• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 38 subscribers
  • Views 1301 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TINA/Spice/LMV7291: Simulation Slows on Subsequent Simulations

Samuel Jaffe
Mastermind 25590 points
Part Number: LMV7291
Other Parts Discussed in Thread: TINA-TI, OPA2684, OPA2683, OPA683, OPA684

Tool/software: TINA-TI or Spice Models

Hi E2E,

My customer is running a sim on TINA-TI version 9.3.150.4 SF-TI

He has run his simulation a couple times and it's taking 2-3x longer now than it did the first time.  Are there any files that need to be cleaned up, or some other solution, to get these sims running faster?

Thanks,

-Sam

  • 0
    TI__Mastermind 25590 points
    The same happened with PSPICE. It ran fine the first time and after a few simulations, it takes much longer. Could it be to do with the models?
  • 0 in reply to Samuel Jaffe
    TI__Mastermind 25590 points

    Update from the customer:

    Restarting does not help but I think it has something to do with the parts I use.  Right now I just ran the simulation three times and it runs with valid outputs.  However if I change the op-amp in the post amplifier circuit it would not run.  So now my question is:

    If our design uses the OPA2683 in the pre-amp and we use the OPA2684 in the simulation are we losing anything.  

    Working simulation uses OPA2684 pre-amp with OPA683 post-amplifier.

    Slow simulation uses OPA2684 pre-amp and OPA2684 post-amplifier

    Slow simulation uses OPA684 preamp and OPA684 post-amplifier.

    Fast simulation with invalid output uses OPA683 pre-amp and OPA683 post amp.

     

    I'm baffled.  Thoughts?

    -Sam

  • 0 in reply to Samuel Jaffe
    TI__Expert 3900 points

    Sam,

    Simulation/convergence time depends greatly on the model used, as well as the Simulation software (i.e. TINA-TI, PSpice, etc.).

    Is the wait the only issue or is this preventing him from finishing to run the simulation? if it is preventing you from finishing the simulation you could try something like what is suggested here by Britt.

  • 0 in reply to Victor Salomon
    TI__Mastermind 25590 points

    Victor,

    Thanks for the reply.  Yes, different models take different durations to simulate but the strange part of this issue is that it's identical simulations that take different durations.

    We will take a look and revisit this after the holidays.  

    Thanks again,

    -Sam

  • 0 in reply to Samuel Jaffe
    TI__Mastermind 41310 points

    Sam,

    With the information provided, I do not see 3 identical simulations that are being run. Each one listed uses a different set of parts and simulates differently. Please correct me if I am incorrect. That would NOT surprise me.

    A couple of things to note about TINA-TI. It uses a dynamic load sharing feature which by default selects all processor for use. If you are running something processor intensive at the same time as your simulation, your simulation will be slower. I have seen this behavior before, but I do not recall it being a 2-3X hit. TINA-TI will run differently if you change the parts out even if they are similar. It uses an adaptive time step by default and this is very dependent on the circuit and models being used. Most of the time this will not be an issue, but I have seen cases where changing out an "equivalent" part required a different set of simulation parameters for the most efficient run. Once again, I cannot be certain of the issue.

    Can your customer provide the .TSC file he is running and the exact steps taken to reproduce the problem?

  • 0 in reply to Britt Brooks - Online Design Tools
    TI__Mastermind 25590 points
    Britt,

    Yes, the post with blue text will simulate at different speeds due to the different parts. The identical simulations occur when one of these simulations is run multiple times in a row. This was the original problem. Without changing parts, subsequent runs took long and longer to complete.

    I'll ask for the .TSC file and attach it if I can get it.

    -Sam
  • 0 in reply to Samuel Jaffe
    TI__Mastermind 25590 points
    It looks like the customer will be using another simulator for other reasons. Thanks for the support.