Too much camber! 3

[jerseyshore]

Came across a new one. A local registered design professional sized a EDIT: W18x175 to span 32 ft clear across a new house. The problem was they also specificed an additional camber of 1/2".

Well, guess what, they put it up and it was bad. Not sure if the 1/2" camber was in addition to any natural beam camber (I'm not camber expert,very very rarely do I specify it), but it was humped pretty badly in the middle.

Normally we get called in to reduce deflection, not increase it. They framed the 2nd and 3rd floors and that thing didn't budge. Yes not all of the dead load was on it, but even with the full 15 or 20 PSF at each floor I don't think that's going to make a difference.

Any creative thoughts on how to get this thing close to level?

 

[jerseyshore]

Good info. Thanks. Learning a lot about camber. We've only done some cambering on concrete slabs, can't remember the last time I saw it for steel beams we designed. We do regular building structures so it's not something that comes up much if ever.

I try to tell these architects that I have little interest in designing roof rafters and headers. Call me for the complicated beams, shear walls, foundations, etc.

But when things get out of their league they just write down whatever the program says...

 

[driftLimiter]

I try to tell these architects that I have little interest in designing roof rafters and headers. Call me for the complicated beams, shear walls, foundations, etc.

Sounds good to me sign me up.

 

[dik]

It's not very often... usually spec that natural camber is up... and that's it.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[jerseyshore]

Looks like they are just going to live with the hump and shim the walls to match the curve basically. The 2nd floor will have gypcrete so they plan to level it with that.

The camber was certainly more than 1/2". I thought the framer said it was closer to an inch or maybe 1.25" off in the middle. Definitely why it was noticeable.

They even asked me if they should frame below as normal and assume the beam will sag back down to flat once all of the dead weight is on it. That thing is definitely over designed so I said I doubt it moves much.

I looked at the BeamChek manual and it says it recommends 1.5x dead load as the camber (you can see that in DL's image above). And Josh makes a great point above about not over doing the dead load when it comes to camber. So this program should be reducing DL deflection, not increasing it for camber. I guess that's what you get for a shitty program that costs less than $200...Scary stuff.

 

[sandman21]

I looked at the BeamChek manual and it says it recommends 1.5x dead load as the camber (you can see that in DL's image above). And Josh makes a great point above about not over doing the dead load when it comes to camber. So this program should be reducing DL deflection, not increasing it for camber. I guess that's what you get for a shitty program that costs less than $200...Scary stuff.

The 1.5 is for GLB as it accounts for long-term creep, which GLB should most definitely take into account. Anything over 7/8" for the steel beam and the fabricator screwed up and exceeded the tolerance. It sounds like people are taking the word of a contractor and not looking into what really occurred.

 

[jerseyshore]

Any idea what it uses for steel beams then?

I think this screw up was a combination of architect over-sizing a steel beam and specifying camber, then the fabricator messing up that camber.

 

[gte447f]

Jersey, sounds to me like the main mess up was in the specification of the beam. You just don't spec a shallow beam that is enormously heavy and also spec camber. Just spec a deeper (obviously preferable as step one) or an even heavier beam. Since this beam was already comically heavy, who cares how much heavier it gets? Whether the fabrication was also in error depends on what the actual camber ended up being. For 1/2" of specified camber on a 32' span steel beam, I think the acceptable range of camber would be 1/2"< camber <1" based on acceptable tolerances of -0" and +1/2". But, again, I say that no camber should have been specified in the first place.

 

[RWW0002]

I guess that's what you get for a shitty program that costs less than $200...Scary stuff.

For what its worth, some of the scariest stuff I have seen has come out of some pretty expensive software.. Just users that do not know what they are doing and/or lack of independent verification. Even some very capable users can really screw things up if they do not understand and verify software output, let alone users that do not have the means or ability to verify output even if they wanted to..

 

[jerseyshore]

I think we're all in agreement with that. But hey, it's good sometimes when architects try to tackle something way past their expertise, just makes us more valuable.

 

[phamENG]

just makes us more valuable

In a perfect world, perhaps. But the people who do it are unlikely to learn or, if they do, it's probably costly enough they won't have a chance to hire you in the future. And for everyone else, they only notice us when we screw up. Everyone will vilify the engineer that looked at the Davenport building, but the one who may have looked at the one across the street and forced them to fix it? He's just the dumb engineer cursed by everyone that inconvenienced everyone by making them move out for a few weeks while repair work was completed. (The latter is completely fictional and created to prove a point through juxtaposition.)

But I guess we don't need to go down that rabbit hole....again...