Some background: I'm getting ready to start a new home project in Boise Idaho. 2200sqft on the main level, 400 sqft daylight basement (TV room). The main level will all be ground polished concrete, 2.5", basement, slab on grade, probably carpet. The lot has a steep hillside in the backyard facing west (almost no windows facing west), most exposed windows facing south east. Fairly ideal setup with respect to good solar heat gain in the winter, minimal in the summer (key reason I picked the lot). I built a house about 12 years ago with an almost identical lot situation. It was 4800 sqft (2400 main, 2400 daylight basement) and had all concrete floors (3.5" on the main floor). The house had a LOT of thermal mass. I used a Polaris hot water heater for heat and built my own chiller (I think it was a 3 ton condensing unit that I plumbed to a heat exchanger). I used a basic thermostat control to kick on the condensing unit and all the radiant pumps (the upstairs pumps on high) as a crude way to try to keep the water temperature high enough to prevent condensation. I had no active humidity control. I also had a whole house fan to take advantage of the cool nights we have most of the summer. Heating system worked great. Cooling system worked fairly good. The whole house fan did a big share of the cooling for all but maybe 5 to 15 days a year when night time temps just didn't get low enough. We'd open windows and turn on the fan late at night and shut the house up in the morning. Even on very hot days, the chiller didn't kick in until late afternoon and would cycle into the evening. However, humidity in the house was noticeable in the late afternoon without any active humidity control making the house less comfortable than it could have been at the same temp. Yes, I got condensation sometimes in my mechanical room on the pumps and heat exchanger but it really wasn't a problem. Never on the floors. Humidity control was going to be my next project but starting a new company sidelined me and then a problem in my control circuit allowed the condenser to kick on without the pumps running which froze and cracked my heat exchanger. I called an HVAC guy and had him install an air handler to my condensing unit and ducting in the attic upstairs and that was the end of my radiant cooling experiment. And frankly, the house was a bit more comfortable in the heat of the summer. However, I think that if I had added some dehumidification to the radiant cooling system, I would have been there. Now that I'm building a new house with concrete floors again I naturally want to go with radiant heat. And it just seems wrong to have radiant heat and a whole separate AC system. But that's probably just a personal problem that I could get over. Boise is a fairly ideal climate for radiant cooling. My house design and lot situation should make my cooling loads fairly low. With a well designed system, I think it would work. Whatever I do, it should make good financial sense too. Last time around, I did the whole radiant design/install myself. Not going to do that this time. I've never owned a house that I liked my traditional forced air HVAC system. Too much temperature variation in the house, too much dust, too dry air in the winter. My questions are the following: Assuming a proper analysis says I can effectively cool the house with chilled water through my radiant floors with a little humidity control using one of those Altherma-like systems to heat and chill the water, is the system likely to cost significantly more to install and run than a dedicated radiant heat with a Polaris water heater (I have natural gas) and a mini-split system (or even traditional forced air AC) for cooling. How good are these mini-split systems? They seem to address one of my primary concerns with traditional forced air: better temperature control by room. Maybe an ERV would solve the wintertime lack of humidity from forced air? This definitely seems like the lowest risk (and perhaps the lowest cost) option. Does anybody make a "system" for taking advantage of cool nighttime air automatically? When temperature outside is 5 degrees cooler than inside, start pumping air inside with an automatic vent so you don't have to go around opening windows all over the house. I'm getting lazier as I grow older. Thanks in advance for any and all advice.

Forced hot air doesn't create any more or less humidity than radiant heat.

I agree with you on the need for a system that will automatically do whole house ventilation to outside air when it makes sense humidity and temperature wise. Open/closing windows and turning on/off the whole house fan is some trouble.

With a well insulated house and a mild climate, the HRV air vents/ducts may be enough for AC. Or close enough that upsizing them wouldn't cost much more.

Radiant floor cooling doesn't make much sense to me.

The first step should be a load calculation for the envelope of the building and then a load calculation for each interior zone.Ventilation requirements need to addressed as per building codes.Is your home subjected to power outages in the winter and what fuel is available.The beauty of radiant heating is obviously the comfort level.Radiant cooling becomes much more complicated .Controlling condensation is where you can get into trouble.Mini splits are very affordable and offer a means to have cooling without spending a lot of money such as in a radiant cooling system.To answer your question about outside cooling at night, what you are describing is an economizer cycle.Out side air is forced into the building typically by an air handler and a set of modulating dampers.The controls regulate the quantity of outside air usually based on enthalpy.To comply with modern ventilation codes ,your home is going to need either an HRV or ERV to deal with bathroom exhaust, kitchen exhaust,and the minimum fresh air intake as per ASHRAE

Have you done any air sealing in these houses or checked for leaks with a blower door test?

jonr: Yea I get that forced air heating shouldn't be any different than radiant when it comes to dryness but it sure feels different (and my daughter seems to get way more bloody noses).

So there's a need for an automated whole house ventilation system but you don't know of any off-the-shelf controls or dampers? Seems like there's a large swath of the western US with hot days, cool nights and low humidity that this would be perfect for.

acwizard: load calcs should happen next week. Electricity and natural gas readily available, few power outages. Seems like condensation control shouldn't be so hard assuming the load is low enough that I wouldn't need really low water temps, just need a system to keep humidity low enough to guarantee we don't get condensation. Seems like it all comes down to the load calcs.

ICFHybrid: this is new construction still in the design phase

The easy solution is the mini-splits, just not sure I'm going to be happy with cold concrete floors in the winter.

> off-the-shelf controls or dampers?

It's probably easy enough to put together, but it's hard to equal open windows in each room. You need high air flows when the outdoor temp is just a few degrees less than the indoor temp.

Maybe there is a smart HRV that opens dampers and switches to direct flow when the outside temp is beneficial.

Posted By jonr on 01 Jan 2013 09:31 PM
> off-the-shelf controls or dampers?

It's probably easy enough to put together, but it's hard to equal open windows in each room. You need high air flows when the outdoor temp is just a few degrees less than the indoor temp.

Maybe there is a smart HRV that opens dampers and switches to direct flow when the outside temp is beneficial.

So maybe just a good, insulated damper in each room that automatically opens when the whole house fan goes on and a control that detects the temperature differential, disables the AC and fires up the whole house fan. Part of the reason this setup was so effective in my last house was that we would get the house cold overnight. Temps are often in the high 50s to low 60s at night and in the 90s during the day. We'd run the whole house fan all night and the house would be down in the low to mid 60s by morning. With all the thermal mass in the floors it would stay cool almost all day long.

So, how tight are you planning to build this home?

Posted By youngwerth on 02 Jan 2013 12:06 AM


So maybe just a good, insulated damper in each room that automatically opens when the whole house fan goes on and a control that detects the temperature differential, disables the AC and fires up the whole house fan. Part of the reason this setup was so effective in my last house was that we would get the house cold overnight. Temps are often in the high 50s to low 60s at night and in the 90s during the day. We'd run the whole house fan all night and the house would be down in the low to mid 60s by morning. With all the thermal mass in the floors it would stay cool almost all day long.

Sounds like my building climate (90s daytime - 60s at night). How much electricity does it take to run that fan all night long? Couldn't an ERV do the same thing but at less power consumption?

I am leaning towards just opening windows at night and then closing them in the daytime. In an open design like mine, air circulates quite easily from end to end.

Couldn't an ERV do the same thing but at less power consumption?

An ERV is designed to conserve the building heat, not get rid of it. How low you can go on fan cfm/power depends on the temperature difference and how many btu you need to move.

ERVs maintain a differential of heat and humidity, whatever direction that differential is in. it helps in both heating and cooling modes. We use/sell the UltimateAir ERV, which is very high efficiency, and it also has an economizer function for night cooling.

Radiant cooling, IF you are suitable for it, in the situation you describe should be similar in price to the minisplit and boiler option, or close to the same. Basically the comparison would be:

Boiler/water heater/minisplits vs Altherma or other heat pump and chilled water coil for the erv ductwork to handle dehumidification. Plus a bit extra for controls... dew point based mixing control is a very good idea.

radiant cooling in floors is weak, but if you have managed your loads the floor you describe can do 10-12 BTUs/sq ft of cooling. not including the carpet areas, if you want to cool them you'll probably need supplemental help.

If you are talking about minisplits doing heat/cool entirely with no radiant component, then they will probably be a bit cheaper. remember you need the ventilation either way though.

Posted By youngwerth on 01 Jan 2013 09:01 PM
jonr: Yea I get that forced air heating shouldn't be any different than radiant when it comes to dryness but it sure feels different (and my daughter seems to get way more bloody noses).

So there's a need for an automated whole house ventilation system but you don't know of any off-the-shelf controls or dampers? Seems like there's a large swath of the western US with hot days, cool nights and low humidity that this would be perfect for.

acwizard: load calcs should happen next week. Electricity and natural gas readily available, few power outages. Seems like condensation control shouldn't be so hard assuming the load is low enough that I wouldn't need really low water temps, just need a system to keep humidity low enough to guarantee we don't get condensation. Seems like it all comes down to the load calcs.

ICFHybrid: this is new construction still in the design phase

The easy solution is the mini-splits, just not sure I'm going to be happy with cold concrete floors in the winter.

If both your house and ducts aren't VERY air tight and the duct system balanced, with low impedance return provisions for all doored off rooms, ducted hot air heating forces a LOT of un-intentional ventilation with the room-to-room pressure differences.  This air-handler driven infiltration makes for drier air in winter than in houses heated with radiant/hydronic or mini-splits.

Getting a "right" system does indeed depend on the load calculations.  In low-load homes it's often fairly easy to heat with point sources  (like one mini-split head per floor), but in the cooling season east or west facing rooms with big windows can drive cooling load balances crazy if trying to do it point-source.  But even then the notion that you'd need a mini-split per room is just bonkers, and it would be both expensive & inefficient to do it that way. (You can heat the entirety of some high-R houses with a single 3/4 ton mini-split, which typically put out ~12,000BTU/hr @ +5F.  Many homes can be heated with 2-head 2-2.5 ton multi-splits)  But if you can limit the amount of east or west-facing glass in doored-off rooms you can do pretty well with point-source cooling too.

If the concrete is insulated, it won't be cold- it'll be pretty much room-temp.  In a high-R/low load house heated with radiant floors the concrete will be slightly above room temp, but barely noticeable untill the heat load is 5BTU/hr per square foot.  In superinsulated houses it may never reach that level more than a few hours out of every heating season.  Again, it all comes down to the load calcs.

Decent quality mini-splits do pretty well in your climate, running a seasonal average COP of about 3 or bit better see Table 24, page 39 (p52 in .pdf pagination.

Personally, I'd be satisfied with radiant floor heating in just the kitchen and bathrooms. And since they are open rooms, a little higher btu rate there is OK.

Decent quality mini-splits do pretty well in your climate, running a seasonal average COP of about 3 or bit better

I believe they are only reporting the COP for the ~60% of the time/load that the heat pump worked. So say more like COP = 1.7 if you make up the shortfall with resistance heat.

ICFHybrid: I plan on building the house tight but could use some help identifying state-of-the-art cost/benefit insulation/sealing techniques to call out on the plans. The plans call for 2x6 exterior walls, has a "butterfly" roof (16" thick, no attic space). In the roof, architect is calling for 4" polyurethane spray insulation @ outside seal and 10" wet-spray cellulose insulation. Walls, 2" spray @ outside, 3.5" wet-spray. Floors R-30.

After thinking through the comfort thing some more, I definitely want to go with radiant heating.

Has anyone here looked at using a cooling tower for a chiller? Boise's mean wet bulb temperature in the summer is in the mid 50s and assuming that indoor outdoor RH is similar via an ERV should take care of all condensation worries, correct? CTS makes a 3-Ton chiller that uses a 1/6HP fan for about $1K. I found an interesting articles that suggest a cooling tower can significantly reduce the AC load: http://gundog.lbl.gov/dirpubs/strand03.pdf. Might be a cost-effective add-on to reduce the AC load.

Mean in the mid 50's means that you are often at other values so no, you're not safe with an ERV and no active dehumidification.

evap tower might do sensible but on A GUT HUNCH, not real math, it would seem unlikely to reliably provide dehumidification quality water temps (about 45 degrees f). but I'm not yet up to speed on the physics and capabilities of evaporative towers and could be wrong.

Dana1: I didn't see your post as I was writing mine. My experience in my last house was that bare feet on the concrete floors felt noticeably warm in the winter and cool in the summer (when I was doing the radiant cooling).

The first floor is 2200sqft, all finished concrete (no carpet), all suspended (about half crawl space, half over family room/garage below) all insulated (architect calling out R30). The house has no west facing windows and also has a steep hillside to the west. Cooling loads should be low. The house is U shaped which makes distribution with a mini-split perhaps a little trickier.

I'm probably letting a poorly designed HVAC system in my current house bias me away from the low-cost mini-split solution that I'd probably be perfectly happy with. However, given the U shaped layout, and the 400 sqft family room in the daylight basement, I'm thinking I would need a unit with 4 heads.

Yeah, I will note all this stuff about slab comfort in superinsulated buildings is a bit... off the mark. I stand barefooted or in socks here in my superinsulated shop every day, and I can most definitely tell the difference between an actively heated floor and one that isn't. floor temp oscillates from about 67 to about 73. 73 is very nice. 67 is kind of chilly and not as comfortable.

so as much as it's a meme in green building to think that minisplits provide equal comfort in tight buildings... I disagree. it may provide comfort people deem acceptable or even pretty good. but it's not the same as a radiant floor.

With a cooling tower, best case cooling is about 5F above wet bulb or 5F above the evaporative water source. In my case, I would probably be limited by the water source (guessing around 55F in the summer). I could probably reasonably get 60-65F outgoing water. Would probably want really good thermal conductivity in the floor.

Why would I need more than an ERV for dehumidification? Mean humidity during the summer months is about 40%. Wont an ERV be able to maintain that indoors?

Best ERV will only exchange about half of your humidity differential between the airstreams.