• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 83 subscribers
  • Views 417 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.

TMS320F280049 Temperature of DSP?

user4565873
Expert 1250 points

Other Parts Discussed in Thread: TMS320F280049

Hello all,

Do you have any data concerning the temperature of a TMS320F280049?

We are worried because we 'burnt " our fingers touching the opertaing DSP with activated ADC, CMPSS, PWMs at a frequency of 15kHZ.

We expect the temperature is around 50-60°. Can you confirm please?

 

  • 0
    Guru 14640 points
    Just a quick suspicion on my part - one time I worked with a customer who was using the ADC outside of the normal range (but only for a short time...) and the heat generated was enough to change the values of some of his flash memory. You may want to make sure that your ADC inputs are in range.
  • 0 in reply to Todd Anderson78572
    TI__Guru 60865 points
    Hi '5873,

    Just anecdotally, I would not expect this device to run very hot.

    The nominal power dissipation in the datasheet is roughly 61mA*1.2V + (26mA + 12mA)*3.3V = ~0.2W

    The DUT to board thermal constant is 24.2 deg. C/watt (also from the datasheet) so this would result in the DUT being roughly 0.2*24.2 = 4.8 degrees hotter than the board (assuming the board is large and a good heat sink).

    I agree with Todd; if the device is hot it is probably because there is an overvoltage condition (now, or in the past that has caused EOS damage) or some ESD or other damage has occurred.
  • 0 in reply to Devin Cottier
    Expert 1250 points
    Ok, Not good.
    It's 40°C minimum. (can't leave the finger on it)

    We only have 3 ADCs connected from 3 (0 to 3.3) LEMs and PWM outputs to buffers to generate motor rotation control.
    This test set does not look like a cause of severe heat.
    But we have previously run several tests with several connection such as (possible erroneous ) -10V/10V analog input...
    Could the heat come from a previous overvoltage?

    Thank you,
    PA.N.
  • 0 in reply to user4565873
    TI__Guru 60865 points

    Hi PA.N.,

    I would definitely expect the device to be damaged if +/-10V was applied to any of the pins via a low impedance source.  This might result in a short to ground or power rail, causing excessive current to flow in/out of the pin when you connect a voltage.  

    If you have a source-meter or power supply with an accurate current read-out at low current you can do the following experiment on each ADC pin:

    • Source 3.3V (or even up to ~3.5V), measure current into the pin
    • Source 0V (or even down to ~-0.2V), measure current out of the pin

    The nominal leakage current should be << +/-2uA, so if you see 10's of mA flowing into or out of one of the ADC pins, you can be reasonably sure that pin was damaged via EOS or ESD.