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.

SM320F2812-HT: Thermal Resistance Spec

Part Number: SM320F2812-HT

Hello,

I am considering using the SM320F2812-HT for an R&D manufacturing application that will require it to withstand a heating cycle taking approximately 2-3 hours from 25C to 180C, and then holding at 180C ambient for 4 hours, about two times a week for several years. I am trying to estimate the lifetime in hours on the datasheet, but cannot find a thermal resistance for junction-to-ambient.

We plan on running on the lowest available frequency (which appears to be 2MHz) and will be logging data at a generally slow rate (2Hz data sampling) from about 14 ADCs at a time. It will send this through a serial interface to a computer outside the 180C ambient chamber. The datasheet seems to suggest (p.88) that the typical current draw for our application will be about 200mW, is this correct?

That being said, I now want to find an estimated temperature for the junction based on power dissipation, but I don't see any thermal resistance value in C/W listed on the datasheet and was wondering if it is possible that you have a number for that. This way we can estimate the average lifetime in hours of our chip.

On another hand I would mention that we are also considering the SM470R1B1MHFQS instead, perhaps you have a recommendation on which to use. However, the 64 pin version of this chip seems to have a really high thermal resistance and at 180C my calculations with 225mW dissipation (from your quote on junction-to-board resistance) implies the junction will on average be consistently over 215C and the operating life will be less than 100 hours.  

Thanks in advance for your help!

Theo

  • Theo,
    I will need to request the thermal models be generated for the F2812, as they do not appear in our database.

    With respect to the R1B1M. The thermal metric for attaching to the board with thermal epoxy represents only a 5.5degree C / Watt above the board temp. However, the R1B1M is an older device with limited collateral and hence more time consuming to design in and program.

    It will take 1 to 2 weeks to generate the thermal metrics for the SM320F2812-HT.
    Regards,
    Wade
  • Hey Wade,

    Thanks for the response, looking forward to seeing the generated thermal metrics for the SM320F2812-HT.

    On another note, do you have any recommendations on soldering/desoldering this chip? We want to use high-melting point solder so that it does not cause problems at our 180C ambient, but I also notice that this chip appears to have some sort of bordered casing, I'm assuming for heat sinking. We have a regular heat-gun reflow station which we've used to easily solder 128-pin chips before and we're wondering if this may be any more challenging. My guess is it would be somewhat similar but I thought it may be useful to ask for any specific advice since I've never seen a package that looks like this before.

    Thanks!

    Theo

  • Hello Wade,

    Any update on the generation for the thermal metrics? If the process hasn't started, is there a chance we could start them soon? The data will be valuable to us. Knowing the thermal metrics and what conditions the test was run in (what kind of heat sinking, how many ground layers, etc.)

    Best regards,

    Theo

  • Theo,
    Thanks for checking back. For some reason I did not get notified that the job was completed.
    Here are the results.
    Theta JA-High K 27.0
    Theta JC, top 10.5
    Theta JB 20.5
    Psi JT 9.2
    Psi JB 18.9
    Theta JC, bottom 2.4

    If this answers your question, please click "Verify it as the answer"
    Regards,
    Wade
  • Thank you Wade! Just one last question to follow-up, what is the setup used to get those numbers? Heat sink or no heat sink, fan or no fan, how many ground layers on the board?

    Regards,

    Theo

  • Theo,

    This application note describes the thermal parameters.

    Regards,

    Wade