RISA Deflections
RISA Deflections
(OP)
Hi all,
Seee the attached RISA-3d model.
I am having incredibly high (24 in. and above) deflections, as well as shears 9.999+. The load is not very large, 100psf and some other point loads that aren't having a great effect. What is the problem with this model?? These are the exact conditions of the structure and it isn't falling down, so it's a modeling error.
Also, when I set the end releases of Member M28A to fully fixed, I get a Pdelta diverging at N27. Why would this be occuring? Any insight is appreciated. Thanks
Seee the attached RISA-3d model.
I am having incredibly high (24 in. and above) deflections, as well as shears 9.999+. The load is not very large, 100psf and some other point loads that aren't having a great effect. What is the problem with this model?? These are the exact conditions of the structure and it isn't falling down, so it's a modeling error.
Also, when I set the end releases of Member M28A to fully fixed, I get a Pdelta diverging at N27. Why would this be occuring? Any insight is appreciated. Thanks






RE: RISA Deflections
My first thought is that a large shear check is often indicative of a large torsional force (which produces shear stresses). And, of course, your high deflections are what then cause your PDelta divergence. So, if you figure out your deflections then you will probably also correct your divergence issue.
I can't tell much more from just a picture. But, it doesn't look like you have any horizontal bracing in your structure. So, what is tying the various frames together?
RE: RISA Deflections
1) Is the strength axis oriented the correct direction of said members?
2) Are M28A and M21 trying to support the long span beam? My guess is there is a large shear being forced into those short beam from the long beam. If the long beam is relatively "weak" compared to the short "stout" members it could be causing this issue. Sometimes draw order effects RISA models, but that might be in Floor only.
RE: RISA Deflections
At M28A, I don't see a member designation on the main beam between M19 and M21. It looks like it should be very similar to M16 (on the other side)- I would see what was input differently between M16 and that beam.
RE: RISA Deflections
Josh - The horizontal bracing is only at the four corners. Like I said, this is the layout as it currently is. Spans have been shortened, replaced etc. over the years without any documentation.
Badger - yes all the members are oriented correctly. Almost all beams are C8.
Hawkaz - The spans are the same on both sides as far as I can tell. I'm (relatively) new to 3D.
RE: RISA Deflections
Something unrelated, but probably bad: You have three planar diaphragms in each of the three different planes. That's odd... I'm guessing this was a modeling mistake.
Now, back to your original question....looking at your large deflections.
Just start with one simply supported member (M28) and trace its support through the structure. Vertically M28 is supported nicely on one side by M4. On the other side, M16 is supposed to provide support, but is deflecting wildly itself. So, let's look at M16 more closely.
M16 is supported nicely by M15 on one side, but M13 on the other side doesn't provide much support. So, let's look at M13 more closely.
M13 is supported on one side by M5, but on the other side I don't see anything that can support it. What's the support for M13 where it connects to N27. To me, this is the biggest problem in your vertical load path. If you provide some support there your deflections should improve dramatically.
RE: RISA Deflections
Thanks for breaking that down. It makes more sense when put that way and there actually is a column at N27 in the design drawings that has since been removed.
Yes I tried to define diaphragms without knowing entirely what I was doing, and forgot about them. Still not sure the purpose of them.
One thing, I have a few connections (M15 / M36) modeled as fixed... The actual connections are bolted clip angles, not fixed. When I change to pinned, the frame becomes unstable. Perhaps that is understandable as you have two perpendicular members that meet at a pinned connection.
RE: RISA Deflections
RE: RISA Deflections
MEMBRANE Diaphragms are usually used in RISA as a way to quickly model a rigid concrete slab. Something which has a very large in-plane stiffness that will tie together the various lateral force resisting frames. But, which does not have stiffness to resist vertical loads. In that way, they are really good for capturing the diaphragm action of concrete decks as they don't affect the vertical stiffness of the joist / girders / frames.
PLANAR diaphragms are almost never used. That would replace a massively rigid concrete slab (10 ft thick?) which has near infinite rigidity both in plane and out of plane.
RE: RISA Deflections
RE: RISA Deflections
You can have multiple pinned end beams frame into the same top of column node. You will need to leave the top of column set as fixed, and change the bottom of column boundary condition to be fixed in torsion about the vertical axis. Might sound wrong, but it will make the model stable. Run it and check that no (significant) torsion reaction actually occurs. If so, you have other issues...
RE: RISA Deflections
Sorry Josh, not meaning to bother. I just want to learn