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/TPS54620: TINA-TI simulation issue with TPS54620

Part Number: TPS54620
Other Parts Discussed in Thread: TINA-TI

Tool/software: TINA-TI or Spice Models

I am receiving a convergence error when I try to run the transient analysis on the enclosed design. Can someone tell me what I'm doing wrong.

Thank you,3.3VDC.TSC

  • James,

    I made a few changes that solved the convergence problem.

    • I started with the transient TSC file already in the product folder. It had customized simulation parameters and it looks like you started from a new TSC file so all were defaults.
    • I changed the 12V source from DC to Ramp from 0V (This helps with DC convergence)
    • I noticed that your output caps were not grounded. Not sure if you did that on purpose or just an oversight, so I grounded it and the sim passes convergence now.

    I can not comment on the results though so please check.

    slim102_NG.TSC

  • Thank you for the quick response. For future reference, can you explain what simulation parameters needed changed, Also why did the 12V source need to be changed?

    Thank you again for your help in this.
  • James,

    We sometimes tweak the sim parameters to make the models converge or run faster with reasonable accuracy. Hence, it is good practice to

    • Use the TSC file on the product folder as a starting point and modify it for your needs. If you need to look at what sim parameters changed from default, click on Analysis >> Set Analysis parameters (or Ctrl + Alt + P) and notice he ones in Red. These are the ones that have changed from the defaults.
    • For Switching regulators, it is easier to converge for the circuit if all nodes are starting from 0V at time t=0. If you apply a non zero voltage at time t=0, then the convergence matrix may get conflicting inputs and may not be able to resolve the convergence. This is just a good practice and has resolved many convergence issues in my experience (hence I use this as my starting point).