Problem with no convergence - how to define roof bracing?

[Anonymous]

Hi,

I'm doing my graduation project and I have problem with covergance of calculations. I'm doing 3D model of steel hall.
My model was ok and I reviced a resault of calculations, but I noticed that I have a collision in my model - a cross bracing and purlin crossed each other which is very difficult to create so I decided to make a lower column so that the purlin connection with the column is not at the same height as the cross bracing are located. After that I can't get resault od my calculation. I'm asking you to help me with getting answer of the question: the cause of my problem are settings of nonlinear analysis or the reason of my problem is that I don't have any brace in the direction C' so in that the program can't get the covergance? The changes I've written about are in the axis C'. I attach the link so you can download the files with my model,

model 12_08 before the changes: https://www.dropbox.com/s/m8ujtujfloqbnuo/12_08.rtd?dl=0

and 12_09 after the changes:https://www.dropbox.com/s/9ne94ksybuy9yfs/12_09.rtd?dl=0

Please look at this.

Best regards,
Karolika

@Anonymous I edited the topic for better findability and marked my post as the solution. If it did not work as an answer, please post your reply in this thread so that I can unmark the solution.

[Anonymous]

Hi,

I'm doing my graduation project and I have problem with covergance of calculations. I'm doing 3D model of steel hall.
My model was ok and I reviced a resault of calculations, but I noticed that I have a collision in my model - a cross bracing and purlin crossed each other which is very difficult to create so I decided to make a lower column so that the purlin connection with the column is not at the same height as the cross bracing are located. After that I can't get resault od my calculation. I'm asking you to help me with getting answer of the question: the cause of my problem are settings of nonlinear analysis or the reason of my problem is that I don't have any brace in the direction C' so in that the program can't get the covergance? The changes I've written about are in the axis C'. I attach the link so you can download the files with my model,

model 12_08 before the changes: https://www.dropbox.com/s/m8ujtujfloqbnuo/12_08.rtd?dl=0

and 12_09 after the changes:https://www.dropbox.com/s/9ne94ksybuy9yfs/12_09.rtd?dl=0

Please look at this.

Best regards,
Karolina

[Artur.Kosakowski]

Hi Karolina,

Sometimes a better is the enemy of a good enough bearing in mid that analytical model is the simplification and is not intended to be exactly the same as a fabrication one.

I noticed that I have a collision in my model - a cross bracing and purlin crossed each other which is very difficult to create

Which bar numbers are these in the working model? 

[Anonymous]

Hi,

the numbers of bars that I mention:

the same situation is on the other side of the hall. 

Thanks for your reply.

Best regards,

Karolina

[Artur.Kosakowski]

Hi Karolina,

Unless you define a node at the intersection the bracing bars will work separately from the beam.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

[Anonymous]

Hi,

thank you for your response.

I turn on that value so I will recive nodes in the intersection of braces. But now I don't recive the covergance

even when I calculate self weight.

I decide to use first option and do the solution like I show in model 12_08.

I set up this data: 

And for incersection of purlin and braces I make a compatible nodes:

I'm not sure if I did it properly. I read your post about compatible nodes, but it was different situation.

There was no column which support the purlin. 

Before I tick node at the intersection I recive the covergance.

I update my model, now you can download it here: https://www.dropbox.com/s/qshrouudc092dyc/12_09_grupy.rtd?dl=0

Can you help mi with this?

Best regards,

Karolina

[Artur.Kosakowski]

Hi Karolina,

My understanding of your problem was that you moved the bracing as it 'collides' with the roof's beam and this caused issues with convergence which you hadn't got before. My picture was intended to illustrate that you didn't have to move it as it actually doesn't intersect with the beam for the default settings (the check box unmarked). When you marked it (which you shouldn't if I understood your original issue correctly) the bracing started to take forces from the roof's beam.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

[Anonymous]

I have a problem with covergance (I think that was casused by) because I remove the bracing and then my column wasn't supported by any beam parrel to  X axis. I wanted you to check if that make a problem wiht covernegce or a problem with covergence was caused by data of nonlinear analysis.

Ok, I understand now. So if I have a purlin which are crossed by X bracing and i want to make a connection like this:

where bracins aren't below the purlin like this:

I don't have to make nodes in the intersection of the bracing?

Karolina

[Artur.Kosakowski · Accepted solution]

@Anonymous

Hi Karolina,

For the connection like this you should define a node and split bracing elements at this point into two parts as well as define releases for them (unless they are declared as tension only elements).

In the situation as below you model continuous bracings with no node at the intersection.

 

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

[Anonymous]

Ok, it was easy. I shouldn't bother you. Sometimes I wonder too much.

Thank you!

Karolina