Autodesk Robot Structural Analysis 6

[ChaliK]

I would like to know how good Autodesk Robot Structural Analysis is compared to other renowned products like SAP2000, RISA-3D etc.
Thanks in advance!
Charles

 

[OCI]

I have used Robot on a few projects. It seems to be a really good program. I like the interface and it seems to have all the features of the other programs mentioned (non-linear ect.) I am also very interested to hear any other opinions on it from others.

 

[Stazz]

I heard it's unbreakable. Plus its getting integrated with Revit.

 

[OCI]

If you are refering to stability, it's definatly not "unbreakable". I have crashed it a few times. Before Autodesk bought it Robot had the ACIS solids kernel (or part of it anyway) which gave you some powerfull solid modeling tools. Autodesk did not keep the ACIS licence with Robot so this functionality is gone for now. The toolbar for these tools is still there, the buttons are just greyed out. This makes me think Autodesk will add a solids engine back in soon.

 

[OCI]

My only other real complaint with Robot was the lack of support material like tutorials ect. Hopefully this will change now that Autodesk bought it. Should become pretty mainstream.

 

[Bausk]

I made several significant projects with Robot - lightweight steel pavillions, concrete frames, retrofit of old industrial cast-in-situ RC building etc.

The lack of support material is really frustrating,
But, anyway, Robot ME is a great engineering software despite some significant difficulties in setting up correct models.

I hope Autodesk won't spoil it. Since their having bought Robot from RoboBAT, they did absolutely nothing to improve the software.
The Robot's strongest point is the very convenient user interface. I would recommend it for all FEA tasks, and it is especially suitable for steel design.

 

[GregLocock]

" I would recommend it for all FEA tasks"

Thereby revealing a disturbing lack of experience by which to make such a recommendation.

How is it for crash simulation of cars?



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[OCI]

Why would you insult him like that? Autodesk Robot Structural is obviously a structural engineering FEA package.

 

[GregLocock]

Because silly recommendations like that get taken literally and out of context.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[ThomasH]

Hi

I think this to some extent is a question of not speaking the same language.

Robot is as far as I understand it an excellent software and compares well with other softwares in the same category. Software's like SAP, STAAD etc.

But it doesn't compare with softwares like ANSYS, ABAQUS, Nastran etc. And that opinion comes from the guys developing Robot.

I was at a demo and they called Robot a FEM-software. I asked afterwards how it compres to for example ANSYS. And the reply was "We are not in THAT division".

So I would say, "All FEA tasks"? No definitly not.
Structural FEA in terms of steel frames etc? Sure, as long is it doesn't get TO nonlinear or TO dynamic.

Regards

Thomas

 

[GregLocock]

Exactly

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[OCI]

Does anyone know of any resources for help with this program. I have spent hours with the demo version, alot of it is really straight forward. Some of the more complex and important features are really poorly documented. I am really having trouble with the response spectrum cases and settings. I really think this is the program I want to purchase and use but I need to understand these things before I commit. I don't want to hear from a salesman. I'm looking for other users or tutorials. Does anyone use this program on a daily basis? Thanks alot.

 

[JuhaA]

Robot Structural Analysis (RSA) is:
- Linked with Revit Structure already (and works fine), see the Revit Extensions!
- Guys doing RSA are those that worked with RoboBAT before the acquisition: No major change there, I think.
- Robot is for STRUCTURAL FEA, also Non-Linear and Dynamics
- I have been doing and training linear, non-linear, dynamic, beam, plate and solid analysis with RSA over the past 7 years and also been using STAAD.Pro long before Bentley bought it.

I am happy with Robot - even with its problems in some cases.

 

[OCI]

JuhaA,
How do you manualy create an envelope load case in Robot. The load case creation system seems very powerfull, but complicated. It sounds like you get paid for this sort of help. So if you can help me, Thanks. Otherwise I would be willing to pay for some training or for some additional documentation if you have it. Where are you located?

 

[zstone]

As i know jou can display results of some load cases like envelope, but is not possible to use them in analysis and design of members.
But during design of members, Robot use envelopes of combinations in automatic way.
During design of reinforced concrete members for example.
If you want to display results of some load cases like envelope you need to type numbers of load cases in that field by dividing them by space.

I use robot from earlier version, and i think that very nice software.

 

[OCI]

Thanks, I know how to display multiple load cases and the upper or lower bounds(envelope). I was wondering if any one knows how to setup a load case as an envelope that could be combined with other loads in a combination. SAP2000 lets you do this and it's really usefull. I hope that made sense.

 

[masomenos]

I'd like to make a few observations for those doing FEA work who do not have structural products like SAP2000, Robot, or Midas on their radar, because most posting here seem to be very unaware as to what is available with structural FEA.

I'm speaking as an almost daily user of SAP2000, and I have a good friend who uses Robot and he loves it. My observation is that there is a huge unawareness among FE users as to capabilities of structural software, which seems to be described as "frame element" only. this is nonsense

1) SAP2000 can perform linear FEA analysis as well as nonlinear time history using direct integration or FNA analysis for models with frames, and/or shell/membrane and solid finite elements. I understand that both Robot and Midas offer nonlinear time history capabilities but I have no personal experience with those programs. Based on the comments in this thread, it seems there is widespread unawareness about the ability to run nonlinear structural FEA with Robot or SAP2000.

2) My consulting firm worked with a Canadian company, Hatch, about 3 years ago, a company which used ANSYS, SAP2000, and Staad for their structural and FE requirements. SAP2000 had replaced ANSYS for FE dynamic analysis of foundations supporting vibrating equipment modeled with FE and because SAP was more productive, and SAP2000 was in the process of replacing Ansys in the application of pushover analysis involving nonlinear material analysis of frame elements. A year later, I was involved in a consulting project in which SAP2000 was replacing Cosmos FE for nonlinear snap though buckling analysis involving large deformations. So the idea that only traditional FE programs like Ansys, Algor, etc should be used for FE or nonlinear analysis is an idea which is wrongheaded in many applications.

3) Face it, basic FE tasks like response spectrum earthquake analysis, load combinations, manipulation of local axis of FE elements, and rotation of local axis of frame elements can be a bitch in many/most FE products compared to many of the structural programs. And of cource, FE products offer little or no design codes.

It's unfortunate that the Robot presenter cited above in this thread made such a blanket statement when questioned about comparison to Ansys, because Robot, or SAP2000 may have been superior choices for many applications involving FEA, depending on the design application.

Having said that, for basic heat transfer analysis, thermal load across a wall, we could not use SAP2000 or Staad. Both programs offer thermal load, but that's it. Also, neither program offer kinematics, or temperature dependent material properties, so I undestand where traditional FE programs offer value. Structural programs offer zero in multiphysics. So please understand that I'm not sugar-coating, I'm just sayin

Bottom line - FE programs are being used in many applications where certain structural programs are far better suited and also more economical. Just open your eyes and your minds

 

[ThomasH]

masomenos:

I don't mean to be rude but when you ask if FEA people are aware of what structural software can do. Obviously I can only speak for myself but I have to ask, are you aware of what general FEM-software can do?

The reason I wrote my post was the statement that somebody recommended Robot "for all FEA tasks". That's plain wrong. I think I could write a list of 5 - 10 analysis types or special applications. And no software in the world, "structural" or "general", could do them all. So I don't thing "all" should be used in that context at all.

Greg mentioned "crash analysis". I don't know of any "structural" software that can do that. And I have a pretty good idea what they can and can't do. I have collegues who use them daily (primarily STAAD but we have others as well). My work can on occasion be to run the analysis they failed to do in the "structural" software with a more "general" choice.

You say that some things can be a bitch in general software, very true. The time required to be efficient is also often less in structural software. On the other hand, the possibilities in a general pre/post processor exceeds the possibilities in the typical structural pre/post processor. And the general choice usually means that it can "talk" to several solvers. The structural choice often means one solver.
Once you have learned the general software it's very often faster to work in that alternative. And the control of the solver is usually much better. But again, it often takes longer to learn.

Obviously there are cases when the structural software is a better choice. It usually cheaper in addition to the other advantages (speed of learning etc).

As for the presenter of Robot, I would say that he knew what he was talking about. Robot isn't Ansys just as Abaqus isn't Ansys. That doesn't mean that Robot is bad. Just make sure that it fits your needs before you buy it. If you need Ansys, don't buy Robot.

Finally, you mentioned SAP200. I don't use it but was part of a test of it some years back. We didn't buy it because it didn't fit our needs at the time but it's not a bad choice.

The structural softwares have a huge undertaking in keeping the codes up-to-date. And for that reason I don't think they will ever catch up with the analysis capabilities in the general FEM-softwares. But if you need codes and don't need certain dynamic or nonlinear capabilities then general FEA might be a bad choice.

It entirely up to you.

Regards

Thomas

 

[johnhors]

A colleague of mine writes help manuals and gives assistance to the helpdesk of a well known structural FEA code. He gets calls from the helpdesk when the guys there have trouble responding to a user problem or enquiry. Recently a user had trouble tracking down the root cause of an unrestrained freedom causing the solver to fail, the helpdesk was not able to fix it and incredibly my colleague stated a lot of users would simply restrain the offending point. My comment was why not perform a natural frequency analysis and the mode shape of the zero frequency modes(s) will indicate where the unconnected region lies. It then transpired that the structural code is not capable of performing "free-free" modal analyses. My colleague put a request in with the developers to enable this function. Then unbelievably the developers did not understand the request! He even told them the mathematics involved (which is very simple using an eigen shift on the stiffness matrix). That just confused them!

This is something that is taken for granted in a general FEA code.

 

[masomenos]

masomenos:I don't mean to be rude but when you ask if FEA people are aware of what structural software can do. Obviously I can only speak for myself but I have to ask, are you aware of what general FEM-software can do?The reason I wrote my post was the statement that somebody recommended Robot "for all FEA tasks". That's plain wrong.
~~~
Please re-read my post, since I never claimed otherwise. I specifically responded to your comment about the Robot presenter who said that Ansys, Abaqus, etc. were on a different level, suggesting a superior level. In many cases yes, other cases no. And btw, as you know, many general purpose FEA programs are similarly not well suited for crash simulations either. Another commenter mentioned nonlinear time history analysis, which indicated to me an unawareness that some structural software products can perform nonlinear TH analysis. I was specific in listing applications where certain structural programs may offer advantages over general purpose FEA programs, and I listed specific shortcomings. Along these lines, it would be nice to know of the specific reasons why your company did not purchase SAP2000 at the time. Were there multiphysics requirements? Heat transfer needs? If so, that's completely understandeable. In other applications, that may not be the case. I agree with your observation that structural programs are limited to one solver, although for many applications that is not a problem. Application programming interfaces are available with several structural analysis programs, facilitating their integration with other programs

I would ask Johnhors to explain what a "free-free" modal analysis involves, and further to name the structural software program he refers to as having limitations so that we can be talking apples-to-apples. In SAP2000 I can run a modal analysis based on an unstressed state, modal analysis based on multiple pre-stressed cases, and Ritz vector analysis all in the same run. I'm curious as to whether most FEA programs can do the same. Do FEA programs even offer Ritz vector analysis? Because we take it for granted

I stand by my statement that in many applications general purpose FEA programs are used in applications where certain structural software such as Robot or SAP2000 would be far better suited.