12 Replies Latest reply on May 25, 2010 9:05 AM by Joe Galliera

    Algor - Worth The Switch?

    Ben Floan

      At my company we are heavy into Autodesk products, Inventor / AutoCad, so naturally our reseller is pushing hard for us to switch to Algor.  Anyone had experience with Algor and some insight in how it might compare to SW Sim?  We use SW Sim. Premium w/ FlowWorks......

        • Re: Algor - Worth The Switch?
          Bill McEachern

          Well, I would suggest you get an eval of the products from Algor and run your stuff on the products proposed. I am heavily involved with a SolidWorks reseller so take all this with a grain of salt - most of my knowledge comes from replacing Algor.  Not too current with things Algor but previously and maybe still you have a three program interface, the pre processor, the solver manager and the post processor - the embedded stuff inside Inventor maybe better - I have no direct knowledge on those products. Their advertising on complex problems is a lot better than their delivery in my experience with customers. Their flow product is completely pathetic compared to Flow simulation. I replaced their flow product at one customer and they told me that it would take days to get an analysis to run and changing things and re-running was very labor intensive. In the demo I did what would take them many man weeks - yes it is that dramatic even on simple problems. Part of the issue is that you u have to manually manged the impulsive start at BC's - you need to ramp them up at a rate that maintains stability which varies with the problem - apparently. That is a couple of years ago.

           

          Try them out and let us know what your thoughts are.

            • Re: Algor - Worth The Switch?
              Ben Floan

              Thanks Bill, I am getting an eval of the premium Algor product; my understanding is still require pre-processor (I will use SW) then switch over to the solver/post-proc. interface.  I am in agreement that they are good at advertising niche complex problems.  In our case, most complex work is with contact and some basic GNL.  Where I am interested most is mesh control and how easily I can set-up and solve shell models for linear and GNL.  Interesting to hear your take on the flow product...that may be a deal breaker since we have had some success with Flow-Sim. and a crippling of that functionality would be not acceptable.

               

              A bit off topic, what is your opinion of the meshing control within FlowWorks.  This I believe in some cases is the biggest headache, but also for some complex scenarios can be very useful.  Pre-proc. for Fluent simulation with complex geometries would take quite a while to get nice mapped mesh profile, but FlowWorks and careful use of Partial / Trimmed cells can yield almost the same result.  Is this what you have also noticed?

                • Re: Algor - Worth The Switch?
                  Bill McEachern

                  Flow sim is a great product. Some of my top reasons, in no particular order, are:

                  1. If you don't have any invalid contact (even if you do they are pretty easy to find and remove) - you can always get a mesh. Full Stop. Most body fitted codes (fluent, CF Design, CFX  to name 3) have major issues here as the little sliver faces and other small geometry can cause enormous grief and flow Sim will just pave over that stuff. The analyst needs a bit of care here to make sure his cut cell approximation is good to go for what is intended. I have not run across many problems where it isn't and the fix is just up the mesh density locally. The stuff is actually brilliant in handling geometric complexity not that I recommend geometric complexity unless it is needed.
                  2. As one user put it to me very succinctly with respect to the manual mesh controls: ".....they are sweet. It was like it was designed by somebody who actually uses the stuff. I hardly ever use the automatic meshing anymore".
                  3. Given the low cost of compute power and multi core support using the brute force and ignorance approach is typically a pretty safe bet.
                  4. Goals: get a nice neat table for all the info you really want to know with convergence or time trajectories. Monitor results as they occur.
                  5. Pretty good set of capabilities for the dough, though it would be nice to have true multi-phase (whatever that means), free surface flows (VOF), moving mesh, moving body object model & chemical reactions but then the code would cost an awful lot more than it currently does even with the latest price increase (it was too cheap before for what you get - still might be).
                  6. User interface: intuitive and really easy to interact with.
                  7. Pretty decent post processor and reasonably fast and considering it is running inside of SolidWorks, maybe blazingly so isn't a stretch- some quirks here but nothing fatal. volume plots would be a nice addition (color all cells below some threshold value - the iso plot not as informative on that front - at least for me).
              • Not really except for $$$
                Bill Lail

                We have Algor at our company.  We are considering a switch to SW Simulation even though we have another 18-months on the Algor subscription. Algor is not very easy to use. I don't believe the linear static results are very different between the different FEA vendors if the model is setup correctly. What we do have is a very clunky and hard to use GUI for the meshing and analysis with Algor. Also, the value of support to us from Algor changed when they became an Autodesk product. Autodesk changed our subscription terms we had with Algor then didn't work very hard to encourage us to stay their valued customer.

                 

                Algor/Autodesk did offer us a better deal dollar wise if we sign up now but the ease of use and time to adjust meshing is a pain and cost us $$$ in man-hours.

                 

                And if any reseller offers you an ROI for the change over to show you the buckets of money you will be saving, just smile and nod in agreement. An ROI is not very sensible for a task like this where the parameters that cost money like man-hours are very difficult to quantify.

                • Re: Algor - Worth The Switch?
                  Joe Galliera

                  Is there really a reason to switch if you already own a product that works?  I have many stories of companies moving from Algor to SW Simulation, and I do not know of one that went the other way.

                   

                  Should you always do what your reseller is pushing you to do?  They are a sales organization and have more to gain from your switching than your company has to gain.

                   

                  Please let us know if there is a technical reason why SW Simulation cannot handle your problems.

                    • Re: Algor - Worth The Switch?
                      Ben Floan

                      Joe,

                       

                      For me it is primarily mesh control in which I believe Algor will be an upgrade.  I also believe it may be more robust in handling non-linear materials such as plastics though this does not have impact on what we do for simulation.  Often when checking stress on pressure vessels it is useful to review SCL's at key locations... the non-mapped Tet mesh is not the best at this..... 

                       

                      Ben