GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

How to heat small first floor rooms?

davidsmartin | Posted in Energy Efficiency and Durability on

I am building an addition with two rooms above a garage in Vermont that will be too far from my otherwise open first floor to get sufficient heat from my current wood stove or a possible mini-split that would be centrally located. I might use the air conditioner about three hours per year — it seldom gets hot here so that is not part of my calculations.

What are good options for a heat load that often might be only 3K BTU/hour? So far I have not found the ideal heating solution:

– single zone split units can have low minimum heat output, but buying three separate single zone units (one for each of the small rooms and a bigger one for the main part of the house) would be expensive upfront and seems wasteful.
– multi-zone split units appear to have rather high minimum outputs and could cycle on and off very frequently, thus losing one of the advantages of modern units.
– pellet stoves also seem to have high minimum feeds that would also require frequent cycling on and off
– many experts advise against trying to vent air from a comfortably warm room to another room only a few degrees cooler. It would take lots of 70 degree air to bring a room up to 68 degrees.
– propane and electric heaters of course can provide low levels of heat but use lots of expensive fuel per BTU

Am I missing an ideal solution? Or should I just bite the bullet and put in three separate single-zone mini-splits?

Thanks,

Dave

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.

Replies

  1. GBA Editor
    Martin Holladay | | #1

    David,
    You didn't list the fuels available to you. I'm guessing that you don't have access to natural gas -- but if you do, the simplest solution would be a forced-air gas furnace. Even if you have to use propane or oil as fuel, that still might be a solution worth considering.

    Depending on how the two rooms above the garage will be used, they might be served by a single ductless minsiplit or a ducted minisplit.

    Another possibility is one or two propane-fired space heaters with through-the-wall venting.

  2. davidsmartin | | #2

    Martin,

    Thanks for the prompt response. Your guess that I do not have access to natural gas is correct.

    Probably a single mini-split in one of the rooms is the way to go. One will be a bed room and the occupants will close the door at night, but hopefully leave it open during the day. The two rooms share a long wall. I was going to insulate it to reduce sound transmission, but if I leave it uninsulated would that let some heat migrate to the other room?

    I know you are skeptical about ducting air from one room to the other, but could a duct that went through that one wall, taking warm air from near the ceiling and blowing it out at the floor of the other make a significant contribution to equalizing the temperature in the two rooms? If so, would a return vent also help, or would it be sufficient to let the air go under the door to the hallway when it is closed?

    Is there a rule of thumb about now many cubic feet per minute of air that was say 10 degrees warmer would be needed to offset 1.8 KBTU of heating loss (at a 60 degree temperature differential between inside and outside)?

    I would rather avoid the propane space heater vented through the wall if possible in the room without the mini-split, but if it was used only as supplemental heat when the occupants are chilly then it probably would not use much propane. Right?

    Thanks for all your insightful answers and blogs.

    Dave

  3. GBA Editor
    Martin Holladay | | #3

    David,
    The best way to keep fuel costs low is to build an impeccable thermal envelope -- with close attention to airtight construction methods and thick insulation.

    I don't recommend that you experiment with fans between the rooms. For one thing, they make acoustical privacy impossible. For another, you will be disappointed in the performance.

  4. mfredericks | | #4

    David, I had a similar scenario where I have 2 bedrooms that don't get enough heat from our wood stove. I considered the options you're thinking about and decided to avoid dropping several thousand dollars on a mini-split, and instead, purchased 2 of these $100 heaters.
    http://www.homedepot.ca/product/econo-heat-400-watt-wall-panel-convection-heater/954362#BVRRWidgetID

    These are a low watt electric heating option that simply mounts to the wall and plugs into an outlet. We installed one of these in each of the bedrooms furthest from our wood stove, and they've worked well at adding some warmth to each room. I figured with the purchase price of $100 each, and such a lot wattage this seemed worth a try, with not much to lose if they didn't work out, but they've worked really well.

  5. Expert Member
    Dana Dorsett | | #5

    Without the actual heat load numbers for each room (which are in your control, if the place has yet to be built) and your actual 99% outside design temp, it's hard to say what's appropriate. Design temps in VT vary from very low negative single digits up to negative double-digits, which has an effect on air source heat pump capacity.

    The layout also makes a difference, which is also in you control if it has yet to be built. It's dead-easy to mount a mini-duct cassette in the top or side of a closet that happens to be between the two rooms, that would adequately conditions both without compromising privacy. The Fujitsu -_ _ RLFCD series have fully rated output down to -5F, and can be mounted vertically in a fat partition wall if that made it any easier/better. The minimum output @ 47F for any in that series is 3100 BTU/hr, and it will deliver less at min-modulation at lower outdoor temps. The -9RLFCD or -12RLFCD may be a better solution overall than a single wall-coil head in one room, counting on convection through an open door to fully condition both. But without better estimates of the heat loads and design temps that's little better than a shot in the dark.

  6. davidsmartin | | #6

    Thanks Dana for getting me thinking about a ducted solution. I guess if the unit is put in a closet between the two rooms and the ducting is very short then it would not suffer from the efficiency issues associated with longer ducts. Right?

    Both rooms are small and should be relatively well insulated. At a 60 degree differential I estimate that one will need 2600 BTU/hr while the bedroom might need 1800 BTU.

    I am not sure what the design temperature is here. My elevation is 2000 feet which means that it seldom gets as cold at night as it does in nearby valley while daytime highs can be lower. It might go below zero half a dozen nights per year and -10 only once a year. In any event, it seems clear that a 9K BTU unit will be more than sufficient even at my coldest temperatures. More often I will need less heat than the minimum the unit produces and it will start cycling. Right?

    I don't really understand how these ducted units work and I do not see as much online info about them as the ductless models. I assume they are all one zone, right? If the mini-duct cassette were put in a closet between the two rooms could I hook things up so that 60% of the hot air normally goes to one room and 40% to the other? And are there baffles so that if the room without the thermostat was not being used the amount of heat going that way could be reduced?

    Dave

  7. Expert Member
    Dana Dorsett | | #7

    The lower the duct impedance, the less power it takes to move that air. It's one zone per cassette, but it's possible to support multiple cassettes with a single compressor (but not usually desirable.) With ultra-short duct runs just size the cross sectional area of the ducts with the approximate difference in load (40/60, in your case), and install tweakable vanes for temperature balancing. You can always install operable registers for micro-tweaking it, but don't turn the flow down very much. Simply turning it off doesn't save energy and may even increase energy use despite the cooler room temp by unbalancing the system.

    Be sure to install the cassette where changing filters is relatively easy, and be sure to think about how you're going to route condensate disposal when running it in cooling mode.

    To meet code you need to be able to heat these rooms to 68F at the 99% outside temperature bin, so you should be looking at load numbers with at LEAST a 70F delta-T, not 60F, and probably more than 75F.

    What is the nearest town that might have good outside temperature datasets, and what is their elevation? Got a ZIP code? A crude estimate of ~3F/1000' adiabatic cooling with altitude change would be a good enough estimate of the very-local 99% temperature bin (though ground fog and radiational cooling can sometimes throw that way off.)

    For instance, the 99% temperature bin for Barre VTat 1165' is -11F. If you're near Barre and are at 2000' of elevation, that's 835' above Barre, almost a 1000', so assume it'll be about 3F cooler- use an outside design temp of -14F, which will be close enough (with enough exceptions to prove the rule. :-) )

    And note a -14F design temp is an 82F delta-T to be able to hit 68F, which would put the design heat load at about 6000 BTU/hr, not the estimated 4400 BTU/hr @ 60F delta-T.

    It's more than an efficiency issue, it's a comfort issue. A 3/4 ton 9RLS3 boast an HSPF of 14.2, to the 9RLFCD's 12.2 But with the better heat distribution the 9RLFCD will have less on/off cycling, and you'll be able to adjust the temperature balance. On a pipsqueak design load of as 4400-6000 BTU/hr the difference in efficiency is kind of "in the noise", and easily erased by even modest use of resistance electric heating.

  8. davidsmartin | | #8

    Thanks Dana.

    I am in Mount Holly, VT, 05758. The nearest medium sized town is Ludlow but they are much lower elevation. My observation is that it is much colder at night down in the valleys and they thus have lower design temperatures. Anyway, if it is very cold I will throw another log on the fire! I was using a 60 degree differential as a typical cold night, not the 99% case.

    Fujitsu tells me that they do not sell ducted units with the low temperature capacity. They sent me capacity tables for AOU9RLFC and ARU9RLF. I note that you referred to AOU9RLFCD. Does the extra D at the end mean something?

    Thanks,

    Dave

  9. Expert Member
    Dana Dorsett | | #9

    I'm not sure if the D means anything other that "Duct". They are supposed to have a fully specified output at -5F, but they won't specify the output below that temp (but it keeps running, and keeps putting out heat.) They don't seem to publish the extended temperature capacity tables for those units online, only the submittal sheets, where the nominal heating is a guaranteed (but not necessarily maximum) capacity at +17F. What is their capacity is at -5F outdoors, with a 70F interior design temp?

    Those units are on Efficiency Vermont's rebate-worthy list, and I'm pretty sure people would be using them in their ski-condos, etc, as well as higher-R houses.

    https://www.efficiencyvermont.com/docs/for_partners/contractors/evt-cchp-qpl-bymanufacturer.pdf?v=10

    Ludlow's design temps aren't in the ACCA short-list, but Rutland's is:

    https://www.extension.org/sites/default/files/7.%20Outdoor_Design_Conditions_508.pdf

    But looking at Ludow's datasets in a Weatherspark graphic, -5F is the the lowest 10th percentile for mid-January, which is very close to that of Rutland's which IS on the short list. Rutland's 99% bin it -8F, so that's probably going to be pretty close to the design temp in Ludlow. Ludlow's elevation is 1020', so your 99% bin isn't going to be colder than -11F.

    Actually, Mt. Holly's mid-January low temps are very comparable to Rutland's in the Weatherspark datasets, so maybe -8F is closer to the right temp for the 99th percentile bin:

    https://weatherspark.com/#!dashboard;a=VT/Mount_Holly

    https://weatherspark.com/#!dashboard;a=VT/Rutland

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |