NRT rob ,
first the comparison was a propane boiler with a 45 KBTU output + upgraded insulation to a geothermal system with 90 kbtu.  My guess is there is more then 10K in differential cost to install.

My next door neighbor built a 5000 ft^2 house last year, heated basement with 1 full wall walk out, no underslab insulation no insulation on concrete walls, fiberglass insulation in walls (stapled to the inside of the stud bays so the drywall would lay flat), code min insulation in ceilings, double hung cheep builders special windows.  Duct work for the second floor in the un heated attic seams of the duct work taped, but no mastic.  The only air barier was house wrap very poorly detailed.   The house was built on the extreeme south end of their lot in a wooded lot with extensive clearing to the north for a majestic driveway and clearing to the west for a heated pool.  I think you would be hard pressed to do a stupider job of building a house.  I am willing to bet you could have knocked the heat load of that building in 1/2 for 40K.  On the other hand, they love their house and it has wonder full countertops and electric heated floors in the master bath. They do admit it seems cold to sit near the windows in the winter though.  That is the state of new construction I see around here, large and shitty with lip service to efficiency and shiny stone countertops.

In my own 1100ft^2  house, I figure I spent about 16000$ over code minimums between ICF basement walls, extra insulation,airsealing, pasive solar,  triplepane windows,  2 pannel PV fan powered solar airheater, solar hot water heater and munchkin boiler +superstore indirect tank.  I would be willing to bet my energy usage is 1/3  of that of a similar sized code minimum house
Now at current energy costs, that is a net savings of around 750 per year so it is a long payoff and marginaly worth it. 

My point is that right now is that it is  basicaly the same monthly cost to insulate well  and pay a higher morgage and lower energy bill.  However if energy costs go up it is a safe bet.

Hey, I like this guy...hehehee

I misunderstood the drop from geo to propane: if that's the case, the savings is much more than $10k in most cases.

I'll concede to you there are many houses out there you could cut the loads in half on for $40k, let me back down on that a bit. In a house already being built somewhat efficiently though, that's a high bar on a big house... and big houses are the only ones that should be considering geo, which is really what we are discussing here, your 1100 sq ft house notwithstanding. It's a lot cheaper to upgrade windows in an 1100 sq ft house than a 5k sq ft house.

So if nothing has been done that is definitely, absolutely, completely the first place to spend any energy dollars.. envelope first, always always always. But assuming a good effort has been made there, it's not the only place to ever put your money (unless you're doing a passivehouse!) and you can hit diminishing returns pretty fast, especially on very big homes... which are never going passive ;)

I think we are on the same page.  For small houses geo does not make sense cost wise.  Insulation does make sense.  For a big house geo makes sense when used with good envelope upgrades.

A couple of years ago I went to a talk at a clean energy show in clinton CT.  The geo expert said he thought the best aplication of geo was to older houses were  it was cost prohibitive to  retrofit with sufficient  insulation.  His point was that only when heating costs were high did geo make sense.  I think it is a tougher sell here with high electricity costs.

to me what is shocking is how much people pay yearly for energy.  My sister regularly runs natural gas+ electicity  bills  of 1000$ /month on a 3000 ft^2 house in west hartford.  

Actualy she would be a great fit for geothermal, she wants to put in AC anyway and  all summer she could  dump the heat into  her pool which she heats anyway. 
Cheers,
Eric

Do you have to put 2" xps under the entire concrete floor or just around the perimeter??  at $25 per 4'X8' sheet - thats 2.5 times the cost of the pex tubing.  One oponion was the earth below was a heat sink which would radiate upwards to the concrete.  What do yo0u think??

You can not compare the cost of insulation to the cost of pex tubing.  Insulation is the most important component of radiant heating, if you skimp on it you are wasting your wasting your money on the whole project. There are excellent products to use but make sure you make the investment.  I sell foil inisulation all day long but it is NOT the product to use under concrete.

Green is right, but is does depend on how warm you want it and when you will be heating it.

Where you live matters. As the average ground temperature is 47° here in Minnesota many of my large commercial designs forego insulation in the middle, as they will be kept slightly above ground temperature (conditioned storage for example). Perimeter insulation is the critical number and must not be overlooked.

If you plan to keep the space at 60F or above, setback temperature or need the space come up to 10 degrees overnight, the whole slab must be insulated.

Professional design is the more sure path to success.

As fond as I am of spouting my own, I don't believe in opinions nearly as much as taking the measurements & doing the math.  :-)  (Arithmetic is not an opinion.)

Unless the soil below already has a fairly high insulative value (probably not if it has much clay or damp-organic content), a big heat sink underneath a warmed slab will suck the heat out of it, not really buffer it (and radiate up into it during the cooling season, eh?).  Unless the sub-soil is naturally over 60F at a 8 ft depth (not likely, unless you live in the a cooling dominated climate or on a thin spot on a volcanic caldera: http://www.geo4va.vt.edu/A1/US-ground-temps.gif ) some amount of sub-slab insulation is usually cost-effective in a 10 year NPV analysis at current fuel prices & interest rates. 

The only time a heat sink (aka "thermal mass") is doing your heating system any good is when it's on the inside of an insulation boundary. Otherwise,  heat-sink is indentical with heat-loss - the only variable is the rate of heat loss, which is a function of the temperature of the slab, the relative R-value of the soil itself (which isn't very high even in the best cases), and the temperature of the sub-soil. 

Unless you've measured & analyzed it carefully,  don't count on the soil delivering better than R0.5-R1.5   between the sub soil & slab (typical soil conductivities tend to range from U0.6-U1.5.  R=1/U).  Lets do a straw-man ballpark calc on a 55F subsoil and an average heating season temp of 75F on the slab:

Taking the middle number, R1 between a 20F delta-T, is a 20BTU per hour loss for every square foot of slab area, or 1000BTU/hour for a modest 500 square foot center-slab area.  Even in a relatively short 3 month heating season that's over 2.1 decatherms, which even delivered with a mod-con boiler will take about 2.4 decatherms (=24 therms/year) of fuel input to deliver.  R5 (1") XPS at ~$0.55/ft^2 ($275 worth of material to do 500s.f.) will cut that by ~85% down to 0.36decatherms/year loss for a net savings of about 2 decatherms/year. Over 10 years that'll be 20 decatherms.  Don't know what fuel prices are in your neighborhood, but in mine it's ~$16, so that's a minimum of $320 in fuel savings even if fuel prices stay flat for the next decade (also not likely).

But my subsoil is cooler than 55F...
















...and my heating season is twice that long...

















...and R1 could be as much as twice the real number for sub-soil insulation performance, making the actual loss (and savings) twice as big (~48 decatherms fuel to support the loss goes down to 0.4 decatherms)...
















...which makes at least R5 (and probably R10 @ ~$550 worth of material ) a no-brainer investment for me, even with all installation labor & overhead factored in.  YMMV.


Maybe you have a more favorable model of soil's R-value, your fuel prices are half  that, or you're gonna flip the place in 3 years and need a faster ROI, have higher temp subsoil and a shorter heating season. Lot's of factors can enter into a cost/benefit analysis, but you can easily over-think it too.

Insulation properties of soils is a very complicated subject with lots of variables & nonliniearities related to soil humidity, actual temperature, the specific heat of the soil materials, etc.  You may do OK with a foot of sand for sub-slab insulation on some sites in some regions.  If your site has dry sandy soils and the slab is 20' above the water table you may be able to use the soil as a seasonal buffer earth-shelter style (read up on earth-sheltered design and investigate the site conditions before making that call.)  XPS is pretty reliable, predictable, and works anywhere. 

You can spend more time & money on the analysis than the insulation.  Or you can just do it "right" and know that it'll work, even if under closer scrutiny it might not always have been the absolute best investment you ever made (no matter what it surely won't be the worst!)