Incorrect bending moment in beam + pinned frame errors

[Anonymous]

I am working on a steel frame structure which I wish to model as a pinned-frame with cross-bracing providing lateral stability (although some beams will have fixed connections into columns). Thus far I have set up the model as follows:

- Beams which are pinned-pinned I have set their releases from Geometry>releases

- I have made bracing members pinned-pinned in a similar way (I did try tension only truss members, but this makes load application a pain as they should be excluded from self weight)

- Floors are one way spanning slabs, forming rigid diaphragms

I am not hugely confident with Robot, so your input on the analysis set up would be appreciated.

I am having the following problems:

- Never ending type 1 and type 2 instabilities. These are typically at the bases, which are pinned. I can go through all of them and then I get some new instabilities somewhere else... how can I get to the end of them??

- The bending moment is incorrect for some beam members. Typically, there is a continuous beam which supports a series of other beams, and I would expect the continuous beam to support these other beams. However, looking at the bending moment, it is as if there is a support at the junction, suggesting the other beams support the main beam (but this is not shown in their bending moment diagram, they are just pin ended). The load seems to disappear.

My model can be downloaded from:

https://mega.nz/#!NIVgVTAI!tGjybJqgyQOUdPpOIwT8KB_NuE6uYS0hcVpG6I8eSrU

[Rafal.Gaweda]

Have you sent us correct model?

There are no instabilities type 1 and 2 in it?

Tell us which bar numbers to check in which load case ?

[Anonymous]

Hi

Thanks for the reply. Try this one:

https://mega.nz/#!QJchQagS!6VuI1DdMPr9d8zmPxH8YKkBjZh2Q5M85H8lwR-wDTy0

I was messing around with the analysis types, changing it to "skyline", based on another post I found. This got rid of type 1 and 2 for the base supports, but now I instead get type 3 instabilities!

For the bar numbers, look at bar 640, bar 59, bar 268 for bending moment issues highlighted.

[Rafal.Gaweda]

Why do you expect no moments there on these bars?

There are no releases on these bars near these nodes.

You have to split these bars with nodes and apply releases correctly to newly created bars

[Anonymous]

Rafal - the long continuous beam should be supporting the adjacent, simply supported beams. The bending moments indicate that the opposite is happening - that the long beams are being supported at the junction with the smaller beams.If this were the case the small beams would be acting as cantilevers, however their bending moments indicate they are simply supported. So there is something odd going on - as if there were a support at these locations.

If it were working properly, then the bending moment would be completely on one side of the beam (except at ends as it is fixed).

[Anonymous]

The bending moment should be more like this attached file

[Anonymous]

The attached shows a test case with a UDL applied to bar 228. As you can see, the load is not transferred to bar 640 - it's as if there is a node at the intersection.

[Rafal.Gaweda · Accepted solution]

Use one of these panel calculation models for floor panels:

[Anonymous]

Thanks! I used the right hand side option.

Are you able to comment on the modelling errors (type 1 and 2)? Do you think the approach I am using is correct?

[Rafal.Gaweda]

Instabilities are related to models.
In 90% of cases - due to wrong definition of releases

http://forums.autodesk.com/t5/forums/searchpage/tab/message?filter=location&location=forum-board%3A3...