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

Minimizing effects of a steel thermal bridge

jimgove30 | Posted in General Questions on

Hi All, 

I have a detail on one wall that has a (4) 3′ steel outriggers on posts, supporting a “floating” canopy over my deck and sliders.  The roof detail on the plan appears to be a generic framing detail and calls for EPDM membrane, 2″ xps and 5/8 sheathing. In order to minimize the impact of my climate zone 6 temps, I’m thinking I should wrap the remaining exposed steel outrigger with foam as well. I’m sure anything is better than nothing, but would 1.5-2″ of xps or polyiso be enough to limit the thermal bridging?

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. Expert Member
    Akos | | #1

    Definitely wrap it in rigid. 2" sounds reasonable as it is not a lot of surface area. Since steel is essentially R0, even getting as little as R5 on there will make a big difference, more is always better.

    What I've done is over the rigid on top is install PT lumber on flat attached to the steel with self tappers. You can attach the rest of your framing on this lumber instead of having to deal with any more steel connections. The holes for the self tappers in the wood need to be predrilled otherwise the drill point will break when it tries to go through the steel.

  2. DC_Contrarian | | #2

    How do those outriggers end on the inside?

    You need to think of steel as essentially a hole in the building, for practical purposes it has no insulating value. You need to enclose it on all sides.

  3. jimgove30 | | #3

    So, the outrigger is welded to a post that runs from the foundation up to the top plate over those transom windows. I was thinking the zipR9 on the outside could be sealed to some rigid wrapped around like Akos mentioned. Sounds like I'm on the right track. Thanks for the tips!

    1. DC_Contrarian | | #4

      Unless you can insulate the entire outside, exterior insulation won't do much good. Similarly, unless you can insulate the entire inside, interior insulation won't do much good. It's like only waterproofing half of the bottom of a boat.

  4. Expert Member
    BILL WICHERS | | #5

    I agree with the suggestions to insulate that steel. I would try to get it insulated on the exterior as much as possible, and as DC mentioned, you really need to FULLY enclose it in insulation to get a real benefit here.

    I also recommend you prime the steel first (which may have already been done by the steel fabricator), to limit future rust. Rustoleum's "Rusty Metal Primer" is good for this. They make a "Clean Metal Primer" too, but I almost always just use the "Rusty" version. You can stick with primer only, or put a coat of one of their enamels over it. The idea isn't to make it look nice though, you just need the heavy coat for rust prevention.


  5. Expert Member
    Akos | | #6

    We can caluculate lot of this. For simplicity I'm assuming the outrigger is 4x3, so 4' of it has 4.7 sqft of area, for a total ~19 sqft for all of them.

    Assuming the supporting post is complete inside conditioned space and no insulation between it and indoors, the post will stay pretty much at room temperature. Since the steel is pretty conductive, the outriggers will a bit colder but be near enough to this temperature for a ballpark calculation.

    So 70f indoors, 0f outdoors (70F delta), bare outrigger has maybe an R0.25 because of exterior air film.

    70F*19sqft/R0.25=5320 BTU/h.

    Cover the outriggers in in R5:
    70F*19sqft/R5=266 BTU/h, so pretty much squat.

    The rest is mostly a ratio, if you cover only half the steel, you get 2793 BTU/h.

    So definitely wrap it in something, even a layer of thicker lumber is way better than nothing.

    1. DC_Contrarian | | #7

      But the important point is if you can't completely enclose it, it's not much better than nothing.

      We talk a lot here about thermal bridging, and my opinion is we probably talk about it too much. But when you're using steel it's the entire game. When you have an assembly with parallel heat paths, the formula for r value is 1/(1/r1+1/r2). If r1 is zero, 1/r1 is infinity, and anything plus infinity is infinity, and one over infinity is zero. So if any part of the assembly has r value of zero, the whole assembly has r value of zero.

      Now steel doesn't quite have an r-value of zero, but it's close, so close that any source I've been able to find says just assume zero. And yeah, you get a little something from the air film, but not much. He's got a complicated assembly where the outrigger is welded to a post and a header over the windows. That entire assembly needs to be insulated, and it all needs to be on the same side of the building envelope.

      1. Expert Member
        Akos | | #8

        With steel, there should always be rigid between steel framing and outdoors, I'm hoping this is the case for the OP for the posts and lintels.

        It doesn't take a lot of steel to loose 3/4 to a 1 ton of heat. The biggest culprit around here is a big wide flange lintel above a wide opening. You can feel the cold in the floor of the level above the opening.

    2. gusfhb | | #9

      R numbers below 1 become irrational
      You have a R value for the outside air layer, and for the inside air layer, and actually the steel is a non zero number, so you have a minimum of 1. It would not be a solid steel bar, but most likely heavy wall tubing

      Oh, and if I stick a 1 mile long 1 inch square bar through the wall of my house it transfers the same amount of heat as a 1 inch long bar.
      The real problem of steel or other low R value materials is condensation damage.

      1. Expert Member
        Akos | | #13

        The only real way to get the actual heat loss is by setting it up something like Therm. Not hard, but not easy either.

        The hand calculation gets you into a ballpark, which is to say the losses are not small.

        1. gusfhb | | #18

          3 supports total .5 square feet of surface area 50 Delta T R1
          50 BTU/hr

          IF you could fill the void inside the bar where it penetrates the enclosure with foam, it should help greatly, cuts it down to few sq inches. Maybe drill a 1/4 hole in the bottom and spray some high fill can foam in.
          Again condensation not literal heat loss is the biggest issue I would think

  6. aaron_p | | #10

    Hindsight is always 20/20 so I think the insulation suggestions are probably the most cost effective, but here are some other thoughts.

    Sometimes in existing conditions we will specify some Areogel enhanced coating/paint - it isn't a miracle product, but the full coverage and adhesion can usually help with the condensation/corrosion risks. (

    Not sure what your roof detail looks like, but if you could insulate the end, sides, and bottom, the last concern would be the bearing of the roof deck. If it is metal roof deck you could look at going to a wood fiber roof deck (Tectum is the common named product) that would minimize that bridging or going to a more complicated thermal clip system ( or another thermal spacer system.

    Probably for future reference: Commercially, we would try to isolate this steel with a bolted thermal break products like this: ( and Between the thermal pad and some stainless bolts this can be a simpler solution than wrapping the steel. These typically are engineered, so harder to do in a retrofit.

    1. Expert Member
      Akos | | #11


      There are always great commercial solution out there, the problem with a small project like this is they are simply impossible to get. Add on the additional engineer costs, wrapping a steel in rigid (say couple of hours of labor) seems like a good deal.

      1. aaron_p | | #17

        Oh I know - I understand that these products and distribution systems can be quite frustrating. Even if the product is expensive - in residential there may be so few instances that it could be worth if for these companies to have a residential distribution method or prescriptive designs. I snoop on here as sometimes there are better residential energy performance design processes than the commercial side, so I think there is a benefit to sharing designs from the other side to potentially come up with better solutions.

        I don't know what the roof system attachment looks like and wrapping in rigid may not be possible due to connection/compression issues with the canopy. Trying to offer some out of the box alternatives in the case that can spark a good solution. The zip on the sides and bottom with lumber on top may be the best value solution depending on structural requirements.

        1. jimgove30 | | #21

          Thanks for the suggestions! I really need to check with the architect and see if he really had any plan at all?

  7. jimgove30 | | #12

    The post has zip r9 to the outside. There is an air space between the inside of the zip polyiso and the post that I wish I had filed with another 1.5" of polyiso but they buttoned it up before I found it. Maybe I can still do that. I'll see if we can wrap the outrigger portion with rigid and seal it to the zip. But if I understand all the suggestions, it's messy to do whatever I can to wrap it up now. Thank so much.

  8. Expert Member
    Michael Maines | | #14

    Can you insulate the steel post at the interior? Even if it means making your wall slightly thicker, even R-5 (~1" of foam) would make a big difference in heat loss.

    Is the steel filled with foam? There is a risk of condensation inside the steel members which can lead to rusting and failure, or at minimum, rusty water weeping out.

    1. jimgove30 | | #15

      Hi Michael, are you suggesting to insulate both the interior and exterior? I guess I was thinking that by keeping the insulation on the exterior, the lack of insulation on the interior would help the warmer side of the post control the condensation. Perhaps I was wrong on that. I'll take a look at the interior side, and maybe I can sneak an inch or so in there. Thank you.

      1. Expert Member
        Michael Maines | | #16

        Jim, because the projecting arms are such large thermal bridges, I don't think insulating the exterior of the post will do as much for heat loss or condensation as going with interior insulation only, or a mix. Most of the steel is going to be very close to the outdoor temperature, with or without exterior insulation.

        I'm assuming you can't fully insulate the projecting arm. If you can fully insulate the steel from the outdoors, then that's probably the best approach.

        1. Expert Member
          Akos | | #19

          I'm going to disagree with this.

          From the OP, the posts have ZipR9 to the outside, they will be pretty warm. The outriggers, even when exposed, are a much smaller area than the posts, so the outriggers will be about the same temperature as the posts.

          With steel in cold climate, you always want to keep it either fully inside or fully outside conditioned space. The in-between ones are always issues, I've see steel beams sweat in older commercial construction in the winter. Unlike lumber steel starts sweating right away when its temperature falls bellow dew point, instant water damage. This is why when the steel is inside walls or roofs, the safest is to have it close to conditioned space to avoid the risk of condensation from the inevitable air leaks.

          1. Expert Member
            Michael Maines | | #20

            I'm assuming the outriggers won't be wrapped in insulation. How would they be the same temperature as the posts? They would be at the outdoor air temperature, and where they meet the posts they will draw heat from the post. How far up and down the post the low temperature extends I can't say.

            I don't like using terms like "sweat" because they introduce confusion. Steel will condense water out of the air if the steel is below the dewpoint temperature. I agree that it's best to keep all of the steel on the interior of the thermal barrier, and the next best would be to keep all of the steel on the exterior of the thermal barrier. I do my best to avoid situations like this one because it's hard to predict exactly what will happen.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |