Post Go back to editing

AD7173-8 input buffer max voltage exceedance

Category: Hardware
Product Number: AD7173-8

I am curious what the expected behavior is when exceeding the maximum input voltage for the AD7173 analog inputs with the input buffers enabled.

I have an AD7173 configured with the input buffers enabled. The signal I am measuring has a max differential voltage of 3.75V and the negative side of the differential input is connected to a 1.25V reference. I am powering AVDD1 and AVDD2 with 5V and IOVDD with 3.3V. This means I may exceed the max input voltage of AVDD1 - 1.1 = 3.9V. I would imagine I would be railing the input amplifier output at 3.9V when I exceed this limit. I would expect this would result in a max sense voltage of approximately 2.65V (one input amp railed at 3.9V, other operating nominal at 1.25V). What I'm seeing is that the max sampled voltage is in the range of 3.06 to 3.17V. Is this behavior expected? If so, can someone explain what's going on?

Thanks so much for your help.

  • I want to add that I am not exceeding the absolute maximum rating for analog inputs of AVDD1 + 0.3V, just the max analog input voltage per the Specifications section of the datasheet.

  • Hi  ,

    Is 1.25V a fixed value? When input buffers are enabled the absolute input voltage limit is AVDD1 - 1.1 = 3.9 V. You are correct. Having the negative input voltage fixed to 1.25V then the maximum sense voltage you could get would be 2.65V when the positive input is railed at 3.9 V. One thing to note is that the ADC is only capable of accepting differential input voltage range of +/-Vref. Therefore, the differential input voltage range is limited by the value of your reference voltage. 

    May I know what is the value of your voltage reference?

    When you say max sampled voltage, are you referring to the voltage input AIN+ or this the differential voltage?

    We also have this tool that you can use for simulation. Virtual Eval | Analog Devices

    Regards,

    JC

  • JC,

    Thank you for your response. My ADC reference is 4.096V.

    When I referred to max sampled voltage, I was referring to the code reported by the ADC, generated by sampling the differential voltage. It's interesting that the value I am seeing suggested a differential voltage around 3.06V to 3.17V, implying that the input buffer is actually railing at a voltage above 3.9V.

    Thanks,

    Jeremy

  • Hi  ,

    First, I would not recommended operating in the rails as any instability in your input may cause it to go outside the absolute input voltage limits and the device may not behave on the way it was expected to behave.

    Since you are using a 4.096V reference it should be able to accept differential voltage of +/- 4.096V. Let us further debug the issue. Could you try probing on the actual Reference pin of the ADC just to check if the actual voltage is 4.096 V.

    What is your actual AIN+ and AIN- input that resulted to sampled voltage of 3.06V to 3.17V? What is the actual code that you are measuring? Did you ensure that the DOUT/RDY pin goes low before attempting a read on the conversion? Kindly check the status register if any error has been flagged.

    Could you try measuring an input that is not near the rails of the buffer? 

    Is it possible to share your schematic so we can verify if there are any errors in the schematic causing the issue?

    Regards,

    JC

  • Hi  ,

    Have you had any progress in your issue?

    Regards,

    JC

  • My apologies for the delayed response.

    I have confirmed my actual reference voltage is 4.096V. When seeing this behavior, I measure 4.85V at AIN+ and 1.25V at AIN-. I am confident that we are sampling correctly, as other channels on the same ADC are functioning as expected.

    My working theory of what is happening is that the output voltage limitation of the input buffers is actually a little higher than AVDD - 1.1V = 3.9V. Part to part variability and temperature effects likely explain why the datasheet spec is AVDD - 1.1V, since this spec is applicable to all parts including all sources of variability. I suspect I'm just seeing the input buffer for my 4.85V input saturate at something around 4.35V instead of 3.9V, giving a differential input voltage at the ADC in the range of 3.06V – 3.17V. I have also observed some variability between parts, with some reporting lower sampled voltages in the range of 2.85V, which is consistent with this behavior.

  • Hi  ,

    Since you are operating beyond the specified Absolute input voltage, there is no way to predict the behavior of the part. The headroom maintains the buffers' performance in the linear region. Thus, exceeding the headroom voltage, it now operates outside its linear region. 

    I would strongly recommend using the AD7173-8 within its specifications to ensure correct performance of the device.

    Regards,

    JC