Saving .log file or your own file with commands may not work from one version of ANSYS to another if you do subsequent Boolean operations. Long time ago, I built a model in ANSYS5.2 that didn't work in ASNYS5.4. I got the well known message "Boolean operations failed. Try loosening.....". It took me a while to realize that the real reason for failure was that the output entity numbers were different between those two versions and so the subsequent Boolean was trying to, let's say, add two volumes that were not touching. Similarly, my model created in ANSYS5.5 doesn't work in ANSYS5.4. So I usually save database of the whole assembled model and then I have a bunch of files that do the remaining work. I haven't had problems with using databases created in lower versions. Abut 6 months ago there was a discussion about how to avoid this by tracking volume centroids rather then volume numbers (or some other ideas), but that's a lot of extra work.
Karol Galik PhD student at ME department University of Pittsburgh, PA kagst34@i...
"Otto, Mike J" wrote:
> Martin,
> You may feel "out of step with the rest of the world" but I am following > right in your footsteps. I concur with your method.
> I would only add that I'd probably remodel and reanalyze any job older than > a few years anyway given the pace of technological change (new elements, > improved meshing and solver techniques, better CAD translators, etc). My > technique changes given the resources, requirements, and knowledge at the > moment and hopefully is more efficient and accurate as time marches on.
> Mike Otto > Eaton Corp > Supercharger Div. > 616-781-0304
> -----Original Message----- > From: Martin Liddle [mailto:martin@t...] > Sent: Thursday, November 30, 2000 1:22 PM > To: xansys > Subject: Re: [xansys] Save your work
> In article , Vezina, > Martin writes > > How do you save (store) your work (model) when it is finished ? I'm > > not asking about the save command !! I mean, how do you manage to > > save the important files (database, results file, ...) to retrieve > > them after 6 months for example ?
> As usual I'm out of step with the rest of the world :-( I find the log > file method too tedious for big 3D models. I tend to keep the database > file (note that it compresses very effectively using gzip (under Unix) > or Winzip (under NT)). I do use command files to apply the loading data > and keep them. I also keep results files (these also reduce in size > using compression but not as effectively as database files) for several > years to allow quick access to results to answer occasional customer > questions that crop up long after you have forgotten about the job. If > long term (say greater than 5 years) access to the data is important > then I'd agree with the suggestions to export it to ASCII using CDWRITE.