Modelling Patch Loads in ETABS using Null Areas

[arik79]

Hi

I am modelling a mid rise building in ETABS and I am having some problems with modelling null areas/shells. The reason I am modelling these is to add additional loads in certain areas (balcony, corridor etc.) to a slab element that has already been modelled and has had its load assigned.

I feel ETABS is a little inconsistent when it comes to modelling null areas. For this model, the addition of one, randomly positioned, square, null area causes ETABS to give warnings about instability. Deleting that null area causes there to be no issues at all.

I have done the spot checks we usually do for our ETABS models and nothing seems to be out of the ordinary (modal mass participation, base shear, etc.), so I am unsure why these warnings are popping up with the addition of a null area.

I was hoping someone could provide some guidance about the correct way of modelling null areas in ETABS OR some guidance on how to add additional loads like balcony/corridor to a floor plate.

Thanks in advanced

 

[Ey_2]

Why don't you have different shells for different loads?

 

[Trenno]

I looked into this previously: [URL unfurl="true"]https://www.eng-tips.com/viewthread.cfm?qid=455443[/url]

I think whenever elements with infinitely small/large stiffness or weight, such as null shells & frames, the stiffness matrix gets all out of whack and will start spitting out errors when you try and do things beyond a simple linear elastic analysis.

 

[evo10]

you can split the area object manually why not split at the areas you want and add additional load case there

 

[Trenno]

When automeshing is involved, it's much cleaner to have simple large shells. The more joints you introduce (ie manual meshing), the more chance for joints/shells/frames to become unconnected unintentionally.

 

[arik79]

It's strange because this building has been split up into two models as there is a permanent movement joint. In one model there are almost 150 null shells used to add additional loads, no errors, warnings, or information messages

But in the 2nd model I am able to model 25 null shells without any problems, but for some reason the 26th just causes warnings to popup? Its just inconsistent.

The 1st model is smaller than the 2nd model which may be a cause, not sure :/

 

[evo10]

did you check the error file usually you can see by coordinate what potential issue is

 

[arik79]

That's the thing. There's no errors, there's only warnings and informational messages. And it doesn't point to anywhere specific. I've attached the file in the thread if you'd like to take a look

 

[Trenno]

If you look at the WRN file, it will tell you the labels of the offending shells and the magnitude of the load lost during the calculation process.

 

[arik79]

The WRN file simply shows the warnings from when you check the model using Analyze > Check Model...

There are no warnings being generated

 

[arik79]

Okay, I figured out the issue. Turned out I had assigned the same shell diaphragm to two shells which were disconnected. Assigning different diaphragm's fixed the issue. Silly error!