• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 98 subscribers
  • Views 192 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.

TMS320C6678: Power consumption in extended version.

Sigurd Henriksen8
Prodigy 30 points
Part Number: TMS320C6678

We are currently using a TMS320C6678ACYP in our application. At high ambient temperature we experience thermal runaway in the device. We have considered changing to the extended version of the device (TMS320C6678ACYPA) as it can handle higher case temperature. However, we want to make sure this will have a positive impact in our system. We are concerned about the following:

1) The current consumption is heavily dominated by baseline currents at high case temperature and these grow exponentially with temperature. The power calculation spreadsheet does not differentiate between the normal and the extended version, but does not allow ambient temperature to be set to 100C. Are the baseline currents equal for the normal and the extended temperature devices? If so, can we extrapolate the baseline currents to 100C? If we do so, the resulting baseline currents would be so high, that the power consumption in the device would be much higher than what has been recommend for the package in other threads in here. This seems like a paradox. What baseline current can be considered maximum for the extended version at 100C case temperature?

2) If the extended version does not have lower baseline power consumption, what causes it to handle 100C case temperature instead of 85C in the normal version.

Thanks,

Sigurd

  • 0
    TI__Guru**** 161600 points

    Hi,

    Let me check this with the design team and will update the thread.


    Best Regards,
    Yordan

  • 0
    TI__Guru* 83975 points

    Hi Sigrud

    We are currently using a TMS320C6678ACYP in our application. At high ambient temperature we experience thermal runaway in the device. We have considered changing to the extended version of the device (TMS320C6678ACYPA) as it can handle higher case temperature. However, we want to make sure this will have a positive impact in our system.

    [MB] This is concerning , how are you ensuring that the devices are running within the max temp spec in the datasheet? Are you confident that using the extended temp part will fix your run away problem? 

    1) The current consumption is heavily dominated by baseline currents at high case temperature and these grow exponentially with temperature. The power calculation spreadsheet does not differentiate between the normal and the extended version, but does not allow ambient temperature to be set to 100C. Are the baseline currents equal for the normal and the extended temperature devices? If so, can we extrapolate the baseline currents to 100C? If we do so, the resulting baseline currents would be so high, that the power consumption in the device would be much higher than what has been recommend for the package in other threads in here. This seems like a paradox. What baseline current can be considered maximum for the extended version at 100C case temperature?

    [MB] The power model allows you to put 100C in case temp field - you need make sure that the frequency is set to 1250 MHz, we do not offer "A" / extended temp for the 1.4GHz variant. The extended temperature part goes through additional testing to ensure it can meet functionality at that temp. The baseline power for both of these parts will be same for 85C. If you run the part at 100C then the baseline power will be higher (you can see the difference in power using the model).

    The package is capable of handling that power increase (and associated temp increase) but you need to make sure you are running device within datasheet max temp spec of 100C 

    2) If the extended version does not have lower baseline power consumption, what causes it to handle 100C case temperature instead of 85C in the normal version.

    [MB]  Not sure I completely understand this query, but in general several of our devices come in commercial and extended temp and the difference is primarily in the test coverage to guarantee functionality and timings for a given voltage/temperature spec.  To further help alleviate your concerns you can program the power model for 1400 MHz @ 85C, and the total power is ~ 18W (much higher than the 1250 MHz @ 100C total power). 

    Hope this helps. Let me know if  you have more questions or if I missed your concerns.

    Regards

    Mukul