Bridge Weld Fatigue Classification

[BridgeGuyShawn]

I tried posting this on the welding page a few days ago, but I didn't receive any hits. Maybe this is a more appropriate forum for my specific question about welds. Here goes...

I am trying to come up with a fatigue classification for a welding detail I have come across on a bridge. The detail is fairly common for bridges of this age (built in 1950's) but nothing in the AASHTO table (6.6.1.2.3-1) looks quite right to me. The detail is this:

Built up riveted longitudinal girder with several flange cover plates. The cover plates are stitch welded, with 2-3 inch longitudinal welds at their edges, every foot or so.

The closest things I can see in AASHTO are parts 3.5 and 3.6 in the table, but my gut tells me it is somewhere in between a B and an E'. I think it is a worse scenario if the weld termination is at or near the end of the cover plate, but these are just along the edge. Since they do have random termination points though, I don't think part 3.4 would necessarily apply either.

I'm guessing this was done to help hold everything together and make it easier to drill/rivet during fabrication.

Does anyone know what category this would fall into?

Thanks!

 

[Teguci]

E for flange thickness less than 0.8", E' for >0.8". Cracking will begin at fillet terminations.

 

[BridgeGuyShawn]

Thanks, Teguci. So it does fall into section 3.5 of the AASHTO table.

To follow this question up, I assume the 0.8" limit applies to the portion of the flange under consideration and not the total thickness of the flange. In other words, for a flange made up of a pair of 8x8x1 angles with 2- 3/4" cover plates, the cover plates would be e', but the angles would be allowed the e designation.

Is this a correct assumption?

 

[ToadJones]

Why even analyze it for fatigue if you think the welds were for fab purposes only?

 

[BridgeGuyShawn]

I'm thinking that thought the welds are not 'structural' in nature, that is they are not transferring any forces from one plate to the next (that work is all taken care of by the rivets) they are pretty sloppy welds and in some locations relatively thick. Because of this, they may inadvertantly act as a stress riser (especially if there was any slippage occuring between the plates), and the base metal immediately surrounding them was most likely weakened to some degree when the welding was originally performed.

 

[Teguci]

I would say E for the cover plates and angles (angles are not welded).

As for the rivets, they are not activated until the welds fail, so in my mind, they are only being used to connect both plates to the angles (not plate to plate).

 

[bridgebuster]

I'm working on a bridge project with a similar condition - built up riveted plate girders with intermittent welds between plates.

Our welds a categorized as C & D.

Look at Example 7.1 in LRFD Table 6.6.1.2.3-1, or Example 15 AASHTO Figure 10.3.1.C.

 

[BridgeGuyShawn]

bridgebuster, this is pretty close to the detail, the plates are on top of each other rather than side by side, but for all intents and purposes it's the same. Thanks for pointing me to that!

 

[bridgebuster]

BGS - you're welcome.