Instability

[Anonymous]

Hi,

another case of instability problems in the model.

I looked throw your 1h video were you showed problems and solutions. I tried different kinds of them in my model and looked throw the displacements and etc.

But if I managed to fix something another problem always showed up, another instability or something else.

I attached a copy of the model if you would please take a look at it try to explain/show what the problem(s) is.

The structure is made of cold formed profiles except the top beams in one directions. The columns is supposed to be two profiles back to back. They are now connected with rigid links.

Thanks in advance.

//Conny

[Anonymous]

The main problem is that you have created two bars in order to create a composite profile with the aid of rigid links. According to me, if you use the compound section creation like in the following picture, you can solve your problem.

[Anonymous]

Yeah but I'm not able to do that because I'm using cold formed profiles that's made in the user database and those can't be used here.

[Anonymous]

Why don't try to directly create the double section in the user profile creation?

[Rafacascudo]

but you can use "mirror" tool on section definition and create a "double section". Save it to the user database and it will work as one single section.

[Anonymous]

Yeah, at first I did that. But fore cold formed profiles they have to have a "shape type" put in the section database after making and saving the section to the database. And there is no shape type for two cold formed profiles ? (1001-1015) and they are all single profiles. For normal profiles I know there is a shape type for two profiles but I haven't find that for cold formed.

[Rafacascudo]

That does no prevent you from save it to database. Choose "?" Type and save it anyway!!

[Anonymous]

Yes I know, excuse me for repeating myself but then for the Calculations to work (code checking) the profiles must have a shape type (when editing the section database) and yes I always saves with the "?" left there. The calculations then at the code checking is then wrong fore that section because if I understand right how Robot calculations it only sees the shape type and the corresponding measures put for that type. So the properties for the wanted section is wrong then.

I made the columns this way because I watched a video were Artur put two bars together this way and the results seems much more accurate. But yeah I can't get rid of all the instability problems.     

[Rafacascudo · Accepted solution]

Your top beam is free to rotate about X global direction due to not a good modelling.

   For all top beams on both sides , you should extend them to the top node of the last columm profile . Rigid links will not be needed because the 2 columm bars will be already very rigidly linked by a very short top beam bar between them. Moment Releases on top of columms are not needed.

Can be something like this below

Keep the Rx Ry moment releases for the beams.

Also ,

1 - the distance from the columm CG to the back of the profile is 18mm . So if they are back to back and there is no extra space between them , the distance between their center lines should be 36mm at least. Right now it is just 10mm !!!

2- Define just one support for each base , not 2 or 3!!!! . Define a fixed support and create a rigid link for the columms start nodes , and then set the releases you need for the bars that go to this support.

[Anonymous]

Thank you for your advises. I will apply them on the model.

BR,

Conny