• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 4 answers
  • Subscribers 48 subscribers
  • Views 192 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.

INA310B: Quiescent current into IN+/IN-

C P van Teylingen
Prodigy 20 points

Part Number: INA310B

I use the INA310 in a somewhat dynamic configuration, where the sense resistor could temporarily be floating w.r.t. the VS supply and GND. However, when I then connect a voltmeter between either the IN+ or IN- pin and GND, I measure a -0.5V on either pin, suggesting that a curreent flows from GND into the IN+ or IN- pin. Applying a current meter between IN+ or IN- and GND confirms that, as this gives a ~70uA current into the INA310 pins. This seems rather strange, as the device is powered by +12V on VS. Current now flowing into the INA310 implies that within the device a negative voltage has to be present to support this. Is this correct and why is it there, or is there another cause for this phenomenon?

Kind regards,

Cees van Teylingen

  • 0
    Guru 326010 points

    This device has a very wide common-mode input voltage range. This is implemented by powering a part of the input circuit from the input pins. See figures 6-13…6-17 in the datasheet for how the inputs behave.

  • 0 in reply to Clemens Ladisch
    TI__Expert 8220 points

    Hi Cees, 

    The -0.5V and ~70µA current you're measuring is caused by internal ESD protection clamp diodes at the INA310's input stage — not by an internal negative voltage rail or charge pump.

    When your sense resistor floats (disconnecting IN+ and IN- from any defined common-mode reference), the input pins are no longer held within their normal operating range. The internal ESD protection diodes begin to conduct in the forward direction (from GND into the pin), pulling the floating pins to approximately one diode drop below ground (-0.5V aligns with a silicon diode forward voltage).

    Your measurement setup (connecting a voltmeter or ammeter between the floating pin and GND) is actually providing the current path that allows the clamp diode to conduct. In your actual circuit, if the pins are truly floating with no path to GND, the current may be lower or the behavior different — but the moment any path to GND exists, the clamp will conduct.

    Best,

    Holly

  • 0 in reply to Clemens Ladisch
    Prodigy 20 points

    I dare to disagree. The diagrams depict the bias current as a result of an applied Vsense. In my case Vsense is 0v (IN+/IN-) shorted to GND via amp-meter. Furthermore does in my case the current flow into both the IN+ and IN- pins, whereas in the diagrams the direction of the bias current relates to the functionality of the pin. To have current flowing into the device from GND would require an internal negative voltage somewhere.

  • 0 in reply to C P van Teylingen
    TI__Expert 8220 points

    Hi Cees,

    You are correct. The charts show that with the device powered on and no Vsense the IB is ~20uA. 

    It is the ESD diodes inside that cause this current to flow from GND into the pins when the pins are floating.

    Best,

    Holly

  • 0 in reply to Holly R
    Prodigy 20 points

    Hi Holly,

    well, can you then please explain to me to which bar the ESD diodes are connected that allows the current to flow into the device from GND? Because that is basically the essence of my question. To my understanding this bar thus has to have a negative voltage w.r.t. GND to cause the current to flow, so where does this negative voltage come from/is for?

    Best, Cees

  • 0 in reply to C P van Teylingen
    TI__Expert 8220 points

    Hi Cees,

    The INA310 has an internal charge pump. If the IB = 0 we can see from the plot below that we expect to see a VCM of -10V on the input pins. 

    When the device is powered on it will seek to pull the input bias current. If it can't get current from anywhere that is when we will see the -10V on the output. If there is any path to GND from which it can pull current, it will. In your case it looks like there must be some sort of FET or diode on that is providing a path to pull current. If we assume the max IB of 30uA * the two inputs pins we get 60uA which is about what you are seeing. And the bias voltage across a diode would give the 0.5V.

    Best,

    Holly