Whitmore Width Definition on a Single Bolt Row Detail 2

[MGaMart]

I'm curious to get the community's opinion regarding this. Many connection details (let's focus on brace connections) I've made in the past constitute a single bolt row (transverse to the line of action of the force). Whitmore sections utilise the overlap length between the first and last bolt row in defining the effective region of gusset plates for analysis of yielding and buckling. With a single bolt row, the methodology breaks down and resolves the calculations to a judgement call from the designer. Other limit states aside which may/may not govern the design (bolt tearout, bolt shear, block shear, etc.), what rationale do some people apply in circumventing this 'grey' area of the Whitmore Section? I do understand that changing the bolt geometry to include an extra bolt row, or swapping the bolt row to be a bolt column is an alternative, just not a very economical one. Swapping bolted details to welded details also does this, but let's just focus on bolts for this discussion.

 

[JoelTXCive]

Here's some background info from Steelwise in 2011.

I just skimmed it and it does not directly answer your question, but it does have a lot of info about Whitmore sections.

 

[HS_PA_EIT]

I'd dig into some of the AISC journals. I believe Prof. Bo Dowswell at UAB has written some papers on applying the Whitmore section in different scenarios, so that might be someone to ask.
I would consider taking the width to be that of the bolt row, but that wouldn't be very economical.

 

[KootK]

I just skimmed it and it does not directly answer your question...

It's got a few pertinent examples as shown below.

Whitmore sections utilise the overlap length between the first and last bolt row in defining the effective region of gusset plates for analysis of yielding and buckling.

1) I feel that the definition of the Whitmore is pretty well defined in these situations and is as shown Joel's paper.

2) I don't agree that the last bolt limits the Whitmore spread. I suspect that it's just coincidence that it appears that way in a lot of sketches. I feel that the Whitmore can spread until it runs out of material to spread too. Obviously, some measure of judgement is required.

 

[skeletron]

Nice article Joel! Thanks for posting!

 

[HS_PA_EIT]

KootK, regarding your point 2), in many cases wouldn't the section just beyond the end of the bolt pattern would have the greatest concentration of stress (minimal spread and no longer strengthened by the connected member)? I suppose there could be cases with strange geometry that would violate the idea.

 

[MGaMart]

Thanks Joel. I have come across that article in the past and it definitely offers good insight into the Whitmore Section. To KootK's point #2, I too agree that the load dispersion continues beyond the 'overlap' region of connecting elements; the truncation as shown in most design articles is a design simplification that errs on the conservative side. Other load dispersion angles have been suggested in the past (I believe the modified Thorton model suggests a 45° spread rather than the standard 30° adopted in traditional practice). Load dispersion truncation does offer the designer a more well defined means to assess gusset buckling as the cross sectional area is constant, thus the gusset is analogous to a prismatic column and therefore easier to design. Defining the Whitmore as the distance between bolts (width of the bolt row) does seem conservative (so long as the bolts follow a reasonable spacing consistent with industry practice (roughly 3 to 4 inches)). I've typically taken the width to be that of the connection plate(s) that frames into the gusset, but this again requires that the connection plate itself isn't excessively wide in comparison to the bolt spacing.