Ansys or. Abaqus? 3

[drile007]

Hello,

I'm a structural engineer who intend to use Ansys or. Abaqus software. Since I'm a beginner I'll be happy to hear any of your comments about the software. My emphasis is on:
1) User friendliness (e.g. material modelling, contact area modelling, FE mesh generation, results review, etc.)
2) Modelling basic nonlinear and buckling problems (e.g. steel-steel, steel-wood and steel-concrete connections, etc.)
3) Free accessibility to Manuals, Tutorials, etc.

As you can see I'm a beginner who is looking for user friendly nonlinear engine

Thank you for help me chose the right software

 

[TGS4]

Coke or Pepsi. Similar comparison.

If it is user friendliness that you are after, test drive both of them.

Both can be highly customized, so I wouldn't be worried about the "engine".

I have used both. I like whichever one I am using at the moment.

 

[TxAg78]

You would be remiss not to look at MSC's offerings: Marc and Nastran Implicit Nonlinear solver (aka SOL400). I have used all 4 in my career. Although maybe not Coke vs Pepsi, there's a lot of functionality overlap and preferences vary based on subtle differences. Also, don't forget to consider the "human factor" -a vendor's ability to support the application (hotline and/or local support).

 

[ESPcomposites]

Most new users don't have the luxury of choosing programs like this. Instead, they are dictated by the company you work at. How do you happen to have access to both programs but not familiar with what your peers are using?

If you are a working engineer, most people migrate to the software is used frequently in their industry or company. This way your peers can help you out when needed. For example, the aircraft industry tends to use Nastran for linear and ABAQUS for nonlinear solutions. I have used Ansys before and liked it, but it would be pointless to learn more.

Brian

http://www.espcomposites.com

 

[rstupplebeen]

If your company already has both learn both. If not demo both. Companies will generally solve a toy problem for you so come up with a representative 'hard' problem for your industry and ask them for a proof of concept.

If user friendliness is the top priority then look into some lower tier solutions such as SolidWorks Simulation.

Rob Stupplebeen

https://sites.google.com/site/robertkstupplebeen/

 

[rb1957]

"user friendly nonlinear engine" ... do you want world peace with that too ?

personally, be carefull with "user friendly". it often gives users an inflated impression of their knowledge. thread carefully !

 

[GenericUser]

I have used Ansys until 3 years ago, and been using Abaqus for about 4 months now. Abaqus seems more user friendly but once you start getting into it you begin to realize there is so much you don't know. Its a bottomless pit in some ways. The documentation is good for the regular UI stuff, but I have come to realize that the command line options and scripts offer so much more than anything you can do in the UI. And so I find myself wanting to learn Python, where as with Ansys I never felt any need to so. In addition, though Abaqus can be installed with windows, after 4 months I get the distinct feeling that unix installations is where users have the most versatility.

With regards to MSC products, since you're post mentioned 'user friendly', I strongly suggest you avoid MSC entirely. We tried MSC/Marc both in its native interface and through the MSC/Patran UI on an evaluation license and it was a nightmare. Doing the simplest tasks was very time consuming and I would assume anyone touting the 'user-friendliness' of MSC products is using a highly customized version.

 

[irq]

GenericUser - beside of ABAQUS, MSC/MARC is a leader in nonlinear solutions - it is very very strong in this area! The problem well known is the poor documentation, therefore in my opinion dedicated for users with any expirience in CAE. Support is working very well, recently MSC is organising many webcasts and you can master their products much easier than before. In addition the MSC.MARC developers are using very actively the user forums. What was most difficult for you to master MSC.MARC via Patran UI?

ESPcomposites - NASTRAN dominates in aerospace since ages. Why do you think, that Abaqus is used in aerospace as a standard nonlinear solution? MSC.Nastran has very strong nonlinear capabilities through SOL400 (similar potential to MSC/Marc, but more powerful) and you do not need to buy any another software.

 

[ESPcomposites]

irq,

NASTRAN has dominated for linear analysis and is especially good for loads models. It is also straightforward to run SOL106 for simple nonlinear analysis.

That said, many of the nonlinear specialists migrate to ABAQUS because it is a more robust solver and has more capability. I have seen this at both Lockheed Martin and Boeing.

When compared to SOL400, I believe ABAQUS has been around longer and has a much larger user base. SOL400 is an additional cost as well.

Brian

http://www.espcomposites.com

 

[irq]

SOL400 includes most of the functionalities of MSC/MARC, which has been released earlier than Abaqus. I personally place Abaqus and MSC/MARC in the same line. A big advantage of SOL400 is that it has Nastran input deck and you can easily interchange your data. Unlike linear analysis I see no clear favourite in nonlinear area, e.g. Airbus uses both Abaqus & Marc; in field of aero engines you can see ANSYS very often.

 

[ESPcomposites]

Perhaps SOL400 has been around longer than ABAQUS, I am not sure. But either way, it does not appear to be as popular other other nonlinear solutions (at least in the US aircraft industry).

MARC has been around for a while, but its implementation in SOL600 is relatively new and hence tends to have less of a following. And though I like Patran/Nastran, I too find SOL600 to be a clunky implementation (but also having some nice advantages). I found the contact solution to work well before when ABAQUS was giving me a problem.

Perhaps I should have better qualified my statement and restricted it to the US aircraft industry. It would still seem to the tendency (not an absolute) is to use Nastran for basic linear solutions, eigen solutions, loads models, etc. The preference does seem to be ABAQUS for more challenging nonlinear problems. In my experience, this combination tends to work well.

That said, there is obviously no absolutes and preferences will vary depending on the industry and in this case, even the region.

Brian

http://www.espcomposites.com

 

[teamohthee]

I have recently had to teach myself ABAQUS for some projects that I am working on and I would have to agree with the comment that ABAQUS is a bit of a bottomless pit, I feel like I have only just scratched the surface of what it can do and there are definately some things that frustrate me about it (but this is probably because I am still unfamiliar with it...).

I do think that it is very easy to pick up though - depending on how complex your model is - I had simple beam and frame test models up and running very quickly when I first started using it. One issue that I do have with ABAQUS is that although there is an extensive online help manual, it isn't very useful, i.e. it explains in detail each option/menu and its theoretical background etc but it doesnt really give you any help or guidance on how to actually USE the feature.

I have also used MARC and I find that very difficult to use, you really need to be taught how to use it and spend a good few hours running some test models before you can become comfortable with the software. I have also used solidworks, but I have to say that I prefer ABAQUS.

 

[AtlanticCoast]

Most people I know prefer ABAQUS for nonlinear analysis.

I find the geometry tools in ANSYS cumbersome, so if you have to create geometry in it I would not recommend it. The solid meshing utilities are nice. Many people who perform dynamic analyses prefer ANSYS.

NASTRAN SOL 106 is fine for many problems if you don't have large strains. SOL 600 is very good but harder to learn. It doesn't have the same strain limitations as SOL 106.

I recommend you consider the whole package (geometry editor, meshing utilities, number of load cases which can be run easily, etc), not just the nonlinear analysis solver.




John

http://www.AtlanticCoastAerospace.com