Unbraced Wood Framed Headers

[bigmig]

One more post, on another subject. I keep getting feedback from my framers that they would much prefer to put their structural wood headers (like the header over an exterior door way) down low enough that the header forms the rough opening for the door. This allows them to use less framing, as opposed to keeping the header all the way up under the double top plate. It also gives them a nice nailing surface for door trim.

The problem I was taught, is that headers built like this are unbraced along their compression edge, compared to a header that is high, and securely fastened to the top plate, which is typically braced by perpendicular rafters, trusses or joists.

For those who keep their headers low, do you simply adjust for the CL factor for wood beam stability with every header and swallow the capacity reduction?

The low header makes sense from a constructability standpoint, but again, it is definitely not braced on its compression edge. This is something I want to accommodate vs the ideology that says "because that is they way I have always done it".

Again, any input would be appreciated.

 

[a2mfk]

I thought this was pretty much the standard way of doing headers. But unless you are talking about over a garage door or sliding glass door where the span is considerable, what kind of loads and spans are you dealing with where bending is an issue? I know if you have high uplift, it is advantageous to put the header right below the top plate so you can eliminate some of those connections. Of course in uplift your bottom edge is not braced either.

But to directly answer your question, if your header is unbraced it is unbraced, and that is your capacity.

 

[XR250]

I have never seen an issue in practice with a header that was framed with a cripple wall on top - even in garage door openings.
Another instance of things performing alot better than they should in wood framed construction.

 

[bigmig]

I agree with a2mfk. Low headers without compression edge bracing is a design condition that can be allowed, assuming the engineer properly designs it that way, and the obvious....that the member will work with this reduction. Regarding what we have all seen and not seen regarding failures is irrelevant, as I have witnessed in court. What matters is the calculation you did that says it will work unbraced.

 

[AELLC]

The reduction in CL is usually very small except when you have say, a 3-1/8" GLB spanning 16' for a garage door. When you get to those spans, deflection or allowable span with standard glulam beam camber only becomes significant.

 

[a2mfk]

To touch on Excel's point- I think residential wood-framed construction (USA) performs well and predictably as it should in regards to supporting GRAVITY loads, but this should be no surprise. Mostly because they are overbuilt and overdesigned for the loads they are most likely to encounter during their typical life span in terms of strength, and it seems about right in terms of serviceability (deflection). Gravity load path is usually pretty simple and there are no uplift connections to screw up. Before real structural engineering and codes there were wood framed buildings, and many of them have been around a century or two and have had no failures.

I have always said that residential roofs are way overdesigned for gravity (outside of snow country, and flat roofs with rain ponding problems). The only live load they ever see is when it is getting re-roofed, or the occasional point load from someone doing maintenance or cleaning the gutters. So that 16psf LL you designed your header to support is almost never going to be actually applied to the roof. And the dead load is most likely overestimated. The bigger the tribututary area of your member, the bigger the overestimation of these loads.

In all my forensic engineering work, I have only seen one gravity roof failure, and it was during re-roofing. It also turned out that this section of the roof, for some bizarre reason, did not use pre-engineered trusses but utilized some very undersized conventional framing. The roofers stacked shingles up along the ridge and the ridge beam failed.

Same thing with floor loads for second stories and crawlspaces. 30psf live load for bedrooms, when bedrooms are mostly empty space. I just moved and have a pretty good grasp on what furniture weighs, and I will tell you that it is nowhere near 20'x15' (30psf)= 9000lbs.

I am not advocating any changes to any of these loads or to the codes in terms of gravity loads, and I fully understand why they are as high as they are, and because we cannot predict where the heavy furniture or loads will be placed so we design the whole structure for a uniform load. I am actually pointing out the code loads are doing their job and probably cover a LOT of mistakes in design and construction. It is one of the many reasons why you are not seeing headers blowing out of the wall in lateral buckling failure mode...

Now wind and seismic- totally different beast, and I am nowhere near convinced that most houses being built even today are where they should be in terms of design and construction.

 

[msquared48]

Well, I think that it is indirectly braced due to the nature of framing itelf, although not directly, as would usually be the case here.

Although end grain nailing would be involved, there should be at least one, if not two king studs at each end of the header. This should supply some rotational fixity near the end. Framing anchors top and bottom could be added to assist in the resistance here too.

Additionally, with the combination of the studs above and the plywood sheathing (assumed here) on the exterior, the nailing in the plywood/ stud combination, and the stiffness of the plywood and studs should provide additional rotational resistance.

This effect would be far less of a factor for longer headers, but certainly apply to shorter ones.


Mike McCann
MMC Engineering

 

[nac521]

Check out the attached 'Dropped Header Design Guide' for additional information.

 

[dhengr]

Bigmig:
I agree with MikeMcC’s. last post, and would just add the following:
I would be more concerned that cripple studs and jack studs are cut true, and square, and all to proper length, so that the header has good bearing and support. I think that end nailing at the king stud, sheathing, cripples and sht. rk. all provide some added stiffening, in most cases. Except on longer, heavily loaded, highly stressed headers, I don’t know that I have ever seen (C_L) bracing being an issue on a header. The deflection has likely made me change the design before bending stress reductions become an issue.

Many framers will pick a header size for the bldg. (with special exceptions being few) and make these at a single work station, off a table of headers/cripples/&jack studs. While there may be a slight waste of material in using only 2x8's or 2x10's for all headers, the material handling and labor savings out weighs this, and makes length calcs. on all studs much easier to keep track of. These are then delivered in bundles to each area of each floor. There are a bunch of variations on making up the headers, and while many instances do not req’r. a double 2x header, blocking, insulation and nailing of sheathing and sht. rk. again, tend to out weigh the cost of the second 2x.

 

[a2mfk]

good points mike and dgengr