• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 62 subscribers
  • Views 241 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.

TPS54318: TPS54318 loop compensation - Datasheet formula vs TINA vs Webench

Sunset
Genius 13385 points
Part Number: TPS54318
Other Parts Discussed in Thread: TINA-TI

Hello Team,

we are currently checking a project to step down the voltage from 5V down to 1V @ 1A.

We are using the datasheet, TINA and Webench, and we are getting different results.

Some are quite different, thus the questions:

  • What is the most reliable, theoretical way to determine the components for the loop compensation? (datasheet, TINA, Webench)
  • How much is expected to deviate the result computed out of the datasheet formula from TINA and Webench simulations?
  • Which model is more reliable for verification - the step response in the transient model, or the average model?
  • Is it Webench using the same spice model as the one given for TINA?

Thanks,

SunSet

  • 0
    TI__Guru**** 191445 points

    Mauro,

    The datasheet uses simplified ideal current mode control equations that ignore the effects of slope compensation.  Since TPS54318 will generally have relatively high duty cycle operation, the slope compensation effects can be significant.  The datasheet calculations are best suited for a rough estimate.

    The average model engine for the Pspice/TINA models and Webench should be the same.  The method Webench uses for creating compensation is iterative and should produce valid results, but they may not adhere to the typical compensation schemes that are commonly used such as pole/zero cancellation or k-factor.  As of now Webench does not offer any easy way to adjust compensation or support certain mixed output capacitor scenarios.  But those features should be available in the future.  I use and prefer Pspice average modeling.  TINA-TI uses the same model, just a different user environment.

  • 0 in reply to JohnTucker
    TI__Genius 13385 points

    Thanks for your quick feedback,

    SunSet