Here it is a 2 X 8 exterior wall with R-27 FG in the wall cavity. I have a heat sensor between the exterior plywood and the Hardi-Board and another on the wall cavity side of the drywall. A/C is set at 80 degrees, outside sensor is reading 107.5 degrees (south facing wall at 5pm in Fl) drywall sensor reads 82.4. Does the 25.1 degree difference sound about right?
Discussion Forum
Discussion Forum
Up Next
Video Shorts
Featured Story
A white piece of paper taped to the back of the bubble window on a spirit level makes the bubble easier to see in low-light conditions.
Highlights
"I have learned so much thanks to the searchable articles on the FHB website. I can confidently say that I expect to be a life-long subscriber." - M.K.
Replies
Roughly right. The A/C setting of 80 F is just so close, and you've got a small amount of R represented by the drywall layer and air film coefficient between drywall surface and bulk room air. Expected temperature rise from bulk air temp, whatever that is, and cavity side of the drywall will be (roughly) the 25.1 degrees across the wall cavity and sheathing times the R value of that drywall/air film layer divided by the total R across which you get the 25.1 degrees.
If the inside layer gives, say, R1, then 25.1 times 0.5/27= 0.037 or only 0.5 degrees. Now, that is not the 2.4 degrees from 82.4 down to A/C setting of 80, but I doubt the temperature at the inside surface of the wall is at 80. It's going to be somewhat warmer, or you wouldn't be having heat transfer from wall to the bulk air. Then, too, I doubt you really know that the room air is at 80 F either. The A/C will cycle on and off and the temperature of the air will swing around some temperature that is in the vicinity of 80 F, whatever that might be.
I doubt you'd be able to use your measurements to determine anything better than "ballpark" ok.
Edited 8/25/2009 1:33 pm ET by DickRussell
I guess that is the one reading that I left off. The room temp at the drywall face is 81 degrees.
Well, now we're down to comparing roughly half a degree "predicted" from overall delta T and relative R of the whole wall vs. the drywall and the "actual" of an apparent 1.4 degree. You can see that it won't take much error in either the nearer sensor or the room-side surface measurement to result in a big difference between predicted and actual.Next I guess we'd have to ask where the sensor behind the drywall is - mid-field vs. next to a stud. If it is next to a stud, then perhaps the "cavity" total R is more like just the roughly 8 inches of wood at R1 to 1.2 per inch vs. whatever the drywall gives. If the temperature difference ratio is 1.4/8 or 0.18, then the drywall R would be almost 5, which clearly is too high for a half or 5/8 inch piece of gypsum. But the point is that the location of the sensor is important too. Then we could get into tightness of the wall cavity and whether you have convective currents bringing hot outside air into the cavity and distorting what the temperature profile through the cavity otherwise would be. The bottom line is that in only crude terms, the temperatures you are seeing make sense at some crude level of accuracy, but clearly there is enough uncertainty in the performance of the wall under the conditions experienced to make the analysis rather fuzzy.
The sensor on the cavity side of the drywall is mid point of the cavity, both horizontally and vertically. The exterior sheating was fastened with 8d nails and PL construction adhesive. All penetrations are sealed, so for all practical purposes the insulated bay is sealed for air movement. The FG was installed with the tabs to the inside of the stud thus giving a 1/2'-3/4' air space between the paper and the drywall.
If I am following your calculations correctly, the reading that I am getting are very close to calculations for a FG R-27 2X8 wall cavity. That is what I wanted to confirm. Three bays over I have a 2X4 wall cavity; I have used a nominal 1" of closed cell foam, with 2" of FG pushed against the foam and a layer of Astro-Foil against the FG and stapled to leave a 1/2" air space between the Astro-Foil and the drywall and I am getting close to the same readings. Actually the 2X4 wall is running .4 degrees cooler on the wall cavity side of the drywall. These calculations holding true I have achieved a minimum of R-27 in a 2X4 wall cavity. I impressed myself.
From this site: http://www.allwallsystem.com/design/RValueTable.html
The R of a half inch of drywall is 0.45. That material is fairly uniform, so I wouldn't expect much uncertainty in using that value. Use 0.56 if the drywall is 5/8" thick.If we use the 25.1 degrees across the insulation and sheathing and the 1.4 degrees across the drywall, then the effective R of the insulation and sheathing would be (25.1/1.4) * 0.45 = 8.06, assuming the temperature profile is steady and there are no convective currents. If the drywall is 5/8", the FG/sheathing R comes to 10.1.While FG batts have been flogged here and elsewhere as to actual performance relative to the lab-tested R value claimed on the paper, I wouldn't think that performing at only R8 instead of R27 is very believable. R13-15 I could accept more easily.At this point, I can suggest only the following conclusions.1. Since the R of the drywall is small relative to that of the insulation, the temperature difference across the drywall will be similarly small compared to that across the insulation. It is, although the actual numbers still don't calculate out as well as hoped.2. With only one set of sensors to use, and with temperature difference across the drywall of only 1.4 degrees, it wouldn't take much error in measurement to produce a large change in the R calculated for the FG batt. Too bad there weren't more sets of sensors, to rule out something funny going on local to where the single set are.3. Generally good advice for installing FG batts is to have no voids within the wall cavity. With an air gap between the FG paper and the drywall, providing an open cavity from ceiling to floor, the obvious thing to consider next is the possibility of heated air next to the sheathing rising up ever so slowly, slipping between the top of the paper on the inside and the top wall plate, and settling down through that air gap between the FG paper and the back of the drywall as it cools, to slip under the bottom edge of the paper and return to the warm side of the cavity. This is a bit speculative, I know, but conceivable.Bottom line here? I don't think there is enough data to really pin down what the effective R of the FG really is.Your 2x4 test assembly should give nominally R7 for the CC foam, plus another 7 for the 2" of FG "pushed against the foam" (making it tighter, more dense, somewhat higher R than typical R11/3.5"?). That's a total of R14. What is that really doing? If the temperature difference across the drywall is now only 1.0 degree, and the delta across the insulation/sheathing now 0.4 higher, we would calculate the effective R of the insulation to be (25.5/1.0) * 0.45 = 11.5. If that drywall is 5/8", the calc would give R16.You don't really have lab conditions, so you can expect just so much accuracy in what you back out of the data, but to some extent the comparative results do line up. We commend you on your efforts.
Thanks for all of the input. The base question is; am I getting reading that are close to real for R-27 FG insulation? I have a 22' 2X8 wall and I have sensors in three bays and they are reading within .1 degrees of one another. When the outside temp is 105 or higher should I see a difference of 25+ degrees? The sensors are all mid point horizontally and vertically. I have a piece of aluminum tape attached to the cavity side of the drywall and then the sensor is attached to that and cover with another piece of aluminum tape.
I'd say it is roughly correct. You can calculate the temps at any surface/point in the construction if you want to ... using R-values. It's a relatively simple calc.
Now you are comparing it w/ another type of construction that uses a radiant barrier as a way to control energy transfer ... that isn't as easily done mathematically.
Before you get too excited about the thin wall high performance, remember that as soon as that radiant barrier starts to get dirty, your performance drops dramatically.
I agree w/ the other poster, too. You are starting to make scientific comparisons and accuracy may become very important. Are you applying your sensors per 'laboratory standards' (whatever that is, but I'm sure there must be some such)? If not, then your measurements may not have the accuracy required to make the comparisons you want. Is your outside temp use a standard shielded sensor? If not, that temp may be inaccurate.
It's one thing to take a few measurements ... and say they are generally accurate. Quite another to begin to compare two systems that may require a greater level of precision in measurements.
I have used digital thermometers attached as I have understood they are attached in testing. I am not trying to publish scientific papers, just trying to find the expected temp differential using FG R-27. As far as the reflective barrier getting dirty, I don't know how it would. It is the last product in the wall cavity and then the drywall was applied. Time does not seem to make a difference in a wall. Last spring I removed a wall section that I put in 8+years ago and the Astro-Foil looks the same as what came in the new roll.
Every time you reply you're adding additional info that impresses me more about how thorough you've been in trying to collect good data. Three sets of sensors, all with consistent results, is good confirmation.Well, if it is 105 out and your wall is at 80, you do have a 25 degree delta T across that wall, by definition. the distribution of temperature drops as you cross the drywall, FG, and sheathing will be determined by their respective R values, in the absence of any convective or radiative heat transport.For steady-state conductive-only heat transfer, the fraction of the total dT you get for any one layer is that layer's R divided by the total R across the wall. This follows from Q per unit area = U * dT, or = dT / R. At steady state, with no heat accumulating in any layer, the Q across each layer must be the same, so dT/R is constant for each layer is the same as total dT/total R across all the layers.If the overall dT is 25 degrees, the drywall R is 0.45 [OK, is it half or 5/8" drywall?] and if the FG/sheathing R is the hoped for 27, then the dT across the drywall alone for purely conductive flow only should be 0.45/27.45 or 0.0164 of the total, or 0.4 degree. But you are seeing 1.4 degree, IIRC. That is what doesn't line up. If you accept the R of the DW and the dT across it, the effective R of the FG must be a low lower, or about 0.4/1.4 times 27, giving about 8, as I posted before, which seems low. This is why I started groping for other explanations, challenging the assumption that you have only conductive heat transfer, with no convective transport.It shouldn't take much of a convective flow of warm air crossing the top edge of the paper VR of the FG batt to boost the temp of the air in back of the DW by one lousy degree, or less if you have 5/8" DW. Now, if you come back and tell us that top and bottom edges of the paper were sealed well against the framing too, so that even a little convective looping is highly unlikely, then I'm out of explanations.There is one flaw in my still trying to explain the higher than expected dT across the DW as being due to convective looping. You report getting almost the same readings for all three sets of sensors. If the amount of flow is limited by how big a space there is between top and bottom edges of the paper and the framing, then I wouldn't expect exactly the same results in all three bays. But if the flow isn't really limited by the size of that gap but by the 8 feet of FG batt that creates a pseudo-dead air space, then I may still be ok.[Edit:] There is one other possibility: stack-effect leakage of warm outside air into the cavity, part of which may slip into the space behind the DW, up and out through cracks in the framing and into the attic. This would be subject to the same comments about getting the same results in all three bays where you have sensors.[Edit:] As I said earlier, about that other wall construction you reported on, the insulation layer R calculated from the dT across the DW comes out a lot closer to what you get by adding up the R values of the two layers, suggesting that convective transport in that wall is not significant. As to the value of the foil surface, since the foil temperature and DW inside temperature are nearly the same, I wouldn't expect much help from the foil surface as a radiant barrier, positioned there.
Edited 8/27/2009 4:36 pm ET by DickRussell
1/2" dw. The bays are sealed as best I could do without going crazy. All penetrations for electrical are sealed. Sheating to framing members are sealed. Drywall is not glued (my son had a dw'r start the job while I was on the road). The top and bottom of the paper are not stapled, but the batts were cut long and the top 1/2 of the batts are held up with insulation wire. I have seen enough FG where the staples let loose and the 8' batt sags. I guess next step will be to give you the reading when the A/C is fully functional. I think that the 80 degree set point of the A/C, giving me a 81-81.5 degree inside surface temp does not give me enough of a spread. You have helped me understand the installation of the various forms of insulation. Thank you very much. The sensors (all 11 of them will become part of the house, so I can keep track of the readings. I will also be getting a thermal image next month of the entire home.
I think your calcs may be off. You say R-27 plus the drywall ... that's it. But you have the exterior sheathing and the exterior siding plus you have two air films: inside and out. Inside might be predictable ... depending on the airflow characteristics of the space. Outside ... hmm ... is it still air or moving. Then you have to find an air film U-value that fits the condition ... could estimate or interpolate between inside and design/assumed outside air film value.
Am I reading you right? Maybe I missed something.
I wasn't trying to pin down total R value of the wall. In his first post, the OP said he had the outside sensor between sheathing and siding, so I ignored siding and any outside air film. Similarly, I ignored the inside air film, since the sensors are on the DW surface, if I understand the info correctly.All I was trying to do was to see if the numbers made sense. In the case of his second assembly, with foam and FG, the R of the insulation backed out of the dT across the DW and across the insulation actually was in the ballpark of what it ought to be. However, I don't think his thought that the second assembly achieved R27 of the first. I think it's more likely that the first simply isn't performing at R27 under those conditions. That's why I started looking for ways to explain what could be happening, knowing full well that the data available probably aren't telling enough of the whole story.It has been nice, though, to see some actual data and have fun play around with seeing what lines up and what doesn't, then try to explain it all.
Definately interesting discussion.
One note I thought about but didn't mention. Given the same outdoor and indoor temps, the delta T across the insulation will be roughly the same ... it's the energy use of the A/C unit that would be different in the two cases. That is ... given the same overall delta T, the lower R-value (or lower performer whether R-Value or radiant value), will transfer more heat across it in a steady state situation.
In this sense, I'm not sure that casual measuring of temps is going to show anything. It would take a lab to show the temp difference on the back side of drywall given the same static indoor/outdoor conditions.
I'd have to think about this, but the difference in temp on the back side of sheet rock given two insulated, but significantly different insulation levels may not be that much. Given the air film assumptions, this can easily be calculated (note the 'assumption' phrase).
Ok, hoping to get some better reading next week. A/C is up and running and hoping that the weather will cooperate and heat up.
Attached how? Digital don't mean much. What are your sensors? Have you considered that your method of attaching the sensors may be affecting the very values you are trying to get?
Remember ... one insulated wall to the next ... the drop across the drywall will be pretty similar. VERY small differences relative to the overall drop in temp across the insulation.
You can't simulate a hot box test using real world conditions. They are different animals. And you are trying to compare the two under identical conditions. While this is a very interesting scenario, I suspect that you may not be able to get the information you are looking for ... well as easy as we've discussed here. Your measuring equipment may not be up to the task, either. If you are e.g. using a common digital thermometer w/ thermocouple/thermister probes, you may not get the accuracy needed. You also have to have the right thermocouples.
But I'm no science data geek, but I think you usually need some pretty sophisticated equipment to really do what it is you are trying to do.
Once again, I am not looking to publish a scientific paper. I have some experience in comparitive studies and I am looking for real life results. No, this is not a ASTM lab, and I do not have endless $ in equipment, and I am not trying to bad mouth FG insulation, I just want to compare two "as built" wall section in my son's house and I believe that Dick Russel has assisted me in meeting that goal. It was also a matter of 45+years in the building business and I have not heard of anyone actually measuring wall cavity temps to see what insulation is doing. Oh, yes they do it in the labs, but I am talking of a living, breathing home.
Yes, I understand. And you are right ... actual applications can be good learning tools and it's great you're sharing it. I'm just adding two cents for what it may or may not be worth. Take what works and discard the rest.
I appreciate you input, especially now that you now my intentions. I never discard anything, maybe it will work someplace else.