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.

INA220: INA220 work abnormally after a few weeks

Part Number: INA220

hi expert,

Need some advice, find some odd behavior on the INA220. Initially under normal operation it works fine but after some time like few weeks or months it will freeze and go low on on the data signal line and it needs a reset (power cycle) in order to get it to work again? Let me further explain how we use it:

 We use the 2pcs of INA220 in a Solar Charged Battery System; one INA220 is placed at Battery and 1 at Load side and with these 2 units we can determine:

  1. Energy in/out of Battery
  2. Energy to Load
  3. Energy from Charger ( is Energy to Battery minus energy to Load)

 

We use an MCU to talk to the 2 INA’s through I2C and every second we read out the voltage and energy data and we have this working correctly.

 

The Issue:

The issue we are finding is that sometimes the communication between MCU and INA is lost. (Energy in and out is reading 0 while in reality there is incoming and outgoing current) Also we have found the issue that one of the INA’s get’s “frozen” and gives a value that is a consistent wrong value.

 

One time we experienced the issue to appear after 7 days of operation to be ok ( 1 reading per second x 7 days = +/-600k readings ok before the freeze took place) other times we experience the freeze happened after just half day in operation.

 

 Any possible work around for this case?

Best regards,
kpk

  • Hi kpk,

    Thank you for posting on the E2E foruns.

    I would like to see a logic analyser/oscilloscope waveform of the I2C bus when this issue happens, in particular, I want to see whenever the communication remains active and functional, with the device sending wrong values, or whenever the bus is halted.

    What fail-safe detection method is implemented on your mcu to handle a loss of comunication, if any? The INA220 has a timeout of 28mS, as per I2C spec, and therefore if there is an error in the communication, the software on your mcu needs to reset the INA220 and resume communication, otherwise it will stay in a loop waiting for an acknowledgement that will never be received.

    I remember a similar issue a while ago, where noise was causing the communication to be corrupted and internal register values were being changed randomly during operation. If the bus is still active I will ask you to perform a dump of those registers, to ensure they are set correctly.

  • Hi kpk,

    Any updates on this issue?