Ducted Air-Source Heat Pumps from American Manufacturers

musingsheader image

Ducted Air-Source Heat Pumps from American Manufacturers

Are there any cold-climate heat pumps that can be connected to regular forced-air ductwork?

Posted on Jul 6 2018 by Martin Holladay

UPDATED on July 20, 2018, with new information from Carrier.

Mitsubishi and Fujitsu sell air-source heat pumps (ductless minisplits and ducted minisplits) that work well in cold climates. Many GBAGreenBuildingAdvisor.com readers report that these appliances are providing dependable space heating in climates as cold as Minnesota, Vermont, Maine, and Quebec, where temperatures drop to -20°F or colder.

Traditionally, U.S. manufacturers of air-source heat pumps have favored ducted units over ductless units. Most air-source heat pumps sold by Bryant, Carrier, Lennox, and Trane, for example, are designed to be hooked up to forced-air ductwork, just like a typical furnace.

Here at GBA, we often advise readers that ductless minisplits are a good way to heat and cool a compact well-insulated house. That said, ductless minisplits don’t work as well for large, spread-out homes or for homeowners who want uniform heating and cooling in every room, even when doors are kept closed.

Most U.S. homes have forced-air HVAC(Heating, ventilation, and air conditioning). Collectively, the mechanical systems that heat, ventilate, and cool a building. systems — that is, systems that distribute heat and cooling through ducts. Some GBA readers wonder, “Are there any cold-climate heat pumps that can be hooked up to conventional forced-air ductwork — the type of ductwork found in a typical American house?” We’ll do our best to answer that question.

Note that there is always an energy penalty associated with any heating or cooling equipment connected to ductwork. Equipment connected to ducts will never be as efficient as the best ductless equipment, because it takes a substantial amount of electricity to push air through ducts.

What about ducted minisplits?

Japanese manufacturers of ductless minisplits also manufacture ducted minisplits. Even though these units are designed to be connected with ductwork, they generally can’t be connected to the type of forced-air ductwork installed in a typical American house.

Most ducted minisplit systems lack the type of powerful blower found in an American air handler, so ducted minisplits usually require a carefully designed duct system with low static pressure. Mitsubishi’s SEZ air handler (used on the Mr. Slim models) has a top setting of 0.20 i.w.g. of external static pressure, while the Fujitsu RLFC series has a capability of 0.36 i.w.g. of external static pressure on the smaller air handlers, with an even lower static pressure setting for the larger air handler in the series.

That means that the ducts for these Japanese ducted minisplit systems usually need to be shorter than is typical for American ducts. These duct systems may also have fewer elbows and have larger diameter ducts than ducts connected to a big American air handler.

The approach used by Japanese ducted minisplit designers — choose a smaller blower and require careful duct design — results in a more energy-efficient distribution system than the American approach (which, with only minor exaggeration, can be characterized this way: “We’ll solve static pressure problems with a bigger blower motor”).

American ducted air-source heat pumps exist — but there are asterisks

Available U.S. heat pumps designed for ducted systems have an Achilles’ heel: Unlike cold-climate heat pumps from Mitsubishi and Fujitsu, air-source heat pumps from U.S. manufacturers won’t perform well when outdoor temperatures drop below 5°F. These units also lack the impressive turndown ratios that purchasers of Asian minisplits have come to expect.

American Standard, Bryant, Carrier, Lennox, and Trane each has at least one air-source heat pump that provides useful amounts of space heat when the temperature drops to 5°F. So if you live somewhere where the outdoor design temperatureReasonably expected minimum (or maximum) temperature for a particular area; used to size heating and cooling equipment. Often, design temperatures are further defined as the X% temperature, meaning that it is the temperature that is exceeded X% of the time (for example, the 1% design temperature is that temperature that is exceeded, on average, 1% of the time, or 87.6 hours of the year). is 5°F, you might consider installing a U.S.-made air-source heat pump and hooking it up to conventional forced-air ductwork.

Areas where this approach will work

What cities are warm enough to use U.S.-made heat pumps? Working from the West coast to the East coast, here is a list of some of the North American cities that have an outdoor design temperature of between 4°F and 8°F (using the 97.5% winter design temperatures from ASHRAEAmerican Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE). International organization dedicated to the advancement of heating, ventilation, air conditioning, and refrigeration through research, standards writing, publishing, and continuing education. Membership is open to anyone in the HVAC&R field; the organization has about 50,000 members. Fundamentals). If you live in one of these cities (or anywhere warmer), you can take the approach discussed in this article.

  • Ellensburg, Washington: 6°F
  • Pendleton, Oregon: 5°F
  • Lewiston, Idaho: 6°F
  • Provo, Utah: 6°F
  • Flagstaff, Arizona: 4°F
  • Boulder, Colorado: 8°F
  • Gallup, New Mexico: 5°F
  • Dodge City, Kansas: 5°F
  • Salina, Kansas: 5°F
  • St. Louis, Missouri: 6°F
  • Carbondale, Illinois: 7°F
  • Detroit, Michigan: 6°F
  • Grand Rapids, Michigan: 6°F
  • Bloomington, Indiana: 5°F
  • Covington, Kentucky: 6°F
  • Windsor, Ontario: 4°F
  • Akron, Ohio: 6°F
  • Cleveland, Ohio: 5°F
  • Elkins, West Virginia: 6°F
  • Scranton, Pennsylvania: 5°F
  • Buffalo, New York: 6°F
  • Rochester, New York: 5°F
  • Framingham, Massachusetts: 6°F
  • Halifax, Nova Scotia: 5°F
  • St. John’s, Newfoundland: 7°F

The NEEP list

Before looking at the equipment offerings available from manufacturers, it’s worth mentioning the very useful database of cold-climate heat pumps maintained by the Northeast Energy Efficiency Partnership (NEEP), a non-profit group in Framingham, Massachusetts.

The NEEP database includes heating capacity of listed equipment at an outdoor temperature of 5°F — information which can be surprisingly difficult to pry out of the hands of secretive equipment manufacturers. To be included on the NEEP list, an air-source heat pump must have a coefficient of performance (COPEnergy-efficiency measurement of heating, cooling, and refrigeration appliances. COP is the ratio of useful energy output (heating or cooling) to the amount of energy put in, e.g., a heat pump with a COP of 10 puts out 10 times more energy than it uses. A higher COP indicates a more efficient device . COP is equal to the energy efficiency ratio (EER) divided by 3.415. ) of at least 1.75 at an outdoor temperature of 5°F.

NEEP lists five ducted air-source heat pumps with a heat output of at least 38,000 BTUBritish thermal unit, the amount of heat required to raise one pound of water (about a pint) one degree Fahrenheit in temperature—about the heat content of one wooden kitchen match. One Btu is equivalent to 0.293 watt-hours or 1,055 joules. /h at 5°F. (I mention this capacity because it is large enough to serve most single-family homes.) These heat pumps are:

  • American Standard AccuComfort Platinum18 model 4A6V8060A1 (5 tons)
  • Bryant Extreme model 280ANV048 (4 tons)
  • Carrier Infinity 20 Greenspeed model 25VNA048A (4 tons)
  • Trane TruComfort XV18 model 4TWV8060A1 (5 tons)
  • Mitsubishi P-series HyperHeat model PUZ-HA36NHA5 (3 tons).
  • NEEP also lists a Lennox air source heat pump, the Lennox Elite XP20, but its capacity at 5°F is only 19,000 BTU/h.

    What about the turndown ratio?

    Professionals who design heating systems usually start by performing a Manual J load calculation to determine the peak heating loadRate at which heat must be added to a space to maintain a desired temperature. See cooling load. and peak cooling load at design conditions. Once this is done, they can select equipment that is capable of meeting these loads.

    Since design conditions — roughly speaking, the coldest and hottest days of the year — don’t occur very often, HVAC equipment is usually operating at partial load, not at design load.

    With an old-fashioned single-stage furnace, the only way to handle a partial load situation is for the equipment to cycle on and off. Although this approach works, it is somewhat inefficient (because of energy use associated with frequent start-ups) and may lead to comfort and noise complaints. Unlike a single-stage furnace, a newer inverted-based heat pump from Mitsubishi and Fujitsu operates at variable speeds (in other words, with variable outputs), reducing the number of days when equipment operates with a stop-and-go jerkiness.

    Ideally, an air-source heat pump will have a high turndown ratio — that is, a maximum heating capacity that is a high multiple of the minimum heating capacity. For air-source heat pumps, the usual definition of the turndown ratio is the maximum heat output capacity at 5°F divided by the minimum heat output capacity at 47°F. As Bruce Harley, a Vermont-based energy consultant, notes, the turndown ratios seen in U.S.-made air-source heat pumps are “not nearly what we see in minisplits” from Asia.

    Some Mitsubishi minisplits (for example, the Mitsubishi FH09NA) have turndown ratios as high as 11-to-1. That means that under partial load conditions, the Mitsubishi FH09NA can operate continuously while delivering just 9% of the heat that would be required when the unit is operating at maximum capacity — for example, on the coldest day of the year.

    If you look at a unit’s turndown ratio from the perspective of heating performance alone, it’s easy to make the case that a high turndown ratio improves comfort and efficiency under partial load conditions. But since cold-climate homeowners often size an air-source heat pump to meet the peak heating load rather than the cooling load, a high turndown ratio also affects cooling performance. (In a cold climate, a house might need a 3-ton air-source heat pump to meet the peak heating load, while it only needs a 2-ton unit to meet the peak cooling load.)

    In one of GBA’s “Q&A Spotlight” articles, Dana Dorsett was quoted as saying that the limited turndown ratio (about 2.5 to 1) of the 2-ton Trane XV20i heat pump makes the cooling performance of this unit sub-optimal. “In a new house design, it’s not super-hard to hit 2,500-square-feet-per-ton of cooling, and even a code-minimum house can hit the one ton per 2,000 [square foot] range, even with full sun exposure,” said Dorsett. “The minimum modulated output of the 2-ton Trane XV20i is about 10,000 BTU/h at 95°F outside and even higher at lower outdoor temperatures. That means it can’t really modulate at high efficiency if your [cooling] load at your 1% design condition is only 12,000-15,000 BTU/h.”

    American Standard and Trane

    American Standard and Trane are two brands owned by the same company, Ingersoll Rand. The American Standard AccuComfort Platinum18 model 4A6V8060A1 (5 tons) is a ducted air-source heat pump listed by NEEP.

    Two air-source heat pumps with the Trane brand are often recommended for cold-climate homes: the Trane TruComfort XV18 (which is on the NEEP list) and the Trane XV20i.

    The heating capacity of the 3-ton Trane XV18 (model 4TWV8036A1) at an outdoor temperature of 5°F is about 20,500 BTU/h. This model has a turndown ratio of about 2.4 to 1.

    The heating capacity of the Trane XV20i at an outdoor temperature of 5°F is about 27,000 BTU/h. At an outdoor temperature of 5°F, the Trane XV20i can deliver 63% of its rated capacity at 47°F.

    Attempts to contact representatives at American Standard and Trane for answers to technical questions were unsuccessful.

    Bryant and Carrier

    Bryant equipment and Carrier equipment are both manufactured by the same company, CCS Americas.

    For equipment with the Bryant brand, the usual advice is to consider the Bryant Extreme. Byrant Extreme model 280ANV is on the NEEP list.

    For equipment with the Carrier brand, the usual advice is to consider the Carrier Infinity Greenspeed. Carrier Infinity 20 Greenspeed model 25VNAO is on the NEEP list.

    In a GBA comment, Dana Dorsett wrote that at an outdoor temperature of 47°F, a 2-ton Carrier Infinity Greenspeed 25VNA024 “can still throttle down to the 10,000 BTU/hr range,” but at 25°F the unit has a capacity of “25,000 BTU/hr or more. … In my area [this type of air-source heat pump] is usually more expensive than a minisplit solution, but may be comparable [in cost] to or cheaper than the 2.5-ton Mitsubishi PVA/PUZ -A30.”

    [Author's note: Further information on the Carrier Infinity Greenspeed was provided by Dana Dorsett in a comment (Comment #2) posted on this page. Dorsett wrote, “For both the 2-ton 25VNA024A**30 and 3-ton 25VNA036A**30, the capacity at -15°F is about half its capacity at 5°F, and the capacity at -10°F is about half the AHRI rated capacity at 17°F. The drop in capacity is fairly linear with temperature from 5°F on down.” As Dorsett explains, he obtained this information using an online Carrier tool called the GreenSpeed Calculator.]

    [Second author's note: On July 19, 2018 — five weeks after I emailed some technical questions to a Carrier representative — I received a response from Todd Nolte, a director of product management at Carrier. Nolte wrote, "The Carrier Greenspeed Intelligence heat pump has performance specifications to outdoor temperatures of negative 3 degrees. The unit will operate to negative 15 in heating mode." Note that Nolte does not provide performance specs at -15°F. Nolte also wrote, "The turndown ratio for the Greenspeed Intelligence three-ton heat pump in heating is 26 percent and cooling of 40 percent." Concerning products that are still under development, Nolte wrote, "The next generation of variable-speed heat pumps will operate to negative 5 degrees Fahrenheit and will include published performance data at negative 15 degrees Fahrenheit."]


    NEEP lists just one Lennox air-source heat pump: the Lennox Elite XP20. It has a heating capacity at at 5°F of 19,000 BTU/h — enough for a single-family house with a low rate of air leakage and a good thermal envelope.

    The minimum heating capacity of this unit at 47°F is 10,400 BTU/h, so its turndown ratio is an unimpressive 1.8 to 1.

    Other examples of Lennox equipment include the Lennox XP25.

    I called up Lennox to try to get more technical information about the cold-weather performance of the company’s air-source heat pumps. After several phone calls and e-mails, I was eventually connected with Brandon Chase, a senior product marketing manager at Lennox.

    I asked Chase whether any Lennox air-source heat pumps deliver significant amounts of heat when outdoor temperatures drop to -15°F or -20°F. “They do,” Chase answered. “That capability can be found in some of our higher end equipment.”

    I asked Chase if Lennox manufactured any air-source heat pumps with a capacity of 40,000 BTU/h at -10°F or -20°F. “I don’t know,” Chase answered. “I’d have to check with one of our engineers.”

    I asked whether Lennox publishes information on heating capacity at outdoor temperatures of -15°F. He answered, “I don’t believe we publish that information.”

    I asked how cold-climate builders could be expected to design a heating system without that information. He answered, “What I need to do is see what the rationale is.”

    I asked whether Lennox will be developing heat pumps that operate at the same outdoor temperatures as equipment from Mitsubishi or Fujitsu. He answered, “We have the capability to manufacturer equipment that performs at very low temperatures. We wouldn’t have to do a lot of tweaking to get there. It’s a question of getting comfortable with the technology. We prefer to do more research.”

    I asked Chase why the turndown ratio for Lennox equipment is so limited compared to cold-climate heat pumps from Mitsubishi and Fujitsu. He said, “I don’t have an answer to that question.”

    I followed up our phone interview with e-mailed questions on cold-climate specifications and turndown ratios. I received a written response from Matthew Pardee, an employee at MM2 Public Relations, the company that performs PR work for Lennox. Concerning cold-climate capacities of Lennox equipment, Pardee wrote, “I just reached out to Brandon in regards to your questions about the XP25 and XP20,” Pardee wrote. “He said that is information that Lennox does not currently publish or are willing to make public today.”

    Pardee also responded on the question of turndown ratios. He wrote, “That information now resides in a database and after reviewing it Lennox currently does not have the information in it to calculate the turndown ratio.”


    Since U.S. manufacturers of ducted air-source heat pumps don’t release information about performance at -10°F or -15°F, it probably makes sense to see if any Japanese manufacturers — that is, manufacturers with a documented interest in producing equipment that performs well when outdoor temperatures go below zero degrees F — have developed a large air handler that would work with an American-style duct system. In fact, at least one Japanese manufacturer, Mitsubishi, has done just that.

    If you want to hook up a Japanese air-source heat pump to standard U.S.-style forced-air ductwork, there are two options from Mitsubishi worth considering: the Mitsubishi MVZ air handler for multi-zone systems, or the P-series air handler (either the PVA or PVZ air handler) for single-zone systems.

    According to Mitsubishi, the MVZ air handler is designed for systems with design heating loads ranging from 13,500 Btu/h to 40,000 Btu/h. (The largest MVZ air handler is the MVZ-A36AA4, part of a 3-ton system.)

    One peculiarity of the MVZ air handler: it won’t work with a single-zone system. Since this air handler requires at least two zones, you’ll have to install at least one wall-mounted indoor unit or ceiling cassette — perhaps in a basement or a bonus room — in addition to the ductwork connected to the air handler.

    For a single zone system, one option is the PVZ air handler — for example, the Mitsubish PVZ-HA36NHA. It is a single-zone air handler rated for 40,000 BTU/h of heating. When this PVZ air handler is paired with a HyperHeat outdoor unit, it has a heating capacity of 100% of the rated capacity at 5°F, 90% of the rated capacity at -4°F, and 80% of the rated capacity at -13°F.

    It's also possible to specify a similar air handler, the Mitsubishi PVA air handler — for example, model PVA-A42AA4. It is rated for external static pressure ranging from 0.20 i.w.g. (low speed) to 0.80 i.w.g. (high speed), and for airflow rates ranging from 1,040 cfm (low speed) to 1,485 cfm (high speed). Paired with the PUZ-HA42NKA, it has a maximum heating capacity at 5°F of 48,000 BTU/h, and is rated for operation down to -13°F.

    Minnesota study

    Four researchers associated with the Center for Energy and the Environment in Minneapolis, Minnesota — Ben Schoenbauer, Nicole Kessler, Alex Haynor, and David Bohac — recently conducted a study to evaluate the performance of ducted air-source heat pumps in Minnesota. Their 2017 report is titled “Cold Climate Air Source Heat Pumps.”

    The Minnesota locations where these heat pumps were installed all have outdoor heating design temperatures that are considerably colder than the 5°F cut-off mentioned earlier in this article. For example, Minneapolis has a 97.5% outdoor heating design temperature of -12°F.

    The researchers described their study this way: “A total of six ccASHP [cold-climate air-source heat pumps] were installed. Four of the systems were centrally ducted whole-house units and two were ductless systems. All of the systems used variable-speed compressors, often described as inverterDevice for converting direct-current (DC) electricity into the alternating-current (AC) form required for most home uses; necessary if home-generated electricity is to be fed into the electric grid through net-metering arrangements. driven technology, allowing the system to change operation speeds and modulation rates depending on temperatures and heating loads. This allows for increased capacities and colder temperature operation.”

    These were the ducted units used for the Minnesota study:

    • Bryant Extreme, 4 tons, model 280ANV48
    • Carrier Infinity with Greenspeed, 3 tons, model 25NA036A003
    • Carrier Infinity with Greenspeed, 4 tons, model 25VNA048A003
    • Trane XV20i, 3 tons, model 4TWV0036A

    In addition, the researchers looked at two ductless heat pumps manufactured in Japan:

    • Mitsubishi ductless Hyperheat, 1.5 ton, model MUZ-FH18NAH
    • Mitsubishi ductless Hyperheat, 1 ton, model MSZ-FH12NA

    As is typical in cold-climate locations, the heating load determined unit sizing. The researchers wrote, “Cold-climate air source heat pumps were sized for each home’s heating load (as opposed to the cooling load), which typically led to an increase in capacity (or ‘tonnage’) of the system by one ton. This meant that where a home sized for cooling would install a two-ton heat pump, the same home sized for ccASHP heating would install a three-ton system.”

    Because the air-source heat pumps that the researchers studied had low (or non-existent) heating capacity at the below-zero outdoor temperatures sometimes experienced in Minnesota, all of the homes enrolled in the study included a propane furnace to provide backup heat. The systems were set up to switch to propane heat when the outdoor temperature dropped to 10°F (or in some cases, 5°F).

    Here’s how the researchers described the setup: “Ducted systems typically use a propane furnace as a backup. These backups take over the load of the system at an outdoor air temperature when heat pump capacity is no longer sufficient. This temperature, about 10°F, is based on the house load and system size.”

    Having two separate heating appliances can be complicated. The researchers wrote, “The original intent of this project was to integrate ccASHPs [cold-climate air-source heat pumps] with the existing heat source as backup. However, there are issues that make integrating a ducted ccASHP with the existing furnace complicated. The two primary issues are 1) the furnace and heat pump require communicating capabilities and 2) a multi-stage fan is necessary to achieve the full benefit of the ccASHP. To deal with these issues, manufacturers and installers specify that the furnace and ccASHP must be the same brand. This ensures that the controls for the ccASHP and the furnace can communicate …. Unfortunately, most 80% AFUEAnnual Fuel Utilization Efficiency. Widely-used measure of the fuel efficiency of a heating system that accounts for start-up, cool-down, and other operating losses that occur during real-life operation. AFUE is always lower than combustion efficiency. Furnaces sold in the United States must have a minimum AFUE of 78%. High ratings indicate more efficient equipment. and older condensing furnaces have single-stage fans. While it is expected that a wider range of options will become available, at the present time only recently installed and higher end furnaces would have the controls and fan characteristics desired for integration. … In the Minneapolis/St, Paul metro area, a homeowner would pay about $4,250 for a condensing furnace and only $1,875 for the same size non-condensing furnace. With a properly sized ccASHP, it is expected that the furnace would have to meet less than 30% of the heating load, and this percentage can be reduced further for homes with lower heating loads. Given that the furnace would only be running for a small portion of the heating season, it is likely to be more cost effective to install an 80% AFUE furnace.”

    While the cost of an air-source heat pump is usually considered prohibitive for homeowners with access to cheap natural gas, a heat pump may make sense for homeowners who depend on expensive fuels like propane or oil. The researchers found that compared to homeowners who burn 100% propane, the air-source heat pumps “in all cases … saved homeowners and renters significant amounts of energy and money.” They concluded, “The greatest potential for ASHP [air-source heat pump] adoption is in cold-climate regions where natural gas is not available for space heating because ASHPs can offset the use of more expensive delivered fuels.”

    The researchers noted, “The project has concluded that the measured performance of ccASHP installed in real homes confirms the potential to provide significant energy savings (39% to 65% of space heating energy use) and cost savings (14% to 29% of space heating costs).”

    I called up one of the Minnesota researchers, Ben Schoenbauer, and asked where their team obtained performance data on the cold-weather capacity of these heat pumps. “The manufacturers don’t provide a ton of information,” Schoenbauer responded. “Nobody is required to report performance below 5 degrees.” In short, the researchers consulted the NEEP list.

    I asked whether the low turndown ratios of the American-made equipment was problematic. He answered, “We didn’t see any problems with short cycling.”

    I asked Schoenbauer whether he thought that there would be an increase in sales of ducted air-source heat pumps in very cold climates. “If you need an air conditioner, this approach makes sense,” he responded. “But for heating only, the cost is a lot to swallow.”

    Equipment cost

    Is it possible to generalize about the cost of ducted air-source heat pumps — especially compared to ductless minisplits from Japan?

    The Minnesota researchers reported that the average cost the install an air-source heat pump and a propane furnace was $14,000. The same source reports that the residential equipment database from the National Renewable Energy Laboratory (NREL) reports that the installed cost of a new ducted air-source heat pump (presumably, not including ductwork) is $6,340.

    Unresponsive equipment manufacturers

    As a final note, I feel duty-bound to express my journalistic frustration at how difficult it was to get basic performance information on air-source heat pumps from equipment manufacturers. After three weeks of attempting to contact equipment manufacturers via email and telephone, I often found myself stymied. Most of my simple questions met a brick wall.

    I found no evidence to suggest that HVAC contractors or architects would have been treated with any more respect when seeking answers from manufacturers than I was.

    There seem to be several issues here:

    • Having worked as a journalist for over 20 years, I am familiar with a basic rule: the larger the company, the harder it is to obtain technical information. A family-run business with 20 employees almost always hires an intelligent person to answer the phone, whereas large corporations handle phone calls with voicemail hell.
    • In some cases, equipment manufacturers have two websites: the obvious website (for example, www.mitsubishicomfort.com) and a “secret” website with the real information (for example, http://meus1.mylinkdrive.com). This deliberate obfuscation can be maddening to anyone looking for answers to technical questions. (That said, it’s worth commending Mitsubishi for doing a much better job of supplying technical information than American Standard, Bryant, Carrier, Lennox, or Trane.)
    • Heat-pump manufacturers infantilize homeowners. Because of this infantilization, these companies’ marketing departments have decided that product brochures should be free of basic performance information like heating capacity at 5°F. Instead, the manufacturers provide links to glossy brochures that brag about “comfort” but are thin on technical details.

    For now, here’s my advice: While it makes sense to try to contact equipment manufacturers with technical questions, don’t be surprised if the response is unhelpful. Until the attitude of the equipment manufacturers change, designers should turn to the NEEP list as a guide.

    Martin Holladay’s previous blog: “Fasteners for Concrete and Brick.”

    Click here to follow Martin Holladay on Twitter.

Tags: , , ,

Image Credits:

  1. Carrier

Jul 6, 2018 11:15 AM ET

Ecologix Heat Pump
by Yupster

You may want to include the Ecologix Cold Climate Heat Pump in this list. They have excellent support and respond to questions quickly. Any data not available on their website (which is very poor) is only an email away. Here is a link to their spec sheet. http://ecologix.ca/products/product-selection-tool/SpecSheet/?productcat...

Jul 6, 2018 1:37 PM ET

Other info on Carrier Greenspeed
by Dana Dorsett

Even though they don't provide the information in written form, they have a mathematical model for the Carrier Greenspeed, and an online tool that can allow you to estimate the capacity (but not the efficiency) at temperatures as low as -30F, at least for the 2 ton and 3 ton models:


Click on the Heating Capacities tab, then adjust the Heating Design Temp down to the temperature of interest.

For both the 2-ton 25VNA024A**30 and 3 ton 25VNA036A**30 the capacity at -15F is about half it's capacity at +5F, and the capacity at -10F is about half the AHRI rated capacity at +17F. The drop in capacity is fairly linear with temperature from +5F on down.

Regarding propane as "backup", that is more expensive to install than auxiliary resistance heaters (available for all the US manufacturers' goods, as well as Mitsubishi's MVZ and PVA air handlers), and may be more expensive to operate too, if the heat pump's capacity can cover the load down at least down to the 90-95th percentile temperature bin for most areas. (In areas with cheap electricity and expensive propane it's always cheaper to use the electric heat strips.) At +10F most heat pumps are running a COP of at least 2, and even at -10F most heat pumps are running a COP of about 1.5, and burning some heat strip to make up for just the capacity shortfall for just 10% of the load isn't going to destroy the average seasonal efficiency. When propane is being used for the "backup" when the heat pump is out of capacity 100% the heat is from the propane, none from the heat pump, so there is no COP leveraging the energy inputs.

Even in fairly expensive electricity areas it rarely pencils-out in favor of PROPANE dual-fuel systems (but it can for natural gas, in some areas.) Given the volatility of propane pricing, and the fact that electricity rates are regulated, it's a safer bet to skip the dual-fuel. Install enough heat strip to cover the shortfall in capacity down to maybe 15F below the 99% outside design temp, not more. (There are usually between 2-5 different sizes of auxiliary heat strip available for most models.) By not oversizing the heat strip you won't run into excessive cycling of the heat pump at low ambient temperatures, which could take an additional toll on efficiency almost as big as running the heat strip.

Jul 8, 2018 7:26 AM ET

Other Asian mini-split ducted manufacturers???
by Antonio Oliver

Are Mitsubishi and Fujitsu really the only players in the game for mini-split ducted systems that can be connected to typical ductwork in US homes? I seem to recall Dana Dorsett pointing me to a system by Panasonic on this website. Is my memory faulty or should I go back and check the static pressure requirements for that Panasonic air handler?

Jul 8, 2018 11:49 AM ET

New home construction
by James

Is there anybody that would recommend a whole house mini split system for a new construction home?

Jul 9, 2018 2:07 PM ET

Other ducted Asian heat pumps @ Antonio Oliver
by Dana Dorsett

I don't believe Fujitsu makes full-sized air handlers that easily use pre-existing ducts designed US style AC & gas furnaces, but they make the most powerful mini-duct cassettes on the market. Almost all mini-split vendors sell mini-duct cassettes similar to the Fujitsu xxRLF "Slim Duct" units, but most of the competition's mini-duct offerings have fairly wimpy blowers compared to Fujitsu's.

LG, Daikin (who now own's Goodman), Samsung and several others have heat pumps with large-duct type air handlers, some VRF, others single-speed. I don't recall Panasonic having anything like that.




Gree makes similar products sold into other markets, but I'm not sure if they've been marketed in North America yet (at least not under the Gree nameplate.)

James: "Is there anybody that would recommend a whole house mini split system for a new construction home?"


Jul 10, 2018 7:38 AM ET

Response to Yupster (Comment #1)
by Martin Holladay

Thanks for the link to the page with information on the Ecologix heat pump.

I look forward to hearing reports from GBA readers who have installed or lived with this equipment.

Jul 10, 2018 7:55 AM ET

Edited Jul 10, 2018 8:00 AM ET.

Response to Dana Dorsett (Comment #2)
by Martin Holladay

Thanks for the further information on the Carrier Greenspeed heat pump, and for the link to Carrier's GreenSpeed Calculator. I have edited my article to include that helpful information.

Like you, I don't recommend installing a propane furnace as a backup heating source for a home with an air-source heat pump. That's why I provided a list of cities with outdoor design temperatures of 4°F or above. If you're going to install one of these American-made air-source heat pumps, you should live in a climate that is warm enough that you won't need a second heating system as backup. If you live somewhere colder, you should probably install Mitsubishi or Fujitsu equipment -- unless the GreenSpeed Calculator convinces you that Carrier equipment can cover 100% of your load at design conditions.

The propane-furnace-as-backup approach was used by a group of Minnesota researchers -- but it's not an approach that I recommend.

Jul 10, 2018 8:49 AM ET

Edited Jul 10, 2018 8:49 AM ET.

Response to Dana Dorsett comment #5
by Antonio Oliver

My mistake, Dana. It was LG, not Panasonic. That said, Martin is still only recommending Mitsu and Fujitsu as reiterated in comment 7.

Jul 10, 2018 9:05 AM ET

Response to Antonio Oliver
by Martin Holladay

I mention Mitsubishi and Fujitsu because I've heard many reports of successful installations with these two brands. I know much less about LG equipment.

That said, if LG provides performance specs at -13°F or -15°F, or whatever your design heating temperature happens to be, and if you can find equipment that meets your design load, I see no reason not to use LG equipment if that's what you want to install.

Jul 10, 2018 4:44 PM ET

Thanks again Martin
by Antonio Oliver

Thanks, Martin.

I found a variety of data here for an LG outdoor unit for the air handler pointed to by Dana:

Operating temperature down to -4F, apparently. Good enough for where I live, but probably not for people requiring operation down to -13F. Although I recall reading that for a particular Mitsu unit, people had been satisfied with performance below the lowest recommended operating temp.

See page 10 of this document:

Engineering Manuals Outdoor Units Multi V™ S

Jul 11, 2018 6:40 AM ET

Response to Antonio Oliver
by Martin Holladay

Thanks for the additional information -- and good luck with your project.

Jul 11, 2018 9:05 AM ET

The 5 tonner is good down to -13F (WB) -12.6 (DB)@ Antonio
by Dana Dorsett

See the capacity tables starting on p.97 of the manual- the 5 ton LG Multi-V is good for 43-44K @ -13F outdoors, 70F indoors.

Capacity tables for their smaller units all stop at -4F.


Jul 15, 2018 3:31 PM ET

Mitsubishi System
by Kevin Camfield

Thanks to a lot of help from Dana and others on this forum, we decided to go with a Mitsubishi zoned system using a MVZA18AA7 multi position unit downstairs and SEZKD12NA4R1 horizontal ducted unit upstairs for our new house. I found the "mylinkdrive" website very helpful when making that selection.

The price for the Mitsubishi system installed with ducting is $17,050 using the Hyper Heat MXZ-3X24NAHZ2-U1 external unit or $15,600 using the Non Hyper Heat MXZ-3C30NA2-U. We live in the Seattle area and so will likely go with the non-hyper heat external unit. The price for the Carrier Green Speed 3-ton, 25VNA036 Infinity Series was $18,600. Our design peak heat load is 22,000 btu/hr in our mostly code built, 2,200 sq. ft. house. We will have back-up propane heat only because we wanted to have a cozy free standing propane stove in the kitchen.

The Mitsubishi unit had a number of advantages important to us including better sound performance inside and out, high efficiency DC motors, and 3 speed fan speed on the internal units. We were able to design around the low static pressure performance of the horizontal ducted unit. We plan to run the units continuously and connect our HRV supply piping to the intake ducting.

Jul 18, 2018 2:02 PM ET

NVH, etc. From Ducted System
by Joe Dwyer

We've been using a Carrier 2.5 ton, 8.5 HSPF, 13 SEER, single stage heat pump since Dec. 2001. Installation cost was $4,575, which included a new variable speed air handler, 5 kW + 10 kW staged strip heat package, programmable thermostat, and large pleated air filter that is changed yearly. It still works great with minimal maintenance. A defrost circuit board failed at year 10 ($700-dealer) and thermostat's humidistat lost accuracy in last year and was replaced with new Carrier COR TS ($75-me).

With a combination of greater home insulation and leak sealing over the years, plus learning the ins and outs of the unit (hidden settings), I can happily report that the HP provides over 98+% of necessary heating and the backup strips only come on (as auxiliary- for short time) when the outside temperature is between 0F to -5F. Some years this doesn't happen. Our design temp. is 9F and home is 2200 sf split foyer.

As the unit is now on year 17 and uses freon based refrigerant (never leaked), I'm keeping abreast of better and more efficient units for future replacement. One that has caught my eye is Daikin SkyAir Inverter Ducted. 1.5 ton, 20 SEER, 12 HSPF, and capable down to -4F, with strip kits also available. Downsizing is the goal for efficiency sake, plus modulation with the inverter and better turn down ratio most of the time. Let the strips handle the 2-5% on the left side of the curve. Also, it has very low NVH with swing rotary compressor and horizontal mount. Or maybe go with 2 ton unit, no backup strips, and less turn down ratio? Need to do some math on this.

NVH- our unit is moderately loud and turns on and off all the time. I congratulate Copeland Scroll Compressor engineers, as this part is old-school durable. I took care of the 'rattle trap' nature of the outdoor Carrier unit with self installed sheet rubber bushings between metal panels and proper leveling, as it had settled a bit over the years. I desire an outdoor unit that is quieter- 50 dB range. Unless Carrier has improved them, the GreenSpeed units sound loud (higher pitch) in severe cold at high operating capacity.

Finally, our single speed unit is excellent at humidity control, especially with the new, more precise COR thermostat. It lowers the air handler fan speed , combined with 3 degrees "cool to dehimidify" range, to keep RH around 50% all summer long. This does lower the daytime set point from 78F to 75F or 76F, especially in the morning or after a period of heavy rainfall. It's back up to 78F and 50% RH in the late afternoon and/or during dry periods. I'm willing to use a little more energy to keep the indoor air dry, especially downstairs which is 50% below grade.

So my last question is- Do ducted mini-splits, especially smaller more efficient units, have less ability to dehumidify the air as the older units?

Jul 18, 2018 3:57 PM ET

It depends on the cooling mode @ Joe Dwyer
by Dana Dorsett

>Do ducted mini-splits, especially smaller more efficient units, have less ability to dehumidify the air as the older units?

Ducted mini-splits have a "dehumidify" or "dry" cooling mode that can be selected, which will do most of much what your current thermostat is doing when set up for "cool to dehumidify", but without over-cooling. A mini-spit in "dry " mode automatically adjusts the latent-to-sensible cooling ratio by running the coils a bit cooler via a combination of refrigerant volume & blower speed adjustment, but some still operate to the temperature setpoint, others not. Operated in "dry" mode the SEER is lower and it's sensible cooling capacity is lower, but that's the nature of physics, eh? (Cooling to 3F colder than you normally would just for better latent cooling has an even bigger penalty.) When the latent loads are low you can always run it in the normal cooling mode. See the ultra-brief discussion on p.8
& 9 of the manual regarding DRY mode for the SkyAir units. If I'm parsing the English correctly it still operates to a temperature setpoint, but you can't adjust the setpoint while it is in DRY mode:


Like most large air handler mini-splits the 1.5 ton Daikin SkyAir has a limited turn down ratio of barely over 2:1: 9,000BTU/hr min, 20,000 BTU/hr max in heating mode). The 2-tonner has the same 9000 BTU/hr minimum, but a 3:1 ratio (27,000 BTU/hr max), which is better than most! Depending on your actual outdoor lows it may have sufficient capacity to dispense with the heat strip kits altogether. Since it's HSPF was tested at 27,000 BTU/hr (instead of 20,000 BTU/hr for the 1.5 ton unit) and has the same 9K min-modulation as the 1.5 ton version, the as-used efficiency will probably be higher than the 1.5 tonner, since it'll be running closer to it's mid-range sweet spot most of the time. That is despite having slightly lower SEER and HSPF numbers at the "rated" modulation level.


The capacity tables only go down to +14F, but the 2-tonner delivers a bit over 24K @ +14F outdoors, 70F in, the 1.5 tonner delivers 20K. See p88:


You can safely probably infer it's capacity at your design temp of +9F and +70F indoors by the capacity numbers at 75F indoors, +14F outdoors (5F warmer than both the indoor & outdoor design temps, for the same delta-T), which is 18K for the 1.5 ton, 24K for the 2 ton.

Do you happen to have the capacity tables for your aging 1-speed 2.5 ton Carrier handy?

Jul 19, 2018 1:07 PM ET

Edited Jul 19, 2018 1:50 PM ET.

@ Dana Dorsett
by Joe Dwyer

Thanks for your thorough reply. I previously looked at the engineering manual, but not the brochure showing the turn down ratios. The 2 ton model seems the way to go, having the same low end capacities for heating and cooling and a higher top end heating capacity for really cold weather events (-5F common to -10F max). It will probably also work without heat strips, so we can get back a little money up front to offset the higher cost of 2 ton versus 1.5 ton system.

I'm not familiar with how SEER and HSPF are calculated for variable speed units, as it's easier to understand with a single speed like ours. So you are saying that a larger inverter compressor can be more efficient in real world conditions due to running more often at moderate speed instead of high speeds, which isn't accounted for in the ratings for such units?

Our unit is Carrier 38YRA030320 outdoor and 40FKA002300 indoor. The company that installed it claimed 30K BTU and 13 SEER, with no HSPF rating. When I said 8.5 HSPF, this is really just a best guess for similar era units and it could be lower. I couldn't find a combo rating, but the 38YRA unit with slightly newer air handlers (circa 2004) does come up in a search:


The balance point table goes down to -3F, but it uses a 3 tons air handler with 2.5 tons HP. Our AH is only 2.5 tons. Anyway, our current balance point is around 0F to possibly -3F, based on observation of the unit a few years ago operating on a very cold morning with old thermostat and "aux" symbol+ higher fan speed+ hot air from register. With the new thermostat, such data is logged to a web portal!

I've set our AH to operate in 'comfort' mode, which means 700 CFM versus 875 CFM for cooling and 785 CFM versus 875 CFM for heating. Thus, both SEER and HSPF need to be down rated slightly versus factory rating. Finally, I did a search for the rating of aging units and came across "Building America Performance Analysis Procedures for Existing Homes," which has a formula:

EFF = (Base EFF) * (1-M)^age

EFF= 100 * ((1-.01)^16.5)

EFF= 84.7%

I need to pin down our heat loss at balance point more precisely, but I think it's around 8K-11K BTUs/hr at -3F. I also need to extrapolate 2 tons SkyAir HP capacity at very low temps. Does this sound reasonable?

Sorry, wrong Carrier 38YRA link. Here it is:


Jul 19, 2018 4:57 PM ET

Yes, lower speed means higher efficiency (mostly) @ Joe Dwyer
by Dana Dorsett

Manufacturers of modulating equipment get to choose the modulated level at which efficiency is tested at 47F for the AHRI "rated" or "nominal" heating output as long as it has enough capacity at +17F to hit that range. Manufacturers play marketing games a bit, and can even inhibit the unit from exceeding a given output at any temperature, which is what Daikin seems to have done with the SkyAir products:

If you look at the capacity charts in the engineering manual, the nominal 20,000 BTU/hr heating capacity AHRI conditions for 1.5 ton SkyAir is also it's total capacity at 70F indoors at all temperatures, from +14F to +50F a curve flatter than a skating rink!

The chart for the 2 ton unit indicates that at 70F indoors the total capacity is almost the nominal 27,000 BTU/hr at +23F, but only 24,000 BTU/hr @ +14F, but limited to 27,000 BTU/hr from 32F through 50F. That's not quite as flat a curve, but 20% more capacity at +14F than the 1.5 ton unit, and probably 20% more (unspecified outputs) at 0F or -5F. They could have kept it a flat 24K from +14F to 50F, but probably needed to call it 27,000 BTU/hr for marketing purposes, to distinguish it more from the 1.5 tonner.

But since they both can modulate down to 9000 BTU/hr @ +47F the 2 ton unit is probably going to be operating in the high efficiency sweet spot over a wider range of temperatures than the 1.5 ton unit. I doubt that it's the same unit, just being controlled differently- the coils.)

Looking at the capacity tables for your existing Carrier you have ove 18K max capacity @ 17F (17K net defrost), just shy of 13K @ -3F (12K after defrost), and if it's running the heat strips at around 0F it means your load is probably something like 14-15K @ 0F. It's not clear if the 1.5 Daikin would be able to deliver that much at 0F (since it's unspecified) but maybe. It's almost certain (though still unspecified) that the 2 ton Daikin would, (with margin!), if it's delivering 24K @ +14F.

A 2 ton GreenSpeed also puts out about 24K @ +14F, which drops to about 17K @ 0F, and 16K @ -3F. Click on the Heating Capacities tab on this page, then play around with the outdoor design temperature setting for eyeballing the capacity:


If it's about the same proportionally the 1.5 ton Daikin's 20K @ +14F would become 13K @ -3F, 14K @ 0F, which could be marginal. But the 2 ton would be delivering about 17K at -0F, 16K @ -3F, which should cover it without engaging heat strips.

So, if comparison shopping, compare the up front cost of the 1.5 ton + heat strips to the cost of the 2 ton with no heat strip. I suspect the difference is pretty small.

Age derating curves have wide error bars, so I'm not sure you should be using the derated capacity value for your heat load analysis, then specify the new equipment based on that derated number. The new equipment will age too, perhaps at a different rate, or maybe the same. But if you use the fully rated value of the old equipment in the analysis you'll do just fine specifying the new equipment without needing to upsize from there.

Register for a free account and join the conversation

Get a free account and join the conversation!
Become a GBA PRO!