In the house that I am currently working on with an architect and engineer, I've already decided that I'll use Warmboard radiant.  High heating climate, either electric or propane, no nat gas available.  The heat load on the house at design temp is about 55,000 BTU.  I had been considering propane modcon boiler, but I saw a post on placement of the ground loop coils recently that caught my eye.  I'm building into the side of a hillside.  The backwall and side walls of the lowest two floors are about 26 feet deep.  As this is all excavated and then backfilled, why not slip the coils in a vertical orientation (coils on edge) before I backfill?  But I need to know if I have enough area to supply what I need.  Frost line is ~42"  That would leave me at least 20 feet of depth.  Wall lengths total about 120 feet.  Would that be enough to get me 5 tons worth of coils in?   I'm thinking that would save all the excavation costs and possibly make it affordable.

Any thoughts ?[script removed]

Very, very, generally speaking you want aywhere from 300-800'/ton of pipe in the ground. You also do not want to extract heat too close to the house.
When you say "affordable"; how 'bout installing the only system that lats you deduct 30% of the radiation system. Warmboard, in floor, baseboard all these things are expensive but with geo you get them for 70 cents on the dollar.

Ie, maybe 1 or 2 tons with the risk of foundation damage. Once the equipment is there, you might find it doesn't cost much more to get loops properly buried. The other reasonable option (other than more insulation) is an air source heat pump.

how far can you keep the coils from the foundation?

Posted By jonr on 13 Dec 2012 09:04 AM
Ie, maybe 1 ton with the risk of foundation damage. Once the equipment is there, you might find it doesn't cost much more to get loops properly buried. The other reasonable option (other than more insulation) is an air source heat pump.

Getting 55KBTU/hr out of output on a hydronic air-source heat pump can be a difficult spec to hit in truly cold climates.

But a 42" design frost depth reads more like a US climate zone 5 or cool edge of zone 4 location:  http://www.decks.com/images/Article...th-map.jpg 

The Daikin Altherma can hit pretty good efficiency with WarmBoard in zones 5 or even zone 6, but the raw output isn't going to hit 55,000BTU/hr at truly cold temps.

What is your 99% outside design temp, and how many square feet of WarmBoard? The finish-floor R and the peak & average water temps will make a difference on efficiency  whether GSHP or ASHP, and have a significant impact on the capacity of an ASHP.  If you need 140F water at a design temp of -5F it's not in the cards, but if you need 90F water at a design temp of +25F there's probably an air-source solution.

In any case, cutting a 55K heat load down to a more manageable size with thermal upgrades to he building might be more cost effective than 4 tons or more of GSHP.  It takes a sharp pencil to find the perfect balance, but a reduced ton of insulation & air sealing has no operating cost, and provides more comfort than the ton of heat pump it displaced.

The Daikin Altherma can hit pretty good efficiency with WarmBoard in zones 5 or even zone 6, but the raw output isn't going to hit 55,000BTU/hr at truly cold temps.

Agreed, but you have electric heat for the coldest times.

Where are you located? I usually like to stay at least 3', if possible 5ft away from any building wall. Sounds like you would need a 5 ton Water to Water unit. However, if this is all build withi a hill, 55 KBTU/H sounds high. You would also need about 4000ft of pipe slinky, for a total lenght of "trench" of about 500 ft. Yes, you could stack a vertical slinky vertically over 26 ft height, but it sounds like an uphill battle, and you don't sound very experienced to overcome onsite issues easily. At the very least I would excavate wider around the house and put the slinky out in layers horizontally, having at least 3-4 inches between them.

Wow, this forum hasn't see so many responses so quickly in a long time [script removed]

I'm not sure where all teh rest of my reply went to!  Just saw now a couple days later that only the first sentence got posted.

Dana The design temp is -25F. you helped me decide on that about a month ago when I had the ID LarryT.  (It seems I forgot my password and tried every combination of things it could be.  Finally hit the forgot passwordbutton but it sent it to an email address I closed long ago!  Getting old causes crazy things ;-)  )

The warmboard area of the house is about 6000 ft split between 2.5 floors. I guess I'll make someone's sales quota. The lower living level is subterranean on three sides as is the garage below that. 

The heat load was simply guesstimated using the calulator from Borst.  Assumed R 32 walls (8" Kama- EBBS foam) R45 ceiling and window U of 0.25.  Used air change of 0.5/hr  As this will be a very tight house, ICF and EBBS and proper sealing between panels and penetrations, is that a reassonable air change coefficient?  Air change was 1/2 the load. 

I was simply trying to determine whether it wass a 2 ton or a 10 ton load.  Very rough calcs for the moment so I could post and get opinions back.  They obviously will be done properly when the plans are finished - soon.

Regarding how far I could set the pipe from the house, -~8 feet.  Don't want to go any further or go horizontal as this is stone, not dirt.  It's a mountain, not a hillside.  With a 42" frostline I would really have some excavating/blasting.  Not worth it.

If I don't have the space to get enough coils in I understand and will go back to propane and modcon unit, unless there is truly a air/water heat pump that can operate efficiently at this design temp.  Other option would be to uses air/water for the range it can (average winter low might be -10F) and propane for backup?  Is there an air/water unit that can operate well at -10F?

You might also consider open loop geothermal or oversized air source heat pumps plus a water tank so that it could run only during the day (how much warmer it is then depends on your climate). And of course there is solar heating.

If you can be 8' from the foundation, then you don't have to worry about robbing heat from the house. It would take some extra care when backfilling, but you could put a lot of pipe in the are that you have.

Be sure to look at soil volume, not just how much pipe you can squeeze into an area.

Rock sounds good, that makes working 26' down easier. Again, where are you located? If you have 8' from the wall, you could put in a 3' horizontal slinky with an 18" pitch and still be away 5' from the wall of the house. This would also increase your length of your slinky going "around the corner" by 10ft on each side, giving you 140' length total. If you put your header vertical on the rock in the middle of your back wall, you could branch off with 600' of 3/4" pipe to each side of the house (1200' total per layer), separated by 3 ft of backfill soil. Assuming an average conductivity of the backfill material, this should give you about 0.8 ton per layer. A slinky at 26', 23', 20', 17', 14', 11' and 8' would support roughly 6 tons of heat (your heatloss indicates 5.5). Your reynolds number will look ugly, and don't know your soil conductivity, diffusivity and moisture content, but this is calculated with some safety margins, so this can work nicely. Given the uniqueness of the design, you need those safety margins since your are drawing not much heat from the house side and not much heat from above and below most of the layers, only the rock side is there to bring heat towards the pipes. Since you are likely way up north, you will be very heat dominated, and there is a long term memory effect reducing the average annual temperature of the loopfield, since the pipes are close together and it is an unbalanced heat load if you plan no A/C (again, where is this project located? Need average deep ground temperature and A/C load).
Again, it depends on the conductivity of the backfill (clay would be good), and you need to insulate the back of the wall very well. And yes, reduce the heatload as much as you can.
Then we should talk equipment selection and flowcenter pumping needed, 2 inch header pipe would be good, but it depends on the final choice of the heatpump, which depends on the final load.
Domestic hot water via the geosystem? How many people in the household?

Posted By LarryT on 12 Dec 2012 10:21 PM
  The backwall and side walls of the lowest two floors are about 26 feet deep.

Any thoughts ?[script removed]

Could we see a plan side elevation view?

I'll try to post some elevation views. next post

Doc, I'll have to reread your last reply again.  I need to put all of it together in my mind.

I'm in Northern NM at 9000'.  No AC required..   Unless there is company visiting it's just my wife and I.  I won't explain why the large house  for two of us here, but there's a valid reason.  Actually, the lower living area will be closed off much of the time, or at much reduced heat load as well as the upper two bedrooms.  we are trying to isolate sections when not occupied to reduce loads.  That said, I was planning to just run a propane water heater for domestic, or run off the modcon if that was being used.

As soon as the construction docs are finished, I'll do a full manual J and determine exactly what my load will be.  Regarding air changes per hour, what is a good number to use in the calculations for a very tight house.

Obviously, you should try to get the load as low as possible. You appear to have enough area to make your ground loop work. You should also consider making part of your hot water (or all) via the geo system. If you stay below 55KBTU you get be with one hydronic heat pump, making this quite economical thesible. However, you need help onsite.

Are we certain that winter design temp of -25*F applies...in New Mexico? The lowest design temp I could find in my weather database for NM is low single digits at 6500'. Allowing for lapse rate to higher altitudes, I still can't imagine a design temp much below -10*F.

Systems are NOT designed for record cold... a ton could come out of this system if the winter design temp is 15 degrees off.

One of the things to consider is the capacity of the heatpump, which is significantly higher at lets say 90F supply temperature than at 104F rated temperature ELT temperature. On the flip side, back up strip heat does not exist in that kind of hydronic application. If you are using an electric tank as your buffer, you could activate the electric heat element as supplement heat on extremely cold days.

Given your uncertainties, constrct the house applying the best insulation you can get to to reduce your load. You know that the loopfield area can cover it, then you figure out what the actual load is the heatpump needs to handle.

Curt, when you say systems are not designed for record cold do you mean for heat pump systems? And if so because you can have a backup source? The -25 design is truly that. Going back as far as I could, which was 6 years of recorded data, that was the average yearly low, with a couple of those 6 years in the -35 range. Yes the other years wer considerably "warmer", with lows in the -15 range. That being said, what would be a recommendation for a geothermal design temp?

Systems are normally designed to meet load 99% of the time via a manual J calculation. Stated another way, the system will be slightly undersized for 40-50 hours each winter or each summer. The 40-50 hours won't happen continuously, but rather in multiple 2-4 hour intervals (such as from 2 am to 6 am during a cold snap or from 2 pm to 6 pm during a heat wave). A combination of Manual J safety factors and building thermal inertia carries through the short intervals.

Design temps are published for hundreds of cities - use the one nearest your home. That said, don't be afraid to adjust for micro climate issues, such as being 2000 feet above the nearest city for which design data is published. THAT SAID, don't take carte Blanche to arbitrarily add or subtract 5-10 degrees from a reasonable design temperature - that is poor practice leading to oversizing and lost efficiency.