Wind Load Calculation with Partial Diaphragms

[bones206]

I noticed a while back that when you have multiple partial diaphragms (such as an industrial building with a high roof and low roof), the wind load generator incorrectly calculates the tributary heights using half the distance between diaphragms, instead of separately considering the diaphragms within each plan area.

I've notified RISA and they put it on their future development to-do list, but in the meantime has anyone else noticed this and found a good workaround? I'm currently applying all my wind loads manually and it's really tedious.

 

[JoshPlumSE]

We're talking about the diaphragm loading generated by RISAFloor that come into RISA-3D. Correct?

There just isn't a good work around for this. In cases like this, the wind load are best entered manually. I know this is a large percentage of low rise buildings. But, keep in mind that the wind loading concepts in RISAFloor were originally developed around the concept of mid-rise buildings where you have a common floor plan repeated for 5 or 10 stories.

My guess is that similar issues exist with RAM and ETABs. But, it would be good to know if anyone has run into this issue with them. If those programs have come up with ways to deal with this issue, then the quickest way to get the attention of the higher ups at RISA is usually to point out to them that their competitors do a better job at this.

Note:
One thing that makes this so difficult in RISAFloor is that the program knows Floors and elevations. It knows the extents of the diaphragm at each level. But, it doesn't know about the "skin" or "shell" of the building. I suppose the program could (or should) figure it out if there were exterior walls around the building. But, what about non-structural cladding supported by steel or such? That never shows up in the RISAFloor model, making it difficult for the program to know enough to get a good wind load.

Now, rather than provide a "perfect" solution for these cases, it may be easier for the program to provide a user 'bypass' of the wind load calc. And, just give users control over the numbers that go to each diaphragm. This would be a lot easier to program. I would think and it would be an effective work around for most cases.

 

[bones206]

Thanks Josh. I’m actually using manually created diaphragms in a stand-alone RISA-3D model. I totally agree with allowing user overrides for as many variables as possible. I have an irregular building in a hurricane zone, so I’m trying to figure out the best way to apply the eccentric wind load cases without the wind load tools at my disposal. As far as I know, you cannot manually apply joint loads to the diaphragm joints.

 

[JoshPlumSE]

As far as I know, you cannot manually apply joint loads to the diaphragm joints.

For semi flexible diaphragm that is absolutely correct. For rigid diaphragms it's totally allowed. Semi-rigid it's not as clear. If the diaphragm meshes to that joint, then it will accept load. But, it will be a concentrated load at that point.

For what you're doing, I might create a "dummy beam" along the front of each diaphragm (using semi-rigid diaphragms) and apply a distributed load (or partial length distributed load) to that dummy member. Then I can have a good visual representation of the load I'm applying to the diaphragm. But, carefully review the force distribution of semi-rigid to make sure it's behaving as you want it to (which I'm assuming is flexible).

 

[bones206]

I'm using rigid diaphragms. The kind that creates a cluster of 5 floating joints at the diaphragm centroid when you do the automatic wind load generator. The problem I've run into is that you can't override the magnitude of those generated loads. If you delete the automatic wind loads, those floating joints disappear. If you manually create your own cluster of floating joints, you get an unstable joint error message. So that's what I meant about not being able to manually apply joint loads.

I have been applying distributed loads along the roof beams as a workaround, but was just wondering if there was an easier way. I was also just curious how many people have actually noticed this issue, since it probably happens in a lot of models and isn't readily apparent unless you are diligent in back checking the generated wind loads.

 

[JoshPlumSE]

Bones206 -

This is actually much easier than you think at least with rigid diaphragms. Set your wind code to "none" and regenerate the wind loads. Those joints should disappear along with the RISAFloor calculated wind loads.

Then create the load points yourself. I usually use individual joints like RISA would have done. But, I add in the following boundary conditions to those joints (Y translation, X and Z rotation) whenever they land in space without connectivity to other members or walls or beams or such. Then you can apply whatever Joint loads you want to them.

Note:
The joints generated by RISAFloor actually internally generate instability warnings too. It's just that the program is smart enough to know that these are inconsequential and suppresses them.

 

[bones206]

Interesting. I thought that adding boundary conditions to any joint in the plane of a rigid diaphragm essentially locks the entire diaphragm in those degrees of freedom. I'll have to experiment with that.

 

[JoshPlumSE]

It's okay as long as the boundary condition isn't in the plane of the diaphragm. Out of plane the diaphragm has no stiffness, hence the joints connected to the diaphragm may report an instability in the out of plane directions.

 

[bones206]

Sorry, In my haste I misread your previous statement as "Y, X and Z translation" instead of "Y translation, X and Z rotation". What you said makes perfect sense. I think this is the workaround I was looking/hoping for. Thanks a lot