Hi Steve,
The message indicates that the version of the model used by the solvers (the 5 databases located in the DS.MOD folder) is out of synch with the user interface version of the model (the .fem file). There are two things that you can try:
A. The software will automatically create parts when some feature are added to the model. Delete any 1D springs, 3D springs. or any type of surface contact that you added to the model.
B. You can try the following steps to manually repair the "AGSDB" database files:
If the error message gives a part number, that will help to locate the lines to delete from the text file. If the part number(s) are not given in the error message, it could be one of the parts that was deactivated and deleted, or it could be a "part" that the software creates automatically -- such as the 1D springs, 3D springs, and surface contact in linear stress.
Good luck, flyboy.
Thank you John for the suggestions.
Removing the spring elements didn't work.
The error mesage does not identify which elements are causing problems:
** Model [C:\Users\flyboy2160\Desktop\FEA fuselage 2014-11-18 low horz\fus and horz wires.ds_data\3\ds] opened
--- 721 nodes
--- 168 element parts ...
More activated element parts in AGSDB (167) than FEM (156)
With the springs removed, I translated the files out as you suggested. But I have 2 problems: it's not obvious what to do with the .csv file and when I try to read it back in, the "Original Format" pulldown window has no choices. The "Output Format" pulldown alllows choices.
Even though I'm laid off, I'd be willing to pay someone a small fee to actually show me how to fix this. This is the second time i've had some fatal error with this complicated model. The last time I rebuilt it from scratch, cussing the whole way. This time I've tried merging the "bad" model into a new model. The elements come in, but their definitions are stripped away. And even when I fix up a couple and try to run the model, I get a different error.
Wow. Only 721 nodes for a model with 168 parts? That's probably a record for the fewest nodes per part.
Basically, you know which "real" part numbers are in the model since they are listed in the browser. Any part numbers in the model.csv that doesn't correlate to one of those parts can be deleted. Note that the model.csv file uses the terminology "group" instead of "part". So lines like "LinearStress.Egroup(1).Type" (the element type for part 1) and "Material(1).General.Library" (the material properties for part 1) can all be deleted if part 1 no longer exists in the model. There is probably even just one line whose value could be changed (not deleted) to indicate the part is suppressed (probably "LinearStress.Egroup(1).Status"), but I do not know what the value should be to indicate the different statuses.
The model should be relatively small if you wanted to create an archive of the model and attach it to a reply. See "Create, Post, or Provide an Archive of your model". Also indicate which version of the software you are using (e.g., 2015) since people with older software will not be able to open your archive, and you would not be able to open an archive with the fixed model if you have an older version.