Launched Q&A site for people seeking specific software recommendations.
^ This site graduated from beta a couple months ago and is now having moderator elections. I think it's interesting to note that we have the same rate of answers per question and a significantly higher rated of answered questions, with about half the userbase and traffic.
For the civil engineers around here, I've been following this question since it was asked and it keeps bringing a question to my mind, especially after CableStay's answer (with their great images as always), and that is: does anyone think offsetting is really worth it?
Obviously there's cases where you kind of have to, but is it worth it in simple cases such as offsetting a beam from a slab as in CableStay's answer? All you get is a force binary replacing some of your bending moment, which as far as I'm concerned is more of a hassle than it's worth. Or do you really calculate your slab to flexo-compression and your beam to flexo-tension (assuming a simply-supported beam)?
@Air I know. Which is why I'm about to post a link to my question here below CableStay's answer asking for their opinion here. Or would that be bad form (this is actually my first time here in the chat as well, so I'm a part of the civil-folks-are-antisocial problem :p ).
It's now been one month since Engineering.SE entered private beta.
In that month, you've helped to collect nearly 250 questions and 400 answers on engineering topics ranging from steering devices and bridge loading to sustainable housing and metallurgical history. Well done!
But as fundamental ...
It's pretty barren, anyway. Needs someone more dedicated to using the Twitters than I and my fellow mods, or to be automated like many of the other SE Twitter accounts are.
I was never that big on Twitter and getting a smartphone didn't really help. Strikes me as a huge tide of noise, to be honest, but I can understand why some folks prefer it.
@Wasabi As I understand it, the offset is not really desirable but happens because shell meshes are defined by the neutral axis (midplane)- Therefore, an offset exists from the midplane to the outer fibre.
@theNamesCross Yes, I understand why one "technically should" use offsets. Your beam's centroid is not aligned with that of the slab, so there is indeed a force-binary with the slab in compression and the beam in tension which consumes some of (but not all) of the bending moment you'd get without offsets. My question is if you think these sorts of offsets are worth it, though. So basically, not "should you" (technically you should), but "do you".
I'm not sure I'm quite following you, @Wassabi, but my answer to 'do you or don't you' is that in practice I usually don't. It's really quite rare for me to do anything using plates or shells. Maybe for a curved girder bridge. But in general, I would simply define an equivalent composite beam element.
It seems to me there are two offsets being discussed: 1)The offset of the centroids between plate and shell, calculated by the parallel axis theorom. 2)The beam nodes do not align with the shell nodes (and must be transformed to nodal forces)
@theNamesCross I translated from Portuguese, just realized the term in English is force couple.
@theNamesCross I don't see any difference between those offsets. When you shift the beam away from the shell the parallel axis theorem is going to kick in as well.
@CableStay How do you do load distribution on multi-beam bridges without the slab? Do you use the methods of old (of which I cannot remember a single name right now)? Or do you work primarily with box girders?
@Wasabi Maybe they are mathematically the same, but I would think of them as separate if I were programming a solution because the offset of centroids changes the section modulus by virute of geometry, whereas the offset from mismatched nodes is simply a meshing problem.
@CableStay I actually will take a look at this. At my last job I worked almost exclusively with bridges and had to make a FEM model for each one. Only trivial bridges (double beam, basically) could be done by hand. I live in Brazil, so AASHTO isn't strictly available, but this would at least be handy for first approximations and verifications.
@theNamesCross Ah, okay, I see what you mean now. Yes, those are distinct problems. I was just thinking of what happens after meshing (with load distributions and whatnot), so once your second case is no longer an issue.
It's now been one month since Engineering.SE entered private beta.
In that month, you've helped to collect nearly 250 questions and 400 answers on engineering topics ranging from steering devices and bridge loading to sustainable housing and metallurgical history. Well done!
But as fundamental ...
