• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 513 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.

ADC128S052Q input protection pulse load capability

Sebastian Wue
Prodigy 215 points

I am going to measure some voltages in an avionics application.

Due to some external constraints, the design is not optimal.

 

The voltage is only filtered with a passive low pass filter.

 

VA of the ADC is 5V (0,2%).

Under normal conditions, no input exceeds this voltage.

 

Accordign the datasheet the input protection diodes are capable of 10mA for unlimited time without destruction.

During a lightning event the double exponential pulse is filtered and the input voltage at the ADC will be exceeded.

 

My question:

Is the input protection capable of an input current of ~40mA for a limited time of 500us up to 1ms?

 

I already thought about other solutions:

  • Input protection via external diode to VA: The drop of the diode has to be very low to conduct before the ADC (i measured ~300mV sufficient for the internal diodes to conduct). And the leakage current will decrease the accuracy of the measurement whitch is not accceptable.
  • Input protection via line resistor after the low pass filter: this may lead to tracking errors when switching the input channels. This could be solved by SW, but i prefer a robust design.

 

Thank you for your answer

  • 0
    TI__Genius 16890 points
    Hello Sebastian,

    When the input voltage goes over the supply voltage it turns on the ESD diode on the input pin. Although the Abs Max section says that it can take 10mA, one of the notes says that it could affect device reliability. Our normal recommendation is to keep the current through the ESD diode below 1mA so that the ESD does not have the possibility of being damaged. This can be done, as you mention, either through a series resistor or and external diode between the input and VA.
    If you use an external Schottky it will turn on before the internal ESD diode and take the current. You will need to make sure that you select the Schottky diode to be able to take the current that you will see.

    Mike
  • 0 in reply to Michael Stout
    Prodigy 215 points

    Hello Mike

    Thank you for your reply.

    I am fully aware that an input current >10mA at one pin can affect reliability.
    In my design, under normal conditions (>99% of the time), the input voltage at any MUX-Pin will stay below VA. Only during a "standard" lighning even, the input voltage can be exceeded and then the ESD diodes will conduct for a very short time (typical 200us, therfore I initially asked for 500us to 1ms).
    That is the reason I would like to know a pulse load (or pulse energy) capability of the ESD diodes.
    Is there any information available?

    If I select a proper external schottky diode, I would need a guaranteed drop voltage of the internal ESD protection diodes.
    I assume, that it is 0.3V mentioned in "Absolute Maximum Ratings" -> "Voltage on Any Pin to GND" -> "-0.3V to VA+0.3V".
    Can you confirm?
    Please note: An external diode will add too much leakage current especially at high temperature >85°C. This decrease of the accuracy is unfortunately not acceptable.

    Sebastian

  • 0 in reply to Sebastian Wue
    TI__Genius 16890 points
    Hi Sebastian,

    The limit listed in the abs max section of -0.3V to VA+0.3V is specified so that the internal ESD diodes do not turn on. A Schottky diode will turn on at about 0.3V which will let it take the current instead of the ESD diodes.

    We do not have information on how the ESD diodes take pulse energy. We know that they will take 10mA and survive but it may affect device reliability. We have recommended to customers for years to keep the current through the ESD resistors below 1mA normally by inserting a series resistor. This is normally how our customers protect the input, they select a series resistor value to keep the current below 1mA. This may be the best way for you also.

    The document slaa593.pdf on ti.com has a circuit for a limiting ADC input driver. Maybe this would help.

    Mike
  • 0 in reply to Michael Stout
    Prodigy 215 points
    Hi Mike,

    I have a freezed Layout, and it is not possible, to implement an suitable opamp in front of every ADC channel (the cost and space will also increase too much)

    I understand now, that it is very hard to give a clear statement.
    So i will Keep the line resistance to limit the current during the lightning with the performance degradation because of the higher tracking error.

    Sebastian