Can I put a ductless mini split condensor/compressor in an unconditioned cellar?
okapi222
| Posted in Mechanicals on
We are renovating a 125 year old masonry building in NYC. We are heating and cooling it with ductless mini splits. They run more efficiently in a warmer environment (in winter) and cooler environment (in summer). So why not install them in the cellar which should stay about 50 degrees all year. Any thoughts on this? Possible hazards? Thanks Daniel
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
Let us know how this idea works out for you.
It won't be a constant 50 degrees once the heat pump is running. Even my heat pump water heater cools my basement significantly. If you had a mini split compressor running all day and night, you might wind up with a frozen crawlspace.
Daniel,
Nick is right.
There are two problems with your suggestion:
1. An air-source heat pump needs access to huge volumes of exterior air to operate. (That's where the heat comes from.) Since your basement isn't directly connected with the exterior air, all the heat pump can do is to continue to lower the basement temperature. Of course, as the basement temperature drops, so does the efficiency of the heat pump.
2. During the winter, it's no fun to live directly above a walk-in freezer.
I beg to differ, I think it depends on what is in the walk in freezer. If it were full of ice cream, or sirloin, I think it would be a fine place to live...............
There is about R28 and an air barrier between the cellar ceiling and the ground floor conditioned space so I don't think it's necessary so bad to live above a walk in freezer if there is a thermal barrier there. The cellar is also not 100% sealed off from the outside. There are covered 2 coal shoots and ventilation could be installed in them. But I question that the cellar temperature would actually drop below the ambient air temperature in mid-winter. I would expect the thermal mass of the earth around the stone walled cellar to moderate the temperature.
Daniel,
The volumes of air that need to be moved across the condenser coil are very large. One source reports, "The outdoor condenser fan draws a large quantity of air (2,000 to 4,000 cfm)."
Even if we take low end of the range (2,000 cfm), all of the air in a typical basement (30'x40'x8') will need to be cycled through the condenser fan every 2.4 minutes. In an hour, the air in the basement would need to cycle through the unit 25 times. Each time the air is cycled through the unit during the winter, it gets colder than it used to be.
Thanks, Martin. I am convinced. Just trying to brainstorm cheap ways to hack together a ground source heat pump. But i guess that idea will have to wait. We will put them outside.
One additional question, are there efficiencies to be had by locating the condenser/compressor outside at ground level next to the dwelling as opposed to on the 3rd story roof? My thinking is that it in winter it would be a couple of degrees cooler on the roof (because of wind) and in summer significantly warmer then in space at ground level, protected by the building.
Daniel,
The volume of outdoor air that moves through the outdoor unit is so large that it doesn't matter where you put it. The outdoor air is at the same temperature everywhere -- expect for a small, thin film of air near surfaces that have been warmed by the sun. The volume of air affected by the sun is so small that it won't affect heat-pump performance.
This question seems to come up a lot. The air-to-surface heat exchange between basement air and the deep subsoil to manage the peak heating rate of a 3/4 ton mini-split requires the temp of the the basement air to be well below the operating temp of a mini-split.
Even if the ground were make of highly conductive copper (basically R0), the deeps subsoil temps in NYC is about 52F( http://www.earthrivergeo.com/img/geothermal-article/geothermal-subterrainean-temperature-contour-map.jpg). With say 1000' of basement to get 10,000 BTU/R out of the copper floor you'd have to chill the basement to about 47F- not too bad.
But the real dirt is far less conductive, on the order of at least R10 by the time the hemisphere of subsoil near the house is wide enough to support 10KBTU/hr of heat transfer continuously, and the basement would have to be well below zero F, a temp at which you'd be luck to get a COP of 1.8 out of the mini-split (if it could run at all.) The only way you get those heat transfer rates out of ground source heat pumps is due to the fact that it's in direct contact with the soil, and doesn't have the insulating properties of air to contend with in the heat-transfer from dirt to air. But even there it's common to have loop temp requirements below freezing to make it work, and anti-freeze in the ground loop.
The good news is, at NYC outdoor temps a better class mini-split (HSPF greater than 10) operates near the seasonal efficiency of a typical ground source heat pump installation, and will BEAT it during the shoulder seasons hands-down. When the manufactures spec the ground source heat pump as having a coefficient of performance of say 4.5 or 5, that does not include all the ground loop pumping power or the heat distribution power, only the pumping at the water-water or water-air heat exchange. The HSPF of a mini-split may be measured only at 47F, but it's an all-in number counting 100% of the power use, and at or near it's max compressor & blower speed where it's operating at a lower efficiency than at part load. An HSPF of 12 implies a COP of 3.5, which is where the all-in numbers for ground source heat pumps come in when you have 52F subsoil. (They'll do better than that only if sized perfectly using best practices, and low-temp water-to-water hydronic output such as low-temp slab radiant floors.)
Sized correctly a mini-split runs at part load nearly all the time, and full-out only at or near the 99% outdoor temp (which is +15F in NYC). While it may only have a COP of 2.2-2.5 at full speed at +15F, it'll be well over 3 running at part load at NYC's average winter temp of ~33F, and well into the 4s when it's just idling along at mid-modulation (rather than full-blast) at 45F outdoor temps. This has been widely tested in-situ in comparable & colder climates than yours in the Pacific Northwest under the NEEA (a utility consortium) Northwest Ductless Heat Pump Project (http://goingductless.com/consumer/about-northwest-ductless ).
http://neea.org/docs/default-source/reports/ductless-heat-pump-impact-process-evaluation-field-metering-report.pdf?sfvrsn=31
See also the bench-tested performance modeling across load & outdoor temp:
http://www.nrel.gov/docs/fy11osti/52175.pdf
Based on their data I'd hazard a seasonal average COP of 3.5 in NYC, maybe a bit higher if oversized slightly (but not by more than 50%, or you lose efficiency to cycling during the shoulder seasons.) In my neighborhood (central MA) a COP of 3.0 - 3.2 is a more likely average, which is somewhat below ground source heat pump all-in efficiencies, but the installation cost delta is high enough that the difference in power use is sometimes more cost-effectively made up with rooftop photovolatics than going with the more expensive heat pump system. (This is becoming more true every year, with the continuing decline in PV costs.)
Mounting the compressor at ground level is not a great idea in a place that sometimes sees snow-storms with drifting snow. Wall mounted on brackets, well above the historically deepest-ever snowpack, protected by roof overhangs (build a shed roof if the building doesn't have deep overhangs.)
I was exploring the idea of putting the heat pump condenser/compressor in an unfinished, poured cement basement (80' by 20' by 8') in coastal Maine. Thank you so much for the advice. Very helpful. Never thought much about the amount of air circulating over the coils! Guess I'll put it outside.