jonr - The foam is held by temporary form ties and doesn't float up. I have my own set of forms (about 200) that I built for about $3500 (HDO plywood) before I started construction. I used them on my basement (completed) and am now getting ready to pour the first floor this month.

I find ICF a very cool technology (that's why I'm reading this board) but was intimidated by the prospect of pouring myself. This whole building project is my main hobby (I totally love doing it) and thought that wood forms would be more forgiving w.r.t. blow-outs (especially using an internal vibrator). I also found it easier to embed conduit and plumbing in "traditional" forms but this may be easy to do with ICF. I don't know if ICF would have been cheaper. Apparently I am getting a good price from the foam company for EPS panels (and that company is also really easy to work with and delivers within 3 days (I purchase the foam for each floor, which is about 1500sf per order)).

arkie6 - I won't have metal form ties protruding the EPS and the foam is continuous on the exterior. The siding (4 ft x 2 ft glass panels) will be screwed to a steel frame that is attached to the top of the house (flat roof) and rests on the bottom (brick ledge) so there are no thermal bridges through the EPS. It's a contemporary house.

The glass siding will collect solar heat?

I suppose one could stagger foam seams with multiple sheets to avoid any leaks.

For some hypothetical small single level house walls, might be $3500 for forms, $3000 for foam and $3000 for the concrete. ICFs might be $4500 + concrete if you can get them for $3/sq ft. Sounds like a wash at two stories, ICF wins at one. As far as I can tell, renting forms is outrageously priced.

TexasICF writes, 'The interior mass system has to have 5" of foam to the exterior and 6" of concrete to the interior' to be comparable to ICF, to which I answer, bogus point.

You can get to the same place by increasing mass rather than insulation -- to the point of skipping insulation altogether in the desert Southwest. Add too much insulation, in fact, and the dynamic benefit stops working. Mass provides time shift -- a thermal lag that means that intense afternoon heat won't be felt inside until early morning. (Actually, heat rarely penetrates adobe or European-style high mass structures, but, rather, ebbs and flows in a diurnal flux.) Insulation increases thermal lag. Judging by ORNL's modeling of ICF in Phoenix's climate (see earlier link), its foam sandwich increases thermal lag to the point that it no longer buffers heat and cold there on a 24-hour basis. Afternoon heat that manifests itself on the following afternoon isn't getting you anywhere. Add to this ICF's interior insulation, which is like queuing up Augie Meyers on your CD player and then tossing a quilt over the speakers. The result for ICF is a modest bang for major bucks if energy savings is the criteria. That would not necessarily be true of hybrid ICF systems like Apex, Rastra, Durisol, or of AAC, except perhaps for the major bucks part.

Here is a good explication of high-mass dynamics: http://home.vicnet.net.au/~oversite/house.html

As the article suggests, mass offers little benefit in places like Minneapolis or Miami, where the thermometer rises or plunges to uncomfortable levels and stays there day after day. UCLA has a free tool called Climate Consultant 5 that will help you sort through a hundred years of San Antonio weather and come up with an appropriate energy strategy. http://www.energy-design-tools.aud.ucla.edu/ But if you read the fine print, CC5 is making assumptions about what constitutes high mass that could skew your analysis. Super insulation is more predictable and usually cheaper. Truth be told, mass is a gimme only in the high desert unless you have bigger goals in mind, like passive solar. Start with your local weather in any case. One suspects that cooling matters a great deal more in San Antonio than heating, such that ICF with night-time setback may recover its premium in, say, 100 years.

I do appreciate your opinion and the link to the report from a few posts back.

I contributed 15 Texas houses to the recent MIT study.  MIT reported ICF buildings to be 20% more efficient than conventional.  I believe this 20% better number is low (at least for Texas). 

Why?   (1) Many of our customers regularly communicate their utility bills.  (2) I’ve lived in an ICF home for several years myself and know approximately what neighbors are paying and (3)  About half of the Texas ICF houses contributed were built with conventional BAT attics.  

All 15 Texas homes I provided to the study had sealed foam attics.   Why?  Because as most of you know, ICF is not ‘nearly’ as effective without taking the attic or roof into consideration.  Imagine a screen door on a submarine.

I believe mixing these different ICF construction approaches knocked the final data down to 20%. 

Regarding the link provided to the ORNL report from ten years or so ago.  Why does the ORNL report say that ICF performs less than half to one third as well as the 20% just reported by MIT?

1^st this ORNL report is actually a compilation of several reports and if you’re not careful you might read a section and think it’s talking about another area of interest.   For example directly from the report:

“The steady state R-value traditionally used to measure energy performance does not accurately reflect the dynamic thermal behavior of massive building envelope systems. “

And they continue…

 “Since the majority of U.S. residential buildings are built using light-weight wood-framing technologies, all energy performance comparisons in this paper are made against light-weight wood-framing buildings. “

Clearly, this report is a compilation of several reports…

“An overview of several historic and current U.S. field experiments are discussed.  These experiments were performed in a wide range of U.S. climates utilizing several building sizes and shapes.  Theoretical energy performance analysis is presented for a series of four wall assemblies. “

Is the report now talking about ICF and the other three mass walls types at this point?  No.

“Burch built four one-room test huts at the National Institute of Standards and Technology (NIST) to compare seasonal energy performance of wood-framed, masonry, and log construction.”

 “It was observed and documented that heating and cooling energy in massive houses can be far lower than those in similar buildings constructed using light-weight wall technologies”

At this point they’re still talking about wood-framed vs. mass.

Often misquoted, the portion about Burch’s research says:   “Significant energy savings were noted for the house with a higher internal thermal mass”.   Are we talking about the four theoretical mass walls at this point?  Any type of ICF?  No, although this statement establishes the superiority of mass walls we are not yet talking about internal and external mass just mass versus low-mass.

The report continues:

“Robertson and Christian investigated eight one-room test buildings that were constructed in the desert near Santa Fe, New Mexico, to determine the influence of thermal mass in exterior walls.  The buildings were identical except for the walls (adobe, concrete masonry, wood framed, and log)”.

So ICF wasn’t included in part of this part of the study?  No.  It wasn’t.    The section on Robertson and Christian continues…

“This study demonstrated that on small windowless massive test huts, energy consumption can be up to 5% lower than in a light-weight building.  It is important to point out that during this study, the massive walls had about three to four times lower R-value than the wood walls (wood-framed wall R-value was about R-13 vs. R2 to R5 for adobe, concrete masonry, and log walls)”.

Just in case you missed that, the report continues…

“This gives a completely different meaning to the 5% energy savings that were reported.”

One might make the claim correctly at this point the internal mass is better but the report is talking about adob, concrete masonry and log walls, not ICF (not yet).  At this point we’re still taking about adobe, concrete masonry and log walls compared to wood-frame.

Then the report discusses the 1999 NAHB study:  “This suggests that most likely thermal mass related energy savings during the NAHB ICF study were in the neighborhood of 11%. 

MOST LIKELY?   Interesting?  The report continues: 

“Masonry or concrete walls having a mass greater than or equal to 30 lbs/ft2 and solid wood walls having a mass greater than or equal to 20 lbs/ft2 are defined by the model energy code [MEC-1995, Christian 1991] as massive walls. “

What?  Solid wood wall that has mass of 20 lbs/ft2 is considered a mass wall by this study? 

“They have heat capacities equal to or exceeding (6 BTU/ft2 F).   The same classification is used for this work.”

This is a very low heat capacity to define a mass wall since the average density of concrete is 133 lbs/ft2 with a heat capacity of well over (20 BTU / ft2 F). 

Regarding ORNLs BTC 1995 Dynamic Benefit for massive systems model (DBMS).  According to the report:  “DBMS should be used only as an answer to the question: What wall R-value should a house with wood frame walls have to obtain the same space heating and cooling energy consumption as a similar house containing massive walls?”

There is some groundbreaking work here but putting it into one mixing bowl has confused the intent of the research and causes many to misquote it.  And folks continue to misquote the report.  Take Bakersfield (figure 7), for example, sometimes used to say that the report says you only get a 10% improvement with ICF over conventional construction.   What this chart actually says is that you will get a 10% savings with ICF over a conventional wall built to R-25.   Not too many walls are conventionally built to R-25.  Furthermore, regarding mass, if 15% of the density of concrete is considered a mass wall by the report what might we learn if we did a study with a higher standard for a mass wall?  Put another way, how much better is ICF if you use a mass value of 130 lbs/ft2 – 140 lbs/ft2 and then compare it to today’s actual conventional construction of R-13? 

Remember the computer simulation for the four different mass wall types including ICF used the same mass and the same R-value for each.    My hats off to galore as he did not miss this point at all and built a custom system that exceeds what’s presented in the report.

The last statement in this report implies that the report was much more about ICF than most of its content and says:  “It was found that for ten U.S. locations, ICF walls … the average potential whole building energy savings (ICF house vs. conventional wood-frame house) can be between 6 an 8%”.

MIT came up with a huge difference in favor of ICF and I believe it too is low.  If you doubt it, just ask a   neighbor living in an ICF home.  Regards.

Well, yes, TexasICF, but the Mad Hatter offended Time, as he explains to Alice, so he can no longer ask for it to be dinner at any point during the day, which explanation confuses Alice even more. "What if I'm not hungry?" she asks.

I am also disappointed in the work on thermal mass. ORNL's aim was to develop a multiplier so that consumers could calculate an effective R value adjusted for thermal mass and compare high and low mass walls correctly. Then the lab pretty much quit.

But the parties with skin in the game pressed on: "The Energy Performance of Log Homes" http://www.crockettloghomes.com/Pdf/Technical%20Library/The%2520Energy%2520Performance%2520of%2520Log%2520Homes1.pdf; "Thermal Performance" http://www.rastra.com/ThermalPerformance.html: "Thermal Performance for AAC Block" http://www.safecrete.com/products/techmanual/pdf/thermal.pdf

So where is the study from the ICF industry showing that ORNL had it all wrong?

One drawback is that an official number, even if it improved on 6 percent to 8 percent, would be a serious letdown from the wild claims ICF types have repeated for decades. The contractor who installed my ICF stem walls in '09 insisted that the effective r value for Arxx blocks was 45.*

The MIT study is a joke. It compared ICF and stick built in Chicago and Phoenix, but it was measuring much more than dynamic benefit. Its leading conclusion states (duh): "The advantages of higher R-value and lower thermal bridging enable ICF homes to deliver energy savings in heating, cooling, and ventilation compared to conventional wood-framed construction." http://web.mit.edu/cshub/news/pdf/BuildingsLCAsummaryDec2010.pdf It was funded by the Portland cement and ready mix industry trade associations. The 20 percent savings figure is vs. "code compliant" stick built in Chicago. I believe I have made the point that climate is everything in the dynamic benefit of mass. Others here have held forth persuasively on the notion that code compliant is not necessarily a high bar.

So, TexasICF, the answer is sitting right there. You know who to call at MIT. You'll want to ask for multipliers that correct R value for the dynamic benefit of mass in ICF walls in sample U.S. climates. Post pub date, I'll shut up.

* R45 on 30 random days each year during which the meteorological stars are perfectly aligned.