You will have thermal lag whether the insulation is on the interior or exterior surface. Which side it's on just determines what the temperature level of the mass is. With insulation on the outside the inner surface of the mass will be near room temp and the outer will fluctuate above and below that. During summer with insulation on the inside, the entire mass will fluctuate with the outer surface fluctuating through a greater range than the inner surface.

Insulation doesn't affect the thermal lag significantly at all because it doesn't store heat. It only affects the temperature difference across its own surfaces.

You can have two walls provide the same interior comfort. In one case, as you say, it can be really thick walls like the adobe houses. In that case the performance is based more on a longer term cycle. Make the wall thick enough, and with the appropriate outdoor temps, you can have a 12 month heat movement cycle. Make the wall something like 2' of foam and you'll be comfortable inside because of very low conductance of heat, not because of heat absorption.

Actually, it would be quite interesting to see the math on these various situations. I'll look to see what ORNL has for models. Or maybe I'll do some in the next few days. It should be fairly easy with a spreadsheet to create several scenarios. Of course, like all things in life, reality doesn't always equal calculation. I will say though, that the heating/cooling calcs for my house couldn't have been much more accurate. My HVAC nephew calculated the heat load at ~30,000 Btuh with an outdoor temp of 9° and indoor temp of 74°. During the cold spell of the last two weeks my Daikin heat pump ran steady. The factory performance chart shows an output of 27,000 at 5° outdoors and 72° indoors. The heat pump kept the house at 70° with the outdoor temp dropping as low as -5°. Extrapolating the factory chart I estimated it was putting out about 24,000 Btuh. I ran my pellet stove for supplemental heat to keep the house at 72°.

There is no longer-term cycle in the dynamic benefit of thermal mass, dmaceld. If you don't get big swings in temperature in a 24-hour period, above and below "comfortable," thermal mass makes little difference in energy use. And thermal lag explains why. If it takes 10 hours for heat to penetrate massive walls, then the peak ambient heat of 5 p.m. turns up inside at 3 in the morning, or would do so anyway, if heat flow hadn't reversed at some point after sundown and began cooling the walls. There is no one temperature in a massive wall; it follows a gradient from, say, 78 inside to some temperature between 55 and 95 to borrow Lbear's summer amplitude in AZ, depending on what time of day we measure it. Insulation isn't necessary; neither is solar radiation.

Insulation on either side is a complication, in fact. Using DOE2 simulation software and the Phoenix climate, for example, ORNL found the greatest energy savings in a R5 ICF wall. But ORNL's work says clearly that exposed mass is better than interior foam. Given the other heat loads on the interior, you want to reach out and touch that 75 degree concrete.

toddm, dmaceld knows what he's talking about. He lives in an ICF home as I do. The problem is probably in assuming that it takes 10 hours for the heat to penetrate. It must take far longer or ICF wouldn't work as well as it does. Now with CMU or other block type non-ICF system I believe you would be correct. Also the gradient you refer to has a slope very close to zero. A gradient without a slope is well not a gradient. Therefore, the concrete being a conductor is largerly at the same temperature throughout with minor differences inside to out. In fact, in a conditioned ICF building the concrete core never varies from the inside temperature more than 10 degrees (extreme cold or hot). Normally it's very close to the set point (about 2-5 degrees). I've run temperature probes now on multiple buildings and this is always the case.

The ORNL report shows an ICF wall that doesn't exist compared with a "mass" system that does not have heat capacity to count as massive!

If you keep your house at a constant temperature then your interior concrete walls have limited thermal value.

I think it is safe to say the generally accepted view here is that the thermal mass of the outside walls has most effect and value in those climate areas where the daily temperature swings are significantly above and below the desired indoor temp,

dmaceld is correct - plus, if the thermal mass is on the inside, you need to let the indoor temp vary.

in a conditioned ICF building the concrete core never varies from the inside temperature more than 10 degrees (extreme cold or hot).

It will be ~1/2 way between the average indoor and outdoor temps with a typical symmetrical ICF.

TexasICF forcing my hand, I am obliged to repeat that Dmaceld has a fuzzy grasp of thermal mass. He was talking about adobe walls here:

"You can have two walls provide the same interior comfort. In one case, as you say, it can be really thick walls like the adobe houses. In that case the performance is based more on a longer term cycle. Make the wall thick enough, and with the appropriate outdoor temps, you can have a 12 month heat movement cycle. Make the wall something like 2' of foam and you'll be comfortable inside because of very low conductance of heat, not because of heat absorption."

And he is dead wrong. Both walls work because of low conductance. In the adobe wall, it's a combination of heat capacity and low conductance that delays transfer long enough for amibient conditions to change. Heat isn't being stored. It's being cancelled out by alternating fluxes of heat and cold within the wall over a 24-hour period. Thermal lag is not storage; it is a measure of time. If the average daily temperature is "comfortable" and mass is adequate, the inside temp of the wall is also comfortable and the AC and/or the furnace stay off. In four-season climates, the benefit is modest and equal parts comfort and savings. (In my part of the world, folks brag about how late they turn their furnaces on. in the fall and their ACs in the spring.) In the desert SW and Greece, mass was once, and often still is, the ONLY means of conditioning.

Because insulation adds to a mass wall's heat resistance, insulation on etiher side does indeed increase thermal lag. ORNL never really explains why walls with concrete inside-insulation outside outperform insulation-concrete or insulation-concrete-insulation. No matter. I'll take science over TexasICF.

Old hands here no doubt rolled their eyes to hear our favorite carny barker complain again (and again and again) about ORNL simulations of ICF walls that don't exist in the marketplace, in the same way that the Higgs boson obviously doesn't exist because TexasICF can't see it. He is an inveterate maker of claims about ICF mass effect that are not found in evidence. He started this thread a year ago: http://greenbuildingtalk.com/Forums/tabid/53/afv/topic/aff/4/aft/79706/afs/ASC/Default.aspx, the result of which was his fellow ICF sales types urging him to give it up, judging from eye rolls in those quarters.

But, hey, bring it on.

.

Posted By toddm on 27 Jan 2013 04:40 PM

Because insulation adds to a mass wall's heat resistance, insulation on etiher side does indeed increase thermal lag. ORNL never really explains why walls with concrete inside-insulation outside outperform insulation-concrete or insulation-concrete-insulation. No matter. I'll take science over TexasICF.

Old hands here no doubt rolled their eyes to hear our favorite carny barker complain again (and again and again) about ORNL simulations of ICF walls that don't exist in the marketplace, in the same way that the Higgs boson obviously doesn't exist because TexasICF can't see it. He is an inveterate maker of claims about ICF mass effect that are not found in evidence. He started this thread a year ago: http://greenbuildingtalk.com/Forums/tabid/53/afv/topic/aff/4/aft/79706/afs/ASC/Default.aspx, the result of which was his fellow ICF sales types urging him to give it up, judging from eye rolls in those quarters.

But, hey, bring it on.

Here we go again. You point to the former thread in which you made a statement of:

Posted By toddm on 24 Jan 2012 06:18 PM
NO effective R value documentation from forms manufacturers? None?

Then you were proven wrong when you were shown this data:

Energy Efficiency Data & Performance:
* Thickness of the EPS.………………………………………………… 2.625" / wall panel (5.25" total EPS thickness)
* EPS Steady State R-Value (thermal resistance of the material)…. R - 23 (R - 4.55 / inch @ 40 degrees Fahrenheit)
* CTL Group Thermal Resistance R-Value Calculation Report……. R - 23+ calculated in accordance with ASHRAE 90.1
* EPS K-Factor (thermal conductivity of the material)………………. K - 0.22 / inch @ 40 degrees Fahrenheit
* Air Leakage (infiltration rate).….…….……………………………….. 0.05 to 0.10 ACH (average air changes / hour)
* ORNL Thermal Mass Calculator Dynamic R-Value Equivalent…... Greater than R - 32


Why do we have to go round and round with this?

Temp Probe Study

OK, I looked at the ORNL report. toddm, you are right when you say they don't explain why the interior exposure of concrete gives the best performance. It would be nice to see graphs throughout some daily cycles at various times of the year showing the temperatures at each of the surfaces. Also, there is quite a bit of variation in the total number of cooling degree days and heating degree days for the cities used. What is missing in the report is an explanation of how the daily temp cycle affects the flow of heat through the wall.

What I'm thinking is happening with the interior vs. exterior exposure of concrete is this. The outer surface of the concrete in the exterior exposed scenario is in almost direct contact with the sun and air, and because it can swing in temperature much more it absorbs and releases more heat than in the interior exposed scenario. In the exterior scheme the total concrete thickness reaches a higher temp during the day and a lower temp during the night than does the interior exposed concrete. Subsequently, in the exterior exposed scenario the outer surface of the insulation layer fluctuates much more than does the interior surface of the insulation layer in the interior exposed scenario. With a higher and lower temp fluctuation across the insulation layer you naturally have more heat flow going across it.

Another thing I see as quite interesting in their study is that the same thickness of concrete, when installed in two layers with insulation between them has the same performance benefit as the interior exposed concrete scenario. That puzzles me a lot, particularly when there is such a great difference between interior vs. exterior exposed concrete. Another thing they don't discuss is the relative benefit of the 6" concrete + 4" insulation combo (R 17.2) vs the 4" concrete + 3" insulation (R 13.0) combo. The benefit of each vs an equivalent wood frame wall is the same.

You say that in the adobe wall it's a combination of low conductance and heat capacity. You also say heat is not being stored. I disagree. The specific heat of stone, adobe, and concrete is fairly high, meaning it takes a lot of Btus to raise the material temp 1°F. That is the definition of heat storage. It's the capacity to store heat that contributes to the concrete's low conductance and thus to thermal lag. Heat doesn't get from one side to the other very quickly as it's being sucked up by the concrete. In a steady state condition when the concrete has reached its full temp the heat will travel fairly quickly, about 10 to 20 times as fast as through the foam. To be clear I'm talking about heat storage in terms of hours. Like you say, it goes in during the day and out a night.

Given a mass wall thick enough, say maybe 4' to 10', you will have daily thermal lag in the outer several inches, but you can have yearly thermal lag through the entire thickness. Make the mass thick enough and you'll never have temp fluctuations in the inside because the heat and cold never reach the inside. That's the case with caves in the mountain, disregarding what happens at the entrance.

I do take some umbrage at your statement that I have only a fuzzy grasp of thermal mass. I understand it quite well, thank you. I think our issue is mainly a difference of how we explain ourselves, and the fact that I am a bit mystified at the ORNL model results of interior vs. exterior concrete exposure. Their results seem to contradict logic. Just because they are ORNL scientists does not guarantee they didn't make a mistake. I spent 22 years auditing the processes and practices of professionals. Sometimes their mistakes and ignorance were nothing short of astounding.
 

We keep going around because you hope ICF prospects will mistake weasel words like "more than" or "up to" as documentation when it actually says: "We don't have documentation that the FTC will accept so we'll leave the selling to shills like Lbear."

Again, proper documentation begins when the manufacturer builds its wall in an ORNL blessed hotbox. If mass is involved, the point is to determine both R value and thermal lag. (R value is all that matters when ambient temps rise above, or fall below, a "comfortable" daily average and stay there. The difference between Phoenix and Boise is the many more days in a year in the former when mass has an effect.) The manufacturer feeds this data into DOE2 and simulates its performance in the representative climates that DOE2 models. The ICF industry obviously does not like how DOE2 treats ICF walls, which apparently led its sponsorship of the side-by-side test of ICF vs stud wall in Knoxville in 2000. The score: DOE2, 1; ICFA, 0.

Like I say, Dmaceld, insulatiion complicates thermal mass. a ZEH in Tucson, for example, wrapped filled CMU walls in R14 foam and reported, post construction, that thermal mass had no measurable impact on energy use, which was surely news to the residents in the many adobe houses nearby. The immediate failure in this house was inadequate auxiliary heat. As you note, recovery can take a looong time.

We'd understand the process better if ICF manufacturers stepped and gave us some real data. Sad to say, they prefer urban myth.

Sphingers, please ignore this toddm character. He's just an angry troll whose mother weaned him a few weeks too early. He had a bad experience with a disreputable ICF installer and now lays in wait under his metaphorical bridge for unsuspecting normal people innocently asking for advice.

Ray Gladstone, Jan. 24, 2012, on why ICF manufacturers prefer weasel words over proper thermal mass documentation:

"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."

Emphasis added.

Hi Toddy. Nice to see you back putting your words into other people's mouths again. Water getting cold under that bridge?

A March 2012 scientific study released by the National Research Council Canada conducted field testing on ICF walls. So together with NRC/IRC and Natural Resources Canada, they measured the thermal mass of ICF walls in their recent study “Field Monitoring of the Dynamic Heat Transmission Characteristics through ICF Wall Assemblies over a Full Year Cycle of Weather Exposure.”

The findings: (emphasis mine)

NRC Study

The scope of work included the design of the experiments, the installation of test specimens, the commissioning of the instrumentation, the operation of the test facility, the monitoring, data collection & analysis. This research evaluated the dynamic heat transmission characteristics through an ICF wall assembly in FEWF for a one year cycle of exposure to outdoor natural weathering conditions. Correlations between the temperature at the exterior surface of the ICF and the exterior surface of the concrete revealed a buffering effect of approximately 5 days due to the mass of the ICF.

This analysis showed that the ICF moderated heat loss to and from the interior. The interior heat flux through the wall was not following weather changes on the outside instantaneously. This was interpreted as the buffering effect of the mass. The monitored data confirmed that the concrete adds very little to the overall R-value of the wall assembly under steady-state conditions. During the transient conditions, the data showed that the concrete played a significant role in tempering heat loss to the exterior. The thermal mass of the concrete was shown to reduce the peak heat flux through the assembly during cold weather. This research is one of a series of projects that highlight direct and indirect impacts of thermal performance of the building Envelope technologies in houses. This paper provides valuable experimental data to be used for energy simulation models.

Thus, ICF walls have the potential to reduce the peak heating requirement of the furnace, and the peak cooling requirement of the air conditioning system. This may have implications for the sizing and cost of mechanical equipment. This experiment only examined a small section of ICF wall on a west façade. Performance on the whole house level will be affected by other factors including solar gains through windows, and the operating mode of the house (for example: the use of free cooling at night or thermostat setbacks). Whole house modeling would be required to better understand the impact of ICF construction on annual energy consumption.

http://greenbuildingtalk.com/Forums/tabid/53/aff/4/afv/topic/aft/81109/Default.aspx $100 to anyone who can demonstrate that I edited or changed Ray Gladstone's post in any way, except for the emphasis I noted. Honestly, does the ICF industry recruit on used car lots?

Fine, Lbear, now add the the rest of the house, as opposed to a west-facing wall section, test it n another seven or eight representative climates and tell us if it amounts to squat. Oh, wait, ORNL has already done that, and it didn't.

Sphingers, walls are a relatively small contributor to heat loss. And thermal mass effect, except in desert southwest, is an even smaller piece of that, small enough that poor door installs could offset it. Go here for some objective help developed by UCLA: http://www.energy-design-tools.aud.ucla.edu/heed/ Climate Consultant 5 downloads historic weather data from the NWS site nearest you. There is a whole lot of detail there, but also a sort-of recommendations page that lists the number of hours in a year that specific energy strategies will keep you comfortable without hvac. If the answer is insulation, I'd go with ICF. If the answer is thermal mass (doubtful) then the poured wall is still in the running. If you are comfortable with modeling software, UCLA's HEED allows you to build a crude model of your design including windows and doors, orient on it your site and run it through 365 days of typical weather. It has both an ICF and a poured wall option. It's still (sophisticated) guesswork, but at least the software's authors weren't used car salesmen in earlier careers.

Toddy, I was referring to your wise-a$$ed little intro. But then you knew that didn't you? Oh wait, Nurse Ratchet just called to see if anyone knows which bridge you're lurking under today. She says it's time for your anger management group session.

There are two different types of people out there, ones that want to see scientific reports documenting the precise benefits of the thermal mass and insulation. And those that see more worth in talking to people who ACTUALLY live in an ICF home. I prefer to deal with option 2. Pretty easy to venture a guess as to which side of the fence some of the posters above are on.

Can you pour a normal foundation and affix insulation to the concrete, sure. Will it cost less, probably. But then you are left with ZERO attachment points other than anchoring back to the concrete. There is no company to support your efforts, and help you with questions. Setting conventional forms is by far not a task that a DIY can do, and involves more labor with stripping and cleaning. If you are going to argue the fact that EPS does not out perform a stud wall placed next to concrete with some fiberglass stuffed in between the studs, then you just enjoy being difficult and are blinded by being set in your ways.

As an industry, I do not believe you will hear us say very often "change over to ICF, its cheaper". What you get is an easy to build and work with system for forming concrete that leaves you with a finished product ready for waterproofing, finishes, drywall, etc.

Every single manufacturer has a plethora of documented customers utility bills showing a drastic reduction in heating and cooling costs, compare them to whatever you like. Those customers will also tell you about how quiet the house is, the sense of security they get if a storm is coming, and how great the feeling of a dry warm basement is. There is no scientific test for homeowner satisfaction, so I don't have an ASTM report for it. Sorry.

To the OP and anyone else reading these pages of ranting and arguing, don't let the stupid little scientific formulas of thermal mass deter you. Build it once and build it right. I am biased, but feel that ICF can play a strong part in you loving your house for many many years to come. All of my customers sure do, and that is proof enough for me.

Yeah, OP don't let those stupid little formulas sway you. You came to GBT to cheer for your team. Go ICF!

Then again, in the real world, Kia is paying a settlement to owners to compensate them for overstated mileage claims. Publishers Clearing House paid a fine for "Board of Judges" letters that misled consumers. More to the point, the Federal Trade Commission is hot lately on deceptive claims by window companies. http://www.ftc.gov/os/caselist/1123005/120518longfencecmpt.pdf

I'm not saying that anyone here is breaking the law. I am saying that only way for a manufacturer to make a legit thermal mass claim involves an ORNL hotbox and DOE2. I am also saying that the average attentive consumer should look at the official disclosure. If it says R22 than expect to get an R22 wall, cuz if the FTC was cool with "more than R36" that's what the manufacturer would say. Rant? Or common sense?

Posted By toddm on 28 Jan 2013 07:45 PM
Yeah, OP don't let those stupid little formulas sway you. You came to GBT to cheer for your team. Go ICF!

Then again, in the real world, Kia is paying a settlement to owners to compensate them for overstated mileage claims. Publishers Clearing House paid a fine for "Board of Judges" letters that misled consumers. More to the point, the Federal Trade Commission is hot lately on deceptive claims by window companies. http://www.ftc.gov/os/caselist/1123005/120518longfencecmpt.pdf

I'm not saying that anyone here is breaking the law. I am saying that only way for a manufacturer to make a legit thermal mass claim involves an ORNL hotbox and DOE2. I am also saying that the average attentive consumer should look at the official disclosure. If it says R22 than expect to get an R22 wall, cuz if the FTC was cool with "more than R36" that's what the manufacturer would say. Rant? Or common sense?

You just insinuated that ICF makes false claims and then you try and go back and claim that you didn't say that.

Once again, I post this for your review:

Energy Efficiency Data & Performance:
* Thickness of the EPS.………………………………………………… 2.625" / wall panel (5.25" total EPS thickness)
* EPS Steady State R-Value (thermal resistance of the material)…. R - 23 (R - 4.55 / inch @ 40 degrees Fahrenheit)
* CTL Group Thermal Resistance R-Value Calculation Report……. R - 23+ calculated in accordance with ASHRAE 90.1
* EPS K-Factor (thermal conductivity of the material)………………. K - 0.22 / inch @ 40 degrees Fahrenheit
* Air Leakage (infiltration rate).….…….……………………………….. 0.05 to 0.10 ACH (average air changes / hour)
* ORNL Thermal Mass Calculator Dynamic R-Value Equivalent…... Greater than R - 32

If the above "Thermal Mass Calculator Dynamic R-Value Equivalent of Greater than R-32" is a false claim like you insinuated, then report them to the FTC. Your above paragraph claims that if it states "more than R36" then the manufacturer would make that claim. The reality is that they do make that claim but you choose to ignore it and act like it's not there.

The reality of it is that the dynamic R-Value equivalent claim of "Greater than R-32" is a valid claim and ORNL and DOE2 and FTC cannot say otherwise because it is validated by scientific studies.

Here's a graph I put together of the ORNL model. What they are showing by their Dynamic Benefit of Mass Systems (DBMS) index is how much better a frame wall would have to be to equal the annual energy cost of the ICF wall. Glazing is 10% of the total wall area. This is for a wall of R-17.2, 6" concrete, 4" foam.

What bothers me most about their results, as I said earlier, is the dramatic difference between concrete on the interior vs. exterior. The first two points in the graph represent the 6" concrete installed in two separate slabs with the insulation between. Both of those, plus the third point, have concrete on the interior surface. The DBMS is almost flat regardless of the thickness of the concrete on the interior except for Miami, which is cooling predominant. The last three points represent 1", 2", and 4" of foam on the interior side. In 4 & 5 the concrete is sandwiched between the two foam layers. There is a steady drop in DBMS with increased foam thickness on the interior.

According to the model, any insulation on the interior side dramatically lowers the benefit of the mass, except for Minneapolis, but they don't explain why.

The upshot, as I see it, is that even in a heating predominant climate like Minneapolis, or a cooling predominant one like Miami, there is a measurable benefit derived from the mass in the wall. In fact, according to the model 6" of concrete on the interior side provides as much benefit in Miami as it does in Phoenix! Now why is that? That certainly goes against the argument that in a south climate where the heat flow is constantly to the inside, mass provides no benefit.

CORRECTION EDIT: 3/18/13 For data point 3, the exterior foam value should be 4", not 0".



Here's the corrected version, with the data included.

Here is the FTC's R value rule for consumers: http://www.consumer.ftc.gov/articles/0107-home-insulation-its-all-about-r-value. The manufacturer's claims must be based on standard tests. To my mind, a calculator is not a test, but my opinion and a dollar will get you a cup of coffee, In the normal course of events, Fox Block's competitors would be screaming bloody murder at the FTC. It is instructive that the industry apparently is not. Clearly, ICF makers could claim higher r values everywhere, albeit far lower ones than the R45 my contractor claimed based on a 90s industry-funded test that can be best described as scurrilous. (BTW I knew he was woofing me; my criticism here of ICF mass claims predates my building permit by a year. Proof cheerfully furnished on request. And I'd choose ICF again. For perimeter walls in a cold climate slab on grade, ICF is far superior to standard practice CMU and cheaper besides.)
Why ICF manufacturers don't claim mass effect is probably a regional thing, a bloc of distributors (north? southwest?) that would suffer the kind of competitiive disadvantage that Ray Gladstone suggests. Note that Fox' use of mass effect is climate-free.

Now, I will confess that I am not the easiest sale in the world. But the last thing I want in products is a mystery. I repeat, if it says R22 then treat it as R22.

One thing about the ORNL model to keep in mind is that it is based on a set of assumptions that may or may not be accurate for a specific case. How willing you are to let interior temperate vary (and/or various other things) has a huge effect.

At least a partial explanation of interior mass sometimes being better is that it will heat/cool infiltrated air.