Beam cambering 1

[chencfp]

I have a WF beam (W40X324)cambering issues for help. We have a 60-ft span beam and will deflect about 1.5" under dead load. It will be expensive to camber this beam to 1.5". The beam will be under L/240" deflection under total load. Do I need to camber the beam? Any suggestions? Thanks.

 

[sdz]

Judging just on Span on deflection ratio that looks OK but you also have to consider the application. Is there anything that could limit maximum deflection including appearance? If it is an industrial or mining application it should be OK.

 

[civilperson]

What cost was quoted for cambering? My experience has shown it is a small add-on, less than 4% of the total or less than 700 dollars. Well worth the cost to approach level with all the dead loads deflecting.

 

[chencfp]

Thank you. It is more than 2000 dollars for cambering each 60 span W40x324.

 

[SteelPE]

If a flat floor is required other options include increasing the beam size until your camber is less than the camber that is naturally allowed in mill tolerance.

Another option, check the beam and other members for the “ponding” concrete and have the contractor pour the floor flat (not sure how they would do this).

 

[ChipB]

When the shape comes off the mill, it will already have a natural camber to it. Given the length of the beam, it's probably already gotten most, if not all of the camber required. The fabrication shop should automatically fabricate the beam "natural camber up" but you may want to call it our on your drawings just in case.

 

[ChipB]

I stand corrected. It looks like half of your required camber would be taken out naturally:

http://www.modernsteel.com/Uploads/Issues/July_2006/30755_steelwise_camber.pdf

 

[ali07]

Chip B

Excellent thread reference, it explains most of the questions.

 

[FLCraneBuilder]

Interesting that the AISC document stated $50 to $75 per beam, but chencfp was quoted $2000 per beam. However, I can see the $2000 per beam easily - if it can even be accomplished. 40 x 324 is a very large beam to camber.

Another approach is to build cambered plate box girders. Overhead crane box girders are commonly built with camber. This might invoke other problems - but may be worth considering.

Lastly, do not count on mill camber to be consistant! Beams come in with a huge variety of sweeps & cambers. We have not used W40's, but we see a lot of W36's that are dead straight - and have seen a lot that have 2" in 60 ft as well

 

[ali07]

If project has already been awarded, they will charge you $3000, but if u specify them before tender, u'll get it in $100. As we checked form fabricators, they charge 2 hours of labour at the rate of $50 per hour with no limitation of span upto 60ft.

But make sure you will never get smooth cambered beam. They will camber in the middle third part of the beam and u might end up with an odd shape at the centre. If it is for roof beam, then it might be fine.

 

[chencfp]

Thanks all for sharing so valuable information.

 

[BAretired]

Camber is not a strength requirement. Structurally, it doesn't matter whether or not you camber the beam.

You don't say whether it is a roof beam or a floor beam and if floor beam, what is the occupancy of the floor?

You don't say what the beam supports. Is it a slab or open web steel joists spaced at six foot centers?

If the beam supports joists, you could provide variable depth shims under each joist to achieve the same effect as camber.

In any case, 1.5" dead load deflection in a sixty foot span is pretty trivial. If you can achieve fifty percent of that by natural camber, I suspect nobody is going to notice.

 

[slickdeals]

There are some important things to be considered.
1. The rotational restraint of the beams is never accounted for. For a W40x234, I am sure you have plenty of bolts at the ends, which offer a considerable restraint, which is never accounted for when you calculate your deflections in analysis. A true pin was probably assumed.

2. On the contrary, cambering maybe important. If the beam only had mill camber, then there will be some deflection on the beam with only wet concrete. Sometimes this deflection might increase the amount of concrete needed to finish the pour (more concrete that was not accounted for). Generally, a 5 psf addition is used for extra concrete.

3. Camber is not a science and it is never precise. Typically during cambering they let the beam yield a little and then let it relax. Plus, camber is typically required to be measured in the fabricator's shop and not in site. There is variability that needs to be accounted for.

Out of curiosity, is the beam W40x234 from strength requirements or camber requirements? If it is from camber requirements, why don't you use a smaller beam and specify camber?

It am surprised that a W40x34 is deflecting 1.5" only under the weight of concrete. What is the spacing on your beams?

 

[BAretired]

In Canada, a W1000x483 is equivalent to W40x324. Its Moment of Inertia is 10700E6 mm^4 or 25,707 in^4. For a sixty foot span, the load required to cause a deflection of 3" (L/240) is 662#/'. Dead load causes a deflection of 1.5" (L/480) so the dead load is about 330#/'.

Thus Dead Load = Live Load = 330#/'. I don't think there will be much concrete on that. It must be a roof beam spaced at about eight or ten feet on center.

Wouldn't it be more economical to use open web steel joists for this roof? Then you can get camber at no extra cost.

 

[BAretired]

There must be some mistake in the original post. A W40x324 on a sixty foot simple span would deflect 1.47" under its own self weight.

Please clarify the roof dead load, live load and beam spacing.

 

[271828]

The MSC article reminded me of a question I've intended to ask for a while.

Every time I see camber guidelines, there is a dire warning to not camber a spandrel beam. I don't buy this for a second. Here's my reasoning and I'm interested to know why I'm wrong.

Option 1: Say I camber a spandrel beam to 85% of the deflection due to the deck, concrete, and beam weight. WHEN THEY START PUTTING ON THE CLADDING, the slab will be more or less flat at the spandrel.

Option 2: Say I don't camber the spandrel beam and the spandrel has deflected 1.5"-2" and THEN they start to install the cladding.

Can somebody explain to me why Option 2 (what all the "experts" say to do) is better?

 

[BAretired]

Contrary to my previous post, the beam deflects 3" under a total load of 662#/" or 7944#/'. My earlier calculation was evidently wrong. Dead Load equals Live Load = 3972#/'.

This may be a good application for a shored composite beam. You could reduce the weight of steel substantially and achieve the desired camber by adjusting the height of the shores under the beam.

 

[ali07]

271828

I agree with you, but if beams are resting on spandrel, is it easy to camber it.

 

[BAretired]

271828 and ali07,

Spandrel beams should not be cambered because spandrel beams should be supported so that they do not deflect significantly. It is not a particularly dire warning. You may do whatever you like, but a spandrel beam which deflects 1.5" to 2" is too flexible, in my mind. It should be supported by an additional column half way between the existing ones so that the deflection is kept to a minimum.

Large deflections are simply not acceptable on an exterior wall because there can be windows and doors subject to damage from excessive deflections.

An additional column provides the opportunity of using a two span continuous beam in the exterior wall, decreasing deflection even more and making camber totally unnecessary.

In the final analysis, it will be dependent on the judgment of the engineer doing the design.

 

[271828]

BAretired, I agree completely that a spandrel beam should be held to tight deflection limits to ensure no cladding damage. However, this is an entirely different serviceability issue than cambering out wet concrete deflections. I'm typing mostly about composite systems.

Deflections that affect cladding are usually computed using the transformed moment of inertia, which can be 2-4x larger than the bare steel Ix depending on the situation. The uniform load from cladding is often very small compared to the load from wet concrete + deck + steel beam, also.

It is common to select a spandrel beam, using camber to take care of the bare steel/wet concrete deflection and then have the deflection due to superimposed loads be tiny. What's the problem if this is the case?

Also, I could probably count on one hand the cases in which I've been able to talk an architect into giving another column along a spandrel. If only I could get so lucky!

I still don't see it and am thinking the "experts" just throw that out there with no real justification.