Girder Camber

[ToadJones]

I am investigating some crane runway girders that have been cambered excessively (in my opinion).
Some of these girders are 60' in length and have been cambered 2 inches.

This seems simply idiotic to me.
The crane rail had to be shimmed all over the place in order to keep the top-of-rail elevation within tolerances. Not to mention there are many different spans and the camber is different for all... exacerbating the rail shimming fiasco.

The most stringent guidelines use L/1000 for max deflection with a fully loaded crane with no impact. In the case of the 60' girder, this would be an max deflection of 0.72 inches.

The most I can ever see cambering a runway girder is enough to counteract dead load deflection only; in this case, girder selfweight.

Has anyone seen such cambering?
Is this common in highway bridge design?

I think the design-build for this building was a bridge builder.

Any thoughts?

 

[BridgeEI]

We do camber beams for bridge design. Typically we usually have something moderate like 1-2" after all the dead load is applied. I always thought that for building design, if you cambered, it was for permanent dead loads only that way your floor is flat. That way people won't notice a difference in elevation in a couple feet. But yea sounds like a bridge guy did your design.

 

[ToadJones]

so, you have camber with the finished product?
Why?
Just curious.
I have noticed this many times just driving under highway bridges.

You cannot imagine how many problems this is giving this crane runway.
I'm basically SOL.

 

[hokie66]

The CISC design guide by MacCrimmon recommends cambering crane beams of greater than 20 metre span for DL + 50%LL. I've never cambered a crane beam, maybe because I haven't designed one with a 20 metre span.

 

[msquared48]

Too many truckers over-inflate their tires. That extra inch or so is what gets 'em. So, over camber and under-rate the clearance.

If you believe that, you've already had a few too many at your local pub.

Mike McCann
MMC Engineering

 

[ToadJones]

hokie-
I'll have to look through that design guide when I get in the office.I'm pretty sure I have a copy.
AIST also recommends cambering long span girders, but not as much as these are.
In fact, EVERY SINGLE GIRDER IN THIS PLANT IS CAMBERED....even girders 24 ft long.
I cant say anything other than, this is asinine.

I hope it has some recommendations for how to install a level rail on a cambered girder.

I am certain that even for a 20 meter girder with DL + 50%LL still would not even come close to 2 inches...it cant really, because that would exceed allowable deflection by more than two times.

Mike-
Good stuff. This is making me want to go to my local pub...for a good long while.

 

[BridgeEI]

Haha too funny msquared48. Too an extent the above post is true. When we check vertical clearances we assume the bottom of the beam is straight from bearing to bearing. If the beam sags then that's just even more time to check those cals. however we usually do round down a couple inches for what is posted.

The main reason for the camber is to accommodate any future wearing surface that may be applied. If we design for an 8" deck and later it is redecked with a 9" deck or an overlay is added, then that's just more deflection in the girder. It's just not desirable to have a girder that is sagging or almost flat. If the profile grade is in a sag, you could have problems with water ponding on the bridge which would be even worse if the girder is sagging as well. I'm sure a lot of it is a preference thing.

 

[dhengr]

A good fabricator will always orient the camber up on a rolled section when he starts fabing. on that section, same as you did with 2x fl. jsts. when you were a nail banger. We used to cut the camber into the webs on large plate girders, then fit (bend, press) the flanges to the web; and in many built-up sections you can plan your welding sequencing to achieve the camber you want, or prevent camber you don’t want. If your camber is really a problem for you, a guy who really knows what he is doing with an acetylene torch could de-camber those girders in place.

I would have guessed about the same camber as Hokie suggested, DL+ .5LL, but I’ve not spent much time designing crane girders. My thought would be that under an average load, you would be running about level; and under max. load you would not have too much deflection. The camber and deflection would both be within the L/1000 range, one + and one -. Under how much load do you level and shim your rails?

 

[ToadJones]

The rail elevation is measured under no load.

 

[csd72]

I still do not see much benefit in cambering a crane beam, the total deflection under the crane load is still the same you are just changing the starting point.

 

[ToadJones]

csd72-
You are correct.
Cambering typical crane girders, IMO, is asinine.
It causes way more problems than solves.

 

[B16A2]

I've done 30m crane runways and cambered them DL+1/2LL per AISE. Off the top of my head, that was about 3/4" if they're designed stiff enough to meet reasonable deflection limits.

If you have 2" camber on shorter spans, there's obviously a design boo boo.


And yes, cambering is needed on longspans so that the crane doesn't have to climb as much when going towards a column from midspan.

 

[Teguci]

Chamferred shims of varying sizes. The rail clips can be shop welded to the top plate. The bottom plate size will need to vary and you might want to have it trimmed afterwards. I'm thinking the chamferred plates should be 8'-0" long or so. Bottom plate can be angled (in plan) to match the taper.

Just think of this as an extended wear plate system.