• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 4 answers
  • Subscribers 62 subscribers
  • Views 131 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.

Original question:

REF5040: TEMP output

REF5040A-Q1: TEMP monitor pin tolerance

Lloyd Camposagrado
Prodigy 10 points
Part Number: REF5040A-Q1
Other Parts Discussed in Thread: REF5020A-Q1

Currently testing this part on about a dozen boards and we've found the number showing up within the expected +/-15degC to be about 1 in 3.  The abnormal units are showing up as +/- 22degC at room temp

I haven't seen this issue with a Ref5020A-Q1 that i have on the same board.  they are both feeding into sallen-key filters before hitting our ADC, but both an onboard voltage measure confirms our readings are seeing this issue correctly.  what parameters/circuit characteristics could be affecting this and not the ref5020a-q1?

  • 0
    TI__Genius 10951 points

    HI Llyod 

    Temp pin monitors the die temperature not the ambient temperature. The die temperature will also get impacted by the power dissipation  in the reference. 

    Temp pin voltage is defined as typical only. You can expect some variation. 

    Regards 

    Trailokya

  • 0 in reply to Trailokya Rai
    Prodigy 10 points

    That make sense for an increase in temperature, but we registered both higher and lower voltages on different units (±22degC) from ambient (25C) .  All I really need is to define a wide enough range at room temperature to call it good during our initial board checkouts when approving a new assembly.  Is there a recommended range to be used for health/unhealthy ICs given a known load/ambient room temperature.  A max/min value with some margin would be acceptable just for us to know its functioning as expected.

  • 0 in reply to Lloyd Camposagrado
    TI__Genius 10951 points

    Lloyd 

    This parameter is not controlled by design. It can vary from part to part by a large amount so we don't recommend to measure the temp value to approve the assembly. 

    It is useful when you do temperature calibration. You can store 25C temp pin voltage and check the device temperature change on temp pin and add offset for  your temperature variation. 

    Regards

    Trailokya

  • 0 in reply to Trailokya Rai
    Prodigy 10 points

    Ah ok, thanks for that Trailokya.  Is it ok to say that the accuracy is only in the delta temperature from the initial ambient temperature measurement

    so at room temperature, if i were to calibrate the temperature coming from the TEMP pin, it would be measured voltage = Tambient + Trise where Trise is the calculated Power dissiption x Tja?

  • 0 in reply to Lloyd Camposagrado
    TI__Genius 10951 points

    Lloyd 

    Temp pin will measure voltage = Tambent + Trise . But the ambient voltage variation with part to part will be high.

    So this is not generally suitable for accurate temperature measurements, this pin can be used to indicate temperature changes or for temperature compensation of analog circuitry. A temperature change of 30°C corresponds to an approximate 79-mV change in voltage at the TEMP pin. 

    Regards

    Trailokya

  • 0 in reply to Trailokya Rai
    Prodigy 10 points

    right, that was the part i was confirming, once you calibrated to the measured ambient + Trise voltage, you can rely on the delta temperature change of 2.64mV/degC ± XXX (i noticed this also only shows typical) as the general delta rise in temp, and how accurate is it for analog circuitry temperature compensation? or is it basically just hand wavy at if you see 79mv assume a 30degC change, and then next would be at 160mV assume an approx 60degC change and you wouldn't be able to say do the following with any real confidence

    ambient + trise = measured voltage at time 0

    1min later = measured voltage at t0 + 79mV = +30degC ± 15C

    2min later = measured voltaeg at t0 + 158mV = +60degC ± 15C

  • 0 in reply to Lloyd Camposagrado
    TI__Genius 10951 points

    Lloyd 

    Voltage slope of temp pin will be fairly constant across temperature for a particular device. So you can say 

    ambient + trise = measured voltage at time 0

    1min later = measured voltage at t0 + 79mV = +30degC ± 15C

    2min later = measured voltage at t0 + 158mV = +60degC ± 15C

    Regards

    Trailokya