Features of NX Nastran compared to NEi Nastran

[adnanh]

Hi,

I am aware that there was already some discussion on capabilities (speed, accuracy etc) of both solvers. However I am especially interested in the features provided by both programs. I have already searched the website of both suppliers and I have some impressions. I have to say that the Noran' site gives a very good impression on features covered by NEi Nastran. And I am impressed, indeed. I have also checked the site of UGS for an overview of NX Nastran an also there quite some information can be found. However, I have to admit that I have some preferences in favor of NE, based on that what I have seen and read. The issue with NE, however, is that is seems that here in Netherlands nobody uses NE Nastran. Those who use some kind of Nastran, use either MSC or NX. It is not clear form why this is the case.

Basically, my question for you is whether NX has more or less the same features as NEi. Take for example the NEi Nastran demo (

http://www.nenastran.com)

of the nonlinear transient analysis of an combustion engine where probably the complete mechanism-assemply was imported in FEMAP(?) and stresses and strains are visualized in this 'dynamic' mode, thus while the engine is running. So the question is, are these types of analysis possible within NX Nastran combined with FEMAP pre- and postprocessor?

If not, what is actually needed to run these kind of simulations?

Kind regards,

Adnan

 

[adnanh]

Can somebody tell me anything on the costs of both Nastrans with basic module and full package, thus including the possibility of modeling nonlinearity (geometrical, material and boundary condition), various kind of contact problems, fatigue analysis, impact analysis, simulation of welds etc.

Thanks in advance,

Adnan

 

[Ronald62]

I'm sure you get the most reliable costs directly from a dealer.

I've been using both of those - NE and NX. My opinion is that NE is the better marketed product than NX. I'm not trying to say that NE is bad product, but please, do not make a decision on the basis of the internet site.

I give you a good advice: Ask NE and NX for testing use. I know they give. Make different tests and use all type of analysis types and compare the results. This is the most confident way to decide which of them is better. Do not believe all you read or hear. Believe your own tests.


NE includes nonlinear solver and you can solve many types of nonlinear problems using it (nonlinear material, surface contacts...).

NX's "basic package" also includes nonlinear solver (SOL106), but it does not contain surface contact. The basic package (I don't know its real name...) includes linear static solver (SOL101), of cource. This linear static solver includes the surface contact! They call it 'linear contact'. It seems to be a linear solution even when you use the surface contacts. The point is that although the surface contact needs nonlinearity the nonlinearity is "hidden". You don't need to define any nonlinear parameters etc. This surface contact analysis solver is taken from IDEAS. This so-called 'linear contact' is very easy to use and it works VERY FINE!

BUT, the surface contact is the only nonlinearity. You can't use for example the plastic material or large displacements.

You can also purchase Advanced Nonlinear package (SOL601) with NX. Advanced Nonlinear is taken directly from ADINA, which is VERY robust nonlinear solver. You can use the surface contact, large displacements, plastic material etc. I'm not sure but this package may cost as much as NX's basic package...

If the surface contact is the only nonlinearity you need, NX's basic package is sufficient.

NE is MUCH faster than NX. NE has very handy Editor so the process is very easy to follow.

------------------

Here was quite confuced and brief report. You can ask more if you like to.

 

[ThomasH]

MSC, NX or NE ?

It's not a new question. MSC is the oldest and due to that in many peoples minds the best. NX is as a brand name the youngest but it's directly based on MSC, MSC.Nastran v2001 to be exact.

Nastran is traditionally not good at contact. MSC added Marc and called it SOL600, UGS (now SIEMENS) added Adina and called it SOL601. Noran (NE) have it in the core product.

MSC and NX have two nonlinear levels. One simple that is the "old" MSC nonlinear engine and one much more advanced that is Marc/Adina. NE has only one level. The main difference (in my opinion) is that Marc/Adina has more material models then NE but I know that Noran is working on that part. I have also heard that the price for the Marc/Adina-addition is high but I have no actual numbers, sorry.

I would test the software before buying and I know that Noran has free testlicenses, you can probably get the same from UGS (NX). I've heard that MSC has a different policy on this but that might have changed.

I tested both NX and NE a couple of years ago and before that I used MSC for several years. They all have unique features. There are things you can run in NE that are difficult or impossible in the others and vise versa.

I conclusion, I agree with Ronald62, try them both (or all of them).

And ask questions. You are quite welcome to ask them here but don't forget the dealers. They are your future support staff.

Good Luck

Thomas

 

[Ronald62]

"I tested both NX and NE a couple of years ago..."

-> Was there the nonlinearity function from IDEAS inside the linear static solver (SOL101) when you tested NX?


"There are things you can run in NE that are difficult or impossible in the others and vise versa."

That's right! For example, NE has sometimes problem with the surface contact and NX's contact works much better. On the other hand, you can run much bigger models with NE.

 

[ThomasH]

"Was there the nonlinearity function from IDEAS inside the linear static solver (SOL101) when you tested NX?"

Yes, it was there but as far as I remember I never used it. Contact has never been a really big issue for me. I sometimes need it but it has never been "big".

My impression of NX was that it was relativly slow and some models (dynamics) didn't run at all. NX is an older version of MSC in same aspects and it shows. The Adina part is excellent but pricey. Both MSC and UGS added other products to get surface contact, something that is "hot" on the market today. This not meant as a critique of the products, just stating a fact.

I "played" a little with Marc some years ago and have friends who use Adina as a separate product. They are both good products. However, for me it was not an option to invest that kind of money for a product to get a feature (contact) that I rarely use.

I need nonlinear capabilities (plasticity/ large deformations / etc) and NE offers that at a resonable price.

When I stated that NE can do things that are impossible to do on the other products I wasn't thinking of contact. There are other capabilities as well ;-).

But it might be so that adnanh has requirements that only NX can meet. If so, go for it or discuss the problem with the competitors. They might be able to add "your" capabilities. That why I think he should test it and not take the advice given here as final.

But feel free to ask questions.

Regards

Thomas

 

[adnanh]

Thanks a lot for the comments.

I know that the needs of FE user can differ a lot. But as far as the versatility of various Nastrans is concerned, I am getting the impression that NEi covers a lot. I make this statement based on what I have read and seen, mostly on Nei website itself. Especially, the demos on

http://www.nenastran.com/newnoran/animations

give me the impression that there is a sound contact capability as well as geometrical and material nonlinearity. It looks as a very versitile program. However, I have not tried it yet.

Is it possible to do above simulations with NX/Nastran? I have NX v4 installed on universitiy computers as well as FEMAP v9.2. I have played a little with it, but obvously I will need to spend much more time in order to explore the program.

 

[R1100S]

Previously Noran used Femap from UGS for there pre and post processor. Noran sold there own solver, NEiNastran, while UGS sold NX Nastran.

Currently Noran is in charge of marketing, selling and supporting NX Nastran. See their newsletter from last April:

http://www.nenastran.com/newnoran/NEiNewsletter April 2007/

I'm guessing you can call your local Noran representative and they can set you up with a test drive for both. Or at least make a recommendation on which will suit your needs better.

 

[wvillers]

Both MSC and NX Nastran cover as much, arguably even more if you think about DMAP programming. The surface-surface contact in NX Nastran is pretty darn good, and is not limited to statics, it is also able to deal with modes and dynamics response... All the NE animations can definitely be done in NX Nastran, we do that kind of stuff on a quasi-daily basis... Also, I'm not sure if NE is available on 64-bit Linux distributions... The Distributed Memory Parallel abilities of NX Nastran are very useful in a Linux cluster environment... Just my $0.02..

 

[automechanicar]

Hi all,

Does somebody know whether there is a tutorial on how to setup/perform analysis types shown as demo's on the Noran's website? I have not been able to find many examples different that just linear statics en very little on nonlinear statics. I would like to practise myself in nonlinear static forming problems, metal sheet forming in 3D and other. Neither I could find eny example on (nonlinear) dynamic problems, like impact and felxible body dynamics

Kind regrads!

 

[wvillers]

Which solver are you thinking of using? Forming might need a more specialized code than a generic non-linear code like SOL106 (I'm thinking Abaqus, LS-Dyna, Marc or Adina).

 

[automechanicar]

@wvillers,

As far as nonlinear analysis is concerned, I have at my disposal the following:

- Nx Nastran with FEMAP
As far NX Nastran?FEMAP is concerned, we have the socalled basic package. This includes basic nonlinear analysis wih just one source of nonlinearity being, either geometric, material, boundary condition or contact. This means that an analysis including nonlinear material behavior involving large displacements/rotations and surface-surface contact is not possible to perform with this license.

- ABAQUS/CAE with both the implicit and explicit solver.
With these tools, I should be able to do almost any type of nonlinear analysis, including both statics and transient.

- MSC Marc
As pre/processors both Mentat and Patran is available.

- Pro/Mechanica
As we all know Pro/M is linear static FE-program which includes either geometrical nonlinearity or contact, but no cobination of both!

What I am looking fo is a software which allows me to do various types of nonlinear static and nonlinear transient analysis with contact. As you see, various sources of nonlinearity are involved in these types of analyisis, often a combination of a nonlinear material with large displacement and contact.

Can you recommend me some of the available tutorials, either for MSC Marc/Mentat/Patran, ABAQUS/CAE of NX Nastran/Femap. I am not able to find any tutorials which adress nonlinear static and transient problems!

Kind regards!

 

[wvillers]

NX Nastran "basic" would be SOL106, probably the weakest of the lot (except for Mechanica), as you pointed out you'd need the advanced non-linear, SOL601, to perform surface-surface contact, although you might be able to get away with gap elements.

Both Abaqus and Marc would definitely do a better job, with the Abaqus CAE pre-processor probably giving you the edge compared to Patran and FEMAP... As far as tutorials, aren't there a bunch of them in Abaqus' help system?