Now for discussing the source side of the system... I've read countless threads and seems every situation is different so let me give you the facts. I have a great well on the property for drinking water, although I've not had the water tested yet for pH,minerals, alkalinity, etc. It's drilled into granite (about 120' well, bottom 35' in rock). When we had it drilled this summer, the driller estimated the refill rate at 50+gpm so open loop is an option. I'm not sure I like the pump and dump idea, simply because I have no way to make use of the returns... at least not that much (the garden only needs so much water). The house is on 30 acres, so horizontal is definitely an option and I have digging equipment available. Behind the house is flat, with soils ranging from red clay to a sandy mix. About 100' to 150' from the house (downhill side) is an area that stays saturated year round. If you dig a post hole, it's full of water within an hour. Vertical wells, I have an experienced driller that has done 2 water wells for me. For Geo wells, he's quoted me $10/foot. Mr. Shiny truck and sunglasses (see earlier post)told me we'd never be happy with anything but a closed loop vertical system (I guess based on the $22/foot he quoted for vertical). I've been considering digging a pond, but it wasn't in the plan/budget this early in the game as my attention has been focused on building. What say you, GBT?

Posted By engineer on 29 Apr 2013 10:56 PM
Glazing solar gain may be your friend in Buffalo, but most assuredly not in Atlanta.

If the peak loads are indeed so far apart (heat load 2 x the cooling load) shouldn't we be able to reduce the loads at certain days, reduce operational costs and balance the building better if we allow a bit more solar gain?

Lets say you take a 5 ton W-W heatpumps, and your loads are correct, loops should be fairy simple.
If you have saturated soil, you can put it in horizontally at 8', and put in 2400' of slinky.
Or you can drill (2) 470ft holes.
Or you can drill a discharge well and go with an open system. It would be quite efficient giving you 64F water all year around.
$10/ft is quite reasonable. Does that include casing, pipes and grouting?
I am not sure how well a pond would work in GA.

Consider not oversizing to 5 tons of heat pump for a 3.7 ton heat load (before aux heat, less after) and a less than 3 ton total cooling load.

Given stated ground saturation and climate, I doubt 8' deep ground loops are necessary or safe.

Posted By joe.ami on 30 Apr 2013 08:30 AM
Given stated ground saturation and climate, I doubt 8' deep ground loops are necessary or safe.

Of course they are not needed at 8', you can run them at 6 feet, even at 5', you can excavate a pit, etc. Many ways to skin a cat. The point was that he does not need much loop and horizontals are certainly an option to be considered.

Posted By jonr on 30 Apr 2013 07:20 AM
Consider not oversizing to 5 tons of heat pump for a 3.7 ton heat load (before aux heat, less after) and a less than 3 ton total cooling load.

Wasn't his heat load around 44KBTU/H, and wasn't he using radiant? So how do you satisfy 44KBTU/H at 99% design conditions, with a less than 5 ton w-w heatpump? And what is your aux heat solution? Radiant should be designed to carry the entire 100% load, there is no aux strip heat to bail you out.

I hadn't planned an aux heat solution for the HVAC system, other than a woodstove or rumford fireplace. We've talked about adding (down the road a piece) some solar panels or possibly an outside wood boiler as I have a near unlimited supply of wood chips/mill slabs and 30 acres of woods that have to be managed/thinned. I originally planned to heat with a wood gasification boiler and radiant floors, but as I am often out of town for work, didn't want to burden the wife with wood as a primary heat source.

I see that one of the key elements of minimizing latent load, especially for DOAS systems, is control of building pressurization. Keep a room very slightly positively pressurized and latent load from infiltration (which is the cause of almost all of the latent load) goes to zero. Ventilation ducts handle the remainder. With the latent load taken care of, radiant cooling is straightforward.

Side benefits are dry walls (no mold) and no unfiltered air getting in.

Posted By jonr on 30 Apr 2013 11:47 PM
I see that one of the key elements of minimizing latent load, especially for DOAS systems, is control of building pressurization. Keep a room very slightly positively pressurized and latent load from infiltration (which is the cause of almost all of the latent load) goes to zero. Ventilation ducts handle the remainder. With the latent load taken care of, radiant cooling is straightforward.

A side benefit is dry walls, no mold and no unfiltered air getting in.

Do you have any example of "straight forward" radiant cooling working well in the humid southeast U.S., especially with radiant floors? Air infiltration never goes to zero, even with pressurized houses, thanks to wind.
I just got back from ISH in Frankfurt in March, and it was all about radiant cooling and the controls for it, and radiant ceilings. But every manufacturer admitted that that it can carry maybe 1/3 of the cooling load, and reduces the ductwork, but never eliminates it, even in dry European and middle eastern climate.
If you pressurize the house, the makeup air has to come from the outside, and you have to constantly dehumidify it and cool it down, blowing your precious tempered air to the outside, which will likely costs you a significant amount of energy.

Air infiltration never goes to zero, even with pressurized houses, thanks to wind.

It doesn't take much pressurization to overcome the effects of wind - ie, infiltration can go to zero. But how much pressure should change with conditions so that you aren't doing any more than necessary.

If you pressurize the house, the makeup air has to come from the outside, and you have to constantly dehumidify it and cool it down, blowing your precious tempered air to the outside, which will likely costs you a significant amount of energy.

The alternative is to let air infiltrate through the walls naturally (or as caused by a typical poorly balanced duct system). And since it's unevenly spread all over (vs centralized), it costs as much or more to condition it. Ie, compare +1 pascal everywhere to +5 on the windward side and -5 on the leeward side. +1 uses LESS energy.

I don't think anyone is discussing no ducts, small ones are needed for ventilation air. And once they are there, why not use them for some dehumidification and cooling?

I'm sure that DOAS + radiant isn't a good fit everywhere. But understanding the concepts is useful, even for non DOAS systems.

I suspect that the typical residential installer isn't going to find any advanced system "straight forward". Even well sealed and insulated houses and "don't oversize" seem to be a challenge.

Dan, I'm interested in your ACH and CFM figures used in coming up with latent load.

I've since gone through and done some tweaking to the numbers in HVAC calc. I will get the updated numbers and post them here. My gains and losses have come to be much closer together than they previously were, and somewhat higher (mostly my gains). For example, I changed the roof to a dark color (which I may change back as I'm leaning toward a lighter color), and turned the house 45 degrees to the northeast (should I have done this? The house is more closely 20 degrees or so from facing due north).

For the latent load, I left the system default for ACH and added 40 CFM to the infiltration to account for the ERV that I planned (per the ERV manufacturer's recommendation)

Additionally, I tried using some of the various manufacturer's sizing software. Entering as much information as the software allows, I am seeing some 20 to 30% variances in the numbers compared to what HVAC calc produced. That has me wondering what to trust. Are the manufacturer's heat gain/loss figures any good? I want to lean toward HVAC calc, as I can enter in so much more detail. Your thoughts? Should I override the calculations and enter in the numbers from HVAC calc?

I wonder what ACH-nat value it uses. Do you have a target for the ACH50 that the house will be built to? I'm asking because I came up with some higher latent load values.

I don't know about the software. I'd probably keep working at it until things are fairly consistent across several of them. Eventually I'd have a good commercial HVAC engineer review the final plan.

okay... looking at the infiltration in the HVAC-calc, I used the recommended values per the guide in the instructions; 0.2 ACH summer and 0.3 ACH winter. I added 40 cfm for the ERV. With all the adjustments I've made so far, I have the following:

Sensible gain -- 30786
Latent gain -- 8607
Total heat gain -- 39393
Total heat loss -- 56295

As far as the target ACH50 for the house, I'm not sure. The house envelope will be built as follows:

8" ICF basement wall (3" foam inside and out)
I-joist first floor system with rim joist spray foamed.
Timberframe structure with exterior insulated walls (seams staggered and taped) extending to mud sill (exterior insulation extending down to termite flashing)
Exposed beam ceilings with 1'5" T&G, insulated with 2 layers of 2" polyiso (label rating of r-7.5/inch) seams staggered and taped, extending out to and sealed to exterior wall insulation.

That comes out close then - I was using .35 for ACH-nat, guessing at the total house volume and then using "LL = .68 x cfm x grains".

Looks like your numbers went up quite a bit.

You don't need to run the ERV at 40 cfm all the time. Adding the internal gains, you are getting borderline with a 5-ton heatpump.

I still think 53 kbtu is a lot for a 4900 sqf house in your climate for the heatloss. worst case you have to look into the hydrons for example, they put about 5 KBTU more heat out than the WF or CM 5 ton model.

I still think 53 kbtu is a lot for a 4900 sqf house in your climate for the heatloss.

I think so too. The closest city I can find for a comparative winter design temperature is LaGrange, which is listed at 23 F (which I used).

If I do away with the fireplace, I can drop 8250 btu/h. I didn't figure the fireplace would be such a heat loss. the wood stove may just have to take it's place. (I'm showing a total of 22k in the great room, with the fireplace being the worst offender).

duplicate post

duplicate post.

On another note... is going by the heating size in selecting the equipment tonnage going to grossly oversize the system for cooling?