Fatigue Analysis of Slip Critical Bolted Gusset Plates

[MagicFarmer]

Good afternoon,

I can't seem to find any literature on the fatigue analysis of bolted, slip critical gusset plates. My initial assumption was that I would be comparing the stress calculated at both the Whitmore Section and the Block Shear failure plane and compare it against the category B limits in the code... There does not seem to be any literature that supports my assumption... or even mentions checking bolted gusset plates for fatigue.

Is my thinking correct on this matter, or have I missed/misinterpreted something?

Thanks in advance,
MF

 

[dhengr]

MagicFarmer:
Tell us if this is the detail you are looking at? And, is that side more critical than the other end?

 

[MagicFarmer]

@dhenger

I have attached an example gusset plate for the truss bridge I am working on.

I have already completed the fatigue checks for the incoming channels. My confusion lay in what stresses (Whitmore tension only? Block shear? Some other tension area?) should be checked, if any, for the gusset plate, and if would be applying the same Category B detail stress limits.

Thanks again,
MF

 

[BridgeSmith]

The AASHTO LRFD bridge design spec. specifies to check "Base metal at the gross section of high-strength bolted joints designed as slip-critical connections with pretensioned high-strength bolts installed in holes drilled full size or subpunched and reamed to size..." as a Category B fatigue detail.

I know what block shear is, but I'm not familiar with the "Whitmore Section". However, notice it says "gross section", not net.

Also notice that it matters how the holes are made. If the the holes are punched full size, it becomes a category D detail, even with tensioned high-strength bolts.

 

[Agent666]

Hotrod, the Whitmore section is a way of accounting for the spread of the tension load in a gusset, for design purposes you assume the load spreads out 30 degrees between the first and last row of bolts and you calculate the width of plate for design purposes on the resulting perpendicular width of plate.

Link

My gut feel tells me that the whitmore approach might not be appropriate for fatigue as its an averaging technique and is not picking up the true stress range that might be occuring/required for the fatigue analysis.

 

[BridgeSmith]

Thanks Agent666! I agree that for fatigue it doesn't seem that would be appropriate. In the text of AASHTO, the fatigue stress for a fully-tensioned HS bolted connection is applied to the gross section adjacent to the connection. The spreading effect would seem to be a moot point. As I understand the AASHTO provisions, the width of the plate at a straight line across the bolt line should be used.

 

[MagicFarmer]

@HotRod10

Thank you for the reply. The reason I am asking, is that, if you had, say, a vertical, diagonal, and top chord (at the end of the truss), for example, with the diagonal in tension, the top and vertical chords would be in compression. Would that not limit the propitiation of the tension field? Would using the gross width of the gusset plate not be un-conservative? Of course, similar situation can occur, depending on the loading configuration with the 5 member gusset connection I linked above.

@Agent666
I agree that the Whitmore section probably isn't ideal. That's why I was asking here. I picked up a copy of Tamboli's Handbook of Structural Steel Connection Design and Details in hopes of finding a discussion on the topic. Unfortunately, books, the net and this forum, so far, no one seems to discuss the topic.

 

[BridgeSmith]

Now you're really making me think, MagicFarmer. I may not be thinking about this or understanding your concern correctly. I don't see compression from other members limiting the tension field, so much as possibly mitigating the stress. In that case, theoretically, I don't see how it could be unconservative. That said, it does seem to get into a very complex stress distribution, though, which is way beyond my ability to visualize. I think to get the real answer would require a finite element model of the connection.

An FEM is unlikely to be worthwhile considering the minor cost involved in increasing the thickness of the gusset plates, even based on very conservative assumptions of the stress distribution. As the I-35W collapse demonstrated, gusset plates are definitely not the place to skimp and try to save a few bucks.

 

[canwesteng]

How is there fatigue level stresses on the bolt holes? If it's slip critical and not only pretensioned, service loads are resisted by friction betweent the plates. I can't imagine using some conservative assumptions for net section you will have an issue with fatigue, which is probably why there isn't direct instruction on this case in a code.

 

[BridgeSmith]

"How is there fatigue level stresses on the bolt holes?"

Although you pretty much answered your question in the next sentence, the answer from the AASHTO spec. would also seem to indicate that there isn't, thus the use of the gross section.