• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 721 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.

INA240: INA240A2 stuck at ground with Vcm = 0

George Pontis46
Prodigy 100 points
Part Number: INA240

I have designed a board with many INA240A2 for current sensing. Some, but not all of the parts will power up with the output ground. The output should be at 1.024V due to a 2.048V offset applied to REF1, and REF2 grounded.The power supply is regulated to 5.1V. The inputs are across a 0.02 ohm sense resistor and held strongly at ground at startup. If I increase the common mode voltage to 1V, then the output goes to the expected 1.024V. Taking Vcm back to zero, the INA240 output returns to 0. According to my reading of the datasheet, the part should be functioning properly with Vcm from -4V to +80V. The individual chips are all marked I240A2 and 86KG4. What is going on ?

  • 0
    TI__Mastermind 34250 points

    Hi George,

    Yes the common mode range is -4V to 80V.

    Does this happen at random? Is the circuit identical between the bad and good? With the output stuck at ground, can you probe the differential input voltage? Can you also measure the input pins individually referenced to ground? The input pins must have a connection to the same device ground. If you still have trouble, it might be helpful to post the schematic.

    Regards, Guang

  • 0 in reply to Guang Zhou
    Prodigy 100 points

    Hi Guang,

    The circuits could not be more straightforward. It is the same as the in the data sheet figure 30 for the "Mid external reference", except that in my circuit the reference is 2.048V. The inputs are across a very low value resistor sensing the filtered output of a half-bridge converter. The problem is apparent when the converter is at 0% duty cycle, so the current sense resistor has no current flowing through it, and both sides of the INA240 are at ground. I verified this. I have 8 of these modules on each board, and four prototype boards. The layout between all 8 is identical and they all share a common and low-impedance external reference.

    I am only seeing a problem on 2 channels of one board. It would be easy to chalk it up to ESD or some sort of abuse, and I hope that is the case. But the circuit is incapable of driving the output as high as 80V. The output is also clamped to ground through a Schottky rectifier. Also these problem channels spring back into action when the Vcm becomes slightly positive, and return to ground when Vcm is zero. This is very consistent. It is hard to imagine a failure mode that could result in this behavior.

    George                           

  • 0 in reply to George Pontis46
    TI__Mastermind 34250 points

    Hi George,

    Thank you for the additional information. INA240 should function in the guaranteed common mode range. It is rather strange that two out of eight identical channels misbehave. Do you have a way of doing ABA swap? Or remove the problem device A and replace with a new one B to see if the problematic channel can be fixed. Then put A back to see if the problem comes back. This will give us confidence if indeed something is wrong with our device and we can focus further analysis on it.

    Regards, Guang

  • 0 in reply to Guang Zhou
    Prodigy 100 points

    Hi Guang,

    I believe that I found a design problem that could be damaging the parts. It is that if the output of the half bridge is unloaded, then a step change at the input can make the PWM filter ring with peak voltages that exceed the INA240 spec in both directions. As a curiosity, I am now seeing the two damaged channels stuck at the positive power supply where before they were stuck at ground. Also like before, the output returns to the normal offset when the input is allowed to float up a volt or so.

    I will replace the parts with new ones and protect against the out of bounds condition.

    George