Hi,
The last time I used Pro/Mechanica was about one year ago. I think the downfalls of Pro/Mech are the lack of element types (e.g no discrete damper) and solution varities (poor dynamic capabilities), i don't know exactly if this is because it uses geometry element methode (a derivativ of FEA). I think Pro/Mech was designed mainly for designer in stead of "real" stress analyzer, so it need not have the capabilities like other powerfull CAE-packages.
The benefit of Pro/Mech is of course its direct integration to the CAD Program Pro/Engineer, so the Pro/Eng designers can analize their parts without changing the GUI (Graphical User Interface).
May be MLoew can give some more explanations about Pro/Mech.

regards

Its also only a linear code and there is less control of the mesh (good in some respects, but less flexible).  Beyond the surface, it lacks much of the detail better packages have.

Zuardy,

I accept your invitation to contribute. I agree with most of what you posted. Pro/MECHANICA, unfortunately, in my opinion is indeed targeted to the designer, rather than engineer. As an engineer interested in First Order Analysis (FOA) of structures and systems, I take advantage of this capability in my "design" work. Personally, I do not agree with the division of designing and engineering. Anyway, Pro/MECHANICA lets me do the type of design direction and parametric studies very important early in the gestation period of a design. We can start another thread to discuss this topic!

Anyway, the interface is quite well integrated with Pro/E now and PTC has added some non-linear capabilities to Pro/MECHANICA (geometry only: contact and large displacements) enhancing the usefulness of the tool. It is also possible to control the mesh more easily now. I no longer consider this a major liability. The convergence study, also, is a real benefit as well. This topic is addressed in detail in another post (www.eng-tips.com/viewthread.cfm?SQID=23503&SPID=727&page=1&CFID=65894337&CFTOKEN=45490927). I might be a stickler for a convergence study because of my background in Pro/MECHANICA.

For example, I was recently asked to conduct an analysis of a vehicle structure. Using an interpretation of the load cases from an industry specification. I was able to get a sense of the stress magnitudes, highlight areas that needed optimization, and gain insight to what higher order analysis would be necessary to conduct. The limitation of the linear materials model was a handicap for this study. I needed to consider energy absorption (through plastic deformation) and could not do it in Pro/MECHANICA. I got most of what I needed from Pro/MECHANICA and have a good template for more detailed engineering activities.

I would never conduct high-end analysis or final design verification with Pro/MECHANICA, unless I knew I had instrumented test data to validate the model to (another good topic for a thread). Again, the strength of Pro/MECHANICA is in having the ability to quickly create FOA type studies on parametric Pro/E geometry with confidence. I hope this helps.

Best regards,

Matthew Ian Loew

Thank you gentlemen for enlightening me on the matter.  So Pro/Mechanica is not a true FEA software but a derivative?  I guess I don't quite understand that statement Zuardy. I do agree with everyone's statements and a part of me is grateful for the "design" benefits of Pro/Mechanica due to my lack of time and experience for a thorough analysis.  But I am dissappointed that it can't be used for accurate studies.  Say for instance, I'd like to study a pressure vessel with a welded manifold.  Would it be able to study the transition between the thin walled vessel to a thick manifold.  I've read that the weld should be modeled with certain elements.  Would Pro/Mechanica simple assume a "mated" surface (continous surface)?  Sorry for the ignorant questions?  As you can see, I'm quite new to the fea world and do a lot of modeling using Pro/E.  I however would like to expand my career into fea in the future, but as of now I do a lot of design work.  Any links to further literature or great books would be more than appreciated.
Is Cosmos/DesignStar considered a design tool as well or decent for fea analysis?  Their user interface seemed very easy.  I suppose Algor would be a better choice out of the three?  But my experiences with Pro/E and their tech support has been top notch which makes me bias.

Thanks a million.

hydrogen

Hydrogen,

Pro/MECHANICA is indeed a Finite Element Method (FEM) application. It is constructed around using higher order (3 through 9) elements that allow for some large looking element boundaries. It can be used as an h-order package with the p-level set to 3 and convergence turned off (this is the "quick-check" solution option). You must however use elements that are sized appropriately.

Many debate the p vs. h methods but they are both FEM based. Rasna corporation (the original developer of what was then known as Mechanica) coined the term Geometric Element Modeling (GEM) as more of a marketing tool than a new type of solution method. Other methods are finite difference, boundary element, precise solids (the software is Procision www.procision.com, available from Rand www.rand.com) and other methods.

Once you understand the limitations of p-element solvers, they can be a very good development tool.

Hope this helps.

Best regards,

Matthew Ian Loew

MLoew, thanks for your comments. I have though that Pro/Mech implemented p-methode, since it has convergency control and polynomial degree setting. I used the therm GEM just because RASNA uses it also.
Regarding precise solids or external approximation (meshless) from Procision, have you ever tried it? I am interested in such numerical codes, because it eliminates the need of complicated element meshing.

cheers

Greetings,

The core to my theme here is: Pro/Mechanica can get you to a robust, first article solution quicker than any other analysis package out there.

Exploitation of Pro/M optimization and sensitivity study capabilities offers the user the ability to find optimum design configuration within a single run. The possibilities are endless from an engineer’s point of view. The ability to assign user-specific measures and associate model parameters to the analytical model is indispensable. The iterative design process is now made easy, quick, and robust because of Pro/M.  

Single-Pass Adaptive and Multi-Pass adaptive solution algorithms allow for automatic adaptation of the P-element to locally high gradients within the model during a run. Typically, the user will have to go back into the FEA model and manually increase mesh densities at the location found to have high gradients when using other packages and then re-run.

The bi-directional associativity with the geometric model is a plus while in integrated mode. Data integrity is maintained across-the-board, and interchanging between design and analysis is quick and free from compatibility breakdown.

I have used Marc/Mentat, Ansys 5.7, and Pro/Mechanica. I have found in my studies and during my professional career, that most analyses can be performed in the linear world. Having the non-linear capability is nice but, in no uncertain terms can it out way the benefits of having at your disposal a one-space design and analysis package capable of providing you with a solution that meets 90% of what you design in a fraction of the time.

If you need to run a non-linear analysis--farm it out. Let a specialized vendor handle it or, if your company can afford it, get a Non-Linear package and send someone off into the weeds to learn it.

My advice to you is--assimilate the fundamentals. Learn the fundamentals of FEA first. Understand what you are looking at and the how the assumptions that you have made effect your results. I like to call-it the "uncertainty factor"

Final Thought:

Finite Element analysis is a cogent contrivance that is predicated on a permissible degree of uncertainty which educes a solution that is invariably delimited by a subjective elucidation.

                       OR,

               It's a good guess.

Regards,

Njaneer

h2now,

Pro/M is a good tool for a designer to use to become familiar with FEM.  Your pressure vessel problem should work well in this code.  You'd probably want to treat the heat-affected zone as a separate material from the rest of the structure.  You can use "Measures" to determine the peak stress in each group of elements you define.  You will be able to determine how close you are to the yield strength.  However, if you want to see what happens when the stress exceeds the yield point, you're out of luck because plasticity is not an available option.  As you become more advanced, you might start to get frustrated by the limitations of Pro/M and then you can check out a full function code such as ANSYS.

A few things to be careful of when using Pro/M:
1)  Be very careful when using different types of elements in your model or you can get degree-of-freedom mismatches that will prevent the solution from converging.
2)  The way Pro/M sets up its table for temperature-dependent coefficient of thermal expansion is non-standard.  Thoroughly test out a sample problem before proceeding.
3)  Watch out for false positives on convergence.  Sometimes the solution will do another pass without increasing the polynomial order, then mistakenly think it has converged because the peak stress (or whatever measures you're using) hasn't changed.

-- Kirk

I am a new user of Pro/Mechanica, it has proven to be satisfactory to me coz i deal with linear analysis. the one disadvantage i have seen is that it has two modes independent and integrated in independent u have a little more control over nodes, which is very essential . but all this is not available in integrated mode. it would be nice if ptc would have just one mode with node-control capabilities in it.

YM

Kreative,

One can control to some degree the creation of elements and notes by using surface regions, and by creating points (under idealizations) in the integrated mode. Pro/MECHANICA will use these geometric features when creating elements.

Best regards,

Matthew Ian Loew

Gentlemen, I am somewhat new to FEA and especially Pro/Mechanica.  I have come into the need of analyzing an assembly under this package and have run into problems I believe stem from contact regions.  Could anyone recommend a set of guidelines in analyzing a differing materials assembly?  Especially appreciated would be comments on the setting of degrees of freedom with the goal of completing an assembly analysis.  Furthermore, could you recommend a text (besides the .pdf that which comes with Pro/Mechanica) that would go into such Mechanica advanced topics suitable for a college level class?  I thank you in advance of any comments.

CriticalMass,

I would first conduct a search on Google.com and of the PTC knowledge base.  Also check out www.genexisdesign.com/download.htm.

Regarding the DOF issue, are you talking about kinematics assembly constraints? There are ways to integrate kinematics & Dynamic analyses in Pro/MECHANICA Motion and Structure if that is what you are looking for.

Best regards,

Matthew Ian Loew

Hello,

I started doing analysis usin high-end packages such as Nastran and LS-Dyna, IDEAS, etc. Only within the last few months have I started using PRO/Mech Structure.

I agree with some of the comments that PRO/M is more for designers versus full-time analysts. You see that when they push the integrated mode over the independent mode. I strongly disagree with the idea that you can do analysis in a fraction of the time with other software. This is especially true with more involved problems (e.g. modal analysis, frequency response, contact). I ran one linear static model (shell and volume elements) which took about 6 hours in PRO/Mech and it took 30 minutes in ABAQUS. My experience (and the group I work in) is that Mechanica is much slower than ABAQUS or NASTRAN.

Some specific complaints:
1. Modal solutions are very slow and create very large solutions. Ex. PRO/MEch: 9000 elements, 100 modes = 4.5 GB hard drvie space; NASTRAN 100,000 elements, 100 modes = 400 MB. Granted that PRO/Mech goes to 9th order polynomial, but 4.5 GB is ridiculous.

2. Crappy Shell mesh algorithm.

3. Very slow rendering of shaded models (i.e. switching from wireframe to shaded on large models.)

4. Crashing while creating Boundary conditions (don't reclick the arrow to pick more BCs).

5. You cannot exclude elements in dynamic response. So your max over entire structure can be thrown off over a single strss riser.

6. You cannot animate stress at a specific frequency.

7. You don't have control over the sample rate of a dynamic response. NASTRAN will give you about 10 different ways to specify modal frequency response sample rates.

8. Sticky nodes with contacts gives you very unrealistic high stresses.

9. Slow post processing. Example, changing the upper and lower range of contour plots, does it really need to regenerate all the open windows between twice?

10. Stress risers. I observed stress risers which make my results useless. the point of "Geometric" analysis is to remove the need to micro manage the mesh, but you have to do it any way.

11. You cannot import FE mesh from other solvers.

12. Limited surfacing capabilities (as compared to Hypermesh for example) in the independent mode.

13. You don't see the mesh in the integrated mode.

14. Linear only material. Much to my horror the PRO/Mech instructor of the intro class stated that if a part yields
its no good, so why would you need to do nonlinear analysis anyway?

15. In PRO/MECH structure you cannot query a points on an Xy plot. You have to dump it into EXCEL. C'mon this is graphing 101!

Etc, could make the list longer But I have to get back to work.


IN general PRO/Mech s good for simple analysis of simple parts. But if you are doing higher end stuff you probably want to use NASTRAN, ABAQUS, IDEAS (their 3rd party packages ESC, TMG, etc. are very good), LS-Dyna , and some others I haven't specifically worked with.

Hope this helps.

CaptainCrunch,

I generally agree with your assessments. However, I feel strongly that designers are not (or at least should not be) the target user base. Design engineers should be the only ones qualified to take advantage of this functionality in my opinion. The user must understand the limitations and capabilities of the software to use properly, especially considering the limitations of the analysis (linear materials only, etc.) as you mentioned. Even as it is, there is too much power in Pro/MECHANICA for the non-engineers to use safely. There are degreed engineers that I feel are not qualified to use this functionality as well; so don't take my comments as a slight on designers. Any analytical technique must be thoroughly understood to be used safely.

Anyway, regarding comment 13: You can see the element boundaries in integrated mode by setting the "feature angle" to zero.

Best regards,

Matthew Ian Loew

Matthew,

Your points are well taken. When I said designer in my mind I meant "design engineer". IMHO, though I think PTC is *marketing* PRO/Mechanica as a design tool for designers as well as an analysis tool for analysts

Another point, you stated:
"Anyway, regarding comment 13: You can see the element boundaries in integrated mode by setting the "feature angle" to zero. "

But you can only do this after the solution. Wouldn't it be nice if you could see the mesh before you run the job?

BTW, I brought this up at the training class and got brushed off by the instructor. In fact just about every question I raised was answered with "Why would you want to do that!?!"


Regards,

George

George,

I have had a fundamental problem with the way PTC has marketed the Pro/MECHANICA product to "designers." I will, however, as an experienced engineer take full advantage of the capabilities.

Also, yes, you are right about the element boundaries being unavailable to view after a solution is complete. Two tricks to overcome this significant flaw:

1) run a quick check
2) save as copy of the model and open it in stand alone

Both options stink. PTC needs to work continue to work on the mesh generation capabilities in integrated mode. I did see once in a user's group that "pasting" a bunch of datum curves, and points, on the solid (or surface) geometry acted as a controlling feature for mesh creation. I have tried this with varying levels of success.

BTW, your "instructor" might have been one of the un-educated designer that did not understand the theory and limitations of analysis that I was cautioning about.

Best regards,

Matthew Ian Loew

Hi everyone !

Guys, most of your statements are true ! Mechanica is slow with pre/postprocessing issues but things are being changed for the next release.
Integration is about to be completely finished for the so-called integrated mode. Keep in mind you don't need to buy Pro/E to use Pro/Mechanica in Integrated mode ...

Post-processing is going to use Pro/E graphics : very fast indeed ! and finally you'll have access to any AUTOGEM options in integrated mode and be able to check your mesh, free boundaries and everything needed to remain as confident as you were within integrated mode.
On the other hand for the h-side, Pro/Mechanica pre/post capabilities will exhibit the same capabilities with interactive/batch calculations to ANSYS/NASTRAN. Yes you're gonna tell me their 3D-mesher is crappy and you were right !!! they replaced it with the GHS3D one !

By the way to answer the CaptainCrunch point on mixed meshing, links within Mechanica are of total disaster but usual techniques can be used to map shell elements on volume faces to get rid of these undesirable elements.

For modal analyses, CaptainCrunch is right above 50 modes Mechanica is a penalizing software ! and as a matter of fact do not forget to mark mass participation factors otherwise you'll go for another run....

I disagree when saying Mechanica is just suitable for simple geometry : this is also robust for large casted part, thin or thick welded assemblies as soon as you don't have to manage real contact areas.

Well I still think Mechanica is a good predesign tool for everyone : Designers, Design Engineers and analysts, and especially for those that use to use Pro/E with or without the BMX design explorer (Mechanica structure is now fully coupled to it )

Cheers,

Philippe

Hi check out 2002,
Do you mean, in integrated mode Pro/E is tightly integrated to Pro/Mechanica so don't need to buy Pro/E. But how about the price? Does it cost Pro/Mech + Pro/E also? or it is cheaper?
And does integrated mode can accept/run an external application written for Pro/Eng using Pro/Toolkit?

Thx in adv

Yes you can upgrade a Mechanica standalone version or more simply Pro/MESH  with the SSP package ! within this package you'll get the Pro/MECHANICA foundation that is a Pro/E version without advanced surface capabilities, detailing, etc.... but if you recover a model from Pro/E you are able to modify dimensions of these advanced features like any other dimensions... thus no need for having additional Pro/E seats.
As far as I know it was around US$ 12,500... but certainly less when dealing with PTC at end of quarter. PTC sales try to avoid that kind of deal by selling a Pro/E foundation instead...

"And does integrated mode can accept/run an external application written for Pro/Eng using Pro/Toolkit?"

External applications (with BMX) are only available with Pro/E

Hope this helps

Philippe

Over the years, I've become fluent in every commercially available FEA code, Algor, Cosmos, Ansys, Nastran, Mechanica. The single biggest problem (or limitation is a better word) with Mechanica, is one is limited in what one can do with Mechanica. I wrote my masters thesis on the P-method (Polynominal) technology so I beat this approach up for a long time, even writing my own custom code from the ground up.

In general some of the reasons I prefer FEA codes like Nastran over the Mechanica type codes are:
      (a.) Mechanica is not as comprehensive as all the other FEA codes.
      (b.) Mechanica isn't very flexible in what one can do with regard to auto-meshing and element formulation; for example one has more control over Rigid Links I don't believe even the current Mechanica code could handle the Rigid Links as eloquently as say Nastran.
      (c.) Nastran is the defacto standard for FEA analysis results {can be a factor when working with large aerospace companies (Boeing & Lockheed) or auto manufactures (Ford & GM) }. Now I'm not suggesting Mechanica isn't accurate, but in my consulting business I typically end up having to supply solutions to these companies in the Nastran code as well as in Mechanica or any other code my clients might be using.
      (d.) The solution times for a fine grid meshed H-Method (like Nastran) is probably an order of magnitude faster than an equivalent Mechanica P-Method coded model.
      (e.) I can recall several Pro-E CAD models that even experienced users were not able to get Mechanica to develop an auto-mesh. With more comprehensive software tools like the ones found in Nastran we are able idealized these CAD models in a relatively short period of time. I had to due some tune-up of the idealizations but I got them.

Some of the More Obvious Downsides of Mechanica:
(a.) Pretty much stuck with Linear analysis only.
(b.) I don't think the current release of Mechanica can do either Large Deflections, Non-Linear Stress/Strain, or Resonant Vibrations with Load Stiffening. Maybe they can do that now, I would have to check.
(b.) Base Excitation for Dynamic Response [ Random & Sinusoidal ] - Not a big deal for most problems but there are many cases where one would like to subject models to variable input excitation at different points instead of *shaking* the entire model on it's supports {or Boundary Conditions}.
(c.) Mechanica's claim to fame when it was called Rasna was that one didn't need to develop detailed meshes. One only had to crudely mesh regions and Mechanica by virtue it's P-method internally updates the order of polynomials until convergence of the solution is achieved One Downside here is that one still needs to define proper patches [2-D] and hyper-patches [3-D] for the models to converge. Another downside is the length of processing time. Today's automeshing tools make the meshing aspects a moot point. With Nastran one can still do a lot more when it come to manipulating the idealization.
(d.) One aspect of the Mechanica code that I've witnessed is that some users take the idea of not needing to develop proper meshes too far and end up with solutions that aren't as good as they should be. The solution to this problem is to do lots of sample problems and spend a lot of time using the software. [Same with any other code.]
(e.) On the more practical ends, the Mechanica code has a limited element library - Nastran has a lot of flexibility in this area.
(f.) One interesting note is that when Mechanica P-Method [and the other FEA codes as well] list comparisons of their solution accuracy they compare to Nastran H-Method. Typically though every FEA code today gives good results if used properly.

The types of analysis most companies perform are typically -Linear Static's, Resonant Vibrations, and Dynamic Response to Sinusoidal Excitation. Every FEA code can do this. Mechanica included. I suppose the most attractive aspect of Nastran over Mechanica is the Comprehensive Modeling capability [ I know that word *comprehensive* covers a lot of ground], more Flexible modeling tools and Defacto industry standard.

That's about all I can come up with on the spur of the moment. Give me a call if you have any specific questions. I think this might be a good topic for one of my articles for Machine Design's FEA Update Column.

Regards,

David Dearth
Applied Analysis & Technology
"AppliedAT@aol.com"