"Database not found." ????

what type of study?

what happens if you make brand new plots instead of use your existing ones

if you have an issue, duplicate the study and rerun, may be an issue with the intiial run/results

I just experienced the same thing.  After a +4 hour solving time period, the "Database not found" window appeared when I attempted to view any of the three plots in the results section.  VERY disappointing. 

You will find dozens of support SPRs on this issue, many of them claiming the issue was "Fixed in SW{year} sp{number}".  It still happens all the time, and I've not seen a way to reassociate the data.  The model file and results file will be right where you left them, but it refuses to acknowledge the results file.  You can copy the files to another directory to try again; I think I got that to work once, but it's tricky when working in a PDM like I am.

I'll add one data point.  We're on SW2013 SP4 Premium 64 bit, with a single Enterprise PDM vault.  At this time all the model and simulation files live in the same large folder.

Here are screen caps showing my simulation setup with the path to the results file, and the PDM folder showing the result file.  This one only takes about 25 minutes, so I'm going to re-run it now.  The customer was expecting data this morning, but we're just going to have to adjust his expectations.

I am getting a similar issue - I'm getting the "Failed to create temporary file. Please change Simulation work directory" message; here's my situation. We are on 2014, but we use PDM Workgroup at the moment. It is a nonlinear study of an assembly with shrink fit. Originally the study ran, got results, made my report, saved into PDM. I am very diligent in saving in the .CWR file with the same revision as the assembly that it is associated with so I have no issues pulling it out later if I need to. Point is, I know with absolute certainty that I am pulling out the correct revision of .CWR file to use with the assembly file because the revisions during a Save to vault operation match. I also pull out the assembly with "As built" selected.

Anyway, sometimes I can go back and pull a file and .CWR file out and rerun, usually because I want to run with updated material information or something, and it works fine. In fact that was the case with this particular file on a different revision, with a different study and .CWR file - successfully reran it. Now I'm doing the same thing on a different revision / study / .CWR file and it says it can't create a temporary file. It's also not consistent with the way it handles the Simulation tree. Sometimes it will gray out the components even though it's in the correct configuration and the components are not suppressed in the model tree. If I change configurations, then right click on the top of the study and select Load SW Configuration, then the parts will come back in the Simulation tree, or so it seems. When I click ReRun, it meshes, (didn't think it would need, especially since the tree doesn't say it needs to), but it meshes, then when done with meshing it gives me the Temporary file message, then starts solving. It won't solve though; at some point it will stop and give another temporary file message, and then a few more.

***

EDIT - it also crashes SolidWorks after actually attempting to solve the analysis anyway.

***

I have had this issue trying to Duplicate studies before as well - that isn't a foolproof process, although convenient when it works. Don't know if there is something relevant there.

This has got to be a more common problem; there is nothing special with our setup or how we are using the software.

Another slightly off topic question: Some people here are not saving the .CWR files into PDM thinking they are just results files and if someone wants to go back and look at results they can just rerun the simulation. Will they be able to rerun the simulation without the .CWR file, or will they have to recreate the study from scratch and then rerun?

Add me to the list.  I've had this happen often.  I think it's only been in assemblies.  2014 and now 2015, yea!

Linear Static.

SolidWorks should ensure that error messages are adequately descriptive.  "Database not found", OK, what database!!??

I got this message this morning.  Loaded an old assembly model with two configs, each w/ one study.  Config 2/Study 2 loads just fine, I can see the results, etc.  I go to Config 1/Study 1 - loads just fine, I can see the results.  I go back to Config 2/Study 2, everything from Results down is greyed out, try to click on Stress Plot, "Database not found."  Close model, reopen, Config 1/Study 1 loads just fine, Config 2/Study 2 still no luck.  .CWR file is in the same place, and I'm sure I haven't changed anything.  Seems like Sim should have a "Load Results" option the way Flow does so you can just point to the results file to re associate it.  That would solve all the problems on this thread, right?

You must change your part file location to a locale without permission issues. Somehow solidworks can not see the results files it creates because of folder permissions.

Another twist on this file management and access problem is that some of my parts are not loading their studies after I installed SP2 (2015).  Some parts and assemblies do load studies and some don't.  All the CWR files and other files are on a local drive and easy to access.

So basically, when I load a part that had a study or multiple studies, nothing shows up at the bottom (yes, the add-in is turned on).

This is a problem because I have days and days of analysis that I can no longer access.

Additionally, one can't re-point the parts to their studies.  It's start-over time.

you should be able to manually attach (reattach) a CWR to a part or assembly.

In the past, I've never had a problem like this, this is the first time.

Hawkridge has verified my problem and is working on it, but as it looks like now, I'm screwed.

This is still a problem, 2015 sp 2.1.

I "ran all" last thing Friday, can't open any results this morning.  Some IT function caused Solidworks to be closed and me logged out over the weekend (long after the runs were done). There is no good reason for the connection between my model and its coherent results to be lost.

I also got the same problem that my results gave the error "Database not found." while trying to show the results. The problem was that the file was saved in a shared google drive folder, saving it on my local disk and all the problems were gone.

I ran into this problem, and it looks like the source of my issue was that the results files that were created exceeded the 218 character limit for file names. My CAD files were BURIED on a network, and then to compound the issue I was putting the results files in a sub folder. Don't fully understand how the simulation was even allowed to create the files in the first place (and I KNOW it created them - saw them in Windows Explorer, viewed the fringe plots, but couldn't view them again after closing the file and re-opening), but when I directed the results to be saved to a location higher up (but still on the network), I was able to view after closing and re-opening.

It would be good if the functionality existed to allow results to be read onto a model. I'm 100% positive other FEA software is able to do this, because I've had to do it before. If I remember right, there was some warning about the user needing to be sure the meshes matched, but other than that it worked like a charm.

I know, old thread. Get over it.

Drive syncing or backup software, like Google Drive, OneDrive, Box, Dropbox, seem to corrupt the meshing and solving functionality.  Try pausing any of this type of activity while using simulation features.  Alternatively, set your simulation working folder to a local drive or your desktop.