• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 46 subscribers
  • Views 57 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.

ADC128S102QML-SP: Clamping diodes for Analog Input clamp to -0.8V for negative spikes

Arooj Aslam
Prodigy 130 points
Part Number: ADC128S102QML-SP


Tool/software:

Hi,

We have the following circuitry with a zener and regular diode to protect the ADC input. However, the diodes we use are JANTXV1N4624DUR-1 (D10) and  JANTXV1N6638US(D5), and the forward bias voltage is 0.8V with a maximum of 1.1V. To my understanding if we had a negative spike (for example our op amp rails at -12V for 1 ms due to radiation), then we would forward conduct through both diodes and clamp the voltage at -0.8V with about 2.5mA of current flow.

However, this is lower than the absolute minimum requirement of -0.3V.

Would a negative voltage damage the ADC, and is our diode protection not sufficient?

I see in the following section, that there is internal protection as well. Would that help and we can keep the diodes we have?

  • +1
    TI__Genius 16060 points

    Hi Arooj,

    A few comments to serve not just your case, but broadly. 

    The internal protection structures should not be relied on to protect against overvoltage/undervoltage events. If an overvoltage/undervoltage condition is possible, external protection is recommended (which you have done).

    Frequently, we see that protection is not in place for existing designs, or the input pins are susceptible to overvoltage/undervoltage conditions anyways. In case an overvoltage/undervoltage condition exceeds the 0.3V above AVDD or below ground, it is important to limit the current through the input pin. Typically, this is done by placing a resistor in the current path.

    In practice, the internal protection diodes have a ~0.7V forward voltage. If the voltage causes the diodes to conduct, current needs to be limited. The absolute maximum current is 10mA. In reality, this is a conservative figure as well.

    As power dissipation increases with square of current, 2.5mA of continuous current is significantly less harmful than 10mA. The chance to damage the device will also go down with shorter periods of time, so my opinion is that 2.5mA for 1ms periodically would not risk damage to the device.

    Regards,
    Joel