Making Moment Diagram match other programs

[c.netoYXD2Z]

my results from the Robot Model is very different from the one I obtain using another analysis software.

Overall, the moments in Robot are a lot lower. In Robot, the slab is taking moment, even though the slab is considered to be pinned along the edges.

I would like to verify by removing the slab from taking moments (in both directions) to see if the results are closer

is that possible? how can it be done?

Thanks.

 

[Stephane.kapetanovic]

hi @c.netoYXD2Z 

you have chosen to model your beams with offsets.

Doing this you cannot abruptly compare the results.
Delete them and you should have a view similar to this one.

Best regards

[c.netoYXD2Z]

I need the top of the beams to be aligned with the top of the slab. 

If I put the beams in the center of the gridline and the slab also in the center of the gridline, the beams will cross the slab in the middle of each other's section, which is not what is happening

How can I model it properly?

thanks

[Stephane.kapetanovic]

@c.netoYXD2Z 

as you can see on the view that I provided you the results are practically identical to those you were expecting to compare which seems to prove that the model of calculation must not have offsets. You can also change the section of your beams to T to include a slab part if it is the case you want.

best regards

[c.netoYXD2Z]

Ok, so that is the problem, however please look at the figure attached

I modelled (1) and I believe your results correspond to (2).

I need to model (1) and I want to know what is the proper way to do it

Because on the program that I compared my results with, I modelled (1) for sure.

I thought I had to use offsets to do that on Robot but they are creating wrong results.

how to model properly and get the right results then?

I really appreciate your help

thanks again

 

[c.netoYXD2Z]

This is the extruded view of my model in the other program.

I dont understand why using offsets on Robot to do exactly the same thing gives totally different results

I probably did something wrong int eh use of the offsets and I dont know what, because in the views in Robot it shows that it did exactly this as well:

 

[c.netoYXD2Z]

another thing to note is that, removing the offsets does not give results close to the other model because those results in the other model are for DL+LL. The figure you showed is only for DL

I just checked and removing the offsets, the DL+LL combo is actually around 80-100 kN. 

I still don't understand what is going on

any help is appreciated

thanks,

[Rafacascudo]

If you want the slabs to just distribute the loads as if it was a manual calculation , you should use claddings instead.

Using slabs, the closest aproximation to the manual calculation would be not using offsets.. But if the slab stiffness is comparable to the beams , results will also be far from what you want.

.

There are literally hundreds of posts explaining how the beam-slab interaction works in Robot.

Just search the forum

[Stephane.kapetanovic · Accepted solution]

@c.netoYXD2Z 

no significant differences

as explained by @Rafacascudo try to search on forum some thread about beam/slab interaction procedure and a API based soft to determine part of slab to be modelized.  

Best Regards

[Romanich · Accepted solution]

Hi @c.netoYXD2Z ,

Please check this topic:

https://forums.autodesk.com/t5/robot-structural-analysis-forum/api-macro-creation-of-quasi-t-amp-l-r...

[c.netoYXD2Z]

Well, it's close in some locations but quite a few have some significant different values.

I was just able to obtain results close to yours and compare to the other program.

I am going to read the link that Romanich in the comment below suggested and verify other alternatives and compare

given that this is a problem with several solutions depending on what I want to obtain/model, I will accept your answer

Thanks for your help Stephane

[c.netoYXD2Z]

Thanks Romanich

your approach seems to be the used most used one, I will use it later and see how my results change

I am surprised this option is not implemented in the program since it seems a lot of people would like to use it and this discussion dates back to 2011.

Now I understand that the slab may take some of the load itself, but I am not sure why that doesn't happen on the other program, I wonder if they have an option for that that by default assumes that when you have a beam+slab system, the beams have an artificially higher stiffness.

thanks anyway once again