Heat Pumps not performing at low temperatures

[nuuvox000]

I have noticed that some Mitsubishi heat pumps claim to provide a certain amount of heat at -14 degrees F ambient outdoor temperature. However, I have had people tell me that in practice they do not perform and they have had to add auxiliary heat after the project was completed because the heat pumps couldn't keep up. Does anyone have experience with this if they actually perform or not? We have just been providing auxiliary heat but I'd like to stop designing auxiliary heat if we don't have to but I don't want to take a big chance by not providing it if they don't perform. Just looking to get some more experiences from you all. Thanks!

 

[MintJulep]

That's pretty cold. Are you sure that the claims aren't for -14 Celcius?

"a certain amount of heat" may be far less than the need at -14 (regardless of the temperature scale).

 

[nuuvox000]

Mitsubishi calls them "Hyper Heat" units. It's actually -13 degree F and they claim 20,000 Btu/h of heat provided on a nominal 2 ton unit. The COP drops to 1.5 at this temperature though. Anyway, this is the claim.

 

[EnergyProfessional]

I'm sure Mitsubishi tested them before making this claim. Most the time if they "don't work" is when some features are missing 9like low temp baffles etc.), of the ambient temperature is lower, or the designer didn't account for de-rating. different climates also may have different dew points at -13F.

IMHO, if you have -13F design temperature, an air-source heat pump is not appropriate. If the design temperature is -13F, it can dip below that. Where I live ASHRAE has a design temperature of -11F, code requires -15F, but it can go below -20f on some nights. Those design temperatures are more for combustion devices that still operate at low temps, even if the building is a bit colder. This is acceptable for the few hours a year. an air-source heatpump just stops working, which is not acceptable. I wouldn't listen to advice of designers who designed for HP heating only in such cold climate and had to come back afterwards to add heating. this was a totally foreseeable problem.

The inherent problem with air-source heatpumps is at 50F, the may have 100% output. but then you barely need any heat. At -13F they may have 20%. So you need to size the HP 5 times and at 50F you have a really large HP for the job. Assuming the HP also is used for cooling, the size needs to be balanced with cooling load. If the HP is too large for cooling, you don't de-humidfy well. So it will never be perfect. Most HP have some good capacity range / turndown, but still.

We once had hired consultants to design a HP system. They were located in a room drawing air through. the idea was to just supplement heat with gas-fire UH. the HP were a block of solid ice! We ended up having to close the room in winter and only heat with the UH and the HP extract heat from that room. So we ended up heating with 80% efficient gas fired units, and add electricity for the compressors.... a boiler would have been much more efficient.

 

[willard3]

I can't count on my hands the number of times I have had to supply some conventional heat in buildings with air-to-air heat pumps.
Apart from the efficiency, which is drastically reduced at low temperatures, air to air HP are meant for climates where delta T is quite low. If you have -13F, the HP is a poor system choice.

 

[nuuvox000]

EnergyProfessional - they are claiming 80% heating capacity at -13 which is not bad if true. The design temperature where I live is 3 degrees but it dips just a few degrees lower than this every now and then. I live in a dry climate too so humidity is not an issue (we're always adding water here in Utah).

willard3 - that's good to know. I would like to get with a local supplier and test their heat pumps at low temperatures. The problem is I would have to wait for a 50 year cold event to test it out... Maybe if I found a large freezer warehouse but now it's getting complicated.

 

[EnergyProfessional]

At 3F, it probably is fine. But I recommend studying the design documents very well and fully understand what is required and why. Also talk to Mitsubishi rep and see if they have local reference installations. I'm not sure how they de-frost and if that is accounted in the 80%. but there are so many minutes every hour defrost where the unit reverses and extracts heat from the building, or works at half capacity. You need to fully understand what that means and how to interpret the 80%.

And always use the actual lowest ever observed ambient temperature. i assume there are local weather data to see if that is -5, 0, or whatever it is. that would be my temperature when judging the unit. The homeowner doesn't call you the 99.9% the unit works. They call you the 0.1% it doesn't work.

I don't know the hyper-heat units well, but think to recall they have a double-circuit in series to get the low ambient numbers. that likely is at the expense of efficiency the rest of the year and adds probability of failure. It may be better to get a regular HP and use other heat for the few days of low temps. an energy simulation would evaluate the options.

 

[nuuvox000]

Good point on the defrost, I'll look into that more. It looks like it got down to -18 in 1973; it was a long time ago but you're right it would be terrible if it happened again. We're looking at 500 apartment units right now and the client really wants heat pumps but we may have to do backup electric resistance coils after all. I mean, if it got that low again people could at least turn on their electric stoves to get warmth in the space but there's no way I would say that as a professional. I'm also thinking about putting one in at my house but I have a wood-burning stove on each floor that I could use in emergencies. Maybe I'll check for a reference project, thanks for the idea.

 

[nuuvox000]

Just FYI, it looks like they have VRF condensers that go down to -31 F with some special add-ons, sounds like a better route to go.