Equipment ranking
Cold-climate heat pumps: what the federal designation actually certifies
Three published requirements, one open dataset, and the certified units that sit at the top of each measured metric.
Written by HYRE HVAC Research Desk Primary-source research, data analysis and fact checking
The short answer
"Cold climate" on a heat pump is an ENERGY STAR designation with three requirements: a COP of at least 1.75 at 5°F, at least 70% of its 47°F heating capacity at 5°F, and a controls verification test.
HyreHVAC found 514 undesignated families that meet both numbers anyway, so a spec sheet is not proof. The tables rank the 6,263 designated families on one measured metric each.
What are the three cold-climate requirements?
A COP floor, a capacity floor and a controls test. Everything on this page rests on one published specification, so it is worth quoting rather than paraphrasing.
ENERGY STAR’s Heat Pump Equipment Key Product Criteria, retrieved September 6, 2026, sets a Cold Climate designation on top of the ordinary certification.
A unit must first clear the efficiency floors for its type — 8.5 HSPF2 and 15.2 SEER2 for non-ducted split systems, 8.1 HSPF2 and 15.2 SEER2 for ducted splits and single-package equipment.
Then it must demonstrate low-ambient performance on all three of the following.
- Efficiency at 5°F
- "COP at 5° F ≥ 1.75, measured in accordance with Appendix M1 H4₂ test." A coefficient of performance of 1.75 means 1.75 units of heat delivered per unit of electricity consumed, at an outdoor temperature of 5°F. For comparison, the electric resistance strips that back a heat pump up run at exactly 1.00.
- Capacity at 5°F
- "Percent of Heating Capacity at 5°F ≥ 70% of that at 47°F" — the 5°F figure from the H4₂ test, and the 47°F figure the unit’s nominal heating capacity.
- For a variable-speed compressor the criteria specify that the 47°F point is taken at "the maximum speed that the system controls would operate at 47°F".
- This is the requirement that answers the old objection about heat pumps losing output as it gets cold.
- A controls verification procedure
- "Perform a controls verification procedure (CVP) to confirm that the above performance metrics measured at the Appendix M1 low ambient test point at 5° F are achieved by the native controls operating as they would in a customer’s home."
- In plain terms: prove the machine does this by itself, on its own logic, rather than in a laboratory configuration a homeowner will never reproduce. Either the ENERGY STAR CVP or a DOE CVP adopted into the Code of Federal Regulations is acceptable.
Source: ENERGY STAR, Heat Pump Equipment Key Product Criteria, retrieved September 6, 2026. Quoted verbatim; the emphasis and the plain-English glosses are HyreHVAC’s.
Why does the controls test matter most?
Because it is the one requirement a number on a data plate cannot satisfy. The two numeric thresholds are the ones every manufacturer quotes and every competing page repeats. They are also the easy part.
HyreHVAC calculation: among the 2,078 certified families that do not carry the Cold Climate designation and that publish both of the values needed to test them, 514 of them — 24.7% — meet both numeric thresholds anyway.
They clear 1.75 at 5°F and they hold more than 70% of their capacity. They are still not cold-climate certified.
Reading the criteria explains it. The numbers are necessary and not sufficient, because the CVP is a separate requirement, and 140 of those 514 families have no controls verification procedure recorded at all. The pattern holds across the whole file: 0.9% of designated families record no controls verification, against 57.4% of undesignated ones.
HyreHVAC analysis: this is why the designation is worth insisting on in a quote rather than accepting a spec sheet.
A laboratory can coax a low-ambient number out of a machine by holding the compressor at a speed and the defrost logic at a setting that the shipped controls would never choose on a February morning.
The CVP is the requirement that closes that gap, and it is the only one of the three that a manufacturer cannot satisfy by pointing at a number on a data plate.
What the designation is not is a quality tier. 74.0% of certified families carry it.
It certifies a specific low-temperature behavior, not that the equipment is better in any other respect — and on efficiency in the cold, the designated and undesignated populations are much closer than the marketing implies.
The full distribution is in our cold-climate performance study, which measures the whole certified market rather than its top.
What does the 70% capacity floor buy you?
It buys heat on the coldest days, and the floor is not a formality.
HyreHVAC calculation: across designated families the median capacity held at 5°F is 83.3% and even the 10th percentile is 70.4% — the whole designated distribution sits above the requirement, as it must.
Across all certified families the 10th percentile is 56.7%, which is to say that the bottom of the uncertified-for-cold population loses nearly half its output on the coldest days.
That difference is the practical argument. A unit that holds 83.3% of its rated capacity at 5°F is still heating the house at 5°F.
A unit that holds 56.7% has handed most of the job to its backup — normally electric resistance strips running at a coefficient of performance of 1.00, which is the most expensive heat available in almost every state.
Retention above 100% appears in 528 designated families and is physically real rather than a data error: the criteria define the 47°F point as the nominal capacity at the speed the controls would use at 47°F, while the 5°F point is measured at maximum capacity.
A variable-speed compressor with headroom genuinely produces more heat at 5°F flat out than at its nominal mild-weather rating. It is not free — the coefficient of performance falls as it does so — but the capacity is there.
Which ductless units have the highest COP at 5°F?
The ductless ranking below is ordered on one thing: the certified coefficient of performance at the 5°F test point, among designated ductless families. It is the metric that decides what the equipment costs to run on the coldest days.
It is also the metric on which the certified market is most tightly bunched. The median designated family sits at 2.00 and the 90th percentile at 2.24, so the 2.89 at the top of this list is genuinely unusual rather than a rounding win.
| Outdoor unit | Sold as | COP at 5°F | Capacity held at 5°F | HSPF2 | Rated cooling |
|---|---|---|---|---|---|
| DLCPRBH18AAK | Airquest, Arcoaire, Comfortmaker, Day & Night +8 more | 2.89 | 120% | 15.5 | 18,000 Btu/h |
| 38MPRBQ18AA3, DMA18HOS42230E4, DMB18HOS42230E8B +2 more | Bryant, Carrier, Goodman, MDV +4 more | 2.89 | 120% | 11.0 | 18,000 Btu/h |
| M3OX430-27HFN8-M1, MOM27H2XE3A | Innovair, MDV | 2.80 | 79% | 10.0 | 27,000 Btu/h |
| MCY-MAP0367HS-UL* | TOSHIBA CARRIER | 2.70 | 79% | 10.9 | 36,000 Btu/h |
| AOUH12KTAP1 | Fujitsu | 2.62 | 106% | 13.7 | 12,000 Btu/h |
| PAS-18BUFASDQ1-LAC | Hitachi | 2.60 | 83% | 11.0 | 18,000 Btu/h |
| PAS-18BUFASDQ1, YPAS18BUFASDQ1 | Hitachi, York | 2.60 | 81% | 11.0 | 18,000 Btu/h |
| RAC-GJ09WHAA, YRACGJ09WHAA | Hitachi, York | 2.57 | 93% | 11.4 | 9,000 Btu/h |
| 4C25SCH24H21, ACIQ-24ZS-HP230C, CH-RES24-230VO +2 more | 1HVAC, ACIQ, CanAir, Cooper&Hunter +2 more | 2.54 | 78% | 11.7 | 24,000 Btu/h |
| BMS2423YC, BMS500-AAS024-1CSXRD, BXS3A24SC0 +11 more | AC PRO, BLUERIDGE, Bosch, Custom Comfort +7 more | 2.54 | 78% | 11.7 | 24,000 Btu/h |
US EPA, ENERGY STAR Certified Air-Source Heat Pumps (data.energystar.gov, dataset w7cv-9xjt), retrieved September 5, 2026. HyreHVAC calculation over 8,465 deduplicated outdoor-unit families. Designated ductless families only; ties on an identical certified rating profile are shown as one entry.
Ordered on coefficient of performance at 5°F alone. Not on price, reliability, warranty or noise — no sourced data for those exists on this site. The top two entries tie on 2.89; they are separate outdoor model numbers with different HSPF2 ratings, so they are not collapsed.
Which ducted units have the highest COP at 5°F?
Ducted and ductless equipment are separated because they are not comparable. A ductless head blows into the room it serves; a ducted system pays for a fan pushing air through a duct network, and its certified numbers reflect that.
Ranking the two together would put ductless at the top of every table and tell the reader nothing. The ducted designated median coefficient of performance at 5°F is 2.00, against 2.00 ductless.
| Outdoor unit | Sold as | COP at 5°F | Capacity held at 5°F | HSPF2 | Rated cooling |
|---|---|---|---|---|---|
| ACIQ-12-EHPB, MOX330U-12HFN1-MR0 | 1HVAC, ACIQ, DIYCOOL, MDV +2 more | 2.60 | 85% | 9.0 | 12,000 Btu/h |
| FOP1560TJAZNA, MLP15AZ60AJ2N, SP15AZ60AJ2N | Fujitsu, MAINLINE, RUSSELL, Sure | 2.60 | 60% | 8.1 | 57,000 Btu/h |
| AC060FCUPCG | Samsung | 2.58 | 79% | 9.0 | 55,000 Btu/h |
| AC048FCUPCG | Samsung | 2.52 | 84% | 8.5 | 47,000 Btu/h |
| AOH-48U3T24P, AUWR-4824P, GRID48OC +6 more | Endura, Gridless, HAXXAIR, Hisense +4 more | 2.52 | 77% | 9.0 | 48,000 Btu/h |
| AUWR-60U3TH5 | Hisense | 2.52 | 74% | 9.0 | 57,000 Btu/h |
| FOP1548TJAZNA, MLP15AZ48AJ2N, RP15AZ48AJ2N +2 more | Fujitsu, MAINLINE, RUSSELL, Rheem +3 more | 2.50 | 60% | 8.1 | 45,500 Btu/h |
| TZPLAD1560AJN | Thermal Zone | 2.50 | 59% | 8.1 | 57,000 Btu/h |
| MOE30-36HFN10-M3X | MDV | 2.46 | 100% | 10.0 | 36,000 Btu/h |
| AOH-60U3T24P, AUWR-6024P, GRID60OC +6 more | Endura, Gridless, HAXXAIR, Hisense +4 more | 2.46 | 74% | 9.0 | 57,000 Btu/h |
US EPA, ENERGY STAR Certified Air-Source Heat Pumps (data.energystar.gov, dataset w7cv-9xjt), retrieved September 5, 2026. HyreHVAC calculation over 8,465 deduplicated outdoor-unit families. Designated ducted split and single-package families only.
A ducted rating is measured at a defined external static pressure. Real duct systems are frequently worse than the test condition, which is one of the largest single reasons an installed system underperforms its certificate.
Which ductless units hold the most heat at 5°F?
A different question from the last two, and for a cold-climate buyer often the more important one: not how efficiently the unit runs at 5°F, but how much heat it can still make.
Every entry here exceeds 100%, which — as above — is variable-capacity headroom rather than an anomaly. These are the units least likely to hand the job to backup strips on a design day.
| Outdoor unit | Sold as | Capacity held at 5°F | 47°F output | 5°F output | COP at 5°F |
|---|---|---|---|---|---|
| AOUG09LZAS1 | Fujitsu | 171% | 9,000 Btu/h | 15,400 Btu/h | 2.14 |
| AOUG09LZAH1 | Fujitsu | 171% | 9,000 Btu/h | 15,400 Btu/h | 2.00 |
| A-KZP06SA-1, A-ZHP06SA-1, A-ZKP06SA-1 +6 more | AI AIR CONDITIONING, BLUERIDGE, Comfort Aire, Comfort-Aire +5 more | 150% | 9,000 Btu/h | 13,500 Btu/h | 2.00 |
| D5CPRAH06AAK | Airquest, Arcoaire, Comfortmaker, Day & Night +7 more | 150% | 9,000 Btu/h | 13,500 Btu/h | 2.00 |
| 37MPRAQ06AA3, MOX430-06HFN10-M1H | Bryant, Carrier, MDV, Midea +2 more | 150% | 9,000 Btu/h | 13,500 Btu/h | 2.00 |
| AOU9RLFF | Fujitsu | 149% | 9,000 Btu/h | 13,400 Btu/h | 1.90 |
| AOU9RLFFH | Fujitsu | 149% | 9,000 Btu/h | 13,400 Btu/h | 1.90 |
| 4TXK3809B10NU**, AWH09ACD-D3DNA1A/O, BMKH0938/O +5 more | AC PRO.COM, ACD, American Standard, BLUERIDGE +5 more | 144% | 9,000 Btu/h | 13,000 Btu/h | 1.91 |
| AOUH09KTAP1 | Fujitsu | 142% | 12,000 Btu/h | 17,000 Btu/h | 2.48 |
| MUZ-FX06NLHZ***, MUZ-FX06NLHZ2*** | Mitsubishi Electric | 140% | 9,000 Btu/h | 12,600 Btu/h | 2.38 |
US EPA, ENERGY STAR Certified Air-Source Heat Pumps (data.energystar.gov, dataset w7cv-9xjt), retrieved September 5, 2026. HyreHVAC calculation over 8,465 deduplicated outdoor-unit families. Designated ductless families, ordered on capacity held at 5°F.
High retention on a small unit is not the same as high absolute output. Every entry here is a 6,000 to 9,000 Btu/h head — read the 5°F output column, not the percentage, if you are sizing for a whole house.
How do you use this on a real quote?
A certified rating belongs to a matched outdoor and indoor pairing, not to an outdoor unit. Swapping the air handler changes the numbers and can lose the designation entirely.
The reference number on the quote is what you look up in the AHRI Directory to confirm the pairing you are actually buying was the one that was tested.
514 certified families clear both numeric thresholds without carrying the designation. If a salesperson offers a spec sheet showing a COP above 1.75 at 5°F as evidence of cold-climate capability, that is a necessary condition presented as a sufficient one.
Equipment is quoted at 47°F. Your county’s winter design temperature is the number the system has to meet, and the certified 17°F and 5°F capacities are public for every model. An installer sizing on the nameplate has not sized for your winter.
Below the point where the heat pump can no longer meet the load, resistance strips take over at a coefficient of performance of 1.00.
Whether that is a rounding error or the dominant line on your bill depends entirely on your electricity price against your alternative fuel — our breakeven study computes the threshold for all 51 jurisdictions.
It is a laboratory metric, ordered. It contains no information about installed cost, reliability, parts availability, dealer coverage or noise, and the combination you are quoted may rate below the one that earned the ranking.
How was this list built, and what does it leave out?
The ENERGY STAR file lists one row per certified outdoor + indoor combination.
One outdoor unit sold against forty air handlers is forty rows, and the same physical outdoor unit sold under a group’s other labels is forty rows again per label.
Ranking rows would return the same machine over and over. HyreHVAC calculation: 281,975 rows describe far fewer machines than that.
Key on outdoor unit brand name plus outdoor model number. This is the rule already published on this site in the cold-climate performance study, kept unchanged so the two cannot disagree. 281,975 rows collapse to 12,516 distinct outdoor units.
Key on the outdoor model number alone and carry every brand name it is sold under.
1,394 model numbers are sold under more than one brand and 6 of them appear under 17 (Airquest, Arcoaire, Bryant, Carrier, Comfortmaker and 12 others all sell D5CUHAH18AAK).
Without this stage a top-ten list is one machine repeated. 12,516 units collapse to 8,465 rebrand families, absorbing 4,051 duplicate listings.
Stage 2 cannot collapse the same platform sold under different model numbers by different equipment groups, because ENERGY STAR records no manufacturer field to join on — icm_outdoor_unit_manufacturer is populated on 2 of 281,975 rows.
At the top of a certified list that omission dominates. So entries whose full certified rating profile is identical are displayed as one tied entry listing every model number and brand in it.
Across the whole file that is 2,521 distinct rating profiles. An identical profile is consistent with one platform under several labels; we do not assert it, because the dataset gives us no field with which to confirm it.
Within a family, different indoor pairings rate differently. The figure shown is the maximum certified value achieved by any certified matched combination of that outdoor unit — a defined, reproducible quantity, and the one a manufacturer would quote.
Your own installed pairing may rate lower: Only the matched combination is certified, which is why the AHRI reference number on your quote is worth checking.
A test artefact in the federal file — brand "BRAND10", model "TEST-COND41966-16(2)", reporting a coefficient of performance at 5°F of 10.0 — would otherwise top every cold-weather ranking. It is not a product. It is the only record excluded on judgment rather than on a missing value.
Price, installed cost, reliability, warranty claims, service network, noise, or how the unit behaves in a real house. HyreHVAC holds no sourced data for any of those and does not rank on them. There is no "best overall" here and there will not be one.
Every figure is a certified steady-state measurement under DOE Appendix M1. A real installation gives performance back to defrost cycles, duct leakage, oversizing, undersized returns and incorrect refrigerant charge. Installed performance is below rated performance, sometimes substantially.
ENERGY STAR certification is voluntary. Uncertified equipment exists and is generally less efficient, so this describes the better end of what can be bought. It is also a census of models, not of sales: a model that sells once counts the same as one that sells a million. No public dataset supports a sales-weighted version.
133 of the 6,263 designated families — 2.1% — do not reach 70% capacity retention on the capacity fields the dataset publishes, on any of their certified combinations.
The criteria define the 47°F point as the nominal capacity from a specific test, and the file’s 47°F column is not guaranteed to be that point for every record.
We report the count rather than smoothing it, and we do not assert which of the flag and the fields is authoritative.
The 514 figure is a HyreHVAC calculation over the federal file, not a statement by EPA.
It compares the two published numeric thresholds against undesignated families that publish both of the values needed to test them.
Families missing either value are excluded from that comparison, which is why the denominator is 2,078 rather than the full undesignated population.
Questions
What does the ENERGY STAR Cold Climate designation mean?
Is a heat pump with a high COP at 5°F the same as a cold-climate heat pump?
What is the highest COP at 5°F among certified cold-climate heat pumps?
How much capacity does a cold-climate heat pump lose at 5°F?
Do you rank these on price or reliability?
Why does the same unit appear under a dozen brand names?
Are these ratings from the AHRI Directory?
Written and audited by
HYRE HVAC Research Desk
Primary-source research, data analysis and fact checking
We are a research desk, not a sales floor. We read the federal microdata file, the statute or the manufacturer data sheet ourselves, and we publish the figure with the document it came from and the date we retrieved it.
Where a number cannot be traced to a primary source, we publish the shorter page and say what we could not verify.
The counts below are generated from the published pages themselves, last counted September 28, 2026, and they are what we have actually published rather than what we intend to.
- 13
- studies published
- 12
- federal sources read and cited
- 8
- studies published with their full dataset as CSV
- 51
- jurisdictions reproduced against EIA’s own tables
How this desk works
- Primary sources only. Federal data comes from the agency that collects it, in the file that agency publishes. We do not cite an article that cites a source; we download the source and compute the figure ourselves.
- We validate against the agency before we publish. First, we use each federal microdata file to reproduce the agency’s own published tables. Our cooling research reproduces EIA’s state estimates and standard errors for all 51 jurisdictions. That check caught a variance formula that was off by a factor of four.
- Every estimate carries its uncertainty. These are survey figures, not counts. Standard errors are computed from the replicate weights the federal file supplies and printed beside the estimate. An estimate too imprecise to publish is reported as such rather than printed.
- Nothing is typed by hand. Prose, tables and charts all read from one dataset built by script, so a number in a sentence and the same number in the table below it cannot disagree.
- We publish the data, not just the conclusion. 8 of our 13 studies offer the full computed table as a CSV download on the page, so you can check the analysis or disagree with it. A study without a row-level dataset gets no download link and claims none in its structured data.
- We correct in public. Where we have published a figure wrongly we fix the figure, rewrite any analysis that rested on it rather than patching the number underneath it, and leave a dated correction note on the page.
- We do not install or sell HVAC equipment, and we take no payment for placement, ranking or a favorable mention. Nobody buys a position on this site.
Data as of ENERGY STAR certified product data, retrieved September 5, 2026. Authorship on this site is organizational: the analysis belongs to the desk rather than to a named individual, and we do not publish credentials we do not hold. Our editorial policy sets out how we source, date and correct what we publish.
Sources & retrieval dates
The rating is the easy half
A certified number describes a laboratory. The load calculation, the ductwork and the commissioning decide what happens in your house on the coldest morning of the year.