Hello all,

I'm quite new to geothermal and in fact to HVAC in general.  I'm a mech engineer by trade, but had to sleep sometime during uni and regrettably chose to do it during any discussions on HVAC.  Thus I understand thermodynamics just fine, but not so much the specifics of HVAC.

My question is that I live in a very mild climate (January 2.5% @ -8C [18F] and have never had a need for cooling with July 2.5% @ 29C), local soil temperature quoted at around 10C (in news article however reliable that is), number of degree days below 18C is 3200, so is there likely to be any net benefit to going ground source instead of air source in my region?  What are the evaluation parameters I should look at?  Can I make up most of the energy deficit of air source by investing the capital cost difference in improved insulation and maybe a HRV?

This would be for new construction on a roughly 3000 sqft house.  As noted, I would plan to optimize the heat loss load for the system to the heating type, ie. cheaper air source pump means money for increased insulation and less heat loss, but likely to have above average insulation in either case.  House is well situated for passive solar gains if that makes much difference.

Some direction would be appreciated.

Where are you? C temps suggest Canada, but mild design temps suggest Vancouver or thereabouts.

You might be right in that the payback for geo compared with an ASHP in your circumstances might be decades. House design loads coupled with temperature bin data for your area should yield a reasonable estimate of total millions of btu needed during an average heating season; then evaluate system options up front and operating costs delivering that number of btus annually

You might also consider multisplit ductless units by Daikin or Mitsubishi. A single ASHP condenser feeds up to 4 wall mounted evaporators, and they are said to produce useful heat at decent COPs at and below your design temps.

Beware that provisions for good passive solar gain in winter don't result in unwanted overheating in summer.

Yup, the house will be in Vancouver when I get back there. Impressive deduction.

Thanks for the advice. That gives me a good idea of where to start.

I'm thinking extended eaves over the windows should work to prevent overheating. I imagine I can determine the sun angle for my area and that it shifts enough summer to winter for that to be effective, but I will need to make sure.

Do manual J calc programs have that level of detail in the input parameters, or would I need to try to estimate solar gain inputs myself? I'm hoping to size the AC accurately enough that I cover only 90% of the need and install an efficient, but mainly decorative gas fireplace to cover the last 10% when required.

You may wish to have someone run a more detailed energy analysis to help get a better handle on the passive solar and the whole house simulation. You're trying to do more than just size the HVAC, you're really designing your entire home, and that's going to require more than a Manual-J calculator.
There are several programs that can be used that will do full, hour-by-hour load analysis that take into account overhangs, solar input, etc. Depending on how deep you want to get into it, you may want to learn one of the free programs so you can play with all the parameters yourself.

OK, Here is my take on it.
 
The number one most important thing is to reduce the need for energy input. It is usually cheaper to reduce the demand in the first place then to supply that demand with a more efficient source. This is done primarily by increasing airtightness, insulation, quality of windows, and tuning the windows to increase or decrease solar gain as needed.

Start looking for books on passive solar design, I recommend the following. The Solar House: Passive Heating and Cooling by Daniel D. Chiras, Solar Water Heating: A Comprehensive Guide to Solar Water and Space Heating Systems by Bob Ramlow, Passive Solar House: The Complete Guide to Heating and Cooling Your Home by James Kachadorian.
 
The hard part is to do this without spending a lot of money. Because you live in a moderate cold climate, I don’t think you need to go to the extremes to build an efficient house that you would in a very cold location. My guess is that you could go with something like R28 clear wall, R60 ceiling and windows ~r5 in a tight structure and end up with a fairly low heat load. I would think a good goal would be a total yearly heating load of 20-30 million BTU(200-300 therms) for a 3000 ft^2 house in your climate.
At 1.10$ per them and 90% efficiency this would be ~ 350 $ per year at current rates. On the other hand since electricity is 6.7 cents per kwh in your area, resistance heat would be ~ 700$ per year in this scenario. A GSHP might drop that to 250 per year, but you have to add back increased maintenance costs. It would be hard to justify the cost of GSHP in your case.

Assuming heating but no AC, I would think that gas heat would be the least expensive overall, but you would have to calculate the cost of installation. Don’t forget that costs of heating hot water can become greater then that for heating in moderate climates. You will want to optimize for that also. I think the simplest solution might be Passive solar plus a small gas mod con boiler + baseboard and indirect storage tank for domestic hot water. Add in an HRV for fresh air and you have a moderate cost, workable system. The next step up might be a solar DHW system, and or solar air heaters, but I doubt at the cost of energy where you are either of these will be that advantageous.

Again reducing demand is always first priority. As to the trade offs in Air vs ground source heat pumps, I can’t give you any advice, but plenty of others on this board will be able to.

Cheers,
Eric

What eric anderson said- in Vancouver you can probably go all PassiveHouse on it at the design phase, and skip the heating "system" entirely for a comparable or lower upfront cost than a minimal ground-source heat pump system.

Your clear-wall R values would probably need to be more like mid to high -30s than his R28 guesstimate to meet PassiveHouse standard (R28 works for much of coastal CA/OR, but Vancouver is quite a bit cooler- like ~5-10C cooler.) Wintertime solar gains are also going to be pretty low due to local climate factors, but with a super tight home well insulated (including the foundation), and heat recovery ventilation the heating load can go near-zero, and even using resistance heating elements or a hydronic coil (tapping off the hot water heater) in the ventilation airstream to cover the shortfall during true cold snaps would be more than sufficient. Most of the time the indoor temp would be controllable by the ventilation rate.

High efficiency heating systems at the single-family house level are only for low-efficiency building envelopes. It's possible to build 'em better at reasonable cost in new construction. I'm guessing R19-R20 batts in 2x6 construction is still the "typical" new construction in your neighborhood, maybe with 1.5cm of foam on the outside of the sheathing, maybe not, with uninsulated poured concrete basements (maybe with R19 batts under the first floor joists in some), and nothing under the slab but gravel. With an upgrade going with R25-R30 insulated concrete forms for basement walls, at least R10 under the slab (thermally broken at the edges, and 3" of polyiso on the exterior, and using blown cellulose or the better superfine high density blown fiberglass like JM Spider instead of batts get you most of the way there. (Thats ~ R40 clear-wall in a ~25cm thick wall) Then it's a matter of specing & sizing the glazing & doors correctly, etc. The PassiveHouse software may be useful for tweaking the details, and the details become very important- at some point the thermal bridging of framing elements, plumbing, etc become a significant portion the total heat loss. Designing the roof structure or trusses to accomodate R75ish of blown insulation (again, no lossy-leaky batts here), all the way out to above the framing without thin spots is just one of the bigger details, but it's not very expensive to build.

In any but the mildest winter climates air source heat pumps are justifiable over resistance strips because they are substantially more efficient and not too expensive since millions are made per year.

The jump from ASHP to geo may be much much harder to justify in mild climates.

As to PassiveHaus, it is true a home can be constructed so as to be able to be heated by 2 candles and a night light, but the materials and labor to do so don't come cheap.

Before going crazy with geo and / or PassiveHaus, look at the numbers (first cost and operating cost) for a conventional but well built frame house constructed to a reasonable standard such as the Canadian "2000" code.

I had built a 3400 SF ICF house with open loop geo whose heating, cooling, and hot water costs combined are less than $50. The incremental cost of it over conventional methods may well have cost $100k. I also benefit from increased resistance to storms, termite resistance, decreased noise, and reduced humidity owing to thermal mass effects. Whether all those combine to be worth the incremental cost is an open question. I'd do (most of it) again in a heartbeat, but energy savings alone can't justify all the choices we made.

Posted By Dana1 on 30 Jul 2010 01:03 PM
High efficiency heating systems at the single-family house level are only for low-efficiency building envelopes. It's possible to build 'em better at reasonable cost in new construction. I'm guessing R19-R20 batts in 2x6 construction is still the "typical" new construction in your neighborhood, maybe with 1.5cm of foam on the outside of the sheathing, maybe not, with uninsulated poured concrete basements (maybe with R19 batts under the first floor joists in some), and nothing under the slab but gravel. With an upgrade going with R25-R30 insulated concrete forms for basement walls, at least R10 under the slab (thermally broken at the edges, and 3" of polyiso on the exterior, and using blown cellulose or the better superfine high density blown fiberglass like JM Spider instead of batts get you most of the way there. (Thats ~ R40 clear-wall in a ~25cm thick wall) Then it's a matter of specing & sizing the glazing & doors correctly, etc. The PassiveHouse software may be useful for tweaking the details, and the details become very important- at some point the thermal bridging of framing elements, plumbing, etc become a significant portion the total heat loss. Designing the roof structure or trusses to accomodate R75ish of blown insulation (again, no lossy-leaky batts here), all the way out to above the framing without thin spots is just one of the bigger details, but it's not very expensive to build.

Thanks guys.  Great food for thought.  I had imagined insulation similar to what you describe, but I hadn't imagined I could get away from some kind of primary heating entirely.  That would be great!  I was thinking R-40 walls and R-60 roof if I can save all the money of the heating system.  I'll  have to check my spreadsheet to see if R-75 roof makes much of an improvement to heat loss.  That brings the heat loss through the roof and walls substantially below the losses created by current window capability, so I'm thinking about reducing window areas on all but the south side as well.  Not too sure how much trouble I'll have getting the wife to agree to that, but we'll see.

I'm unclear on how I achieve the R-40 walls.  My current thought is to build an offset double framed wall (standard 2x4 load bearing inside, then seismic sheathing, then a 2x3 framework standing six inches off, all filled with dense pack cellulose, and dense or wet cellulose on the inside load bearing wall.  However, I'm really struggling with some of the details.  I don't think I'll be permitted to skip the vapour barrier and being nervous and conservative, I don't think I'd want to.   I'd like to put it on the outside of the sheathing so it can be vertically continuous and unpenetrated by utilities, which would remain inside.  It would be within the first third of the r-value, so should conform to the rule of thumb.  This could make for far less failure points and provide some redundancy in the air sealing.  Should be super tight and have an R-value before windows of about 48.

However, this means I would have to get the inside and outside insulation blown separately.  It also means the outside would be complicated.  Open bays, blind blowing through holes cut in the outer shell, and some structural details I haven't figured out yet.  For instance, I can't figure out how to support the outer wall at the bottom without causing a bad thermal and air bridge in way of the sill plate and first floor joists.  Also, I know just enough about dense pack cellulose to know that you can't blow into open bays and blowing it blind is asking for trouble unless you have a really good installer.  Anyone have any ideas of how that could be done?  Alternatives?  I was thinking of stringing nets between the bays and maybe netting the outside during the blow and closing afterwards, but that seems expensive, complicated, and potentially wet.  There has to be an easier way.  I've read about it being done blind and open in similar Larsen truss systems, but I can't imagine an installer around here having the skill or experience to do it reliably.

I don't know how well I've explained it, but I like my idea so far other than these problems.  It provides a thermal layer bridged only by a plywood layer at each floor, with only windows and doors as air barrier penetrations, and potentially much easier electrical installation as there's no need to seal off all the outlets and no risk to nicking the barrier.  It also uses only wood and old recycled wood (cellulose) both of which are cheap and in plentiful supply around here.

Sorry this is getting long, but one more point:  I want to be fully electric (except for one small gas fireplace, mostly for decoration) as electricity here is 80% renewable, and that percentage is likely to go up.  Increases in electricity prices are unlikely, while I don't think the same is true of gas long term.  Also, I prefer renewable energy route.

For the record, I'm aiming more for the most energy efficient I can get within rationale bounds, rather than the most bang for my buck. As you've described engineer, I'm willing to pay more capital costs to get that. 100k might be too much more, but certainly more.

Consider a hybrid insulation system.

Closed cell spray foam has an aged R-value of 6 per inch, higher than pretty much everything else (the word "aged" is important). It also stops infiltration, adds structural strength and serves as a vapor barrier. However it is really pricey compared with other materials. A more economical way to use it is to apply an inch or so to pick up the infiltration and vapor barriers and then use fiberglass or cellulose for the bulk insulation

Things we don't know off the top of our heads that impact the answers to your questions are:
Actual Manual J load
Available local/national tax credits
Design preferences (radiant vs forced air)...........

I agree that geo is not for every application.
Installers that court your job will give you operating cost calculations to help with decision.
Good Luck,
Joe

A 2x6 24" o.c. studwall with high density batts or sprayed/blown cellulose has a clear wall R of ~R20. Adding 3" of rigid iso (2 lifts of 1.5", seams taped on both layers works better for air-tightness) adds another ~ R20, for a ~R40 clear-wall. Double studwalls filled with dense-packed cellulose is another way, but it's a lot more framing labor. If you go the double-studwall route, using 2x3s 24" o.c. on the interior works, with lower thermal bridging.

You might want to cruise through these for some ideas; http://www.buildingscience.com/resources/high-r-value

Closed cell spray polyurethan foam has about the same R as rigid iso (polyisocyanurate), and makes it easier for perfect air-tighness, but unless there are lots of exterior bump outs, bay windows, nooks, & dormers etc, the installed cost of rigid iso will be lower.

Even thought the foil facers are vapor barriers, and on the "wrong" side for heating-dominated Vancouver, if the iso is more than half the total R the studs wont' have mold or rot issues. But that also means you can't use poly vapor retarders on the interior without creating a moisture-trap. The needs to be discussed in advance with both the builders & inspectors. It can be explained easily with a psychrometric chart in front of you. The dew point of 20C 30% relative humidity interior air is 2-3C. With half the R on the outside of the studs, the studs will still be above the dew point even on "design day" (the 1-3% coldest hours of the heating season), so it'll be just fine being inside the water-vapor envelope of the structure. (In Vancouver you'd be just fine with only ~30% of the R outside the studs.)

If you go with the double-studwall/cellulose you may run into issues with a poly vapor barrier on the interior, since drying times of a thick cellulose wall will be long- longer if it only dries toward the exterior. You'll only get substantial drying toward the exterior when it's above 5C. Without a vapor barrier the interior studs will be protected, since they'll stay above the dew point of the conditioned space air. The outer studs are protected by being in the exterior drying zone. In the middle there may be some wintertime accumulation of moisture, but cellulose can buffer quite a bit without damage, and will wick water away from structural elements that may penetrate. Using vapor-retardent latex primer on the interior will A: slow, but not stop the diffusion of water into the structure, but B: Still allow drying toward the interior for much of the spring/summer/fall. This approach has been used successfully in the US midwest, with a much higher moisture-accumulation potential than Vancouver. In Vancouver there are significant fraction of hours even in January where even the OUTDOORS is above the dew point of the interior air, and the assembly will be drying in both directions it isn't blocked by a true vapor barrier like poly. Even standard latex is probably vapor retardent enough (and may perform better), but vapor retardent latex might be the best compromise you can make with the code inspectors. (Interior poly vapor retarders make sense in the much colder Canadian midwest & eastern maritime, or even Kamloops/Kelowna, but are as likely create rather than cure problems in temperate Vancouver.)

Whatever the matirial stackup, DO comply with the 10mm rainscreen requirement on the siding, which will give the assembly far more drying capacity toward the exterior than it would have without it, and limit the amount of bulk-incursion from wind-driven rain.

It's probably worth springing for the Passive House tools to tweak the details of the design (you may indeed be able to do this with R60 attic, not R75, though the cost difference isn't large if blown-cellulose.) http://www.passivehouse.us/passiveHouse/DesignTools.html Download the demo version. Window placement is as important as window sizing, and you'd be able to use substantial sized windows on the south side. To use larger windows elsewhere you'd likely have to spring for very-high performance glazing to keep the heat loss low enough. Passive Houses don't tend to be dark little cubes- one aspect of being able to meet their stringent energy use standard is to make use of natural light.