So, I have spent many nights on the entire forum over the past few weeks. And I have searched. Alot of the info and threads are older. So we kinda know the high effective R value early claims are inflated with icf. I completely understand the why they cal tthem effective # since fiberglass is rarely installed correctly, and basically does Not perform to the advertised R value. Full disclosure I own a Spray Foam insulation company. So that is why I understand why company's are trying to level the playing field with"effective" numbers. Now,. My first exposure to icf is when a customer asked us to do some additional air sealing on their icf structure, we did some stick framed dormer areas. Stick framed garage walls, and along the icf tops to the vented roof deck.... I thought about how tight you could make the structure if they would have done the complete roof.... fast forward a few months. And we were on another icf build, they decided to do a encapsulated roof at the roofline, it was a lower pitch roof and it made sense to do the roofline since there was a air handler in the attic with all the ductwork , putting it in a semi conditioned space. I contacted them later and they were insanely happy with their HVAC operational cost. Normal 90%+ furnace and a/c. I do not know the size or brand... Now our current house, it is 7+ years old, I custom built it. It is 2x6 walls. 9 foot tall walls. It has a spray foam encapsulated crawlspace. Open cell spray foam on the 2x6 walls, and a open cell encapsulated attic and bonus room. It is full brick dark red in color. Has pella low e double hung windows off the shelf from a big box store. 2 direct vent gas fireplaces. Power vent gas water heater. 90% carrier furnace and a/c. House is~ 2500 sq with the bonus room we have a 1.5 ton AC for downstairs which on the hottest day last summer(well over 100 *) I could keep inside @ 69. And my a/c system kept up very nicely. Upstairs system barely ever runs. And the upstairs system never runs for heat. Our biggest gas bill was a couple of years ago, and we barely broke $100. That's for heating the house and hot water. We rarely use the direct vent fireplaces. Today I could do a few things better than I did 7+ years ago, I used a lot of lumber.... So what to do on the new house? I want crazy efficiency,with common sense in mind..... I like the strong aspect of icf, I like the forever aspect if icf. In 2 different personal homes I have never considered a addition. We completelyremodled my first home which I bought existing, then built our current home. We "think" this is our legacy home. We are gonna slightly downsize to 1500 -2000 sq ft single story. No kids in the forseeable future. The reason we want to move is we purchased 7 acres wooded with terrain. What we always wanted, a little further outside town. Without many neighbors. it seems pretty hard to compar wall structures, some seem to be opinion and not experiences. Most comparisons are to typical house construction. Saying this or that % of savings.... I obviously would spray foam any structure, if it is stick built I would at least go to 2x6 again, maybe the next step.. Foam roof or ceiling. And foam crawl. If I go with icf I would go with a foam roof or ceiling. I believe not foaming a roof on any foam structure is like leaving the lid off of your cooler on a hot day and expecting ther to be ice 2 days later, dumb. Sips.... I have remodled a ton of homes and over the lifespan of the house it seems there is always some sort of water damage.... and I personally don't like the aspect of the OSB sandwhich structure. If it gets wet the repair is compreimised at best.... I know it is probably the best off the shelf thermal performer. But I don't think it is for me... So icf vs 2x6 maybe 2x8 smart framming with spray foam any real world comparisons? I know the argument for thermal bridging. But the cement has little r value and I am confused and read conflicting info on the whole mass argument.. would you be better using the new inserts some block manufactures have? ...thanks for any input to help clear up things

If you plan a passive solar home the thermal mass can help but only if you remove the interior foam and apply it to the outside. Wish there was a way to do that. Look at Passive Haus designs and LEED Platinum Designs. There is a new Passive Haus in Cincinnati. Also in Cincinnati are these two smart framing LEED Certified homes http://www.usgbc-cincinnati.org/?mid=70&mid2=166&mid3=22/. Not sure it you like traditional design or are willing to do what it takes, this Cincinnati home is completely solar hot water heated and passive cooling http://green-cincinnati.com/category/kinsman-residence/. It uses ICF on the first floor which is also a retaining wall. Get in touch and they will give you a tour. Good luck.

ICF construction should work well in your climate.  Another way to build is to use removable metal forms to place solid, reinforced concrete walls and then spray closed cell foam on the exterior.  Extra thick insulation on the exterior is not a problem if you plan to brick veneer the home.  Just use longer brick ties.  My understanding is that the only way to end up with a uniform thickness and acceptable smoothness with closed cell spray foam is to make repeated light layers.  Installing the brick ties before spraying allows the operator to judge thickness.  Placing all of the insulation outside the structure is a great way to benefit from the mass.  I mention this way primarily because you own the spray foam company and can do the work.

Yes it is going to be brick/stone exterior. What is the performance difference in the mass system? And how do you compare it to another wall system? This may sound dumb, but since concrete is a poor insulator, would the ground temp work its way up the wall from the footers? If it had no direct solar gain /contact? Again I have looked around on the web. Seems one person says one thing another ones says something different... pretty confusing.. and I wonder who has actual experience or if it is just a theory.. Also my lot is heavily wooded so I wonder how much passive solar I can get. I usually work within given paramaters... House is already up, architect has it drawn a certain way, etc. Being able to start fresh on my own house is way harder..

I'm kinda surprised that other isn't much input on here.....

If you plan a passive solar home the thermal mass can help but only if you remove the interior foam and apply it to the outside.

I am not sure the foam needs to be moved. Thermal mass is thermal mass, whether on the inside or the outside, or in the middle. Maybe I am wrong but I looked for data to support keeping the thermal mass on the interior and I gave up because I didn't find any. Obviously if the sun shines directly on the mass it will change temperatures faster, but I have yet to find actual data to support the "mass on the inside" theory that I used to have myself. Has anyone found any studies to support this?

Wayne,
I think I understand your question, but would like to clarify. When you are talking ICF vs. 2x6 or 2x8, are you talking price or performance?

Based on my experience on my own home, the 2x6 spray foamed doesn't quite perform like the ICF wall right beside it.

While concrete doesn't have much R-value to speak of, it does offer a few things: strength, air barrier, and thermal mass.

The strength of concrete over a wood framed house is inarguable. The longevity of concrete is inarguable. The lack of rotting, warping, and mold with concrete is inarguable.These are big factors when people decide to use ICFs. Energy efficiency is just an added bonus to most people.

The fact that a solid concrete wall does not allow air to penetrate a wall is HUGE. That, coupled with the great insulating property of foam, creates a very energy efficient wall.

I do believe that thermal mass does get a little more hooplah than it really deserves, but it is quite effective! In 2001 PCA did a study comparing wood framed walls, ICFs, sandwich concrete panels, concrete walls with interior insulation, and concrete walls with exterior insulation. The result (all 49 pages) was quite interesting.

http://stuff.mit.edu/afs/athena/dept/cron/project/concrete-sustainability-hub/Literature%20Review/Building%20Energy/Concerte%20Industry%20Reports/PCA%20articles/SN2518%20-%20Energy%20Use%20of%20Single-Family%20Houses%20with%20Various%20Exterior%20Walls%20_Thermal%20Mass.pdf

Do I believe you can achieve the same energy performance with wood and foam? Sure. But how thick of a wall you would have to go, I don't know, but I would venture to say it would be well over 6". On page 32 of the PCA study, they came to this conclusion: As can be seen, the flat-panel ICF wall has a performance essentially equal to or better than the 2x12 wood frame wall with R-38 insulation in all three locations. Granted, they were comparing it to fiberglass...

I hope this helps in some small way...

Also, as a side note, there quite a few ICF homeowners in Indiana that rave about their homes. PM me if you would like to talk about it further.

The Romans started making concrete more than 2,000 years ago, I started last year.
I am happy to have ICF house. It will live more than me.

Posted By dwangle on 10 Apr 2013 01:36 PM
Wayne,
I think I understand your question, but would like to clarify. When you are talking ICF vs. 2x6 or 2x8, are you talking price or performance?

Based on my experience on my own home, the 2x6 spray foamed doesn't quite perform like the ICF wall right beside it.

While concrete doesn't have much R-value to speak of, it does offer a few things: strength, air barrier, and thermal mass.

The strength of concrete over a wood framed house is inarguable. The longevity of concrete is inarguable. The lack of rotting, warping, and mold with concrete is inarguable.These are big factors when people decide to use ICFs. Energy efficiency is just an added bonus to most people.

The fact that a solid concrete wall does not allow air to penetrate a wall is HUGE. That, coupled with the great insulating property of foam, creates a very energy efficient wall.

I do believe that thermal mass does get a little more hooplah than it really deserves, but it is quite effective! In 2001 PCA did a study comparing wood framed walls, ICFs, sandwich concrete panels, concrete walls with interior insulation, and concrete walls with exterior insulation. The result (all 49 pages) was quite interesting.

http://stuff.mit.edu/afs/athena/dept/cron/project/concrete-sustainability-hub/Literature%20Review/Building%20Energy/Concerte%20Industry%20Reports/PCA%20articles/SN2518%20-%20Energy%20Use%20of%20Single-Family%20Houses%20with%20Various%20Exterior%20Walls%20_Thermal%20Mass.pdf

Do I believe you can achieve the same energy performance with wood and foam? Sure. But how thick of a wall you would have to go, I don't know, but I would venture to say it would be well over 6". On page 32 of the PCA study, they came to this conclusion: As can be seen, the flat-panel ICF wall has a performance essentially equal to or better than the 2x12 wood frame wall with R-38 insulation in all three locations. Granted, they were comparing it to fiberglass...

I hope this helps in some small way...

Also, as a side note, there quite a few ICF homeowners in Indiana that rave about their homes. PM me if you would like to talk about it further.

The whole-wall R of a 2x6 wall with a typical 25% framing fraction is about R14 for open cell, about R16 for 5" of closed cell, so HELL no it won't perform like an R20-R22 ICF, independently of the thermal mass.  For timber frames it's always better to put the foam (especially the high-R foam) on the exterior, eg:

A 2x4 wall with open cell in the cavities comes in at roughly R10 whole-wall, but adding 2" of closed cell on the exterior brings that up to R22, and it's VERY air tight. Prior to spraying the closed cell foam, pre-installing 1x or 2x furring 2" off the sheathing using chunks of 2" XPS for spacers every 24" through-screwed to the studs by 1.5" with timber screws can support the weight of even the heaviest siding and the closed cell foam itself qualifies as a weather-resistant barrier.

And yes, an R20 ICF would outperform an ~R22 wall describe, due to the mass effects, but it's at least a more reasonable starting place for making the comparison by which you'd tease out the thermal mass component, since the steady-state U-factors would at least be comparable.  Comparing center-cavity R on a timber frame to continuous insulation R on an ICF just the ICF marketeer's straw-man. "It's the thermal bridging, stupid!"

The 2x12 with the low-density R38s pencils out to about R27 whole-wall R at a 25% framing fraction in a simple 2-D model using R1.2/inch framing timber. It's not exactly a stunner that an R20-ish continuous-insulation wall with a bit of help from mass makes it's energy use comparable and peak-loads lower. But it kinda matters that only an idiot would ever build a 12" thick wall that way.

Most folks building 12" thick timber framed walls are going either double-studwall or Larsen Truss, and use something denser &  higher R than bottom-of-the-line batts. Doing that saves some in costs by using far fewer board-feet of lumber, had a higher center cavity R, and a fraction of the thermal bridging. A typical 12"  Larsen Truss comes in at around R36 whole wall, a double studwall that thick would be comparable.  See cases 4 & 5 in table 3, page 13 (.pdf pagination), as well as the ICF cases (7a-c), all thermally modeled in THERM (a more sophisticated 3-D thermal modeling tool.)  It's unlikely that an R20-R22 ICF would reach the thermal performance of those walls on a fuel use basis in any climate, but an R28-30 ICF probably would.

Whether an equivalent-performance ICF would be cheaper or more expensive than a 12" double-stud or Larsen Truss approach is an open question with lot's of local-cost drivers (like the cost of timber framing labor & blown cellulose vs. the cost of concrete & EPS). In New England it's been fairly strongly tilted in favor of a timber framed approach, and I've yet to see a real-world example of a Net Zero Energy or PassiveHouse with more than the foundation being done in ICF. The 2x6 + 3-5" of exterior foam approach is even higher performance than the double-stud/Larsen Truss walls and is a pretty common approach for the regional Net Zero builders. The foam-overs usually cost a bit more than the double-studs, but still not as much as a high-R ICF.  (There's a guy near me building Net Zero houses for under $200/square-foot for the whole shebang, including the photovoltaics on the roof, and he's not even using ICF for foundations, though I've suggested it from a cost/speed point of view.)

The mass effect has a far more pronounced effect on mechanical system sizing than it has on energy use but at R20+ whole-wall U-factors it's really not a very big initial-cost savings for residential size buildings, and the lowest-output range of standard furnaces/boilers & central air conditioners are already defacto way oversized for true high-R houses, but mini-split air source heat pumps often fill the void.

Thanks for the link to the PCA/MIT piece- I'm saving a copy for reference!

Wow, that was quite a read! I really enjoyed it. In fact, I saved it for future reference. All I got to say is you are much more thorough than I am.

Dana, I always enjoy the info you share!

Wayne, I would recommend that you google TFSystems and consider their vertical ICF products. They have ICF products that puts all of the foam on exterior side and zero on the interior side...which is preferable in general and especially for a passive solar / thermal mass design.  I would also recommend considering using Helix in the ICF pour to eliminate the majority of rebar.

PCA did a study comparing wood framed walls, ICFs, sandwich concrete panels, concrete walls

I'd prefer this wasn't done by a biased source (the Portland Cement Association) and was an actual study, vs modeling based on the author's often poor assumptions. In any case, R38 (~R27 actual using Dana's figure) stud walls were about the same (sometimes better, sometimes worse, always close) as ICF. So how much less does a R27 stud wall (not necessarily double wall) cost?

Posted By sailawayrb on 10 Apr 2013 07:04 PM
  I would also recommend considering using Helix in the ICF pour to eliminate the majority of rebar.

And that's coming from an M.E.(;=)). I sure wish we could come to some consensus on Helix. My one P.Eng. says it a nice add on but doesn't meet any of the code requirement (Canadian) for a reduction of actual steel section in the walls. Some others agree that it doesn't meet the code but still meets the intent therefore cut the bar by as much as 50%. On the one hand the Helix would appear to hang up on the webs more than concrete without, on the other hand, without all that bar in the wall, there's less for the concrete to hang up on and the vibrator passes much easier. Any one that has good data on Helix please post.

I seriously doubt that we will come to a consensus. However, Helix now has ICC approval and is starting to get used for commercial buildings in several states. It is typically only used to replace the majority of rebar in walls and slabs, and rebar continues to get used for the lintels. I would say that the structural engineers that have taken the time to learn how to design with it are beginning to use it more frequently. Yes, it is still a relatively new product.

Posted By jonr on 10 Apr 2013 08:42 PM

PCA did a study comparing wood framed walls, ICFs, sandwich concrete panels, concrete walls

I'd prefer this wasn't done by a biased source (the Portland Cement Association) and was an actual study, vs modeling based on the author's often poor assumptions. In any case, R38 (~R27 actual using Dana's figure) stud walls were about the same (sometimes better, sometimes worse, always close) as ICF. So how much less does a R27 stud wall (not necessarily double wall) cost?

The 2x12 R38 approach to an R27 wall has to be more than a 24" o.c. 2x6 cellulose wall with 2" of exterior iso at about the same performance point, otherwise that's the way people would be building them.

The tables of U-factors used in tables 4 & 5 the PCA study are a bit of a joke too. You'd have to go all the way back to the 1950 version of the I=B=R manual (see table 1 part 2 section e. on p.6) to find U-factors that implied performance equal to than the center-cavity R-value (sheathing & siding & gypsum included.)  For a wood-clad R11 batt wall to have an average U-factor of 0.078 (~R13 average) at a 25% framing fraction wood sheathing & siding would have to be about 2.5-3" thick.  With more relistic 1/2" sheathing and ship-lap siding and half-inch sheet rock it would adds about R1.8 to the stackup, making the center-cavity R about R12.8, which is U0.78 center-cavity. 

I'm sure DOE 2.6 correctly accounts for framing fractions- mayhaps they were just mis-labeling the tables as "Actual Assembly U-factors" rather than U-factors at center cavity, but it doesn't exactly lend cred to the quality of the documentation, even if they actually got the DOE modeling inputs right. A  wood clad U0.078 wall is a 2x6 16" o.c. R19, not an R11 2x4 wall as indicated in the tables and elsewhere.

Picking a nice shoulder-season 1-2 April period for Boulder CO for the Figure 1 graph in the PCA analysis without specifying the outdoor temperatures (or the actual year dates, so that weather data could be looked up) demonstrates the mass-effects at it's most obvious, but obfuscates the actual average performance in Boulder or anywhere else. Shoulder season diurnal swings in Boulder are volatile and swing through the balance point. According to Weatherspark data 1-April 2011 saw a low of 54F in the AM followed by a high of 70F, but the next day it swung from a low 37F to a high of 79F, falling to 32F by 4pm on the third.  This is prime-season for mass-wall performance, the season where the energy use savings are maximal. But it's more interesting for some of us to look at what it does for peak heating & cooling loads on days near the 99% & 1% design conditions.

Also lost in translation is the fraction the wall structures account for in the total heat loss of the house. It's a substantial fraction, but rare is the house where it's the overwhelming dominant feature.  At R20 whole-wall the windows alone account for as much or greater heating/cooling load as the walls in most designs, even on a low-mass wall.

It's definitely a biased source, but their biases are pretty well understood, and not tough to filter. It's a far cry from some of the over-the-top marketing put out there by the ICF industry, but it's not exactly science it pretends to be either.

I agree, a cellulose filled 2x6 stud wall with 2-4" of rigid foam and well sealed air barrier(s) (preferably one on each side) is the interesting comparison for ICF.

Posted By Dana1 on 11 Apr 2013 06:20 PM
Shoulder season diurnal swings in Boulder are volatile and swing through the balance point. According to Weatherspark data 1-April 2011 saw a low of 54F in the AM followed by a high of 70F, but the next day it swung from a low 37F to a high of 79F, falling to 32F by 4pm on the third.  This is prime-season for mass-wall performance, the season where the energy use savings are maximal.

40 degree diurnal swings is the norm for Northern AZ areas. It is pretty much a daily occurrence during the summer to see temps around 90F during the day and then for it to be in the lower 50's by 5AM. Take for instance tomorrow, the high in Prescott will be 75F and the low will be 35F. The high elevation (4,500 - 6,000 feet) and dry desert climate makes these temperature swings very common.

And that's exactly why building with high thermal mass is an appropriate way to build in a moderate to high altitude dry places like northern AZ. (If it worked for the cliff dwellers of the region 1000 years ago who are we to argue?)

I once bivvied next to a stream in the Santa Catalina range near Tucson on a midsummer day when it hit the mid-90s (even at altitude), yet found skim-ice on one the shallower pools in the AM (due to the radiational cooling.) Air temps were still above freezing at dawn, but not by much. I was pretty cold, but the rocks on the less-shaded bank were still warm enough to sit on comfortably. (In retrospect I would love to have had an infra-red thermometer to compare the surface temps of the shaded vs. unshaded rocks vs. the vegetation, etc..) It got back up to at least 80F again by noon.

The mass benefit in an Indiana climate are still there, but it's NOTHING like the benefits achievable in the desert SW (especially at altitude.) Shoulder season air in IN is much more humid & heat-trapping, with correspondingly more moderate diurnal temperature swings than in N. AZ.

Posted By jonr on 11 Apr 2013 07:22 PM
I agree, a cellulose filled 2x6 stud wall with 2-4" of rigid foam and well sealed air barrier(s) (preferably one on each side) is the interesting comparison for ICF.

Having recently been involved in a retrofit on a 2x4 balloon-framed building in central MA adding ~4" of exterior iso and U0.18 windows, any difference in performance between that and any ICF version would be a "who cares?" situation.  The annual average heating & cooling bills the place is coming in well under $50/month (using mini-splits @ ~15cents/kwh). Comfort is high, and nobody inside can guesstimate what the outdoor temp is without a thermometer or looking it up on their smart-phone.  The whole-wall R came in at over R35 (low framing fraction & full-dimension 2x4s), maybe even pushing R40.  But if an R30 ICF would beat it on performance, it's not as if anybody would notice. With more favorable shading factors & roof pitches it wouldn't be too difficult to make it net-zero.  An R20-22 ICF couldn't touch it on performance in this climate, but an R30+ ICF might.

I suspect a 3-story + full basement even an R20 ICF wouldn't come close to touching it on construction cost at local material & labor rates, even if they had started the timber-frame from scratch rather than re-working the 1890s antique.

The mass benefit in an Indiana climate are still there, but it's NOTHING like the benefits achievable in the desert SW (especially at altitude.)

Looks like PCA report has somewhat similar data with wood frame vs wood frame + external CMU results. Albuquerque (7% savings) and Springfield (5% savings). Sure, different cities and internal mass will change these numbers, but for almost every place in the US, I'd probably just add more insulation to a stud wall and end up with lower cost and less energy use.