It has been done, in fact, Lbear, and several times. Trouble is, the ICF crowd doesn't like the results. The whole mass-effect-claim issue was resolved many years ago for every wall system except ICF. Again, ICF manufacturers could have defensible effective R value numbers in a mater of months. Same problem: They aren't that impressive.

I know you're not looking for school data but you might take a hard look at the K-12 50% solution document I linked to which specifically mentions ICF. You can also gain significant insight from the 50% solution for commerical buidings (ICF is not specifically mentioned there -- but mass walls are).

These documents are worth the read because this is where your code is going in the next few years -- e.g. IECC 2012 will be appoximately 30% tougher than IECC 2009 (which is now in effect in many or most states).

The reason this document is important to us ICF folks is that it's not from the ICF manufacturers -- it's from the guys that actually control how you and I build:

American Society of Heating, Refrigeration and A/C Engineers - ASHRAE
The American Institute of Architects
U.S. Green Building Council
U.S. Department of Energy

What you may be missing here is that I don't need to validate myself to you (or continue explaining that old mass wall report) if these guys are holding ICF to a new level and performance standard that is going to get a lot tougher. Regards.

ASHRAE’s Advanced Energy Design Guide for Small to Medium Office Buildings (50% Energy Savings) Revised September 2011 also references ICF construction and features an ICF building in its case studies – the offices of CMTA Consulting Engineers in Louisville, KY.

CMTA is an engineering firm that specializes in energy efficient and net zero energy design. They have been instrumental in Kentucky’s net zero K-12 initiative. Perhaps a call to them might yield some insights for the doubters. 502.326.3085.

It isn't ASHRAE describing ICF's mass effect as cavelike. It's you, TexasICF. So let's see some documentation that the FTC would bless. At this point, I'll accept ANY form manufacturer's effective R value report.

No you don't have justify yourself to me. Even so, you might pause to consider whether readers of this thread have recognized by now, to borrow a Sherlock Holmes theme, why the dog hasn't barked.

Posted By toddm on 24 Jan 2012 11:52 AM
It isn't ASHRAE describing ICF's mass effect as cavelike. It's you, TexasICF. So let's see some documentation that the FTC would bless. At this point, I'll accept ANY form manufacturer's effective R value report.

No you don't have justify yourself to me. Even so, you might pause to consider whether readers of this thread have recognized by now, to borrow a Sherlock Holmes theme, why the dog hasn't barked.

You're picking on his semantics?

Todd, You do make it tough to focus on what matters sometimes. I don't really care what the FTC thinks about "cave like". I suppose I could just refer to ICF as "better" than all other wall systems. Nevertheless, if you don't understand the term then you haven't been in an ICF home or saferoom (or a cave) -- either or. It's actually quite descriptive because even in a building with no roof the walls will take on the average temperature over several days and be very slow to change. And yes, you can actually feel the difference. BTW, cold does not exist -- only absence of heat -- but I don't think we should go there.

Furthermore, the walls change even slower still because of the insulation. And yes, they change extremely slowly and behave with a greater lag than some other mass walls.

At any rate, I can't make you read the ASHRAE report any more than I was able to convince you a year or two ago to read the small print on the ORNL report you are so fond of. Regards.

NO effective R value documentation from forms manufacturers? None?

It is not semantics, Bruce Polycrete, to ask TexasICF to back up his claims with the research that every other mass wall system has done to the FTC's satisfaction.

Toddm, it sounded like you were upset with his use of the term, "cave like," and it's not clear why you're concerned with the FTC. "Cave like" is a subjective term, if Tex likes it, he can use it. It would seem that this analysis is not of critical importance. The truth is that 2-1/2" of Type II EPS is what it is. ICF manufacturers spend significant amounts of money for required testing on a regular basis. To spend another $80,000 or $100,000 on a test that will benefit all their competitors does not seem like a prudent way to deploy marketing dollars.

Posted By toddm on 24 Jan 2012 09:15 AM
It has been done, in fact, Lbear, and several times. Trouble is, the ICF crowd doesn't like the results. The whole mass-effect-claim issue was resolved many years ago for every wall system except ICF. Again, ICF manufacturers could have defensible effective R value numbers in a mater of months. Same problem: They aren't that impressive.

What were the results? The CliffsNote version?

Lbear, I linked to the ornl study above. Briefly,ornl and icfa built icf and stickbuilt houses in knoxville in 2000 and measured their performance over 11 months. Occupied, the icf house used about 9 percent less energy. Unoccupied, the difference was 7 percent, while DOE2 said 6.8 percent. Climate has a bearing on mass effect, so your results will vary.

Ray Gladstone, form companies are free to ignore mass effect documentation if they want, but that would require them to ignore it. There is a reason the FTC polices insulation claims. To ignore it officially while TexasICF blows smoke on these forums is duplicitous at best.

Posted By toddm on 25 Jan 2012 08:43 AM
Lbear, I linked to the ornl study above. Briefly,ornl and icfa built icf and stickbuilt houses in knoxville in 2000 and measured their performance over 11 months. Occupied, the icf house used about 9 percent less energy. Unoccupied, the difference was 7 percent, while DOE2 said 6.8 percent. Climate has a bearing on mass effect, so your results will vary.

When I look at the ornl document, I see that this is not based on raw data. The authors have made a number of corrections. Also he crawlspaces are not treated the same. Further the occupied and different climatic conditions are simply modelled not actual.



When we look at the the air infiltration rates we see a large difference between the tests. Almost double the rate in test 4 for icf from March to June. The author assume that this is because of winter conditioning of the truss material. One would have to assume that the rate would double again compared to January. Why did they pick these three months to do the test? Where is the results for the other months? If their assumptions are right then if we were to model the differences to north of the 49th were summer temps and moisture can be close to Knoxville but winter scores a much greater, I would think that a house up here would turn into a wind tunnel.

Then the Insulating Concrete Forms Association employees and designates who did this study with ORNL should give their salaries back, eh?

The DOE2 modeling for the twin houses was in fact validated by the actual measured performance of those houses, with all of the infiltration factors and other known differences duly noted (and modeled).

The measured energy use of the ICF house fell within 1% of what was predicted by modeling. (DOE2 must be pretty lousy at modeling ICF and thermal mass issues to have missed by that much, eh? :-) ) See table 3, p7:

http://www.ornl.gov/~webworks/cppr/y2002/pres/114086.pdf

The DOE2 model proved less accurate on the stick-built, which isn't a surprise given that it was batt-insulated not blown (or sprayed), but was still hit within 5% on the cooling energy, and within 0.5% on heating energy. Yup, as sucky model indeed, not to be trusted...

The ICF house also had a raw whole-wall-R of R15, compared to the stick built (straw man?) 2x4 house's ~R10.6. (This was presumably true for all of the other-climate modeled results.) What, and R15 wall outperforms an R11 wall? Go figure!?!

Yes, the mass was a factor (and duly modeled), but so was the additional R (also duly modeled), and in none of the simulations did the modeled difference end up in double digits. (But it might have had more variation in real houses, if the builders were sloppy on air sealing, and the insulation installations 3rd rate as happens.)

Yes,the mass had an effect beyond the ~40% higher wall-R (which is real performance upgrade even on a low-mass wall) but the modeled and actual performance delta between the real physical houses was still in single digit percentages. Call the whole experiment crap if you like, but it's way more representative than marketing BS. It wasn't designed to do-in or underplay the value of building with ICF (quite the contrary.)

Bottom line, there's more to building an efficient house than a high wall-R or thermal mass.

Posted By toddm on 25 Jan 2012 08:43 AM
Lbear, I linked to the ornl study above. Briefly,ornl and icfa built icf and stickbuilt houses in knoxville in 2000 and measured their performance over 11 months. Occupied, the icf house used about 9 percent less energy. Unoccupied, the difference was 7 percent, while DOE2 said 6.8 percent. Climate has a bearing on mass effect, so your results will vary.

Would the mass effect play a bigger role in a climate that has 30+ degree temperature swings within a 24hour period? If so, what would that effect be?

LBear, mass works best in dry climates with extreme swings in daily temperatures, such as the desert Southwest. Humidity matters because it can make you uncomfortable even if the mass effect keeps air temps at otherwise reasonable levels. IIRC, ORNL said that interior insulation improved results in Miami -- in the only exception to best- practice mass exposed to the interior.

You wouldn't want to pay much extra for mass in Pa, where I am building, but my house is full-out passive solar and needs mass to buffer periods of extreme insolation.

Posted By Lbear on 25 Jan 2012 04:35 PM
Would the mass effect play a bigger role in a climate that has 30+ degree temperature swings within a 24hour period? If so, what would that effect be?

Maybe this ridiculous example will help you understand the impact of mass in the wall.

Think of two boxes (houses) with no windows, one made with a 1/4" thickness of paper, the other with a 10 ft thick wall of rock. Now think about the desired inside temperature being at 70° with the outside temperature swing going from 40° to 100°. For simplicity let's assume the temp swing is a perfect sine curve.

It should be fairly easy to comprehend that with the paper house when the outside temp drops below 70 you will have to add heat, and when it rises above 70 you will have to cool the house. Without heating or cooling the inside temperature will track the outdoor temp very closely because heat moves through the thin paper wall very quickly. Now, if the temperature outside swings 0 and 60 you will always have to add heat to keep the house at 70. But you will add less heat at an OT of 60 than at 0, so the demand on your heating system will be fluctuating up and down. But the total heat load will be near what it would be if the outside temp stayed at the average temp of 30°.

Now, let's add insulation to the wall and you will decrease how much heat moves back and forth through the wall, decreasing both the total heat load for the house and the fluctuation of the heat load. Carrying this thought to an extreme let's add 10 ft of foam insulation to the wall. With the outside temp swinging from 40 to 100 you would probably not have to add heat to the house. In fact, you would have to cool it because your body heat will add heat faster than it can travel through 10 ft of insulation. Now, if the OT is swinging from 0 to 60 and you do not add heat to the house, it eventually will cool down to around 30°, the average outside temp, and stay there.

Now let's consider the 10 ft thick rock wall house. The rock will soak, or absorb, heat just like a sponge soaks up water, whether on the inside or outside. If the OT is swinging 40 to 100 heat will move into the wall at temps above 70, and move out at temps below 70. The heat that moves in and out will probably not move more than 2 or 3 feet at most into the rock. That means that over time the rest of the rock will sit at 70, and the house neither gains heat nor loses heat. That is, if it sits empty and the the 40 to 70 temp swing lasts for many, many, months. Now think about the 0 to 60 temp swing. If that is the case for many, many, months, and again the house is empty, it will eventually settle at about 30°, the average of the outdoor temp swing. Now consider the real temp swing over an entire year. The total annual heating or cooling load inside the house will be the net amount of heat that moves all the way through the wall. During the summer heat will move toward the inside, and during the winter move out toward the outside.

You can probably intuitively see that 10 ft of insulation and 10 ft of rock both moderate heat movement, one by resistance and the other by absorption. Back to your original question about a 30° temp swing. What matters more than the extent of the swing is the average temp of the swing and the total amount of the mass and the total amount of insulation. The closer the average stays near the indoor temperature, the less the average amount of heat transfer there is, adding together the heat moving in and the heat moving out. The absolute net amount of heat moving in or out at any moment depends on the total mass and the total insulation. Think about how a 10 ft wall on the seashore holds back water. If it's all a sponge the water movement is moderated by the soaking up and releasing of the water as the waves hit it. It it's all a tight fiberglass mesh the movement is moderated by the resistance to the flow of the water. The fiberglass doesn't hold water, but the sponge does.

And exactly what is the impact of the mass and insulation combination? That my friend is what everyone is debating in this thread! There ain't no simple answer!

Usual heating/cooling load calculations for frame homes take into account only the coldest and hottest temperatures for the climate the house is in. That's because the rate of heat movement from the exterior to interior, and vice versa, is quite fast, on the order of tens of minutes. With 6" ICF walls the heat movement can typically be on the order of 12 hours from one side of the concrete to the other, but the net heat flow still depends on the average indoor vs. outdoor temp. My 2000 sq ft ICF house loses heat in the winter just like every other house does. It loses it constantly (except when the sun shines through the front windows) but at a lower maximum rate. Therefore, my 36,000 Btuh heat pump keeps the house heated quite nicely, just like a 100,000 Btuh gas furnace would. The big difference is the gas furnace would cycle many times a day. The heat pump cycles one time a day, i.e., it turns on late in the evening and off late in the morning.

Those who argue that mass has less impact in the heating dominated climates compared to the cooling dominated climates are partially correct, IMO. That's because in the cooling dominated climates the temperature swing average is fewer degrees above normal room temp than the average temperature swing in the heating dominated climates is below normal room temp. [I could be wrong as I haven't looked at the temps in detail.] They tend, however, to ignore the moderating effect of the concrete mass which leads to a more uniform heat movement and therefore greater comfort.

dm- interesting example. However after all of that, something that stands out to me is....a 2000sqft house and a 36k btu HP?!! Shouldnt the heat requirements be much lower for ICF?....(at least that is what we are being fed)

Posted By dmaceld on 25 Jan 2012 06:37 PM


You can probably intuitively see that 10 ft of insulation and 10 ft of rock both moderate heat movement, one by resistance and the other by absorption. Back to your original question about a 30° temp swing. What matters more than the extent of the swing is the average temp of the swing and the total amount of the mass and the total amount of insulation.

Those who argue that mass has less impact in the heating dominated climates compared to the cooling dominated climates are partially correct, IMO. That's because in the cooling dominated climates the temperature swing average is fewer degrees above normal room temp than the average temperature swing in the heating dominated climates is below normal room temp. [I could be wrong as I haven't looked at the temps in detail.] They tend, however, to ignore the moderating effect of the concrete mass which leads to a more uniform heat movement and therefore greater comfort.

THANKS FOR YOUR EXPLANATION! It made things a lot clearer on how this works.

Here is the climate of the area where I am looking to build an ICF home:

Chino Valley - ARIZONA - WEATHER

As you can see 30+ degree daily temperature swings are very common. In my application, do you think I would see better mass effect results from an ICF home?

Posted By Dana1 on 25 Jan 2012 03:49 PM
The DOE2 modeling for the twin houses was in fact validated by the actual measured performance of those houses, with all of the infiltration factors and other known differences duly noted (and modeled).

The measured energy use of the ICF house fell within 1% of what was predicted by modeling. (DOE2 must be pretty lousy at modeling ICF and thermal mass issues to have missed by that much, eh? :-) ) See table 3, p7:

http://www.ornl.gov/~webworks/cppr/y2002/pres/114086.pdf

The DOE2 model proved less accurate on the stick-built, which isn't a surprise given that it was batt-insulated not blown (or sprayed), but was still hit within 5% on the cooling energy, and within 0.5% on heating energy. Yup, as sucky model indeed, not to be trusted...

The ICF house also had a raw whole-wall-R of R15, compared to the stick built (straw man?) 2x4 house's ~R10.6. (This was presumably true for all of the other-climate modeled results.) What, and R15 wall outperforms an R11 wall? Go figure!?!

Yes, the mass was a factor (and duly modeled), but so was the additional R (also duly modeled), and in none of the simulations did the modeled difference end up in double digits. (But it might have had more variation in real houses, if the builders were sloppy on air sealing, and the insulation installations 3rd rate as happens.)

Yes,the mass had an effect beyond the ~40% higher wall-R (which is real performance upgrade even on a low-mass wall) but the modeled and actual performance delta between the real physical houses was still in single digit percentages. Call the whole experiment crap if you like, but it's way more representative than marketing BS. It wasn't designed to do-in or underplay the value of building with ICF (quite the contrary.)

Bottom line, there's more to building an efficient house than a high wall-R or thermal mass.

If you would like to call the whole experiment crap, be my guest, but that was not my intent. The report is being bandy about as if it answers all the question so my point was simply this. Be careful to read and understand the whole report before quoting extracts.

That said I would draw attention to a few items.



We don't know what effect the saturated crawlspace in the ICF house had. Neither do we know what the different venting regimes of the crawlspaces had on the experiment.

There is no evidence that the report measured energy consumption for the homes as occupied. They just modelled it.

The report does not give us standard infiltration by air change numbers. If you assume an 8 foot ceiling you would come up with 8754 cubit feet not allowing for walls or crawlspace. If you use the cfm's for test 4 you get 3.89 changes/hr for conventional and 3.07 for icf. Remember that all normal air leaks have been eliminated for test four. Windows and doors sealed and supplies and vents sealed. All infiltration for the icf has to come through the ceiling or through the floor from the crawlspace or possibly through a void in the concrete pour. I'm sorry but I find this infiltration rate to high for standard icf construction with window and doors taken out of the equation. Most have reported better than that counting the window and door infiltration. I also question the ratio of the infiltration between the two houses. The test would have been much simpler with a sealed crawlspace.

The authors attribute the difference in infiltration to the exterior wall to foundation joint. If the floor is air sealed and insulated (unconditioned vented crawlspace), this joint would pretty much be taken out of the equation as the air seal should be at the top of the joists or at the top of the floor sheathing.Part of the problem may be the floor to conditioned space not being properly seal. Alternatively the authors speculate that the change in infiltration rates between May and June are due to seasonal lumber changes especially the trusses.

The higher than (I) expected infiltration rate will have a larger negative effect on the icf house as the additional airflow would eliminated much of the gain that the mass and or insulation should have provided.



The report also expresses an opinion on houses in different climates base on these two houses in a single climatic point. Part of the discussion on this thread is the impact of heat and cold on the icf build up. Just using data from one climatic point and than suggesting that a house in Minneapolis will perform such and such base on weather records is a little arrogant. If they had build one house in Whitehorse, one in Knoxville and one in Miami then they could use a model to predict the areas in between but to say foam and concrete behaves like this in Knoxville therefore it will do XXX in Whitehorse defeats the whole discussion.



Every model has its strong points and its weak points. You need to understand these before taking everything at face value. It does not mean the whole report is useless.

Posted By lzerarc on 25 Jan 2012 07:13 PM
dm- interesting example. However after all of that, something that stands out to me is....a 2000sqft house and a 36k btu HP?!! Shouldnt the heat requirements be much lower for ICF?....(at least that is what we are being fed)

The heat pump was sized according to Wrightsoft, which does include an option for ICF, and the available size, 36k or 48k. According to Daikin's engineering data manual the output is about 27k Btuh at 5° outdoor, 72 - 75° indoor, temps. The heat pump has kept the house within a degree or so of the thermostat set point of 74° even when the outdoor temp was down to 0° to -10°F. This would be in the early morning hours when there is limited heat being added to the house from life activities. I don't have the tools available to make a measurement of the actual heat load of the house but looking at detailed meter data on the power company web site it looks like I do max out the HP capacity during the coldest nights.

The best I can determine actual heat load and the calculated heat load at the design temp of 9°F are fairly close in the region of 30,000 Btuh. I really hate to say it but I don't see a very long lag between outdoor temp dropping, or in most cases the sun setting, and when the heat pump kicks on. It's probably about 6 hours. How much windows impact that I don't know. I don't recall the % of wall that is window but it's not at all excessive. From a heat load standpoint I'm kind of questioning the value of ICF, but I tell you, the uniformity of heat throughout the house and day is absolutely the best I have ever experienced. I use the crawl space as the heat plenum and all our floors are wood and tile. During the winter the floor in this house is quite comfortable but not noticeably warm. I have what I set out to achieve in that regard, a quasi radiant heated floor with no cold floors! I do wonder how much heat is going out through the wall area in the crawl space. The thermometer I have in the crawl space shows the max temp to be about 84°. That would when the HP is running. Most of the time the crawl space temp is the same as the living space, about 74°.

Would I build another ICF house? At this point I cannot honestly answer that. I am somewhat embittered at what it cost me to build this house compared to what it's value was appraised for when I converted the construction loan to a mortgage, i.e., 30% less. I started building near the peak of construction activities in Idaho and nationwide so I got no super bargains on material or labor. I'm sure I spent more for this ICF house than a frame one would have cost, but I don't have a handle on how much of that extra cost was the ICF system vs. the extra time, and in some cases money, I spent to build a damn good solid high quality house without useless frills. I got the mortgage after the first wave of the housing recession, so I got whacked twice. It hurt bad. So when it comes to asking if I would do it again I'm not in the position to be real objective. I guess I haven't completely gotten over my anger at the way things turned out.