Pspice EPC 8002 and 8010

I am trying to analyze EPC 8002 and 8010 in pspice for TI. I have followed the released TI video for importing EPC 2023 and would have thought it would be the same steps for all the pspice net list files.
link to video: PSpice for TI: 3rd Party Model Import | Video

I have been success in importing EPC 2023, but get this error message when doing 8002 and 8010

Any advice as to why this is happening would be appreciated.
Patrick Willard

Hi Patrick,
I lead the device modeling team here at EPC. Thanks for the question. I agree that your results are unexpected, and I’m glad that you have shown EPC2023 can be imported successfully.

The most recent versions of the models are as follows:

Can you please confirm that these are the versions you are using?
I ask because they differ only in parameter values, and the actual model equations are identical. So if there is still an issue where EPC2023 imports successfully but EPC8010 does not, we can rule out model equations as the reason why.

Thanks for getting back to me,
I have been using the most up to date files. Today I tried copying 8010’s contents into 2023’s file and saved it as EPC2023.lib and it seemed to import alright. I am pretty new to looking deeply into spice lib files, but is there a better way to import all of EPC’s GaN fets in one library rather than importing them 1 at a time. I think pspice is have trouble dealing with very similar libraries. Thanks

Interesting. That gives me a few ideas. In the video you originally referenced (2.6 PSpice for TI: 3rd Party Model Import), they include a number of key steps:

  1. (@ 0:49) Save all .net pspice model files in the same folder directory, but rename to end with .lib
  2. (@ 1:08) Generate a unique symbol for each model using the “Generate Part” dialog and model import wizard
  3. (@ 3:08) Update the “Configuration Files” section of Simulation Settings by clicking “Add to Design” for each part’s .lib model.

Can you please confirm that you have completed all 3 key steps?
If someone did #1 and 2 but not #3, for example, that would explain why the simulation would be looking for the EPC8010 subcircuit, but failing to find it.