Depth of Crane Beam Bracing 3

[JoshH726]

Evaluating a 5T crane beam originally installed mid-1970s. Running through preliminary numbers, I'm discounting the bracing shown as any type of App. 6-level bracing. Unfortunately that eliminates all capacity in the system. App. 6 says bracing at or near the compression flange, but I don't see it being effective this far down. However, the question I'm going to get pushed back on if I say to reinforce it is that "we've been using this for years with no issue". My gut says the bracing must be doing "something", but I'm not sure how to quantify it. Any suggestions?

 

[KootK]

The bracing creates a condition whereby the beam is forced to LTB about an axis at the level of the bracing. That, as opposed to buckling about an axis below (or above) the beam. This constrained axis bending condition involves more energy -- and therefore more load -- than the usual case. There's an example of this in the AISC seimic manual I believe.

That's the only trick that I can think of so far. And it's a pretty safe bet that it was not the designer's intent.

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.

 

[SlideRuleEra]

Since there is only one beam, the crane probably runs on the bottom flange. Normal LTB calcs assume the load is applied to a beam's top flange which destabilizes the beam. Load applied to the bottom flange stabilizes the beam increasing it's ability to resist LTB... not a big increase, but some.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[TehMightyEngineer]

Funny, I literally just went through redesigning a single girder bridge crane for one of our shops.

Agree with SRE that bottom flange loading helps but it's hard to quantify, I can dig up some papers on this if you like.

While I'm sure you've considered it, how much work will it be to retrofit gussets or some other way to brace the top flange adequately? I imagine your biggest issue is the subsequent OSHA 125% load test (assuming you're in the USA by your mention of AISC). Maybe you can justify waiving this requirement as you're clearly strengthening the crane.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[Lomarandil]

The Guide to Stability Criteria for Metal Structures would indicate that a bottom flange loaded beam would see a Cb bump of about 40%.

TME, do you know of a more substantial increase available? I'm in the middle of a project right now that could benefit from it.

Josh, from the sketch, I think torsional bracing isn't going to calc out -- but if somehow your channel boundary conditions worked out in your favor, you may be able to get enough out of the angle flexure and web stiffness to consider at least some sort of restraint.

----
The name is a long story -- just call me Lo.

 

[JoshH726]

KootK, I'll take a look through the SDM later today to see if anything useful pops out.

SlideRuleEra, yes, I should have clarified, underhung crane.

TME, anything you think would be helpful, please pass along. Yes USA.

While chewing on this last night, I don't see how you can justify the existing system as providing effective restraint. I'm working on developing a series of dog-legged rigid framing <6 feet o/c from the top flange out then 90 down to the C12.

 

[TehMightyEngineer]

Guide to Stability Design Criteria for Metal Structures is probably the best reference. If OP has a copy I believe they may also discuss lateral bracing as well; I don't have a copy on hand to check.

I thought I had another reference but can't seem to find it.

Yura discusses lateral bracing at centroid for OP. Yura briefly discusses centroidal loading vs top flange loading but basically just references other papers.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[KootK]

I don't see how you can justify the existing system as providing effective restraint.

I do. In fact, I submit that it's all but impossible to lateral torsional buckle (LTB) a beam with a) bottom flange loading and b) the constrained axis buckling that I mentioned above. Walk with KootK...

1) When a thing buckles, it needs to result in the loads moving closer to the earth. Otherwise, the system gains energy and that's pretty much the opposite of buckling.

2) As it's name implies, LTB involves beam twist and beam lateral sway. While your bracing scheme doesn't address twist effectively, it eliminates sway completely (assuming a stiff horizontal truss).

3) From #2, we can envision LTB in this case essentially being just pure torsional twist. And pure torsional twist means that the bottom flange, and the load, move upwards away from the earth.

1 + 2 + 3 = No LTB. Granted, as a diligent SE, you'll want to find some calcs to run to back up this fanciful story. I get it.

It's also worth noting that a thing can be only locally stable. Like a ball stuck in the local valley between the peaks of a twin peaked mountain. Conceivably, said ball could be pushed back up one of the peaks and then roll all the way down to the real valley floor and, thus, be considered sort of unstable initially. Here, analogously, it could unfold like this:

1) Beam twists.
2) Load raises.
3) Beam flips to weak axis position.
4) Load deflects closer to the earth than its original position.

It would take a fair bit of energy input to make that happen however. And we don't often take things that far in design office work. It would probably make sense to spot check some of the unbraced segments away from the load. You'd have an interplay between the moment dropping off and torsional flexibility reducing the extent to which the twist would raise the load.

Of course the stakes are pretty high here. If it were me, I'd probably just run some angle bracing straight from the top of the C12 to the top of the crane beam. It's not as though your contractor is likely to be able to tell this same story. Yay arcane knowledge!

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.

 

[SlideRuleEra]

IMHO, it's not going to take much modification to bring the beam into LTB compliance. Even at OSHA test load (say, 13 kips at midspan), bending stress is < 15 ksi and deflection is < L/500. Benefits from existing bracing, explained by KootK, help. These factors together with load applied to the bottom flange all indicate to me that while the beam is probably not (technically) acceptable... it likely does work successfully, as is.

A neat, straight forward improvement (solution) is to create a combination section by capping the W24 with a light C12. Fabricated the full length C12 with pairs of holes in it's web, spaced, say, every 2 or 3 feet. Plug weld the C12 (welding from the top) to the top flange of the W24. Of course, this assumes the top of the existing W24 is accessible.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[MikeHalloran]

I'm flummoxed as to how the hoist travels on the C12's flanges.
It would normally travel on the W24's flanges.
Actually, it looks like the W24 was installed upside down.

Can you provide photos, especially of the hoist trolley/beam interface?



Mike Halloran
Pembroke Pines, FL, USA

 

[TehMightyEngineer]

Mike, that's a plan view OP has sketched.

Here's a isometric view of how a crane similar to OPs looks:

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[MikeHalloran]

Thanks, TME; that makes more sense.


Mike Halloran
Pembroke Pines, FL, USA

 

[JoshH726]

TME is showing the condition probably better than I did.

Special thanks to Kootk, as the "1+2+3=No LTB" will be my opening statement on the witness stand, and it rhymes so it must be true.

 

[Eagleee]

It is a bit difficult to quantify how much better a stabilizing load is compared to a destabilizing one since there are many other parameters involved, but 40% does seem to hit the mark. For instance, the use of an approximate formula for the LTB slenderness gives a 20% increase for a destabilizing load. You can check out

http://www.steel-ncci.co.uk/Clauses/List-NCCIs

for more documents on LTB, maybe they can help to get a general feel. I guess most here design to American codes, so this link would just be to get a general overview.

 

[Settingsun]

"1+2+3=No LTB and, since I made a rhyme, I must do no time!"

Australian Standard AS 4100 gives formulas that include the distance above or below the centroid, and says they are approximations to the results of elastic buckling analyses. Hard to post it since it refers you all around the document to get the whole story. When I have some time, I'll compare top flange/centroid/btm flange loading for your W24 beam and post some results for interest.

 

[Eagleee]

Actually I think the easiest way to check the beneficial influence of a stabilizing load is using the LTBeam software. It is free and very easy to use.

 

[Ingenuity]

Australian Standard AS 4100 gives formulas that include the distance above or below the centroid, and says they are approximations to the results of elastic buckling analyses. Hard to post it since it refers you all around the document to get the whole story. When I have some time, I'll compare top flange/centroid/btm flange loading for your W24 beam and post some results for interest.

AS4100 probably uses Trahair's approximations (from his text Flexural-Torsional Buckling of Structures).

Using Trahair's approximations: For a W24x55 (doubly-symmetric section), I calc a factor of 1.31 assuming midspan concentrated load applied at the bottom flange, compared to a factor of 1.0 for load applied at shear center, and 0.76 if the load is applied to the top flange.

 

[TehMightyEngineer]

I thought I had another reference but can't seem to find it.

Lomarandil, Trahair's Flexural-Torsional Buckling of Structures was the other reference I was thinking of. I couldn't find it in my library because it actually belonged to a past coworker in our office library and I forget that I didn't still have a copy. Good reference for sure.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[KootK]

By "stabilizing load", I think that most of us have been envisioning load delivered at the bottom of the beam. If one considers the physical realities of the crane itself, I would wager that the effective point of load application is really the bottom of the trolley / top of the rope. I wouldn't be too enthused about designing to that but, still, I can see why things are not going awry out in the field.

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.

 

[Lomarandil]

Perfect, thanks TME & Ingenuity!

KootK, that's a good point about the additional stabilizing effect of the trolley. We often design lifting devices (yokes) for precast girders that bring the point of load application some distance above the top flange (same problem, just flipped).

----
The name is a long story -- just call me Lo.