• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 50 subscribers
  • Views 5743 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

INA219 Failure (Overheating, Smoking, etc)

Steve Lindsey
Prodigy 20 points
Other Parts Discussed in Thread: INA219, LM22678

We are using the INA219 to measure the Input Voltage and Current of a control board that we have designed and are manufacturing through a local and reputable electronics manufacturing company.  The INA219 circuit is essentially a copy of the "Adafruit INA219 Current Sensor Breakout".  The Input Voltage is a range between 11.0V DC and 15.5V DC.  The maximum current draw of the entire system never exceeds 2 Amps.

We have been manufacturing about 200 of these control boards over the past 6 months and have noticed an abnormal failure rate of the INA219 chip.  The problem we are having is that the INA219 IC will start to overheat in what appears to be a random manner.  We have tried various power supplies and even 12V batteries but when power is first applied to the circuit board the chip will start to overheat at random powerups.  We're not sure why this is and are hoping an expert who knows this chip can help us figure out the source of the problem.

See the attached PNG Image of the Power Circuit for the control board.

We suspect the LM22678 Switching DC regulator may be the culprit (maybe the LM22678 is briefly supplying more than 6 Volts at VDD during system power up) but would think that the switching regulator would not allow more than the 5 Volts at VDD.  However, I don't know if switching regulators have weird anomalies during power up that would cause the INA219 to overheat.

Thoughts anyone?

  • 0
    TI__Genius 14825 points

    Steve,

    If possible, try removing the connection from the LM22678 to the VDD of the INA219, and try powering up the board with an external 5V supply on the INA219.  If several cycles of this show no issue, then you may be seeing an issue with the converter output.

    Since you suspect that the voltage on the Vs pin may exceed 6V, please be aware that these maximum voltages are process dependent and are very firm; violating them can result in device damage.  If your circuit recovers on subsequent power-ups, then that is lucky, but be aware that the devices may be getting damaged.

    If you have an oscilloscope, try probing the voltage at the Vs pin of the INA219 and look for spikes on power-ups.

    Another thing to try is to use a <6V zener diode for protection from Vs to GND.  Here's a 5.4V zener at Digikey, for example, which you could use if the Vs pin was supposed to be about 5V.

  • 0 in reply to Jason Bridgmon
    Prodigy 20 points
    Jason,

    Thanks for the response. Based on your response, I'm going to assume the LM22678 is not being a very good regulator and is causing all of my problems. Although I have not performed tests with an oscilloscope during bootup, I have seen evidence on other boards deployed in the field where the output of the LM22678 does exceed 6 volts and has damaged the rest of the components on the board. What has me scratching my head is when I do have an INA219 chip that starts to get warm I will test the output of the LM22678 and it reads 4.9996 volts. I guess at some point the LM22678 did briefly go above 6V and damaged the INA219 and then dropped back down to the regular 4.9996 volts. I guess I'll redesign the power regulator circuit and not use the LM22678.

    Thanks,

    Steve
  • 0 in reply to Steve Lindsey
    TI__Genius 14825 points
    Steve,

    I wonder if it isn't some kind of over-voltage stress induced latchup condition? Can you check the current into the INA219 when it gets warm and see if the power it is sinking is considerably more than expected?

    I would consider the zener-diode limiter on the voltage supply pin, regardless of the power source. Zeners have served me well for many years and knock those transients down, protecting everything powered by that supply.
  • 0 in reply to Jason Bridgmon
    TI__Guru 69715 points

    I agree with Jason, that a Zener is good protection if the regulator is causing some kind of overshoot.

    We will need more details in order to help troubleshoot the regulator.  Please scope the output of the regulator to see if the voltage is overshooting.

    Also, we will need the layout of the PCB for the requlator since that is the most important aspect of the design.

    Did you see Webench to design the power supply ??

     

  • 0 in reply to Frank De Stasi
    TI__Genius 15810 points
    Looking at the schematic I do not see an input bypass capacitor. There is one large cap at the 12V rail, but nothing close to the IC pins. There should be something there so as to bypass the noise from the switching action. This cap should be placed close to the Vin pin and the anode of the diode. If the device is not stable there could be oscillations which could disturb the output. This will not be picked up on a DMM. As Frank suggested, you'd need to scope Vout, Vin, SW, the inductor current to see that. Scoping the output during the startup will also be helpful.