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.

LM2901: Not connected input

Part Number: LM2901


Dear All,

I'va a question about LM2901 comparator.

We're performing a failure analysis of our circuit and we would like to know what happen if a input of comparator goes "floating".

What is the voltage of input pin if it became "not connected"?

Thanks in advance for your reply.

Best Regards

Stefano

  • Hi Stefano,

    you must always provide a current path to ground for the input bias currents. So, do never leave an input float.

    Kai
  • Hi Kay,
    thank you very much for your reply.
    Yes, I know that I never leave an input floating, but we are performing a failure analysis on our circuits and we are need to know what happens in the hypothesis that the circuit connected to input is damaged, leave the input "floating".
    Can you help us?
    Thanks in advance

    Stefano
  • Hi Stefano,

    I would add a high ohmic resistor form the input to ground (pull-down resistor) to always allow the input bias current to flow. Choose the pull-down resistor not too high ohmic, otherwise the input voltage can exceed the common mode input voltage range of LM2901.

    An example: 500nA is the maximum input bias current. 3V is the maximum common mode input voltage, if the LM2901 is supplied by 5V. This gives a maximum pull-down resistor of R = 3V / 500nA = 6M.

    Kai
  • Dear Kai,
    thank you for your suggest that we'll keep in consideration for our design.
    Anyway, we are interested in knowing what happens at comparator if its input become "floating" after a failure of circuit connected to it (included the 6M pulldown resistor that I've suggested to insert).
    I hope that my question is clear.
    Thanks in advance
    Stefano
  • Hello Stefano,

    See the Figure 6 internal schematic in the Datasheet.

    The input stage base current flows from V+,  through the Darlington B-E's, and OUT of the input pin towards ground. If you leave the input floating, there is no path for the base current, and that inputs side "floats" towards V+ and the input stage is cut off - just like the "Feature" described in Section 9.2.2.1 when an input is driven above the input voltage range.

    So if one input is still within the valid input range, the output should still be correct.

    But there is a difference between "driven" and "floating" inputs. A "driven" input is forced to a voltage by an external source. A "floating" input is internally driven, usually by internal leakages.

    If both of the inputs are cleanly "driven" above the input range, then the output will be low.

    But "floating" inputs do not have a determinate level and can change, usually depending on temperature, and will vary between between units. And these inputs may "float" to similar levels at the edge of the biasing threshold, which means the "higher" input may still determine the output. This can be random between devices and will vary over temperature.

    So there is no determinate output value for "floating" both inputs, but the output is expected to be correct as long as one input is within range.

    This is expected behavior, and NOT warranted behavior. A "floating" input is not a valid input condition.