GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Why is more insulation still required in roofs than walls in very tight buildings?

matt_in_CA | Posted in GBA Pro Help on

In researching high efficiency buildings, and deep energy retrofits, I am confused why in a tight building, it is still suggested to insulate the roof/ceiling more than the walls.

I understand that generally more insulation is required in the roof/attic than in the walls. My understanding for this is that buildings lose more energy through their roofs due to pressure gradients (hot air rising, creating increased pressure near the roof) and thermal gradients (air near the roof will be hotter, and therefore more insulation is required to limit increased energy loss due to larger temperature differential across the insulation.

I am confused why this rule of increased insulation in the ceiling still applies in very tight building envelopes (particularly ones with controlled air circulation/ventilation). In a very tight building (eg a REMOTE or PERSIST building ) there should be very little energy loss through the roof/ceiling due to pressure gradients (since there should be very little air leakage possible) and there should be very little energy loss due to thermal gradients, since the temperature near the ceiling should be very close to the temperature in the rest of the building (due controlled circulation of air throughout the building ). Can somebody please explain the factors that I have missed?

Thanks very much,

Matt (in Canada)

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. GBA Editor
    Martin Holladay | | #1

    The reason is simple: it's usually cheaper (and easier) to install thick insulation in a roof assembly or on the floor of an attic than in (or on) a wall.

    No one wants 24-inch-thick walls, evidently. But it's relatively cheap and easy to install 24 inches of insulation on your attic floor.

  2. matt_in_CA | | #2

    So, in that case if I were to do a PERSIST retrofit with foam on the outside of the walls and roof, it may make the most sense to use the same amount of foam on the walls and roof?

    Thanks for the answer,

  3. GBA Editor
    Martin Holladay | | #3

    Well, more insulation is always better.

    But once you reach the end of your budget, the limit of available screw length, or R-60 -- which ever comes first -- you're done.

  4. matt_in_CA | | #4

    Yeah, that is exactly why I'm thinking about this. It seems that instead of having R40 walls and R60 roof, you would get better performance with eg R50 walls and R50 roof using approx the same amount of insulation.
    Thanks again for taking the time discuss this.

  5. user-723121 | | #5

    Actually, Passive House promotes a more uniform R-value for the building envelope. I adhere more to the 10-20-40-60 equation but I should probably give this some thought, old habits die hard.

  6. dickrussell | | #6

    Given a certain amount of insulation being used for a tight house, uniform distribution throughout all parts of the shell exposed to the outside air in general does give minimum heat loss. Consider an extreme example, in which all of the attic floor insulation is piled up on half the area, giving very, very high R value there, but none elsewhere. Obviously this is not optimum. The 10-20-40-60 equation is a reasonable balance between results achieved and cost or difficulty of providing good insulation levels in the various parts of the overall envelope. The numbers in this "formula" obviously could be varied according to construction techniques being applied.

  7. Mr_Gerbik | | #7

    Maybe the extra insulation argument would hold more true in a traditional vented attic, where the attic might see some pretty warm temperatures in the summer, potentially much warmer than the walls?

  8. jameshowison | | #8

    My thoughts are similar to Ryan W's. Surely it depends on the delta-T that the surface experiences? And in a cooling climate (or rather season) the angle and constancy of impact of the sun's rays are such that I would have thought the roof/ceiling would be exposed to much higher delta-T.

    Not the case?

  9. jinmtvt | | #9

    I think that what Martin means is that it is much "easier" or "cheaper" to reach R60 on a regular attic/roof, than it is on vertical walls...thus since we want as much insulation as possible, the cheaper gets the most to bring up efficiency.

    on a 2 story house, the walls usually make up for 2-3 times as much area as the roof,
    so high R value on ceiling/attic/roof per sqft is less rewarding than the same added R on walls.

    Dick Russell: the 10-20-40-60 classic insulation scheme proposes exactly that,
    place insulation where it is the cheapest .

    underslab requires dense foam boards which is generally much more expensive than blown stuff and batts installed in the attic.

    Of course, the delta-T of each situation tells where the insulation will have the largest payback.

  10. Mr_Gerbik | | #10

    Sorry, in re reading my post I basically re-stated the part you said you already understood. My mistake!

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |