Hey Home Energy Pro's!

Tim Reynolds here in Des Moines Iowa, I have been able to work with a group here that is trying out a hot roof system in a remodel. Now I can get behind the air sealing properties and understand that the foam has a high r value per inch, but there are a few problems that I am concerned about. 

1. The SP-Foam contractors didn't completely seal the roof decking, thermal bridging is occurring and some wicked ice dams have been created.

2. From what I understand here in Iowa we recommend around R-40/R-60 within the attic space,(after an air barrier is installed) with the foam only covering the decking say 3.5 inches  given that the foam R value is true 6.5 we will only have around a R-22 with a great air seal.

3. The stack effect is still in play with the attic now sitting at a average temp of 62 - 65 degrees (wouldn't we want the heat within the living space?) 

4. What if any concerns should I consider if I bring the air barrier from the roof to the attic floor? What if additional insulation is added upon the floor to bring the attic system to a proper r - value?

Thanks for any assistance.

Views: 1364

Reply to This

Replies to This Discussion

#1 - it is not a properly done hot roof, the sheathing should have all been sprayed / covered & even the joists

#2 - truthfully you can get less than code minimums when done right, but you should have at least 6" total or at least a 36 - 42

#3 - stack effect is meaningless - you will generally end up with the same or reduced heating & cooling loads with a hot roof system especially if you have any mechanicals in the attic

#4 - the air barrier essentially becomes the foam, but you should still air seal between the living & semi-conditioned spaces (you did remove the attic floor insulation right?)


Hey Sean,

Thanks for the response!  I was called into this project after the fact to touch up leaks from SP Foam contractors. 

1. Kind of what I was thinking

2. R-36/ R-42 would be reasonable, but I feel R-22 is too low for our zone. 

3. Maybe stack effect wasn't what I wanted to use there, but you're #4 is more what I was looking for 

4. Yes all existing insulation was removed this was a complete gut and renovation. 

In the above article it states that the attic floor should be sealed to help prevent the spread of fire, and not typically for air sealing - I guess my next question is why would we allow the heat transfer into the attic and not try and keep it within the conditioned living space? 



It isn't going to be easy to fix this, problems as Sean noted.  I see two approaches.

1.  Bring the spray foam contractors back and double what they installed and as Sean stated cover the rafters completely.  Check local codes to see if a fire barrier is required.

2.  Abandon the foam on the roof deck and move the air barrier and insulation to the ceiling below.  BUT, you would need to ventilate the attic space as would normally be done.

The problem with just adding insulation to the attic floor is lowering the temperature of the foam above, along with trapping the moisture as it migrates up there.  Even closing off rooms on the second floor of home can result in a mild farm.

A third approach might be to fill out the rafters with other insulation and then apply a layer of rigid foam over the exposed bottoms of the rafters, but there are guidelines for this which I don't have at my finger tips.

The bottom line is to only use a hot roof where absolutely necessary.  In southern climates with mechanical equipment and ducts in the attic, it is necessary.  In a northers climate it is most often less expensive and better to leave the attic as an attic.


Thanks Bud!!

I think having the foam guys back will ultimately be the best solution. What are your thoughts about sealing the attic floor?

Thanks -Tim

All openings that allow a fire to spread need to be sealed with a fire rated product.  Large bypasses are of course a priority.  One of the issues with a hot roof is the conditioning of the resulting space.  Often they are called semi-conditioned, but some heat and air exchange with the house below is preferred.  Yes, that attic space will get warm, but that ensures that no condensation will form.  Circulating some of the conditioned air from the house controls the humidity.


I got to say Bud has nailed most of it BUT... in a cold climate a hot roof is still a great option especially for those with cut up roofs, or ones where you cant get full depth out to the edge of the wall sheathing, or you want a higher performing structure

As for the "but all the heat loss" - you are thinking volume & not actual heat loss - air is an amazing conductor, easy to heat & cool, but rarely where you lose or gain heat - most comes from the shell & components - you also got to remember that semi-conditioned crawl & attic spaces are great buffer spaces


Home Energy Pros

Home Energy Pros was founded by the developers of Home Energy Saver Pro (sponsored by the U.S. Department of Energy,) and brought to you in partnership with Home Energy magazine.  Home Energy Pros is sponsored by the Better Buildings Residential Network.

Latest Activity

Alfie Davis posted a photo

Handschuhe im Wandel der Zeit

Im Laufe der Jahrhunderte haben sich #Handschuhe zu dem entwickelt, was sie heute sind:…
1 hour ago
Dennis Heidner replied to Rod Fox's discussion Utility Meter Accuracy
"For Rod, second part... When I see really large electric services... my first reaction my be…"
11 hours ago
Dennis Heidner replied to Rod Fox's discussion Utility Meter Accuracy
"Rod,  I did see earlier that you had calculated the loads.  The key is even though a…"
12 hours ago
Grzegorz is now a member of Home Energy Pros
17 hours ago
Rod Fox replied to Rod Fox's discussion Utility Meter Accuracy
"Dennis, that's a ton of useful info and it will take me a bit to fully assimilate those…"
17 hours ago
Ben Haas replied to Rod Fox's discussion Utility Meter Accuracy
"Yup..... 1200 amp service to the compound.....  3 -400 amp panels, all single phase since 3…"
Walter Ahlgrim replied to Rod Fox's discussion Utility Meter Accuracy
"The more I think about it if the problem cannot be the current transformers reading small loads,…"
Dennis Heidner replied to Rod Fox's discussion Utility Meter Accuracy
"Note to readers 0.5A instead of 5A for reading with A/D.... the edit to correct that didn't…"

© 2015   Created by Lawrence Berkeley National Laboratory.   Powered by

Badges  |  Report an Issue  |  Terms of Service