Which mechanical engineering software is the best?

Which software would you choose from Ansys, SolidWorks, GAMBIT, Fluent ... etc.?
Which of them is more important, especially concerning fluid mechanics and energy conversion?

Topics

0 / 0  ·  18 Answers  ·  3341 Views

Popular Answers

All Answers (18)

  • Evgeni Trutnev · Ural State Technical University
    I worked in different software and my choiсe is - Autodesk Inventor. I thought he was more comfortable for "long work".
  • The human Brain, no doubt.
  • One cannot disagree with Erik's promotion of the human brain as the ultimate tool for mechanical engineering software. I have but one comment, that hopefully could shed some more light onto this issue. As I do not design devices but fluid dynamic processes and have no background in technical drawing, therefore I have to rely on design engineers to give physical shape to the devices that spring up from our research. To my surprise it very much matters which software we use. Our judgement is that it is the Solid Works that absolutely beats the rest, because before we complete a prototype or deliver something ready for production, we make variants or replace some elements several times. This software allows us to make such elements (usually on a 3-D printer, or a small CNC machines we operate) as we go with the actual design, that is we can free our brains to concentrate on a device we build and be ready for an unavoidable string of corrections, improvements and other changes, forced upon us. This is a situation where the risk of failure is significant, like R&D, prototyping or use-orientated design activities. When the design, as in terms of a structure to build, is required, all the software packages handle the task very well. Fluent is quite good, for problems others have already solved, but would not advise anybody to make any binding design decisions for a completely new device only on what has been churned out on the Fluent.
  • Human mind software is best
  • I worked by fluent to simulation of fluid flow and heat transfer in different geometry and various flow regime. it is a good software if you give it correct data.
  • I agree with you entirely. Problem is that there are some fudge factors in the flow computations, e.g. for k-e model, friction factors, k's for throughs and bends, etc., so for problems solved on a routine basis everything is fine and dandy, but when you want the Fluent to suggest some novel design solution you are at the mercy of arbitrary selection of models and tuneups. A company that sells in bulk what I design on a retail basis, they pay over 2 million US dollars per year to have the software maintained and upgraded and still have to experiment a lot. The crux is that fluid flow equations change types with compressibility or limited-scope modelling, often have parabolic aproximations, pressure inlet and outlet conditions are known only to some extent, to say nothing about heat exchange on the boundary, and than you get all the partial pressures, particulates and droplets to worry about. My point of view is that computing is subject to the same limitation in engineering everything else is, bang for the buck, or cost and time versus the yield. If the Fluent tells you how your heat exchanger would work to a degree of the accuracy you require, hesitate no more. For a tri-fluid, supersonic nozzle or an incopressible turbuent boundary layer over a deterministic roughness wall, get a rough view and go to a workbench.
  • For me there is no definitive response.
    According with your engineering specialty, or in opposite if you need a diversified purpose.
    If you are working with specific customer or provider it would be better work with compatible program.
    Personally I prefer the Autodesk Inventor than Solidworks because is more instinctive and with much more possibility to use the undo for erases the last change (my brain is very bad). Now I have some friends in different situation. One of them past to Catia and is amazed, he said “it´s another championship level” (above all in complex surfaces application), by the other end, this week I saw a prototype rather complex and working quite good, made by a non graduated engineer, plan it with a elementary Intaglio program for Macintosh.
    The formation and possibilities to share tricks and solutions with colleagues and friends are very important because it's better fully dominate a simplest program than working with another too sophisticated.
    And remember that: what is true today tomorrow could not be. The competition between the numerous mechanical engineering software companies is high and evolutions are permanent.
  • Morco's pointing to the comunity of practice approach probably is the best comentary I saw on this forum to date. Net Hercules contra plures.
  • Nick Hamburger · HN Consulting
    There is no absolute best soft. It depends on the complexity of the work to be done. You mention very expensive packages designed for very complex systems (CATIA was conceived for planes by Dassault) there are a lot more simple systems than complex which deserve attention at a affordable cost. One of the best CAD 3D I met is ALIBRE which offers for a small part of the cost usually asked by main suppliers almost all functions and facilities. The same is valid for fluid analysis all depends on the degree of requested accuracy and what can be done for simplification of the model. A very low cost soft LISA can for some problems bring a hint which could be used for trend analysis. It is simple and so low in price that people cannot think it can do what it does. It si not always a "gain" using complex software.
  • Hi Nick, long time-no see! Never heard of LISA, who makes it? Perhaps, talking of accuracy we should also point to the rate of change for a given phenomenon. I work on failry well organised flow, except that nitrogen bubbles keep on exploding within the jet of water, doing to water what I want to be done to it. No way I can calculate this in earnest, measurement equipment would send me to a poor house, whereas approximate analysys gives explosion time equa to zero.
  • Nick Hamburger · HN Consulting
    I am as the little Lord: every thing was OK so that I had nothing to say! In fact I wanted to make some further comments on the hot air flows but since the interested person did not show any interest any more I thought it was lost time and effort. I thought that both suggestions yours and mine were valid and if you want I can explain how I think it happens.
    With respect to LISA it is a first in Germany developed soft which was bought by a Canadian editor and of course is not as complex as fluent or flower or some other packages but for $60 offers a possibility to "solve" or "estimate" stress, heat, capacity (electric) and flow problems. For many many problems encountered in engineering what it brings is very satisfactory. It accepts an unlimited number of nodes and many other features. If you Google "LISA software" you get address and how to download a limited (1400 nodes) free version. I made some comparisons and I was satisfied by the very small differences. I do not think it could solve problems as yours (2 phases) but incompressible flows can be treated. as you can imagine most problems are less sophisticated so that it can be for 1st approach a good tool. they have a very efficient "help service" reacting very fast (Karl) if you have problems or questions. I used much more complex softs for stress analysis as ALGOR and only a few times did I need their full capability.
  • Nick Hamburger · HN Consulting
    to Jerzy,
    Did you have a look at COMSOL ? It is a multiphysics simulation software. Although I do not use it I know it pretty well and it offers a lot of possibilities.
  • Hi Nick. I am not well versed in commercial packages, because I use my old domain decomposition shcemes and as what I do is increadibly transient I have very little faith in codes I cannot calibrate against anything I trust. How did you come across this package?
  • Nick Hamburger · HN Consulting
    You can validate a soft by running same problem with your soft and the other one if results are witih acceptable differences liöits it is OK. Those packages are avalable ans since I am interested in simulations I look around.
  • Thanks any help would be a rescue for me. I have bubbles of compressed gas exploding inside a water stream close to the free exit of a jet. In most cases it all rotates as a some Gatling gun, to discern between the jet exit velocity and the mist plume one. That is all I can write on a public forum. Nick, where aer you physically located?
  • Nick Hamburger · HN Consulting
    Geographically in France/Paris. I think that COMSOL mentioned once a mixed flow problem. The best would be to get in touch with them.
    As for your problem it is normal that the bubbles explode near to the outlet. It depends on the diffusion of gas in fluid which (Bunsen coef.) decreases when pressure goes down or near to the outlet section pressure drops.
  • J'adore Paris, la ville de ma jeneusse, je visitait tout les quoins. Unfortunately it is not gas dffusion in water, but a transient problem of highly unstable plug flow. Nick, I shall not dwell on this in public, because it took me about 14 months to redesign and reinvent things some crooks stole in part and squandered totally before I resigned. two companies. Private e-mail address and phone numbers are on my researchgate page.
  • Slawomir Zolkiewski · Silesian University of Technology
    Not so expensive is also Working Model and if you can agree with Siemens they give you access to the educational/science version of NX Unigraphics as well

Question Followers (15) See all