MY wife and I have 3.2 acres out in the country ready for us to build on with well in place and septic to come when the house gets started. We have our plan all picked out. The home is a single story with full basement ranch style house with hipped roof. I am pretty much dead set on going geothermal. But we can't figure out what building method we want to use. Our first quote was OC 16'' 2x4 construction with 2.5'' of closed cell insulation. Then we got a quote from a builder who sends the framing off to a manufacturer and is framed in about 3 days but this construction was 2x6 16'' oc. Now we have had some family friends insist that we go with ICFs. We have not got a quote for ICFs but will in the next week or so.

My questions are what is going to be the most efficient home? Should we really be looking at the ICFs?

What about SIPs? I have heard of them but don't know if it would be better than the other two?


I just want the most efficient home I can afford while keeping the plan we currently have because we like it soooo much!

Should I skip the geothermal and go with 95% efficient gas furnace?

WHATS WORTH THE MONEY!?!?!?

before we rip apart your plans :-) tell us where are you located and how big is the house. do you have acces to a lake or do you plan ground loops for the geo. maybe few words about the orientation of the house

adi

WHATS WORTH THE MONEY!?!?!?

I think that is different for every person. ICF was worth it to me for the security, durability, insulation, sound resistance and tightness. Hard to say what that is worth to someone else.

Maybe you could start by letting everyone know what kind of a climate you live in and what your heating and cooling preferences are.

The most efficient home is the tightest & most highly insulated, with the lowest amount of exterior glass. It can be built any number of ways.

2x4 o.c. stud walls with 2.5" of closed cell in the cavity bays is a poor value. The air-sealing isn't guaranteed, and thermal bridging of the studs/plates/headers etc. give it a whole-wall R of only ~ R11, despite center-cavity R of over R15. For about the same cost you could get ~ R23-24 (twice the whole wall R value ) and better air sealing by using wet-spray cellulose in the stud bays, and only 2" of closed cell (or rigid board foam of equivalent-R) applied to the exterior of the sheathing.

Going to 2x6" framing and cellulose fill (no exterior foam) or batts (perfectly installed) you'd hit ~ R14. With an inch of exterior closed cell or (1" rigid iso) that would bump up to R20. With 2" you'd be in R25 territory.

ICFs start at ~ R16 (whole-wall), and bumping it to R20-22 isn't a huge cost adder but if you're looking for high-R (R30+) there's cheaper R to be had. The local cost of the concrete can be the make it or break it for some. As much as the marketers make of the dynamic effects of the thermal mass of concrete, in most places it only amounts to a single-digit percentage reduction in heating & cooling energy use, something more cheaply achieved by diverting some of the difference in concrete vs. stick built to higher-R. The better reason for spending the money on concrete houses are structural, not thermal. In hurricane prone areas that can be a significant & very real long-term factor. They're nice and quiet too (less of an issue in a not-so-windy rural area than next to a highway or in the city or on a windblown ridge top.) While I like concrete construction, you can't usually make the cost-case on thermal performance alone. For insulating poured concrete basements to R16-22 it's cost-competitive with other solutions though.

SIPs start in a similar R-range, and while R30+ SIPs are available, they're not necessarily cost-competitive with stick built. Like ICFs, although air-sealing is easier to achieve, it's not guaranteed, and attention MUST be paid to keeping air-infiltration as low as possible, or the money paid for the insulation package is largely wasted. (A well insulated wind-tunnel is unaffordable to heat.)

Geo vs. condensing gas vs. air-source heat pump depend a lot on your design-condition heating & cooling loads, the climate, and the site-particulars. But in general, that first ton of geo is expensive, and before spending an "extra" $20K on the mechanicals, do at least the napkin-math on what that would buy you in terms of a tighter house with higher R values. In US climate zone 5 or lower, for modest to mid-sized houses there's usually a case for spending the efficiency money on the building, and getting the heat loads down to where R410A refrigerant split-system air source heat pumps can handle both heating & cooling rather than geo or condensing gas + AC.

Less glass, tighter windows and U-values under 0.35 make a real difference when the whole-wall R is north of R20 (or R30). At R30, every square foot of U0.33 glass leaks as much heat as 10 square feet of glass. The Energy Star HERS index for homes uses a glazed fraction of 18% of conditioned floor area as the "typical" paradigm house for comparison purposes. It's not tough to peel that back to under 15% (or even 12%) without living in the dark, and the effect on the thermal performance of even a moderate-R20 house is significant. Fixed windows leak less air than windows that open. Casements & awning windows leak less air than sliders and double-hungs, and provide more ventilation & egress area per square foot of glazing too. Think about EVERY window, and down-size where you can. It's far cheaper to go with smaller/simpler windows of moderate performance than it is to make those same gains with ever lower U-values (triple-glazed, etc.) or bumping the walls to R40.

Doors, same thing- use only insulated doors (R4-5), don't use sliders- they all leak considerable air, and are hard to retrofit weather stripping after the fact.

Consider building to the whole-wall R-values for your climate zone found in table 0.2 p10 of this document:

http://www.buildingscience.com/documents/reports/rr-1005-building-america-high-r-value-high-performance-residential-buildings-all-climate-zones

(Pay attention to foundation R too- even center-slab.)

Home will be in East Central Illinois... we saw negatives quite a bit last winter... and about 2 foot of snow in one storm

I want to use geothermal closed horizontal loop... i got 3 acres so I got the room... no access to lake or pond

House is a single story with full basement 1640 sq feet upstairs only... basement will be finished at some point

The house will be protected on the north and west side of the property with an existing wind break. The house's front door will be looking to the Southeast. Basic length and width of house is 52'x54'... gas fireplace upstairs and basement... 9ft basement walls but 8ft upstairs with a vaulted ceiling in the Great Room.

You say not to use sliding doors.... what doors would you have to go from the kitchen/dining room to the deck outside? French doors? I am trying to understand which one would be more efficient

Double swinging patio-doors/French-doors are easier to get a good seal on initially and maintain over time. Wood sliders have a tendency to bow and irrevocably lose seal over time. But look carefully at the U-values of doors, as well as their size. A dual 36" door may look great, and you can roll your 7' Steinway out onto the deck for an evening concert, but even a dual 24" French door gives you more access than an 8' slider, at half the area, and better prospects for air-tightness. A single door is even tighter, but may not be what you're looking for.

Using Springfield ILas the climate model, your average January temps are ~25F, a temperature at which split R410A air source systems are hitting a total coefficient of performance approaching 3.0, which is about a best-case scenario for geothermal after pumping power is factored in. With R25-R30 walls, R60-65 roof, and R16-R20 foundation walls, in a house that sized it's pretty easy to pull the +0F heat load into the range of a single mini-split. The 97.5th percentile design temp for Springfield is +2F. You may need supplemental heat during the true cold snaps if you under-rate it but if your +2F design temp is under 20K, with a 30-36KBTU heating-rated Mitsubishi Hyper Heat would still be good down to -10F, and should run half the cost of a geo system, delivering similar annual performance. But it's likely that you can get the 0F load down to 15K if you design & build carefully, in which case an 18K unit would handle it. (I'm more familiar with these than their competition, since they were used on a recent project in central MA. But there IS competition- it's important to pick a vendor with good local support. If you limit yourself to models with an HSPF 10 or better you can thin them out pretty fast.)

The installed cost of an 18K Hyper Heat is under $5K. (Quotes for 3 identical 18K units installed in a 3-family came in under $14K.)

Horizontal loop or no, I doubt you'd get a 1.5-2.5 ton (or larger) geo system in for anything like the installed price of a mini-split, and the efficiency delta would be marginal, if any.

Gas fireplaces are tough to make air tight, and unless they're used, they're nothing but a 24/365 energy-liability. And if the fuel is propane, not natural gas, your cost per BTU is high and going ever higher even if you DO use them.


If you model the house as you tweak the design with a decent downloadable freebie like DOE2 or HOT2000 you really CAN get there from here.

Something else to consider in new construction is making the roof lines simple, easy to air-seal (minimize dormers), and designing the orientation & pitch to optimize solar uptake for photovoltatic panels. Even if it's not something you'd do yourself, there are companies offering preferred power rates & lease deals to homeowners to be allowed to use the roof area. PV costs have been crashing for the past decade, and at some point in the next 25 years the economics may be compelling even for homeowners.

You should do some thinking about how long you are going to stay there. Look at insulating shutters before you go too far on super insulation and small windows. Geo is noticeably more efficient than air source heat pumps. Nat gas looks better if you won't be there all that long.

For geo to be "...noticeabley more efficient than..." a decent R410A minisplit when pumping power is factored in requires a lot of diligence in the system design and implementation, (a diligence alas not reflected in the industry average performance) and/or a very low-temp radiant solution (which is also a cost-adder.) With geo-to-air, the difference in performance of geo with low-mid-50s soil temps compared to mini-splits with ~ 25F mean January temps is really "in the noise", but the upfront cost difference is usually quite large. (Where's that nice big deep pond when you need one?! :-) )

Yes, geo beats air source reliably when average air temps are below 10F, and can operate just fine at extremely low outdoor temps since they're not dependent on outdoor air temps the way air source is. But in bang/buck terms, in a climate where the January mean temp is in the mid-20s the overall heating bill is better offset with better insulation plus ASHP than by pouring it all into geo. Even with a best-in-class geo design the reduction in heat load achievable in the building envelope with the same money is far larger than can be made up with a difference in system efficiency, even if you're assuming the ASHP season average COP is 2.0 (which is about where they run at +5F) and the geo is 4.0 (haven't seen one yet, marketing nothwithstanding) rather than mid-2s for either (which is typical, in my neighborhood.) Mind you, my subsoil temps are about 5F cooler than in central IL, but for a horizontal coil application I don't think it that's going to make a dramatic difference in geo performance in IL as compared to MA- the whole system design is a much bigger factor.

To get dramatically better efficiency a mini-split the output of the geo system would have to be a low-temperature radiant floor with very low water temp requirements like 90F (which would also require a high-R structure) , but that too adds expense to the system, and would have to be cost/performance compared against R410A air-to-water heat pumps such as the Daikin Altherma (which also does pretty well with 90F water output.)

Geo has to only hit a COP of 2.8 (WITHOUT pumping power included) to earn an Energy Star rating. Most R410A mini-splits will meet or exceed that at mid-compressor speed @ 25F, and that includes all power. The up-charge for the extra tonnage of geo to be able to provide design condition heat is also significant and results in performance loss at part load due to cycling, so they usually have resistance strips as auxilliary heat, which KILLS the low-temp COP (since the resistance heat fraction has a COP of only 1.0.), whereas the up-charge for another ton of air source isn't usually very big, and with continuously-variable inverter drive compressors the hit in seasonal COP from even 2x oversizing is zero (or even negative, since the compressor never has to run at full speed, where it's COP is lowest.) Designing a geo system to TRULY hit the numbers at any price or load is not trivial. Split system ASHP systems are cookie-cutter designs, the engineering is mostly done, and the risk is low, and up-sizing it a bit has little down side to either system performance or upfront cost.

"out in the country" is usually not consistent with "on the gas grid", but if gas service really IS available condensing gas may be a comparably cheap way to heat, but it still woudn't provide the air-conditioning that a mini-split or geo would. If you added up the cost of separate systems for condensing gas + AC it'll likely be more than a mini-split, but well-under that of geo. From a carbon-footprint point of view, condensing gas at 95% is lower-carb than heat pumps @ COPs of 2.5-3.5 given that IL power is ~48% coal -fired @ ~30% thermal efficiency (but hopefully that will improve over the next decade), and coal has nearly 2x the carbon per source-fuel BTU as gas. In IL you'd need COPs in excess of 5 to come close to condensing gas on carbon footprint with the existing grid-sources, but since the bulk of the heat load is at night when grid load is lowest, at least some of that power use is carbon-neutral, since it is power would have been otherwise spent heat-dumping at the nukes & big fossil plants to avoid having to shut down.

One thing to note too is that if you're putting in a gas fireplace with city connected gas lines, mainly for the ambiance, you're going to be paying monthly service charges just for being hooked up, even if you don't use.
Geo's a waste unless you're in a very cold climate, or you build a mansion... Otherwise, as Dana points out, an ASHP will be far better... 30-35k vs 5k... That's simple math in this case.
Make sure you get a nice HRV/ERV (ultimate air/renewaire/etc), and if you're interested, stick in a small ground tempering loop for pre-heat/pre-cooling of the air exchange.
At the end of the day though, 2x4 walls just aren't good enough. 2x4 walls with some exterior foam... now you're talking. Don't just think in terms of fuel costs. Comfort is a huge aspect. Warm walls & windows are a pleasure to be around.

LP is located on the property already... there will be a fireplace on the first floor and also in the basement...


some of your responses are very detailed!! wow!! you talked way over my head there!!



so skip the geo and go with what?

Posted By petrey10 on 05 Jul 2011 02:05 PM
LP is located on the property already... there will be a fireplace on the first floor and also in the basement...


some of your responses are very detailed!! wow!! you talked way over my head there!!



so skip the geo and go with what?

For the heat delivered, LP is about 3-4x the cost per BTU of running a "mini-split" air-source heat pump using R410A refrigerant in central IL, and comparable in efficiency to geo at a fraction of the installed cost.  But to be able to use it as your primary heat source it's best to have high-R walls/low-U windows & doors, and a somewhat open floor plan.  While there are split-systems that can use as many as 8 indoor heads per compressor,  at $1500/pop  they add up if you micro-zone the place to death.  Ducted air source heat pumps can make zoning less costly, but with ducts comes a hit in system efficiency.

A 1-2 interior-unit split system (one interior unit per floor/zone) can handle quite a few layouts in a high-R house though, since if you the rooms without the interior units to be very low loss/gain, the temperature difference between those rooms and the rest of the house is quite low.   (This works much better at R30+ than it does at R10.) 

Mini-splits are very quiet compared to traditional reciprocating compressors found in most central AC systems too.  A bit of marketing hype vendors to give you an idea:

http://www.youtube.com/watch?v=9knXqO1QnME

http://www.youtube.com/watch?v=_6mDAfWQm1Y

http://www.youtube.com/watch?v=ygUDEPfUpnk


(If you can't stand the look of the interior units, they're easy to hide in valances, etc, and in high-R house it hardly matters where in the space they're located.)

Amateur video of just how quiet the compressors are:

http://www.youtube.com/watch?v=TJ2GAT6GiSU

http://www.youtube.com/watch?v=X4Acny261U8

A 1-head 24000BTU/hr unit will run ~$5KUSD installed, one with 2 interior units would run ~$6-7K.  When you get to 30,000 BTU+ and 4+ heads it starts to add up, which is why spending the money on more insulation and smart thinking about windows & doors pays off, both in efficiency and up front cost.  It really IS possble to get your heat load at +6F to under 24000BTU if you pay attention to the design & construction, and the payoff in both comfort and mechanical systems costs (compared to geo) is there. 

The max heating BTU rating at +5F  & efficiency (HSPF) is more important in your application than the cooling season BTUs & efficiency (SEER).  Some vendors only advertise the 45-47F heating performance which can be twice what it'll actually deliver at 5F.  You sometimes have to dig to find the charts.  Mitsubishi usually posts the 4F heating number as the heating BTU output, but give 17F & 47F numbers as well. (Their "Hyper Heat" units also deliver 70%+ of full rating even when it's -13F.) But the Fujitsu Halcyon or the Daikin Quaternity have comparable performance when comparing apples-to-apples at 47F or any other arbitrary point on the curve.  Key is to have good local contractors and distributor support (and this goes for ANY heating system, particularly geo.)  This is a competitive market, with lots of options, even if you haven't seen them used in heating applications near you (yet).

It may be useful to have the LP fireplaces as backup heat should the heat pump fail and you need to wait for parts, but open flues become 24/365 infiltration-drivers, and in a very tight house an open-style fireplace may be prone to backdrafting when kitchen/bath/clothes dryer exhaust is running.  Air-tight units would  be preferred, sealed combustion (taking the combustion air from outdoors) even better.

the fireplaces will be direct vent fireplaces... those are the ones that use outside air for combustion correct?

That's what they usually mean by "direct vent".

my wife wouldn't go for anything that she had to see that she didn't put out... so if the mini split is like that its prolly a no go... also the quote for the geo is a hair below 11,000 completely installed after rebates and tax incentives.

Which insulation would you get for a 2x6 built home. Wet cellulose? Plain ole batts with high R value? Closed cell? Open cell?

The house will be wrapped and a vapor barrier installed.

thanks for the quick responses!! We are getting to crunch time and need to start making decisions

Posted By petrey10 on 05 Jul 2011 04:34 PM
my wife wouldn't go for anything that she had to see that she didn't put out... so if the mini split is like that its prolly a no go... also the quote for the geo is a hair below 11,000 completely installed after rebates and tax incentives.

Which insulation would you get for a 2x6 built home. Wet cellulose? Plain ole batts with high R value? Closed cell? Open cell?

The house will be wrapped and a vapor barrier installed.

I have no clue how anybody could quote a geo system without knowing the heat load first, but those have to be some pretty serious incentives to get even a 2-ton system down to that kind of money.  If there are similar incentives for high-R buildings that would still be a better way to go.  (It's pretty easy to hide the mini-split units from view, but I understand what it's like to have to explain it to "the boss" ... )

It almost doesn't matter what you stuff in between the studs if it's installed perfectly. Even with 5" closed cell foam you wouldn't hit hit better than ~R16 for a whole-wall R after the thermal bridging of the studs is factored in and with crummy R19 batts it would be ~R14.  With spray foam it sort-of air seals the wall, but cavity fill only addresses the tip of the air-leakage iceberg. Open cell cavity fill also only delivers  ~ R14 for a whole-wall R, as does wet spray cellulose.

The best value is usually wet-spray cellulose, since it's cheap, fills all voids, and has only a low premium (if any) over batts.  Blown/sprayed fiberglass "blown in blanket" systems such as JM Spider or Certainteed Optima have a slight performance edge over plain old cellulose if done to 1.8lbs density, but it's usually more expensive than "dense packed" 3.5lbs cellulose applied simliarly.  The difference in whole-wall performance between the best and the worst blown/sprayed fiber is ~ R1 or less, ergo, open-stud wet-sprayed cellulose usually wins the value argument- spend the savings on air-sealing or exterior foam.

Batts basically suck-  but if you MUST go with batts look at the density, not the R value. Higher-density trumps higher R, since it loses less R value at the temperature extremes than low density goods.  Go with "cathedral ceiling" batts appropriate for the stud spacing- they're  always the higher-density.  (R22 batts are almost always the lowest density, and the worst value- even when installed to the manufacturers' instructions the center-cavity R is only R19, not R22- it's only R22 at higher-than-stud-depth loft in an ASTM C 518 test jig with perfect air barriers on both sides.  Calling it R22 borders on fraud, IMHO.)

Even low-density cellulose fills better than even the best batt-jobs, and has less internal convection and it performs well across all temperatures.  By contrast batts are NEVER perfect, and always underperform spec in anything but a laboratory test.

Foam on the exterior is where the real value proposition is, and where to consider spending some money since:

A: It puts a thermal break over the ~R5 represented by the framing, including the plates, sills & band joists

B: It raises the average winter temperature of the sheathing, making it less susceptible to interior moisture, and in your climate you can & should skip the interior side vapor barrier

C: can be detailed as a primary air-barrier. (especially closed cell spray foam, which is almost inherently an air barrier when applied to the exterior)

An inch of exterior foil faced iso brings the whole wall R up to R20 from R14. Two inches brings it up to R27- nearly twice that of a bare-bones 2x6 wall. Even with geo it's worthwhile to go for at least R25, and the difference in comfort will be something you can FEEL.  You can get there cheaper with EPS (bead board), but it takes 50% more thickness to the foam to achieve the same R with iso or closed-cell foam.  Closed cell foam runs about a buck a square foot per inch of depth, installed price.  Iso is cheaper in material cost but requires more attention to air-sealing details, and can sometimes approach closed-cell foam in installed cost if there is a lot of cutting & cobbling to do, after window-wastage is factored in.

Housewraps are not air barriers unless detailed as such and tested. It's easier/better to use the sheathing & framing as a primary air-barrier, caulking the stud plates to the floors/sills and caulking the sheathing to the framing as it goes up.  An overshot of closed cell on the exterior is sufficient to seal any further seam leakage. If foil-faced rigid board is used 2" FSK tape over the seams makes a good seal, with 1-part foam to seal the edges.  Air sealing is by far the cheapest/best efficiency improvement you can ever make in therms bang per buck- don't skimp.

Interior vapor barriers are best avoided if you can build the assembly to be resilient to interior moisture drives.  It's only on the "right" side of the assembly for a few months out of the year, and moisture drives come from all sides.  Exterior foam is a sure-fire way to achieve that resilience, since it puts all of the structural wood closer to the temperature of the conditioned-space interior, and is resisitant to summertime moisture drives.  Depending on exactly where you are in central IL, you may be able to get away with as little as R4, with 2x6 construction, but R7.5+ is good for the entire state, at which point the only vapor retarder you'd need on the interior is standard latex paint.   That way the drying capacity of the wall is enhanced- it dries easily toward the interior without putting the sheathing at mold risk from wintertime moisture.  The more exterior foam there is, the less vapor retardent the interior needs to be.  See:  http://www.buildingscience.com/docu...quirements  (You're on the edge of zones 4 & 5.)

Foundation walls (even stem-walls for slab-on-grade) are easily and cheaply insulated using insulated concrete forms.  Design your exterior sheathing foam to be co-planer with the exterior foundation foam for the least amount of thermal bridging.  Slabs should be insulated from below with EPS and floated, not in direct contact with the footing or foundation wall.  R16 is the minimum you can get in insulated concrete forms, and R20 isn't much of a cost-adder.  Either is a huge improvement over uninsulated foundation and has long term payback in your climate.  Under-slab  EPS is ~ 10cents/R/square foot, and 2" (R8) center-slab has a long-term payback at your subsoil temps, even with geo or mini-splits as the heat source.  (The improvement in bare-foot comfort is also pretty good.)  This is true for both slab on grade and for basement slabs, but for slab on grade bumping it up to R10 at the outer 2' perimeter and slab-edge is worth it from a long-term payback point of view.

man... im so confused...

the geo was a 3 ton horizontal closed loop system

How did they determine that 3 tons was needed, without knowing the R & U values of the house?

Did you give them a floor plan with window & door details from which they presumed a bunch of stuff (such as code-minimum on all R & U values, and air leakage at ~10 ACH/50) ?

You started the thread wondering about SIPs & ICF, etc, which says to me the R/U values (essential for calculating the heat load) are unknowns, only the legal lower-limit being in play.

Shrinking the windows and making it air tight (at least under 3ACH50) is pretty cheap & easy too, and it makes a real difference in the heat load.

It would be almost a crime to build a new house at less than R20 whole wall R values (2x6, cellulose fill, ~1" of exterior rigid board) in that climate. R25-R30 would be even better, and more comfortable, but the up-charge for going R30+ might only work financially using more aggressive assumptions for discount rates and energy inflation in a net-present-value financial analysis. Spending the money on a higher-performance building envelope results in higher comfort, and is more reliable than high-efficiency heating systems. If a 3 ton geo system would heat a code-min house with your layout, a 1.5-2 ton system would if you bumped it up to an air-tight R25 and cut back on glazed area by 20%.

Where/which state-zone are you on this map- B,C, or D? http://reca-codes.org/pages/codes/IL-IECC.pdf

Note R18 walls & R38 ceiling in the prescriptive list is the center-cavity installed value of an R19 batt or wet-sprayed cellulose. The "whole wall" R value after thermal bridging is include is about R14 for that type of wall, and the ceiling comes in at ~R30. I usually only talk in whole-wall R values, since center-cavity R is (nearly) meaningless, but code prescriptions are usually specified center-cavity.

Slab on grade, full basement, or crawlspace?