D1.5 One-Sided Weld - Hole Closure

[MRhodes]

In an attempt to create a box section, AASHTO Category E, with a required CJP at all locations due to large tensile stresses and stress reversals, we have run into a constructability issue. On another thread, there is a discussion of a one-sided weld and issues presented by this procedure. The discussion has yet to produce a viable solution in our situation and without access, the backing cannot be removed. Therefore a one-sided weld is not allowed by the Bridge Welding Code. In addition, PJPs are also prohibited by AASHTO. As a possible solution, access holes for construction of the box are possible. With the access holes other problems arise. The access holes are in a very tight location on an internal plate of the box.

The 2” vertical plates of the structure are approximately 6’ tall and only about 1’-10” apart.

The access holes are in 2” and greater plate. We are investigating several questions. First, we do not have a large space and the access hole or slot can only be 12 wide or so and up to about 40 inches long. Does anyone have experience with removing backing bars through access holes? If so, what size and shape recommendations are suggested?

By SAP analysis, the placement of the holes creates increased and excessive stresses elsewhere in the structure. Therefore, we are looking for a way to structurally seal the holes while meeting the requirements of D1.5. Any suggestions would be appreciated!

 

[HgTX]

The other thread is thread725-150229, for those who feel like going over to look at it.

As I'm about to post over in that other thread, the solution indeed involved access holes in internal plates, but the access holes are to be left open. I too would like to hear about what size is recommended. The designers built a cardboard model and stuck their arms through, but I'm not sure if they stuck their heads through too so they could see what they were doing in this hypothetical backing bar removal.

If you're already Category E, though, why not leave the backing in place? It violates D1.5, but it violates D1.5 because D1.5 doesn't like Category E. Put a note on the plans saying it can stay in those locations.

Are thse going to be internal or external access holes? If internal, how about some kind of bolted detail to transfer stress across your access holes? If you use twist-off bolts or DTIs you can bolt them from one side.

When you say "PJPs are prohibited by AASHTO", you're talking about a one-sided PJP, right? That's more of a problem because it's one-sided than because it's a PJP; a two-sided PJP is no worse than a double fillet weld. What AASHTO provision are you talking about?

Hg

Eng-Tips policies: faq731-376

 

[HgTX]

Followup:
The designer on my project (subject of the other thread) says they cut a 12" hole in a piece of cardboard and he could get his head and arm through with a grinder.

FWIW.

Hg

Eng-Tips policies: faq731-376

 

[unclesyd]

A skinny person should be able to get through a 14" diameter hole in plate if they are have narrow shoulders.
You might also cut an oval like used in boilers. In my younger days I could get thru an 11x15 but preferred a 12x16

 

[MRhodes]

HG,

I think we are discussing the same problem, same bridge in both threads.

We are currently looking at two 12" access holes that are spaced 28" or so, center to center. In the cardboard cut-out, I was able to get my head, one arm, and shoulder into the hole. It was not comfortable but I was able to reach all of the corners where the backing bar will need to be removed. The removal will be okay but I am not sure about the repairs that will need to be made.

If the holes are placed but are not large enough and later widened, is there a way to structurally close the hole?

In what situations can the backing bar be left in place? Only when the member or weld is classified as a secondary member or non-FCM?

Correct, I am speaking of one sided PJP being prohibited. Since the access we are providing is extremely tight, I do not think that welding in that position will be very effective. It will be difficult enough to perform the repairs.

 

[HgTX]

How bout dat. That explains why the designer just recently told me he heard the suggestion was to leave him inside, even though I'd passed on something of the sort to him weeks ago when the suggestion was first made...

Well, that means your access holes are internal, and there's no reason not to bolt them. No worse than a bolted splice connection.

Check out Chet Guilford's response to me in the AWS Forum yesterday:

http://www.aws.org/cgi-bin/mwf/topic_show.pl?id=9140

FCM vs. non-FCM doesn't help you with backing bars. I'm not seeing any exemptions for secondary members either, as long as you have tension or stress reversal. If you have compression only, or stress parallel to the weld, you're good.

One-sided PJP and one-sided CJP are not all that different in my book. PJP has more of a notch at the root, but you have at best a questionable situation even with the one-sided CJP.

I'm thinkin' with the access holes, my original clever and utterly nonworkable idea of ceramic backing might work after all, but I think your detail is such that the backing would need to be beveled. Does anyone (other than me and M. Rhodes) know if such a thing exists as beveled-edge ceramic backing strip?

Off to the AWS bulletin board...

Hg

Eng-Tips policies: faq731-376

 

[unclesyd]

HgTX,
The links I posted in your post have about all forms that are available on the market. Look under the different welding processes in the headings for all the different cross sections.


Is it possible to use a backing strip and leave it in when you replace the metal in the access holes?

Can you use a V-groove backing strip with or without a flux coating where you will not get fusion to the backing strip and again just leave it place or remove it to planish the weld through the access hole. A V-Groove backing strip would be a to easier to remove.

One other trick is to completely fuse the backing strip to one side of the joint from the outside with TIG. Then TIG the other side of the joint to the backing strip. This makes a very slightly wider joint but has worked well on several occasions.

 

[HgTX]

The problem we have is that a fused backing strip can't be permitted to remain if there is tension across the joint, and an unfused backing strip can't be applied without access to the rear of the weld once the joint is assembled. Looks like the original welds of concern have now been addressed with access holes, change in fabrication sequence, or reclassification, but now they're looking for a way to fill the access holes.

Rhodes--Unclesyd talking about his post of 6 Apr 06 0:02 in the other thread, but I don't see anything there for a skewed T joint.

Also I would not look into a welded solution for the access holes if you have tension or stress reversal in the plates that the access holes are in. That kind of detail was already eliminated at other locations in the structure because a patch welded all the way around is bad for fatigue and restraint conditions. A bolted patch, like a bolted compression splice, might serve to transfer stress across the hole though.

Hg

Eng-Tips policies: faq731-376