Web Reinforcement of Existing Crane Runway Beam

[StLHokie]

Hi folks,

I'm currently working on a project that includes the design of a bridge crane runway girder. At this point in time, I've gotten news that (despite the runway girders already being fabricated and constructed on site) the bridge crane that will be used has increased significantly in size. The original maximum wheel load was around 7 kips, however the new maximum wheel load is now closer to 12 kips.

I originally oversized the beam a bit to provide some margin in case things changed, however that topped out at around 9 kips, the limiting failure mode being web sidesway buckling.

Removing and replacing the beam with a web of larger thickness is not possible, as the roof assembly has already been constructed. Therefore I'm left to pursue web reinforcement, namely transverse stiffeners or doubler plates.

I'm having issues finding information on spacing requirements of transverse stiffeners for crane rails, as transverse stiffeners are usually used for non-moving loads. The limited information on spacing I have found has mostly been with regards to built-up plate girder requirements. The issue with using built up plate girder requirements is the reliance on tension field action, which requires a weld of the stiffener to the bottom flange. From what I've been able to find, it isn't recommended to weld the stiffeners to the bottom flange for runway girders due to fatigue loading issues, so I'm not sure I can follow that line of methodology for spacing of the stiffeners.

Nothing I've been able to find in AISC/CMAA74/etc has gone into much of a discussion about the load distribution of a compression load longitudinally along a beam, other than J10.8 (AISC 14th) where there is brief discussion about web inclusion in the compression capacity. (But still not enough)

Depending on if I can find any other leads on stiffener spacing requirements, my other option would be to pursue web doubler plates. I, admittedly, have not researched this as thoroughly, however I have not found much in terms of standard practice in reinforcing existing beams. From my understanding doubler plates are typically much more useful for shear reinforcement of the web. Since web sidesway buckling is not exactly a shear failure, I'm not sure how effective a doubler plate would be. My thoughts are that the web and doubler plate would both buckle out plane due to the connection between the two only being at the location of the welds on the top and bottom of the plate (rather than the whole interface between the two).

Has anyone worked on anything like this before/have any leads? I can provide sketches and diagrams if what I've explained isn't clear enough. Thanks for the help!

 

[FLCraneBuilder]

Yes - I/we deal with this quite often
Approach 1 - are you using ASD or LRFD? - LRFD is MUCH more lenient with sidesways web buckling
Note that PEMB guys often disregard this calc when they provide runways.. They use extraordinarily narrow flange beams and cap channels. narrow flange beams are much more susceptible to this problem
I do not believe this limiting condition was even calculated until 20 years ago or so.
When we do have to reinforce - we add web stiffs on 5' centers
AISC has "crane girder" calculator - free down load from AISC - it does both ASD and LRFD

 

[Mjkkb2]

AISC design guide 7 talks about this a bit but des not provide guidance for stiffening the web. Id does note that the LRFD will give you a higher buckling stress allowance but based on your crane loads it may still be not enough. I think you just have to use some engineering judgement here.
I think adding a web doubler is an option that you can put numbers to. One sided doubler essentially will reduce the crane load by half (if it's the same thickness as girder web). I think adding the stiffeners is more efficient but not sure how to justify the spacing by numbers. Doubler is basically another web which I'd assume would be cumbersome to add to an elevated girder but you can engineer it easily.

 

[Mjkkb2]

I found one more thing for you, the AISC design examples. Look at examples G8. This is vers 13 but should be similar to 14'th edition. I also attached the design guide I mentioned before. Hope this helps.

 

[Mjkkb2]

exmaples

 

[StLHokie]

@FLCraneBuildier
Up to this point I've been using ASD. I'll look into using LRFD for this particular case and let you know what it results in.

This isn't a PEMB, and I designed the runway beam that's already been built, it's a W24x68 with a C15x33.9 cap channel.

Do you have a source/calcs for the 5' spacing on center?

I'll also try and grab the crane girder calculator you mentioned, although at this point I do already have a mathCAD file that performs all the required calcs (other than stiffener spacing or web doubling)


@Mjkkb2
I've been using that design guide for the majority of the runway girder calcs I've done so far, but as you also mentioned, stiffener info is limited. I'll try to find something similar in the design examples, but from the looks of it that closest thing there is the transverse stiffeners for plate girders. I've been a bit hesitant to go the doubler route because, from what I've read, welding too close to the fillets at the web/flange interface can reduce the capacity of that connection, especially if the weld has to be along the entire length of the girder.

Another thing that needs to be considered is the field work involved using stiffeners vs doubler plates. Everything for this modification will be field welded. This particular bridge crane is roughly 165' long (x2 girders), so even if 5' spacing for the stiffeners would work, it'd be close to 132 stiffeners and 264 welds that would be needed. I'm not sure if this would be more or less efficient than adding doubler plates to each of the 2 runway beams. It gets even more complicated when you factor in that all of the beams have heavy corrosive coatings that would need to be removed before any addition of elements, and then replaced after the modifications.

 

[jayrod12]

I get that it's already been fabricated, but isn't it worth investigating the idea of replacing the runway outright, just so there is a cost comparison. That amount of field prep, welding and re-covering will add up quickly. A straight removal and replacement may come in cheaper, in my eyes it's at least worth throwing a number at it.

 

[bones206]

You could end up having to deal with top flange distortion from the stiffener welding if the welders do not take measures to mitigate it. I imagine that could end up being a real headache for a crane girder. Just something to consider.

 

[Mjkkb2]

completely agree with last two posts. Complete replacement may be most economical approach.

 

[StLHokie]

@jayrod12

I would normally agree with you, but replacing the runway beam just isn't an option at this point. The roof of the structure above the girders has already been completed. It's not possible at this time to remove the roof, as it's a concrete slab on custom stainless steel deck. The pricetag on the deck is significantly more than any modification to these girders. Trying to replace the girders by coming off of the floor also doesn't work, as this is on the 3rd level of the structure and very large industrial process tanks cover most of the floor below. Coming in through the sidewalls is also extremely difficult, as the total spans on the bridge crane are around 165' and there isn't ample space (or lifting capacity) to lift the beams into the structure from the sides.

@bones206
My thoughts are that any top flange distortion will be mitigated by the fact that the girders are capped with a continuous channel. Welds will be made to the top flange, but the channel will not be modified.

 

[jayrod12]

Fair enough. My response to the client when they scream about the number of stiffeners required and the amount of work required to install them would be "go back to the original crane then". They can't increase the loading 70% and expect zero consequences.

 

[KootK]

I feel that many of your answers can be found in that paragraph from AISC commentary. Additional comments:

1) For a stiffener to help, I believe that one of these conditions must be met:

a) The stiffener must be beneath the load and attached to a loaded flange restrained against rotation OR;

b) The stiffener, away from the load, must itself be attached to a compression flange restrained against rotation. Otherwise, the stiffener does nothing to prevent tension flange translation, the whole point of the stiffener in the first place.

Taken together, I believe that this means that stiffeners are useless for moving loads where the loaded flange is not restrained against rotation at all stiffener locations.

2) Clearly, stiffeners need not extend down to the unloaded flange.

3) In a hypothetical case where there were stiffeners away from the load but not at the load, and those stiffeners were attached to a flange unable to rotate, I think that the design checks would simply use the stiffener spacing as the unbraced length. And that would establish stiffener spacing.

4) Where your crane runs on a rail on top of the beam, I might even question whether or not that condition represents rotational restraint at the point of load application. In the absence of that restraint, neither web doubling nor web stiffening would be effective.

5) If it could be accomplished in a fatigue appropriate manner, a cost effective solution might be to attach a channel to the bottom flange. That's one of the few remediation options that doesn't require rotational restraint of the loaded flange.


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.