• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 62 subscribers
  • Views 733 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.

WEBENCH® Tools/CSD87355Q5D: Thermal Calculation

Reto Christen
Prodigy 80 points
Part Number: CSD87355Q5D

Tool/software: WEBENCH® Design Tools

Hi, 

I have a short question about thermal results in Webench.

In a design simulated by Webench they are using an "Effective Power Block Junction-to-Ambient Thermal Resistance" of 20 °C/W whereas the datasheet states 50 °C/W. Is Webench already considering some active cooling? 

Results by Webench: 

- Pd in M1 = 2.09 W

- R_{\theta JA} = 20 °C/W

- TA = 30 °C

=> Tj = TA + Pd * R_{\theta JA} =71.8 °C (Webench: 72 °C, therefore, the used formula should be correct).

Thanks for your help.

Best regards,

Reto

  • 0
    TI__Mastermind 18200 points

    Reto,
    You raise an interesting point. It looks like they used the junction to top thermal impedance instead of the junction to ambient. I believe this is probably a mistake but I'll have to dig a little deeper.

    Of course, this gets at a bigger problem - it is very difficult to predict the thermal impedance of your specific design / board. The datasheet provides a min and max value depending on the amount of Cu and the size of the board, but in reality, the impedance could be anywhere in between these two values depending on the design.

    So trying to predict the final junction temperature based on datasheet thermal impedances is somewhat futile - you can get in the ballpark, but there is a lot of margin for error.

  • 0 in reply to Brett Barr1
    Prodigy 80 points
    Brett,

    Thank you very much for this fast reply.

    Maybe a further question to get it correctly. You mentioned that the real value is somewhere between min and max value.
    But as soon as I am using a heat sink, I can use the "junction-to-case thermal resistance" and then in series add my thermal impedance of the heat sink, right?

    Reto
  • 0 in reply to Reto Christen
    TI__Mastermind 18200 points
    Aha,
    Yes I did not realize you were using a heat sink. In that case, your real RthJA would be the parallel sum of the impedance through the top (RthJC(Top) + Rth(Heatsink)) with the impedance through the bottom (somewhere between 50 and 100).

    So say you were using something similar to the max Cu board, the real RthJA would be (1/(20+Rth(HS))+1/50)^-1.
  • 0 in reply to Brett Barr1
    Prodigy 80 points
    Brett,

    my original question actually was without a heat sink (or maybe if Webench already is using a heat sink since they use a much lower value).

    When I use these 50 °C/W the junction temperature reaches its limit rather fast, therefore, I asked the question with the heat sink.
    Your example was extremely helpful. Thank you very much for that.

    Reto