Bent beam flange repair 1

[danvines]

I have a steel roof truss with top and bottom chord members fabricated from W12x96 shapes (A992 steel). After erection, but before roof decking was installed, there was a crane accident which resulted in the 1-ton overhaul ball swinging around like a yo-yo and impacting both the top and bottom chord flanges.

I have found quite a bit of info regarding heat straightening similar damage on a larger scale when it comes to bridge girders (bracing, jacking, and heating), but I'm not sure about the practicality of these methods on so shallow a member.

The worst top chord damage is a "dip" in one side of the top flange approximately 12" long and 3/4" to 1/2" deep. The worst bottom chord damage is also in the top flange and about 12" long, but it is almost 2-1/2" deep. I have attached photos of each.

Any advice or pointer to literature would be greatly appreciated.

Dan

 

[racookpe1978]

Only the flange is bent, and that bend is only in one place, right? There is another possible dent to the extreme right of the picture.

That is, the web and lower flange are still straight and not visibly distorted in plan nor in section?

 

[danvines]

racookpe1978 - that is correct. There are no visible distortions in either the web or opposite flange in either location. And you are correct, there is a much smaller dent in the top chord photo. I was focusing on the two areas with the greatest damage.

 

[KootK]

If the bottom chord never goes into compression and isn't loaded between panel points, I would let it go. Under load, if it yields, it will just straighten itself out.

I'd also let the top chord go on account of the damage being too minor to worry about in my opinion. A reasonable check might be to see if the flange in bending could resist the upward and downward thrusts created by the kinks.

I do tend to be quite liberal when it comes to dinged up trusses.




I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[racookpe1978]

That high up, you'd have to "aim" a vertical WF from the slab up the the truss, then anchor it against any sideway motion, then "push" the heated upper flange (getting to an "orange hot" evenly across the whole dented area is going to be tough! ) up with the jack mounted to the vertical.

With an orange heat, the steel upper flange will be soft enough to move with little resistance. BUT!!! the beam itself will be softened as well, and so you need to support it with something "cold" from below. A doubled WF under the bent upper beam could work.

A WF brace over the top of the upper flange might be better. Then, instead of "Pushing" the upper flange from below, you pull it up towards the brace WF that is above the upper flange with the equal of large C-clamps. A bit better control that way..

You really don't want to jack "up" the top flange by pushing down on the lower flange just at that one spot - that will get you a bent upper flange and a distorted lower flange.. And, you don't want to straighten the upper flange by all force resisted and "anchored" only with the structural bolts at both ends.

 

[CELinOttawa]

As you have said, bridges with this type of damage are often flame-camber repaired. It takes an experienced crew, but I have seen localized damage like this repaired before.

I don't believe this top chord would straighten itself out under very many circumstances, but such things are possible in a plastic yielding state.

As others have said, be sure this is a problem before you try to fix it.

 

[PittEng88]

Kootk, if you don't mind entertaining me for a minute, how come you are liberal when it comes dinged up trusses?

 

[KootK]

Happy to entertain PittEng. I'm not really sure why I tend to be liberal on dinged up trusses. I've just noticed that I tend to be relative to other folks. And that's rare for me as I seem to land on the other side of the line for most things. Some of my rationalizing:

1) I can rarely envision consequences arising in pure tension members as I've mentioned above.

2) compression member damage that doesn't modify kL/r or alter the location of the centroidal axis location strikes me as minor in a member dominated by compression. If the top chord were a forklift damaged column in a warehouse, most engineers would pass it in my experience.

3) If margins are such that we can't handle a little ding in such a big thing, we're all doomed anyhow. It's a truss, not a piano.

4) I have a general sense that many "repairs" either cause more damage than good or, at best, are neutral and costly. I use them sparingly. CEL and I debate this philosophy from time to time as few share it.

These are all broad generalizations riddled with exceptions of course.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[XR250]

If you had to fix it, what about laying a spreader member on top of the bottom flange and just use a bottle jack off of that?

 

[jike]

We had a few bent chord flanges in 185 foot long roof trusses years ago. Instead of attempting to straighten them, we opted to add a partial stiffener at each significant buckle in the compression chord.

 

[SAIL3]

as others have indicated....determine if it really is a problem in the compression first....personally, it would bother me to leave it as it is right now....a few fixes I might consider...the simplest would be to add a partial web stiff @ the peak of the dent to react the component of the compression load in the fla...another more expensive fix would be to cut out the affected area and add a 1" pl with a CP weld....I would also check for any crack in the web/fla area....also check the connections for any sign of damage and, lastly ,ensure the member is still straight...to me ,it's a borderline problem and I would rather address it now ,in some fashion ,than worry about later...other engineers may have more experience with this type of problem and be more comfortable in living with the uncertainty....

 

[Brad805]

I am a bit surprised by the responses. We have had after the fact discussions about very similar scenarios. Those tend to start off with a comment like, "I did all of this extra work because of a contractor error, and now nobody will pay me." Assuming you are the EOR and not the fabricators engineer, my first question would be if someone is willing to pay me to review this. If not, the next question would be to the contractor, "how can you fix this?" It is not our job to figure out how to complete (bottle jack, heat...) the repair or speculate on the methodology to do so. We are the math guys, not the hammers. If we as the EOR made an error there is almost always a discussion about extras, so unless there is some extenuating circumstance I don't see much reason to be Mr. nice guy. This looks like less than two hours to repair if it is accessible with a 120ft JLG or something else on site. When my father was alive I watched him repair far worse with a torch, hammer, some plate, and welding rod.

 

[rb1957]

i'm alittle stunned by the "leave it" responses ... not my field, and i'm not being "holier than thou".

at a minimum bend (ie heat and hammer) it back to flat; preferribly, since all this plastic working will mess with the allowables, nest a plate/angle to relieve the flange some.

another day in paradise, or is paradise one day closer ?

 

[PittEng88]

Kootk, I appreciate the response. Here is some of my rational when it comes to dealing with trusses:

1.) I completely agree with you about the bottom flange, if it never goes into compression it should not be an issue.

2.) As for the top chord, if there was any distortion whatsoever in the member the P-Delta effects could be pretty significant. Especially if that member was not designed for flexure, and even if flexure was considered the bent flange will result in a reduced I and reduced moment capacity.

3.) Looking at the truss as a whole, if the impact twisted, displaced or distorted the truss in any way globally (not necessary damaging the members themselves), an eccentric load could also be generated that way, causing unwanted bending in the truss.

4.) The lack of redundancy in a truss also causes me to err on the side of caution when dealing with damaged truss members.

Like you said a lot of this is case dependent and there are numerous exceptions to each of these. However, at the very least I feel that when dealing with a damaged truss compression member, one should tread cautiously when looking to "call ok, leave as is"

 

[KootK]

I agree on all points PittEng88, save perhaps the degree of caution to be exercised. I too made a point of stating that the centroidal axes of any damaged compression members should not be significantly altered.

I'm likely desensitized by my previous experience in the wood truss industry. I once went to a site where most of my truss diagonal webs had been removed to make space for attic storage. When I inquired about that, the builder told me that he'd been doing that for twenty years and had it on good authority that the diagonal webs were "just for shipping".

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[msquared48]

Strangely enough, since the bottom chord vertical distortion is less than the top chord, the crane operator actually locally increased the S and I of the truss. Have we stumbled on a new method to modify trusses to increase their sectional properties? Maybe you should give the crane operator a bonus AFTER you patent the your new invention?

I tend to side with Koot on this one if there is no net uplift on the truss.

Mike McCann, PE, SE (WA)

 

[jayrod12]

I'm in the "don't do anything unless required". If the truss chords will never see stress reversals (uplift) then I can't see the bottom chord having any issues as the tension force will attempt to straighten in on its own. The top chord damage at only a 1/2-3/4" is less than I've seen on heavily loaded columns that ended up with no reinforcement so I again wouldn't likely do anything.

 

[CELinOttawa]

Left unrepaired, the behaviour at plastic is not readily predictable. I'd be very happy to stand corrected on this one!

Regardless, you still need to be sure that there is a problem before you start looking for a fix.

The other fact that many engineers fail to understand is just how complex the actual state of stress of the steel in such a section is... If you do repair, don't worry about any residual stress and strains. Once the section goes plastic none of that matters, and the plastic strength is what you are designing to...

 

[bigmig]

I had the same problem about 3 weeks ago. The post is on engtips.

A few helpful references from the guys/gals who gave me a hand:

http://www.fhwa.dot.gov/bridge/steel/

http://www.eng-tips.com/viewthread.cfm?qid=106922

And finally, my post:

http://www.eng-tips.com/viewthread.cfm?qid=379060

 

[racookpe1978]

Scaling from the level in picture 2, I'd call it 1-1/8 to 1-3/8 bend.