With the Ecobee thermostats you can down load your temperature data.

The data is recorded every 5 minutes. Thus you can chart: 1. Temps inside and out (also remote sensors if you have them) 2. Humidity inside and out 3. Wind speed 4. Furnace commands and set points.

I can set our house to heat at a constant rate all day long (aka BTU input being constant). Attached it a 2 day chart of doing this.

You can see there is little if any thermal lag between the outside temp and inside temp changes as the ICF manufactures advertise. I have seen Nurdura claim up to a 6 hour shift.

House is a Build Block house with 6" Global Block on main floor and 6" Standard Block in a walkout basement that is 50% exposed.

Sorry....

Curious what others see.

It isn’t entirely clear to me what you are asking for or what you are saying? We have lots of test data with thermocouples embedded in the ICF concrete core and in the EPS insulation on each side of the core. We used this data to validate our ICF performance software. Are you just saying that the indoor temperature stays relatively constant given a large variation in the outdoor temperature? Or are you saying that the concrete core temperature stays relatively constant given a large variation in the outdoor temperature? Both are true given the relatively large specific heat capacity of the concrete core and given that an ICF building heat transfer is a function of the indoor temperature, the concrete core temperature and the interior EPS R-value. The concrete core temperature is a function of the indoor temperature, outdoor temperature and the interior/exterior EPS R-values. All of this is now precisely determinate using our ICF performance software and the actual indoor/outdoor temperature profiles.

http://www.nudura.com/divisions/builder/r-value-thermal-mass

We have been brought to believe that this thermal lag is what makes ICF so great. This is also been the justification for the inflated R-value.

I don't see any lag. Was wondering if others have seen a lag in temps.

I am also pointing out that the data that is logged in our Ecobee's can be downloaded and manipulated easily (not sure is other therm's have this function also).

Thanks, I do love my ICF house for air sealing and strength though.

There's more to that exercise than just measuring on an entire house. What kind of windows do you have? how many sq,ft are covered by windows? Are the windows facing south? what's the insulation in your attic? The ICFMA did an ASTM 1363 test on a ICF wall assembly, you can download that test here http://icf-ma.org/resources/thermal-study/ Before they start testing R Value they wait for the wall to be at steady state. The 2x6 wall took 60 hours to reach steady state and started consuming energy right from the start. The ICF wall took 324 hours to reach steady state and it took 40 hours before it even started to loose any heat, hence the flat line for the first 40 hours of the test, I will let you read the conclusion of the test written by a third party engineer.

In theory, the 2x6 wall should've been performing at R19 when we calculate the efficiency of the wall the way professionals do, air space, lower value for 2x6 etc, but the end result was R15 which was not surprising for the lab. On the other hand, using the same method, the ICF wall should've perform at R23, but the final result was R24. The engineers conducting the test were blown away since they NEVER before saw a wall assembly were the actual performance exceeded the "sum of all parts" calculation.

That being said the walls can only do so much, remember that R value is calculated with the amount of BTUs being consumed and is inversely proportional. The numbers are not 100% precise but close enough to give you a good idea. R3=32 btu/sq.ft, R6=16 btu/sq.ft, R12=8 btu/sq.ft, R24=4 btu/sq.ft. With that in mind it's easy to see that investing in good windows is key here, if you ad R3 on top of R24 you only improve your wall by 12%, thus saving 0.48 btu/sq.ft. But if you add R3 on top of R3 (double pane vs triple pane windows) you have 100% gain and you go from 32 btu to 15 btu. Let's say you have 2000 sq.ft of wall and 500 sq.ft of windows. 2000 sq.ft of wall at R24 consumes 8000 btu and 500 sq.ft of windows at R3 consumes 15000 btu, only 7500 btu at R6. Keep in mind that if you want to double your efficiency, you have to double the insulation.

I was a contractor for a long time and when I started doing ICF I saw a lot of people thinking that they didn't need to have super efficient windows because they had very good walls and to certain extent, it will work since they would often compare to a very inefficient stick frame construction, but to really see the benefits it's important to do the whole thing right.

Thanks

First off, it is called “thermal lag” and NOT “thermal leg”. There is only an ICF concrete core thermal mass lag effect when either the indoor temperature or outdoor temperature varies with time. The ICF concrete core tends to change temperature slowly because it has a much larger specific heat capacity compared to other typical building materials. This can be a good thing or a bad thing depending on the specific situation and objectives. If the indoor/outdoor temperatures remain constant, there isn’t any ICF thermal mass lag effect at all and the ICF effective R-value is identically the same as its conventional R-value which is about R23 for standard ICF (i.e., 2.5” EPS + 6” concrete + 2.5” EPS).

I would strongly agree that there have been many unsubstantiated ICF effective R-value claims made by ICF companies over the years. Since we do a lot of PE stamped building heat gain/loss analysis, we needed a more accurate and fully substantiated method for determining a given ICF wall assembly effective R-value and for different given building locations (i.e., different outdoor seasonal temperature profiles). This was our principal motivation for developing our ICF performance software. The physics and computational/numerical approach to accomplish this is well understood and well validated. A validated computational tool is more useful for accomplishing more accurate designs than extrapolating limited and sometimes questionable test data.

Borst ICF Performance Software

Here are some excerpts taken directly from the software instructions and some of the conclusions:

This software exercises the fundamental heat transfer equations that govern this ICF heat transfer system at a 0.01 hour (i.e., a 0.6 minute or 36 second) sampling rate over a twenty four hour period using finite element time/temperature numerical differentiation/integration analysis. The software assumes that the initial concrete thermal mass temperature at midnight (i.e., 00:00) is what the steady state concrete thermal mass temperature would be if the indoor/outdoor temperature that is entered for midnight were to remain constant for a sufficiently long enough time to reach this steady state temperature. Doing this initializes the simulation with a concrete thermal mass temperature that is neither “over-charged” nor “under-charged” with thermal energy so as to fairly evaluate only the heat flow generated by the entered indoor/outdoor temperature profiles. However, this calculator has a concrete initial temperature input parameter to allow experimentation with “over-charging” or “under-charging” the concrete thermal mass.

It should be noted that this software solely considers and only accounts for the effective R-value of the ICF portion of the wall assembly and this software does NOT consider or account for the R-value of any additional wall assembly layers (interior/exterior finishes, interior/exterior air film convection, etc), outdoor air infiltration, ground temperature wicking effect, solar heat gain, etc. These other heat transfer mechanisms should be accounted for and are normally accounted for when accomplishing the overall building heat gain/loss analysis.

For standard ICF construction using our Rogue River, Oregon outdoor temperature profiles, we determined our ICF thermal mass effective R-value to be R21.38 in Spring, R66.15 in Summer, R63.42 in Fall, and R21.07 in Winter. This is an average annual ICF performance of R43.07 or 1.86 higher performance than the conventional R23.10 R-value.

When the interior EPS thickness is reduced to 0.1” and the exterior EPS thickness is increased to 4.9” (i.e., essentially putting all the insulation on the exterior side, but keeping the total conventional R-value the same), the summer effective R-value went from R66.15 to R44.05. In this case the concrete temperature stays very close to the indoor temp. The effective R-value remains the same (i.e., R44.05) when all the insulation is put on interior side. However, in this case the concrete temperature stays very close to the outdoor temp. This would imply that both these non-symmetrical ICF designs result in an equal, but lower performance than the standard, symmetrical ICF design for this summer outdoor temperature profile. When this is repeated for the winter outdoor temperature profile, the winter effective R-value went from R21.34 to R21.96, i.e., a slight performance improvement.

When both the interior/exterior insulation thickness is reduced to 0.01” (i.e., essentially removing the interior/exterior insulation), the effective R-value went to R0.77 (i.e., 1.12 times higher than the new R0.69 conventional R-value). This would imply that even an uninsulated, 6” thick concrete wall will exhibit some degree of increased effective R-value. When the insulation is retained, but the concrete thickness is reduced to 0, the effective R-value went to R22.50 (i.e., exactly the same value as the new conventional R-value). This would imply that as the concrete thickness is reduced, the effective R-value tends toward becoming equal to the conventional R-value. When the concrete thickness is then increased to 4”, the effective R-value went to R63.40 (i.e., 2.77 times higher than the new conventional R-value). When the concrete thickness is further increased to 8”, the effective R-value went to R67.79 (i.e., 2.91 times higher than the new conventional R-value). When the concrete thickness is further increased to 12”, the effective R-value went to R69.99 (i.e., 2.95 times higher than the new conventional R-value). When the concrete thickness is further increased to 36”, the effective R-value went to R78.48 (i.e., 3.01 times higher than the new conventional R-value). When the concrete thickness is further increased to 360”, the effective R-value went to R177.12 (i.e., 3.03 times higher than the new conventional R-value). When the interior/exterior insulation is then removed from this 360” concrete thickness, the effective R-value went to R109.25 (i.e., still 3.03 times higher than the new conventional R-value). This would imply that as the concrete thickness is increased, the insulation thickness becomes increasingly less important in achieving the higher effective R-values. Furthermore, for this summer outdoor temperature profile, this would imply that the absolute maximum possible effective R-value is 3.03 times the conventional R-value and we are achieving 2.86 times the conventional R-value with a standard ICF design.

One might expect that higher effective R-values might be achieved by having some portion of the indoor temperature profile be the same as the outdoor temperature profile. For example, having that portion of the indoor temperature profile be the same as the outdoor temperature profile during the summer cooling season when the outdoor temperature is below 70 degrees Fahrenheit to simulate having open windows or using increased ventilation during these times. However, this is NOT the case. While doing this will reduce some of the Thermal Mass Temp output parameters, doing this will result in some of the Conventional Heat Flow Rate output parameters becoming 0 (where they were previously providing a building heat loss cooling effect), will reduce some of the Thermal Mass Heat Flow Rate output parameters (where they were previously providing more of a building heat loss cooling effect), will increase both the Total Conventional Heat Flow and Total Thermal Mass Heat Flow output parameters (i.e., provide increased building heat gain), but will increase the Total Thermal Mass Heat Flow output parameter more than the Total Conventional Heat Flow output parameter, and this will REDUCE the Thermal Mass Effective R-value output parameter. Doing this for the summer profile resulted in the effective R-value going from R66.15 to R39.06. One needs to keep in mind that this calculator only determines the effective R-value performance for just the ICF for the entered input/output temperature profiles and this calculator does NOT determine the HVAC system performance of the building. Or perhaps another way of thinking about this is that when you run the building AC system, the building cools down to a lower temperature than it would otherwise be, however, the heat transfer through the walls increases because of the increased inside/outside delta temperature. More thermal mass heat transfer is equivalent to having a lower thermal mass heat transfer resistance or a lower thermal mass effective R-value. Obviously, opening windows or increasing ventilation during the summer cooling season when the outdoor temp is below 70 degrees Fahrenheit is a good thing to do as this provides increased building cooling with minimum HVAC energy usage.

We've been living in our ICF house for almost a year now, and have seen a lag between the inside and outside temps. But it may not appear that drastic.

We also have an Ecobee, and I've attached a copy of our January graph, which has had a pretty wide range of temps. There's not much difference when the outside temps are constant, but you can see it in the heating hours on the days of a big shift.

For example, we saw a temp dip between 1/13 and 1/14, but the heating hours remained relatively consistent. The same thing can be seen from 1/12 to 1/13, and 1/23 to 1/24, and 1/26 to 1/27.

Conversely, the outside temps rose from 1/1 to 1/2, 1/9 to 1/10, 1/18 to 1/19, and 1/25 to 1/26, but those days called for more heating, rather than less.

Some of this may be due to the time of day that the temps changed and the furnace was running, but it does seem to show a lag.

Regardless of the inside/outside differential, we've been very happy with our ICF house. And one of the advantages not shown here is just how even the heat stays in our house. With just a four or five degree temperature setback, it's rare for the furnace to run overnight.

For a second example, I went back to our records for October. I was looking for a rather quick temperature change, and found one where the outdoor temp went from 51 to 72 and then back to 59 in about 24 hours. As you can see, the indoor temp fell while the outside temp was rising, and pretty much held at that temp while the outside temp fell again. There was no heating or cooling during that time (The white line is the HVAC fan, the purple line is the whole-house dehumidifier). The windows remained closed, but we have an HRV that runs constantly.

Hope you found this interesting and helpful. Scott

I don't see that "lag" at all. How thick are your walls? How much insulation is in your roof?

I do get a lot of solar gain on sunny days that warms the house and then it slowly cools down over my hours, but normally if it heats or cools outside the house does the same by a few degrees. Very happy though.

Thanks for sharing.

Our walls are 2.5" EPS, 6" concrete, 2.5" EPS, and the ceiling is R-44. Maybe I misunderstood what you were looking for in lag. On the second graph, the green (outside) temp drastically rose by 20 degrees from about 9:00 AM to noon, and then held there until about 3:00 PM at which point the temp slowly dropped by 10 degrees. In contrast, the white (inside) temp barely moved until the outside temp started to fall, and then continued to rise or remain stable while the outside dropped. That tracking difference is about six hours.

I went back and looked at the Nudura link you provided. (My house was not built with their system, for what it's worth.) I then used my Ecobee data to average the temp by hour for inside and outside, and I used a room that has all ICF walls and no windows, to help minimize any solar gain. Here's the same graph that shows a lag that's very similar to their graph for the same day I mentioned above.

The peak outside temp (red) was 72.3 degrees at 2:00 PM. The peak indoor temp (blue) was 69 degrees at 9:00 PM, so there was about a seven hour lag between peak temps.

Their graph has a wider spread for both indoor and outdoor temps, but the effect is the same.

You are showing exactly what I was asking for. In my house the temp inside and outside peek at the exact same time, yours does not.

I miss read the Nudura picture. I thought they were displaying inside and outside temps and they were using inner wall core temp as sailawayrb nicely pointed out.

I have R30 walls on my first floor which are Global Block (Build Blocks waffle core, so there is about 35% less concrete but more insulation) the lower floor is standard ICF block at 6 inches(half of it is in the ground half isn't). Exterior of the house is all brick and ceiling is R30 right now, plan on going to R60 later this year when funds come available. Location is lower Michigan.

Thanks for sharing your information and graphs. Still curious what others see in their houses.

Posted By newbostonconst on 30 Jan 2018 10:19 AM
You are showing exactly what I was asking for. In my house the temp inside and outside peek at the exact same time, yours does not.

The graph you posted was while your furnace was running. Do you have any more history from your Ecobee? If so, try to find some dates where there was no heating or AC running, and you might see the results you're looking for.

I have looked at this years heating season and every day is the same. I circulate hot water through my heated floors constantly from my hot water tank so heat input is constant. My geo hot water heater kicks on when thermostat calls for heat which isn't that often(like once a day, if that). It should give a more consistent look at what is going on then if the heat were constantly turning on and off. I have a six inch heated slab in the basement and 1.5 inch slab over all the upper floor so lots of mass.

Thanks

Also, I'd suspect that the GlobalBlock, seeing as it's creating a concrete lattice, rather than a solid concrete wall, wouldn't have QUITE the same thermal lag characteristics is a solid-core ICF setup.

Yes, I think you might be correct. That is why I asked how thick your ICF was. Would like to see data from someone with a 8 inch wall.