Hello internet friends,

I’m working on replacing an exterior door for my 1940’s walk-out basement in NC. The basement walls are brick, with wood frame built on top.

The existing door was framed in with 2x8 “king studs”/door jamb (the studs went the full height of the wall…and were also used as the door frame) and a “header” made of four 2x8’s stacked up to the sill plate on top of the brick walls. There is no brick above the door at all. I don’t have a great photo but here’s a not great one:

not great picture of old door

Because of this - interesting - structural design I assumed the door frame (or lack thereof) wasn’t bearing much load. In hindsight, unwise.

Once I removed the door and frame, I realized the sill plate wasn’t continuous across the span of the door opening and was toe-nailed together with some sizable 1940’s nails, which raised some concerns that I might should be installing a proper header above the new door.

break in the sill plate above the door

The bind I find myself in is this: I was greedy and opted for a 36” door for my 39-1/2” opening, which means I could only fit a treated 1x on either side of the door frame without having to cut brick. This left juuuust enough room to get the door in, but now means whatever load is above the door is being carried down by two measly 1x4’s bolted into the adjacent brick.

So…any thoughts on course of action here?

Should I:

  • Notch the top course-and-a-half of brick on either side and install a steel (concrete? Wood?) lintel?
  • Widen the brick opening a couple inches and frame the opening with 2x8’s and a quadruple 2x4 (maybe even 2x5) header?
  • Some combination of the above?
  • Leave it alone, it’s fine?
  • Something else?

I’ve included some photos of the new door installed (I’ve only screwed it in place and plugged gaps with some insulation for now, since I might need to remove it to fix this issue).

Your time and expertise is appreciated! I’m sure I don’t have to say this but let me know what info I’ve missed and/or poorly explained.

new framing and header

detail shot of new framing

  • 1
    A 2x6 in pockets (options are remove 4 bricks and add a board, or forty bricks and then rehang the door, and still add boards...). Or less than a x6; w/e fits in there but still leaves the 1/2" extra for the rough opening, for so when it gets replaced again in 20y you don't have to cut the bottom of the new door off. - You need a (wood) header not a (stone) lintel; all it does is carry the floor joists; no masonry above. Pull your 2x4, make it longer and put it in pockets. Then add a 2x6. "Leave it alone, it’s fine?" depends on the pitch of the roof. NC gets ~5" of snow per year. Prob fine.
    – Mazura
    Commented Jan 5 at 5:38
  • 1
    Wood bearing on masonry requires at least 3" of bearing length if you decide to notch the brick.
    – popham
    Commented Jan 5 at 5:53

3 Answers 3


Install the header and jack studs against the inside face of the brick wall instead of in line with the brick wall. This is a good opportunity to construct a finished wall along the entire length of that brick wall. Use Table R602.7(1) below IRC R602.7 to size an adequate header and get your jack stud count, where footnote e indicates that you should probably use the 30 psf snow load columns.

Footnote d under Table R602.7(1) is interesting in how close it comes to bailing you out, but it anticipates wood king studs. Maybe there's some sort of ICC approved bracket for masonry that bails you out. Closest I found was Simpson MBHU brackes, but those would require an engineer, and I don't think I'd trust them for any old brick wall.

  • Why do you have 3 separate answers here? Are you trying to address 3 of the OP's options? There are cases where one person posting 2 or more answers is appropriate, but it's really hard to determine why you've done so. Maybe an edit or two to indicate what's going on would help everyone...
    – FreeMan
    Commented Jan 5 at 12:31
  • As luck (or whomever) would have it, I am planning to build a finished wall on the inside of the brick pretty everywhere else but here (for aesthetic purposes - the brick looks nice) but I can easily ditch that plan and add a framed wall here as well. If I were to build this way, I would essentially want a framed opening in line with the door with appropriately sized headers and jack studs per the cited code, then I could leave the door installed in the brick as-is, correct? Or would I for some reason need to install the door in the “new” wall and leave the brick opening alone?
    – JimboSlice
    Commented Jan 5 at 12:50
  • 1
    @JimboSlice, yeah, I really like the look from that one picture that captures a bit of the window. I would hate to bury it assuming that the local climate is cooperative. I don't see why you couldn't keep the door as-is. I imagine it would look awkward out of plane with the windows.
    – popham
    Commented Jan 5 at 14:36
  • @FreeMan, three distinct solution patterns, therefore three distinct answers. Seems perfectly reasonable. Did you not read at least two of them?
    – popham
    Commented Jan 5 at 14:39
  • 1
    @JimboSlice, ah. I've never found a formal statement in the IRC. You can check out the illustrations under IRC R602.7.2 and IRC R602.7.3, where they both illustrate "header span" as measured inside-to-inside. I computed some headers once for this exact reason, finding that the maximum span was measured inside to inside (with some extra complexity that I won't blabber about).
    – popham
    Commented Jan 6 at 5:03

Following the IRC's "Framing of openings" pattern from IRC R502.10, you can eliminate the two joists over the door opening. With the two joists gone, you're free to install your header at the floor joist elevation instead of just below it.

Adding a trimmer joist to the tripled up joist on the right of the opening and adding a trimmer joist to the joist to the left of the opening, you add enough strength at the trimmer joist locations to carry the load from the two joists that sit on top of the opening. With the trimmers in place, you cut away the two joists from over the door at a 12" offset from the wall, throwing away the little chunks of floor joist from between the cuts and the rim board. Using a doubled up header between the trimmer joists, the cut joists terminate to the header instead of sitting on the wall.

Even with the floor joists loads gone, however, you still have to account for the load from the wall above the door's opening. Assuming that you've got a rim board above the door's opening, follow IRC R602.7.2 to size the header. With those deep floor joists, it should take just one header paired with the rim board (unless you get unlucky and find a splice in the rim board).

Since the trimmer joists are loaded so close to the wall by that header located at 12" offset, there's a very good chance that the trimmers aren't actually necessary. The deflection and bending stress caused by that load location is small, where deflection and bending stress are the typical constraints determining a joist's size. By-the-book, an engineer's analysis would be necessary to cut the corner of adding the two trimmers. With the addition of the two trimmer joists, though, the above methodology fits neatly within the scope of the IRC.


Remove the door and buy one that provides sufficient space for the required jack studs. You determine the header size and number of jack studs from Table R602.7(1) below IRC R602.7. The table tends to confuse people who live in regions without substantial seasonal snow, but footnote e says to use the 30 psf snow columns if you have no snow. I've always assumed that this conservatively accounts for wind pressure loads on the roof.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.