measured efficiency would change.

Where an amended test procedure would alter measured efficiency, EPCA requires DOE to amend an energy conservation standard by measuring, under the amended test procedure, a sample of representative products that minimally comply with the standard. In this case, minimally compliant units are those with single-speed technology. Consistent with the statute, DOE has tested a representative sample of covered products that minimally comply with the existing standard. EPCA requires that the amended standard should constitute the average of the energy efficiency of those units, determined under the amended test procedure. As a result of that testing, DOE has determined that there is no change in measured average energy efficiency for single-speed units between the current test procedure and the amended test procedure. Thus, under 42 U.S.C. 6293(e)(2), the amended standard applicable to the amended test procedure and the current standard applicable to the amended test procedure are the same. As a result, DOE does not need to amend the existing standard to require that representations of variable-speed heat pumps be based on the amended test procedure in appendix M.

If DOE were to include this change in appendix M, Goodman requested that DOE allow industry up to two years to re-test and re-calculate SEER and HSPF, by either modifying the implementation date for this provision or by issuing a policy of non-enforcement for this provision. (Goodman, No. 39 at p. 6) DOE notes that this proposal would not

require additional testing. The proposed change only impacts how ratings are calculated based on the new interpolation method, not the data that is measured or how it is measured. If manufacturers have test data that is otherwise valid under the amended test procedure, there would be no reason to retest solely because of the change in the way represented values for variable speed heat pumps are calculated.

Several commenters suggested that because the change to bin-by-bin interpolation for variable speed heat pumps might cause changes in ratings, DOE should not require the new method in Appendix M. Commenters did not explain why a simple change in ratings would warrant a decision to postpone the change in method, but DOE has considered three possibilities. First, commenters may be concerned about the work to comply with the new method. However, as noted above, the new interpolation method is only a matter of calculation; it will require no new tests. DOE believes that the burden of recalculation using existing test data will be minimal; Appendix M will specify how to perform the bin-by-bin interpolation, and relatively simple revision to a spreadsheet would suffice to implement this method as a substitute for the quadratic method required under the prior test procedure. Second, commenters may be concerned about the cost of revising labels and other representation documents to reflect the new ratings. Third, some commenters may object because if the new method results in a decreased rating, that change will make the affected models appear less efficient to potential buyers.

With respect to these second and third concerns, DOE believes that the inaccuracy of the current method warrants the change. As the August 2016 SNOPR explained, the quadratic interpolation method can produce inaccurate results. For HSPF the quadratic method can produce a value up to 7.9% different from what the bin-by-bin method produces (and DOE regards the latter as more accurate). Thus, for some equipment the rated HSPF is overstated, with respect to a fair measure of efficiency, by as much as 7.9%. A buyer using such equipment would consume 7.9% more energy, at 7.9% more cost, than expected based on the rating. DOE believes that amount is a significant difference. By contrast, the regulation requires a represented cooling capacity to be within 5% of the average measured cooling capacities, and it permits rounding of figures to approximately 1% precision (200 Btu/h for a 20,000 Btu/h system). Using 1% and 5% as indicators of what amount of error in a rating is significant, DOE believes it is important to correct an interpolation method that generates, for some models, larger errors. Of course, if a rating based on the old method is still valid—including by being within the regulation's tolerances with respect to recalculated values—a manufacturer could choose whether or not to revise the rating.

For these reasons, DOE is adopting this proposal both in appendix M and appendix M1 in this final rule.

4. Outdoor Air Enthalpy Method Test Requirements

In the August 2016 SNOPR, DOE proposed modifications to requirements when using the outdoor air enthalpy method as the secondary test method, including that the official test be conducted without the outdoor air-side test apparatus connected. 81 FR at 58175-58176 (Aug. 24, 2016)

During the August 26, 2016 public meeting, Carrier suggested that the proposal to require a heat balance only for the full-load cooling test and, for a heat pump, the full-load heating test be extended to other secondary capacity measurement methods, including to use of the refrigerant enthalpy method. Carrier contended that it can be difficult to get an energy balance for some operating conditions, particularly for variable-speed systems, when there is insufficient subcooling or superheat.

10

(Carrier, Public Meeting Transcript, No. 20 at pp. 38-39) Ingersoll Rand agreed with this suggestion; Goodman also agreed and indicated that the issue applies for tests of single-stage, two-stage, and variable-speed systems for the heating mode test conducted in 17 °F outdoor temperature. (Ingersoll Rand, Public Meeting Transcript, No. 20 at p. 39; Goodman, Public Meeting Transcript, No. 20 at p. 40)

10

In this context, subcooling refers to the difference between the saturated temperature associated with the pressure of the refrigerant liquid exiting the outdoor unit (in cooling mode) and the temperature of the liquid. Similarly, superheat refers to the difference between the temperature of the refrigerant exiting the indoor unit (in cooling mode) and the saturated temperature associated with the pressure of this refrigerant. The enthalpy of the refrigerant at these locations generally cannot be determined if these values are zero.

JCI, Lennox, Carrier, Ingersoll Rand, Goodman and AHRI agreed with DOE on this proposal but recommended that the ducted test be a 30-minute test. (JCI, No. 24 at p. 12; Lennox, No. 25 at p. 12; Carrier, No. 36 at p. 7; Ingersoll Rand, No. 38 at p. 4; Goodman, No. 39 at p. 13; AHRI, No. 27 at p. 11-12) Carrier, Ingersoll Rand, Goodman and AHRI also suggested DOE similarly only require balance checks for the A

2

and H1

2

(or H1

N

) tests for the refrigerant enthalpy method. (Carrier, No. 36 at p. 7; Ingersoll Rand, No. 38 at p. 4; Goodman, No. 39 at p. 13; AHRI, No. 27 at p. 11-12) In addition, AHRI and Ingersoll Rand suggested DOE eliminate the five consecutive readings for verifying the primary capacity measurements. (AHRI, No. 27 at p. 11-12; Ingersoll Rand, No. 38 at p. 4) CA IOU and Rheem agreed with DOE's proposal. (CA IOU, No. 32 at p. 4; Rheem, No. 37 at p. 3)

DOE agrees that validation of proper capacity measurement for cooling and heating modes for full-load operation is sufficient to show that the indoor air enthalpy method is being applied properly and gives an accurate measurement. Hence, use of the secondary method and achieving an energy balance for all load levels in each operating mode is not necessary. DOE notes that systems with capacity greater than 135,000 Btu/h are tested without any requirement for a secondary capacity check. (American Society of Heating Refrigeration, and Air-Conditioning Engineers (“ASHRAE”) Standard 37-2009 (“ASHRAE 37-2009”), which is incorporated by reference into the DOE test procedures for both residential and commercial air conditioners, indicates in Table 1 that a single method is used for systems with a cooling capacity greater than 135,000 Btu/h.) Further, DOE believes this modification will help to reduce test burden. The situation discussed in the public meeting and written comments, in which, when using the refrigerant enthalpy method as the secondary test method, a heat balance cannot be calculated for some conditions due to subcooling or superheat being too low, would technically make completion of a valid test impossible, according to the current test procedure, without resorting to an alternative secondary method. DOE recognizes that use of different secondary methods for different parts of the test would significantly increase test burden. Hence, DOE is modifying the test procedure to require use of a secondary capacity measurement that agrees with the primary capacity measurement to within 6 percent only for the cooling full load test and, for heat pumps, for the heating full load test.

DOE has decided to change the names for “ducted” and “non-ducted” outdoor air enthalpy methods to avoid confusion with certain product types. Specifically, DOE is adopting the new name “free outdoor air test” for non-ducted outdoor air enthalpy test, and “ducted outdoor air test” for ducted outdoor air enthalpy test. In this final rule, DOE is also

adopting a 30-minute ducted outdoor air test with measurements at five-minute intervals, and eliminating from section 3.11.1.2 the requirement of five consecutive readings for verifying primary capacity measurements.

DOE's proposed changes to outdoor air enthalpy method requirements in the August 2016 SNOPR included revision to section 3.11.1.2 that removed the reference to section 8.6.2 of ASHRAE 37-2009. 81 FR at 58209 (Aug. 24, 2016). However, the key points of section 8.6.2 still apply for the revised approach for the outdoor air enthalpy method. The finalized test procedure retains the reference to this section.

5. Certification of Fan Delay for Coil-Only Units

In the August 2016 SNOPR DOE proposed to amend its certification report requirements to require coil-only ratings to specify whether a time delay is included, and if so, the duration of the delay used. DOE proposed to use the certified time delay for any testing to verify performance. 81 FR at 58176 (Aug. 24, 2016)

Nortek, Ingersoll Rand, Carrier, JCI, Rheem, Goodman and AHRI suggested that the certification of the indoor fan off delay should not be public information. (Nortek, No. 22 at p. 2; Ingersoll Rand, No. 38 at p. 3; Carrier, No. 36 at p. 7; JCI, No. 24 at p. 13; Rheem, No. 37 at p. 3; Goodman, No. 39 at p. 12; AHRI, No. 27 at p. 12) ADP agreed that the duration of the indoor fan time delay needs to be specified but should be a part of the public product-specific information. ADP commented that making this information public improves the accuracy of ICM AEDM ratings. (ADP, No. 23 at p. 4) Lennox and ACEEE, NRDC, and ASAP supported DOE's proposal. (Lennox, No. 25 at p. 12; ACEEE, NRDC, and ASAP, No. 33 at p. 6)

DOE understands that manufacturers want to keep fan delay setting information private. Given that DOE proposed to require this information in the section of additional product-specific information that would not be posted to DOE's public certification database, DOE has decided to adopt this proposal in this final rule. In response to ADP, DOE will address concerns regarding reporting for ICMs through a separate process.

6. Normalized Gross Indoor Fin Surface Area Requirements for Split Systems

To help ensure that the test procedure results in ratings that are representative of average use, in the August 2016 SNOPR DOE, proposed to include a provision that would prevent testing certain combinations that are not representative of single-split systems with coil-only indoor units that are commonly distributed in commerce. Specifically, DOE proposed to limit the normalized gross indoor fin surface (NGIFS) for the indoor unit used for single-split-system coil-only tests to no greater than 2.0 square inches per British thermal unit per hour (sq.in./Btu/hr). NGIFS is equal to total fin surface multiplied by the number of fins and divided by system capacity. 81 FR at 58177 (Aug. 24, 2016)

In the August 2016 Public Meeting, Ingersoll Rand commented that it did a rough calculation for a micro channel heat exchanger and determined the NGIFS to be 0.81. Ingersoll Rand commented that this indicates that there are problems with looking at today's technology and coming up with a value for NGIFS. Ingersoll Rand further commented that in coming up with a value for NGIFS, it needs to be ensured that doing so does not create issues or loopholes. (Ingersoll Rand, Public Meeting Transcript, No. 20 at p. 45) Rheem commented that there needs to be further study on the 2.0 value of NGIFS before making a decision in order to not limit future efficiencies. (Rheem, Public Meeting Transcript, No. 20 at p. 46) Carrier/UTC similarly commented that there may be unforeseen consequences of limiting design options that manufacturers will have to comply with the efficiency standards. (Carrier/UTC, Public Meeting Transcript, No. 20 at pp. 47-48) Rheem also commented that due to the complexity of the issue, the NGIFS criteria should go in appendix M1, not in appendix M. (Rheem, Public Meeting Transcript, No. 20 at p. 46) Johnson Controls commented that units that are above 2.0 today would need to be retested, and the ratings for these units would most likely change. JCI commented that for this reason, they believe that the proposal for NGIFS belongs in appendix M1, not in appendix M. (JCI, Public Meeting Transcript, No. 20 at pp. 50-51) Allied commented that the values that DOE is proposing are reasonable, but that there are further considerations associated with the different technologies that apply. Allied also commented that, based on their review, future standard levels could be even more stringent and still allow some latitude in design approaches. (Allied, Public Meeting Transcript, No. 20 at pp. 49-50) JCI also commented that usually normalized values do not have dimensions and questioned whether the proposal takes into account fin and tube spacing. (JCI, Public Meeting Transcript, No. 20 at pp. 56-59)

Nortek and AHRI opposed DOE's proposal and commented that DOE does not have the authority to regulate the design of residential central air-conditioners and heat pumps, so all NGIFS restrictions should be removed from both appendix M and M1. AHRI commented that AHRI would like to aid the Department to address this “golden blower” issue in a way which does not put restrictions on design and is both refrigerant and technology neutral. AHRI proposed to develop a solution within 30 days of the close of the August 2016 SNOPR comment period, but they did not provide additional input. (Nortek, No. 22 at p. 10; AHRI, No. 27 at p. 12)

JCI commented that while DOE stated in the SNOPR that the 2.0 limit of NGIFS does not affect 95% of tested combinations, this also showed there are current systems that will not be compliant. JCI expressed concern that if such changes are made to appendix M, standards adjustments would be required. JCI recommended that DOE limit NGIFS in M1 only and the DOE recommended value of 2.5 appears to be a valid target. (JCI, No. 24 at p. 13)

Lennox commented that while it is reasonable to use

3/8

″ round tube, plate fin coil in the NGIFS definition for outdoor units with no match, DOE must revise the definition for other split system products because there are other tube diameters and technologies used across the industry. Lennox recommended that DOE expand the definition to include all tube types and fin surfaces. Lennox supported DOE's proposal on the NGIFS calculation and proposed limit. (Lennox, No. 25 at p. 6-8) Carrier opposed DOE's proposal to limit NGIFS for the indoor unit and preferred DOE not restrict design options as that could impact consumer choices when different refrigerants are used in the future or lessen a manufacturer's ability to optimize for hot dry climates. Additionally, Carrier commented that this proposal does not address microchannel coils or any other coil tube diameter besides

3/8

″. (Carrier, No. 36 at p. 7)

Rheem objects to the limitation of a fixed value for NGIFS and proposed that indoor coil area should be determined by balancing with the outside coil area. (Rheem, No. 37 at p. 3-4) Ingersoll Rand opposed the proposed NGIFS limit because it is only appropriate for 3/8″ tube coils. Ingersoll Rand commented that it would be better to set a limit on coil cabinet volume based on coils sold in the 5 years prior to the elimination of a refrigerant. (Ingersoll Rand, No. 38

at p. 4) Goodman also expressed concern that this requirement on the tested combination may inhibit future designs and did not support the proposed restrictions. Goodman suggested that some requirements in cabinet width might be appropriate and that DOE and AHRI should work together to develop a reasonable restriction. (Goodman, No. 39 at p. 7-8)

ACEEE, NRDC, and ASAP supported DOE's proposal and also suggested DOE should consider the input of manufacturers who may have a few models designed for hot, dry climates where the apparent evaporator surface oversizing can improve rated performance. (ACEEE, NRDC, and ASAP, No. 33 at p. 6) CA IOU and NEEA agreed with DOE's proposal. (CA IOU, No. 32 at p. 4; NEEA, No. 35 at p. 3)

In response to JCI, valid normalized values may have units. For example, energy efficiency ratio is a normalized value representing capacity per electric power input with units of British thermal units (Btu) per Watt-hour (Btu/W-h). Additionally, the NGIFS does take into consideration the fin spacing—the number of fins, N

f

, is a parameter in the equation to determine NGIFS. As an example, consider two indoor coils with the same finned length—the coil with the higher fin density will have more fins and thus a higher NGIFS. It is true, however, that NGIFS does not include the impact of tube spacing.

Addressing the Ingersoll Rand and Allied comments, DOE acknowledges that NGIFS does not provide as good a representation of the heat transfer performance of microchannel indoor coils as that of conventional tube-fin indoor coils, and the development of an appropriate equivalent value for this newer technology will be important in order to prevent loopholes in the requirement. However, DOE is not aware of any significant current market share of systems using microchannel indoor coils, and so good information to use as the basis for development of NGIFS limits for this technology is not yet available. Further, the likely lower value of NGIFS for microchannel coils will mean that imposing a limit based on conventional coil technology would not limit use of microchannel coils before a better approach is developed. DOE has not developed an appropriate approach at the moment, but could consider adopting an NGIFS approach for microchannel indoor coils in a future rulemaking.

Because DOE's NGIFS analysis for coil-only systems does not consider tube diameters other than

3/8

inches and fin types other than plate fins, as well as the units currently on the market that would not meet the 2.0 NGIFS limit (

e.g.

as indicated by the JCI comment), the proposed approach does not resolve DOE's concern while maintaining a reasonable test procedure for units with different designs. Accordingly, DOE is not adopting the NGIFS requirement in this final rule for either appendix M or appendix M1. DOE will consider how best to address this issue in the future.

7. Modification to the Test Procedure for Variable-Speed Heat Pumps

The August 2016 SNOPR proposed changes to the test procedure of appendix M for variable-speed heat pumps to allow more flexibility in the design and testing of these products. 81 FR at 58177-79 (Aug. 24, 2016). The June 2016 final rule imposed restrictions on the compressor speeds that could be used in testing, indicating that full speed must be the same speed for all heating mode operating conditions. DOE adopted this approach based on the observation that extrapolation of performance outside of the range of conditions used for testing can lead to unreasonable results if the speeds are allowed to be different for the different test conditions. 81 FR at 37029 (June 8, 2016). However, the final rule discussed stakeholder comments regarding heat pumps that improve heating mode performance by using different compressor speeds at lower ambient temperatures, and indicated that consideration would be given in the future to test procedure revisions that would better address their operation.

Id.

In the August SNOPR, DOE proposed a test procedure revision that would allow testing of heat pumps whose compressors operate at higher speeds in lower ambient temperatures. 81 FR at 58177-58179 (Aug. 24, 2016). Specifically, DOE proposed the following amendments for appendix M.

• A 47 °F full-speed test used to represent the heating capacity would be required and designated as H1

N

. However, the 47 °F full-speed test would not have to be conducted using the same compressor speed (determined based on revolutions per minute (RPM) or power input frequency) as the full-speed tests conducted at 17 °F and 35 °F ambient temperatures, nor at the same compressor speeds used for the full-speed cooling test conducted at 95 °F. For appendix M, the compressor speed for the 47 °F full-speed test would be at the manufacturer's discretion, except that it would have to be no lower than the speed used in the 95 °F full-speed cooling test. Prior to the June 2016 final rule amendments, the heating capacity was represented either by the H1

2

test (for which the compressor speed guidance was not explicit), or, if a manufacturer chose to conduct what was then the optional H1

N

test, this latter test (using the same compressor speed as the full-speed cooling mode test) represented the heating capacity. Under the proposal in the August SNOPR, heating capacity would be represented only by the H1

N

test, which would be mandatory, while the compressor speed would be at the manufacturer's discretion within a range from the speed used for the 95 °F full-speed cooling test to the speed used for the full-speed 17 °F test.

• The full-speed tests conducted at 17 °F and 35 °F ambient temperatures would still have to use the same speed, which would be the maximum speed at which the system controls would operate the compressor in normal operation in a 17 °F ambient temperature, although the 35 °F full-speed test would remain optional.

• It would be optional to conduct a second full-speed test at 47 °F ambient temperature at the same compressor speed as used for the 17 °F test, if this speed is higher than the speed used for the H1

N

test described in this preamble. This test would be designated the H1

2

test. Because DOE does not expect that an H1

N

test would ever use a higher compressor speed than used for the full-speed 17 °F test, the proposed test procedure would not provide for this situation.

• If no 47 °F full-speed test were conducted at the same speed as used for the 17 °F full-speed test, standardized slope factors for capacity and power input would be used to estimate the performance of the heat pump for the 47 °F full-speed test point for the purpose of calculating HSPF.

• The capacity measured for the H1

N

test would be used in the calculation to determine the design heating requirement.

In addition, DOE proposed that the H1

N

test, at 47 °F ambient temperature, be conducted to represent nominal heat pump heating capacity, but that there would be no specific compressor speed requirement associated with it for appendix M, except that it be no lower than the speed used for the 95 °F full-speed cooling test. Under the proposal, if the H1

N

test did not use the same speed as is used for the 17 °F full-speed heating test, it would affect the HSPF calculation only through its influence on the design heating requirement, since the standardized slope factors would be used to represent full-speed heat pump performance. 81 FR at 58179 (Aug. 24, 2016)

A number of manufacturers and AHRI recommended the proposed changes should be part of appendix M1 rather than appendix M. (Rheem, Public Meeting Transcript, No. 20 at pp. 54-55; Rheem, No. 37 at p. 4; Carrier, No. 36 at p. 2; Nortek, No. 22 at p. 11; AHRI, No. 27 at p. 13; Mitsubishi, No. 29 at p. 2-3) Carrier commented at the public meeting that the proposals may be good, but that there had not been sufficient time to thoroughly review them, adding that a key concern is avoiding any potential need to retest products. (Carrier/UTC, Public Meeting Transcript, No. 20 at pp. 55). Unico recommended moving the slope factor change and the proposal for compressor speed at 47 °F test to appendix M1. (Unico, No. 30 at p. 4)

JCI recommended the proposal that the H1

2

test be conducted at maximum speed should be made optional, and the use of slope factors should be permitted if the test is not run. JCI commented that the standardized slope factors predict performance fairly closely, but can lower the HSPF by as much as 0.5 HSPF, and requested to move this change to M1. Additionally, JCI objected to DOE's proposal on H1

N

test and commented that if a manufacturer wishes to rate the heating capacity of their units at 47 °F at a speed above the A

2

speed, they should be permitted to do so. (JCI, No. 24 at p. 14)

Goodman supported DOE's proposal to require a full-speed test at 47 °F to be designated H1

N

. However, Goodman does not support the proposal to mandate that the compressor speed for this test be equal to or higher than the cooling full compressor speed. In addition, although Goodman generally supported DOE's proposal regarding the standardized slope factors to be used if no 47 °F test is run using the same compressor speed as the H3

2

test, Goodman commented that the datasets DOE's contractor have used to set the standardized slopes are not appropriate. According to Goodman, developing ratios of capacity based on certified heating capacities can lead to errors because ratings might be conservative. Further, Goodman asserted that it would be possible for models to be counted more than once, or that a limited number of an appropriate cross section of representative models would be included. Additionally, according to Goodman, varying technologies could have different slopes. Goodman suggested that DOE work with AHRI and manufacturers to review real test data. Goodman also supported the optional 5 °F test and suggested DOE to take a further step to provide an optional 5 °F test for two-speed and single-speed heat pumps. (Goodman, No. 39 at p. 5-7)

AHRI suggested that a test procedure similar to triple-capacity heat pumps should be made an optional procedure for variable-speed heat pumps. (AHRI, No. 27 at p. 13)

EEI strongly recommended that the 5 °F test and any additional considered test should remain optional. EEI also suggested that DOE should require tests and information be published for all furnaces and boilers at the same temperatures as for heat pumps. (EEI, No. 34 at p. 2)

Carrier supported DOE's modification to allow the H1

N

speed to be any speed between the 17 °F full heating speed and 95 °F full cooling speed. (Carrier, No. 36 at p. 8)

Lennox, ACEEE, NRDC, and ASAP, and NEEA supported DOE's proposals for revising the variable-speed heat pump test methods in appendix M. (Lennox, No. 25 at p. 13; ACEEE, NRDC, and ASAP, No. 33 at p. 7; NEEA, No. 35 at p. 3)

DOE considered the requests to move the proposed variable-speed heat pump test method amendments to appendix M1 and other detailed comments regarding specific aspects of the amendments. DOE revised part of its proposal as discussed later in this section. DOE's intention with the changes to the variable-speed heat pump test procedure of appendix M was to allow the tests conducted previously (

i.e.,

prior to the effective date of the June 2016 final rule) to still be used to represent heat pump performance, while preventing use of extrapolation of the performance below 17 °F using the results of tests conducted at different speeds at 17 °F and 47 °F. For this reason, DOE is not finalizing some aspects of its proposal for appendix M, and instead is finalizing them only for appendix M1.

DOE believes that the standardized slope factors (or use of same-speed tests, if a manufacturer does prefer to retest rather than use the standardized slope factors) would provide more accurate representation of heat pump performance. As discussed in section III.B.3, pursuant to 42 U.S.C. 6293(e), DOE is required to determine to what extent, if any, the proposed test procedure would alter the measured efficiency of the covered product. DOE proposed changes to heating mode test procedure for variable speed units only. For single-speed and two-speed products there would be no change in measured efficiency because they would not be impacted by this change in test procedure. However, variable-speed products would be impacted by this change in test procedure, so the measured efficiency may change.

Where an amended test procedure would alter measured efficiency, EPCA requires DOE to amend an energy conservation standard by measuring, under the amended test procedure, a sample of representative products that minimally comply with the standard. In this case, minimally compliant units are those with single-speed technology. Consistent with the statute, DOE has tested a representative sample of covered products that minimally comply with the existing standard. EPCA requires that the amended standard should constitute the average of the energy efficiency of those units, determined under the amended test procedure. As a result of that testing, DOE has determined that there is no change in measured average energy efficiency for single-speed units between the current test procedure and the amended test procedure. Thus, under 42 U.S.C. 6293(e)(2), the amended standard applicable to the amended test procedure and the current standard applicable to the amended test procedure are the same. As a result, DOE does not need to amend the existing standard to require representations of variable-speed heat pumps to be based on the amended test procedure in appendix M. Therefore, DOE is finalizing aspects of its proposal for appendix M, including the use of standardized slope factors, which might require recalculation of HSPF for variable-speed unit.

DOE believes that Unico's comment about the “changing the slope factors” may have been a comment regarding the heating load line equation slope factor rather than the standardized slope factors associated with the appendix M variable speed heat pump proposal. If so, the change was proposed only for appendix M1. If not, DOE's discussion regarding the standardized slope factors in the above paragraph responds to Unico's comment.

Based on the comments received, DOE concluded that the proposal details that commenters believed would lead to a need to retest are (a) requiring the compressor speed for the H3

2

and H2

2

tests to be the maximum speed at which the system controls would operate the compressor in normal operation in a 17 °F ambient temperature, and (b) requiring the compressor speed for the H1

N

test to be no lower than the for the A

2

test.

To resolve the first of these issues, DOE is adopting this requirement in appendix M1, but not appendix M. However, for appendix M, DOE is amending the proposal to require that

the compressor speeds used for the H3

2

and H2

2

tests be the same (if the optional H2

2

test is conducted), and will require that the compressor frequency that corresponds to maximum speed at which the system controls would operate the compressor in normal operation in a 17 °F ambient temperature be provided in the certification reports. However, DOE will not post this information to DOE's public certification database. DOE has added this reporting requirement in 10 CFR 429.16(e).

To resolve the second issue, DOE is revising its proposal to allow the compressor speed used for the H1

N

test to be lower than used for the A

2

test, provided that the H1

N

capacity is no lower than the A

2

cooling capacity. Goodman's comment regarding this issue states that it is normally the case that products on the market today have heating full compressor speed equal to or higher than the cooling full compressor speed, but Goodman believes this does not necessarily have to be the case. (Goodman, No. 39 at p. 6) While DOE agrees that such a possibility could exist, this is not a very strong statement regarding the existence of heat pumps with lower heating speed. Goodman's comment continues with an explanation that achieving roughly equivalent capacity in heating mode at 47 °F as in cooling mode at 95 °F would likely provide better performance at lower ambient temperatures.

Id.

These statements suggest that a reasonable compromise would be to allow lower H1

N

speed than A

2

speed as long as the H1

N

capacity is no lower, which is the approach that DOE has adopted in this final rule.

Similarly, JCI's comment that the compressor speed for the H1

N

test be allowed to be higher than the A

2

speed is consistent with the previously-stated approach that DOE is adopting in this final rule.

As for Goodman's suggestion regarding an optional 5 °F test for two-speed and single-speed heat pumps, DOE discusses this in section III.C.4, as part of its discussion of amendments to appendix M1.

With regard to AHRI's suggestion to add an optional test procedure for variable-speed heat pumps that is similar to the test for triple-capacity heat pumps, DOE considered this suggestion, but is declining to adopt these optional tests in this final rule because stakeholders have not been given an opportunity to comment on them. However, DOE may consider such an option in the future. In response to EEI's comment on making the proposed 5 °F test and any additional test points optional, DOE notes that it has not proposed nor adopted any new heating mode tests for heat pumps that are not optional, either in the June 2016 final rule, the August 2016 SNOPR, or this rulemaking.

In response to JCI's comment that conducting the H1

2

test at maximum speed should be made optional, DOE notes that this was optional as proposed and is optional in the test procedure adopted in this final rule.

In response to Goodman's comment about rigorous review of test data to develop the standardized slope factors, DOE requested data or suggestions regarding how they should be changed. 81 FR at 58179 (Aug. 24, 2016). However, such data were not provided. DOE notes that the standardized slope factors, which DOE derived from different data sources, some of which must have represented test data, were remarkably consistent. Further, if capacities reported for both 17 °F and 47 °F test points are conservative, it is not clear that there would be a dramatic difference in the calculated slope. Therefore, DOE has adopted the standardized slope factors proposed in the August 2015 SNOPR.

Regarding EEI's comment that furnace performance should be provided at the same temperatures and for at least two temperatures for both furnaces and CAC/HP, DOE is reluctant to impose that additional reporting burden at this time. The capacity and steady-state efficiency for furnaces does not vary significantly as a function of outdoor temperature. Thus, DOE is not convinced that the additional information would be of significant value to consumers.

8. Clarification of the Requirements of Break-In Periods Prior to Testing

In the August 2016 SNOPR, DOE proposed modifications to the test procedure to clarify the use of break-in, generalizing the requirement so that it applies regardless of who conducts the test, indicating that the break-in requirement applies for each compressor of the unit, and clarifying that the compressor(s) must undergo the certified break-in period (which may not exceed 20 hours) prior to any test period used to measure performance. 81 FR at 58179 (Aug. 24, 2016)

During the August 2016 Public Meeting, Ingersoll Rand commented that DOE's proposed rule was unclear about whether a compressor change-out is required if the compressor of a unit operates longer than the certified break-in period during product development or operation associated with test set-up prior to making the first measurement used to determine an efficiency representation. (Ingersoll Rand, Public Meeting Transcript, No. 20 at pp. 27-29).

Many stakeholders commented that changing out compressors during testing is a significant burden. Nortek suggested that DOE extend the break-in period to 50 hours and allow the break-in to be conducted at ambient conditions. (Nortek, No. 22 at p. 11) ADP and Lennox commented that the 20 hour maximum should remain in place for any verification, enforcement or other non-development testing. ADP also suggested that the break-in period should be part of the public product-specific information so that ICMs can use this information for more accurate AEDM ratings. (ADP, No. 23 at p. 4; Lennox, No. 25 at p. 13) JCI suggested DOE allow up to 72 hours of break-in time and recommended allowing break ins to be conducted before installing the compressor in the unit, or to break in a system outside of the test cell. (JCI, No. 24 at p. 14) AHRI provided data from two compressor manufacturers and suggested DOE extend the allowed break-in period to 72 hours and permit the break-in to be conducted at ambient conditions. Rheem supported AHRI. (AHRI, No. 27 at p. 13-15; Rheem, No. 37 at p. 4) Unico supported a 72-hour minimum break-in period and commented that it is easy to run the unit outside the test chamber. (Unico, No. 30 at p. 5) Emerson commented that longer break-in will ensure repeatability and improve stability of compressor performance. Emerson also included data for several compressors. (Emerson, No. 31 at pp. 1-2) Carrier suggested that DOE allow a 72-hour break-in period and allow break in outside of test chamber while running tests on other units. (Carrier, No. 36 at p. 8-9) Ingersoll Rand, Goodman and the Joint Advocates commented that there is no technical reason to establish an upper limit for break-in. Goodman suggested to permit 72 hours of break-in. (Ingersoll Rand, No. 38 at p. 4; Goodman, No. 39 at p. 8-9; Joint Advocates, No. 33 at p.7) NEEA supported DOE's proposed modification of the test procedure. (NEEA, No. 35 at p. 3)

DOE does not intend to require a compressor change-out in the development test. Rather, the establishment of the 20-hour limit is to maintain test repeatability among labs regardless of who conducts the test. DOE notes that there is no requirement in the test procedure that the break-in has to be conducted in the psychrometric chamber, so manufacturers and technicians have an option, if needed, as to where break-in

is conducted. Finally, DOE adopted the 20-hour break-in limit in the June 2016 Final Rule, and the proposal in the August 2016 SNOPR was intended to clarify how this requirement applies for manufacturers and third party testing. Accordingly, DOE will not change the 20-hour limit in this final rule.

In response to ADP's comments, DOE will discuss concerns about reporting requirements for ICMs through a separate process.

9. Modification to the Part Load Testing Requirement of VRF Multi-Split Systems

In the August 2016 SNOPR, DOE proposed to remove the 5 percent tolerance for part load operation from section 2.2.3.a of appendix M when comparing the sum of nominal capacities of the indoor units and the intended system part load capacity for VRF multi-split units. 81 FR at 58179 (Aug. 24, 2016)

DOE received no objections on this proposal, and adopts it in this final rule.

10. Modification to the Test Unit Installation Requirement of Cased Coil Insulation and Sealing

In the August 2016 SNOPR, DOE proposed to remove the statement about insulating or sealing cased coils from appendix M, section 2.2.c, in order to avoid confusion regarding whether sealing of duct connections is allowed. 81 FR at 58180 (Aug. 24, 2016)

DOE received no objections on this proposal, and adopts it in this final rule.

11. Correction for the Calculation of the Low-Temperature Cut-Out Factor for Single-Speed Compressor Systems

Equation 4.2.1-3 in section 4.2.1 of appendix M, used for calculating the low-temperature cut-out factor for a blower coil system heat pump having a single-speed compressor and either a fixed-speed indoor blower or a constant-air-volume-rate indoor blower, or for a single-speed coil-only system heat pump, was incorrectly modified in the June 2016 final rule, in that the “or” initially in the equation was changed to an “and”. 81 FR at 37107 (June 8, 2016). DOE was alerted to this issue in comments received in response to the notice of data availability (NODA) associated with the CAC/HP energy conservation standard rulemaking published October 27, 2016. 81 FR 74727. (Docket Number EERE-2014-BT-STD-0048, AHRI, No. 94 at p. 2; Unico, No. 95 at p. 1) The equation originally used “or”. This modification could have changed the range of temperature bins for which it is assumed that the heat pump function has cut out. DOE has corrected this issue in this rulemaking in appendix M and also has adopted the correct equation in appendix M1.

12. Clarification of the Refrigerant Liquid Line Insulation

In the June 2016 Final Rule, DOE adopted clarifications for insulation requirements for the refrigerant lines in section 2.2(a) of appendix M. 81 FR at 37027 (June 8, 2016). In some cases, these requirements may indicate that the refrigerant lines should be uninsulated, exposed to the air. However, DOE notes that this requirement is not appropriate to apply for every inch of refrigerant line, particularly where it would conflict with the requirements in ASHRAE 41.1-1986 (RA 2006) (referenced in section 5.1.1 of ASHRAE 37-2009, which is incorporated by reference, see § 430.3). ASHRAE 41.1-1986 (RA 2006) requires in sections 8.2 and 8.3 that it is acceptable to use surface temperature measurement for the refrigerant liquid temperature, but that insulating material extending to at least 6 in. on each side of a surface temperature-measuring element should be installed on the line. The liquid temperature measurement may be essential,

e.g.

when the refrigerant enthalpy method is used as the secondary method (see section 2.10.3 of appendix M). Therefore, DOE has decided to clarify in the test procedure (in both appendices M and M1) that the refrigerant insulation requirement in section 2.2(a) does not apply for portions of the lines insulated according to the ASHRAE 41.1-1986 (RA 2006) requirements for temperature measurement.

Because this clarification simply addresses DOE's intention on how to correctly conduct the test procedure, DOE finds that there is good cause under 5 U.S.C. 553(b)(B) to not issue a separate notice to solicit public comment on this change.

C. Amendments to Appendix M1

The November 2015 SNOPR proposed to establish a new appendix M1 to Subpart B of 10 CFR part 430, which would be required to demonstrate compliance with any new energy conservation standards. 80 FR at 69397 (Nov. 9, 2015) In the August SNOPR, DOE continued to propose establishing a new appendix M1. Under DOE's proposal, the appendix would include all of the test procedure provisions in appendix M as finalized in the June 2016 final rule, all of the changes to appendix M that are finalized in this rulemaking as discussed in section III.B, and all of the additional changes discussed in this section III.C, which would be included only in the new appendix M1. DOE proposed to make appendix M1 mandatory for representations of efficiency starting on the compliance date of any amended energy conservation standards for CAC/HP (however, note the phase-in of testing requirements for certain proposed new requirements for split systems discussed in section III.A.1).

1. Minimum External Static Pressure Requirements

Most of the residential central air conditioners and heat pumps in the United States use ductwork to distribute air in a residence, using either a fan inside the indoor unit or housed in a separate component, such as a furnace, to move the air. External static pressure (ESP) for a CAC/HP is the static pressure rise between the inlet and outlet of the indoor unit that is needed to overcome frictional losses in the ductwork. The external static pressure imposed by the ductwork affects the power consumed by the indoor fan, and therefore also affects the SEER and/or HSPF of a CAC/HP.

a. Conventional Central Air Conditioners and Heat Pumps

The current DOE test procedure

11

stipulates that certification tests for “conventional” CACs and heat pump blower coil systems (

i.e.,

CACs and heat pump blower coil systems which are not small-duct, high-velocity systems) must be performed with an external static pressure at or above 0.10 in. wc. if cooling capacity is rated at 28,800 Btu/h or less; at or above 0.15 in. wc. if cooling capacity is rated from 29,000 Btu/h to 42,500 Btu/h; and at or above 0.20 in. wc. if cooling capacity is rated at 43,000 Btu/h or more.

11

Table 3 of 10 CFR part 430 subpart B appendix M.

DOE did not propose revisions to minimum external static pressure requirements for conventional blower coil systems in the June 2010 test procedure NOPR, stating that new values and a consensus standard were not readily available.

12

75 FR 13223, 31228 (June 2, 2010). However, between the June 2010 test procedure NOPR and the November 2015 test procedure SNOPR, many stakeholders submitted comments citing data that suggested the minimum external static pressure requirements were too low and a value

of 0.50 in. wc. would be more representative of field conditions. These comments are summarized in the November 2015 test procedure SNOPR. 80 FR at 69317-69318 (Nov. 9, 2015). Ultimately, in the November 2015 SNOPR, DOE proposed to adopt, for inclusion into 10 CFR part 430, subpart B, appendix M1, for systems other than multi-split systems and small-duct, high-velocity systems, minimum external static pressure requirements of 0.45 in. wc. for units with a rated cooling capacity of 28,800Btu/h or less; 0.50 in. wc. for units with a rated cooling capacity from 29,000 Btu/h to 42,500 Btu/h; and 0.55 in. wc. for units with a rated cooling capacity of 43,000 Btu/h or more. DOE reviewed available field data to determine the external static pressure values it proposed in the November 2015 test procedure SNOPR. DOE gathered field studies and research reports, where publically available, to estimate field external static pressures. DOE previously reviewed most of these studies when developing test requirements for furnace fans. The 20 studies, published from 1995 to 2007, provided 1,010 assessments of location and construction characteristics of CAC and/or heat pump systems in residences, with the data collected varying by location, representation of system static pressure measurements, equipment's age, ductwork arrangement, and air-tightness.

13

79 FR 500 (Jan. 3, 2014). DOE also gathered data and conducted analyses to quantify the pressure drops associated with indoor coil and filter foulants.

14

The November 2015 test procedure SNOPR provides a detailed overview of the analysis approach DOE used to determine an appropriate external static pressure value using these data. 80 FR at 69318-69319 (Nov. 9, 2015). DOE did not consider revising the minimum external static pressure requirements for SDHV systems in the November 2015 test procedure SNOPR. DOE did, however, propose to establish a new category of ducted systems, short duct systems, which would have lower external static pressure requirements for testing. DOE proposed to define “short duct system” to mean ducted systems whose indoor units can deliver no more than 0.07 in. wc. external static pressure when delivering the full load air volume rate for cooling operation. 80 FR at 69314. DOE proposed in the November 2015 SNOPR to require short duct systems to be tested using the minimum external static pressure previously proposed in the June 2010 NOPR for “multi-split” systems: 0.03 in. wc. for units less than 28,800 Btu/h; 0.05 in. wc. for units between 29,000 Btu/h and 42,500 Btu/h; and 0.07 in. wc. for units greater than 43,000 Btu/h. 75 FR at 31232 (June 2, 2010)

12

In the June 2010 NOPR, DOE proposed lower minimum ESP requirements for ducted multi-split systems: 0.03 in. wc. for units less than 28,800 Btu/h; 0.05 in. wc. for units between 29,000 Btu/h and 42,500 Btu/h; and 0.07 in. wc. for units greater than 43,000 Btu/h. 75 FR at 31232 (June 2, 2010).

13

DOE has included a list of citations for these studies in the docket for the furnace fan test procedure rulemaking. The docket number for the furnace fan test procedure rulemaking is EERE-2010-BT-TP-0010.

14

Siegel, J., Walker, I., and Sherman, M. 2002. “Dirty Air Conditioners: Energy Implications of Coil Fouling” Lawrence Berkeley National Laboratory report, number LBNL-49757.

ACCA. 1995. Manual D: Duct Systems. Washington, DC, Air Conditioning Contractors of America.

Parker, D.S., J.R. Sherwin, et al. 1997. “Impact of evaporator coil airflow in air conditioning systems” ASHRAE Transactions 103(2): 395-405.

In response to the November 2015 SNOPR, the CAC/HP ECS Working Group members weighed in on appropriate minimum external static pressure requirements. (CAC ECS: CAC/HP ECS Working Group meeting, No. 86 at pp. 31-128) Recommendation #2 of the CAC/HP ECS Working Group Term Sheet states that the minimum required external static pressure for CAC/HP blower coil systems other than mobile home systems, ceiling-mount and wall-mount systems, low and mid-static multi-split systems, space-constrained systems, and small-duct, high-velocity systems should be 0.50 in. wc. for all capacities. (CAC ECS: ASRAC Term Sheet, No. 76 at p. 2)

In the August 2016 SNOPR, DOE proposed to adopt a minimum external static pressure requirement of 0.50 in. wc. for systems other than mobile home, ceiling-mount and wall-mount systems, low and mid-static multi-split systems, space-constrained systems, and small-duct, high-velocity systems based on DOE's analysis and consistent with the CAC/HP ECS Working Group Term Sheet. 81 FR at 58181 (Aug. 24, 2016)

During the August 2016 SNOPR public meeting and in written comments, many stakeholders expressed support for the new minimum external static requirements that DOE proposed. JCI, Goodman, Unico, AHRI, NEEA, Carrier/UTC, Lennox, Ingersoll Rand, and Nortek expressed support for DOE's proposal to require conventional systems to be tested at a minimum external static pressure of 0.5 in. wc. consistent with Recommendation #2 of the Term Sheet. (JCI, No. 24 at p. 15; Goodman, No. 39 at p. 13; Unico, No. 30 at p. 6; AHRI, No. 27 at p. 16; NEEA, No. 35 at p. 3; Carrier/UTC, No. 36 at p. 9; Lennox, No. 25 at p. 10; Ingersoll Rand, No. 38 at p. 5; Nortek, No. 22 at p. 11)

In light of DOE's analysis results, the Term Sheet recommendation, and support expressed in written comments, DOE is adopting a minimum external static pressure of 0.50 in. wc. for all capacities of conventional CAC/HP products in this final rule.

b. Non-Conventional Central Air Conditioners and Heat Pumps

In response to the November 2015 SNOPR and during the CAC/HP ECS Working Group negotiations, DOE also received comment regarding the minimum external static pressure requirements for mobile home systems, ceiling-mount and wall-mount systems, low and mid-static multi-split systems, spac

This text is long and has been trimmed here. Open the source document for the complete record.