• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 83 subscribers
  • Views 408 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TMS320F28377D: Estimating Junction Temperature of F28377D

Brian Wang0
Expert 5420 points
Part Number: TMS320F28377D


Hi team,

Recently I am supporting my customer on a F28377D second source verification, and they would like to confirm whether the PTPT version (105℃) is enough for their application.

On the current board, the measured case temperature is 89.33℃ and measured power consumption is ~850mW. Thus the junction temperature is estimated as:

89.33+0.85*6.97℃/W =95.25℃

How shall we evaluate whether a 105C version would be enough?

Regards,

Brian

  • 0
    TI__Mastermind 42195 points

    Brian, 

    By your calculation you have a bout a 10% safety margin from the 105C limit. Is this the highest expected operating temp? Also remember when using Theta JC this is typically a conservative an inaccurate estimate, so I would expect that you actually have a little more less margin than you think. Deciding if this is enough margin really depends on your application conditions and expected lifetime. If i remember correctly this is a great appnote on the topic https://www.ti.com/lit/pdf/sprabx4 .

    One other important consideration is localized heating. If the C2000 is not near other heat producing components then you will get good thermal dissipation though the PCB. If you put the C2000 near high current devices you may actually be heating the C2000 device though the PCB. I don't see this issue too often, but its a consideration during layout.

    Regards,
    Cody 

  • 0 in reply to Cody Watkins
    TI__Expert 5420 points

    Hi Cody,

    From the paper you posted, I noticed the if Rθja is used as the thermal resistance, then P is the thermal power conducted from junction to case rather then total power dissipated.

    In stead, we should use Psijt which is only 0.11C/W, much smaller then Rθja which is 6C/W.

    Regards,

    Brian 

  • 0 in reply to Brian Wang0
    TI__Mastermind 42195 points

    Brian, 

    Correct. Psi JT is more accurate. The conventional advice is to use Psi as using Theta is not as accurate. I had mis-spoken above and corrected that now.

    Regards,
    Cody