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.

INA199: Current Leaking through the sensor

Part Number: INA199

Hi,

I've been using INA199A2 to measure current flowing from my 12V battery to a load. It worked fine.

Later I started designing a boost converter (from 12V to 20V) with a INA199A2 sensors. For my first experiments I controlled only voltage, so I assembled INA199A2 sensors but I didn't even apply any voltage to Voltage Supply port. They are on my PCB for further experiments.

First couple of tests run okay. Later I started having problems as voltage  in the circuit just collapsed. I started removing different parts of the circuit and I couldn't find where the problem was. Finally, I saw that whenever I energize the circuit my V+ port of INA199A2 shows the same voltage as the batterym which is 12V. The current is clearly "leaking" through my INA199 sensors.

Can someone tell me what is wrong with the circuit and why suddenly 12V is being generated at the V+ terminal of INA199A2?

Best Regards,

Bartosz

  • Hello Bartosz,

    Thanks for considering to use Texas Instruments in your design. All of our parts will have some current leak through the inputs known as input bias. However, if alot of current is drawn through those inputs, it would suggest that you have damaged the internal ESD structure of the part. This could happen from transient events in which the absolute maximum ratings are exceeded for longer than a typical ESD event. Such events could possibly be from inductive kickback or something else.
    I tested an INA199 in my lab with a floating supply, floating output, 12V common mode, and 1V differential across VIN+ and VIN-. From that setup, I observed that VCC was a diode drop above ground. So I suspect your part is damaged. To facilitate the debug, can you provide a schematic and operating conditions?
  • Hi Patrick,

    Thanks a lot for your support.

    I was obviously expected INA199 to consume some current but in my case it was around 70mA which is very high. The chip was clearly damaged.

    Here is a schematic of the circuit.

    In my experiment only Mosfet Q1 was active and it was boosting voltage from 12V (capacitor C1) to 20V (capacitor C2 and C3). 

    The L2 and L1 parameters are 2mH. 

    I didn't expect voltages higher than 26V which is a maximum for INA199 but perhaps during transients this voltage could be as big as 26V?

    I am going to add a transient protection, just as it is recommended in the datasheet. Let's see what happens.

    Best Regards,

    Bartosz

  • Hello Bartosz,

    I consulted with my team on the issue and your schematic looks fine. However, we suspect that the version of INA199 you are using is the issue. You are currently using the A version. Versions B and C are a bit more robust. While version A may have a wider absolute max, it does not handle fast slew rates well. We suspect that you could confirm this through supplying power with a power supply rather than a battery. The power supply provides a much slower ramp up time than a battery that can quickly dump a lot of charge. If your devices consistently survive with the power supply, then a simple change to the B and C versions should fix this issue for a battery supply.

    Versions B and C will be named INA199B2 and INA199C2.
  • Hi Patrick,

    Thanks a lot for that suggesting how to fix this problem.
    Do you think I still need a transient protection circuit with a zener diode and a resistor? Even if I use INA199B2 or INA199C2?

    Best Regards,
    Bartosz
  • Hello Bartosz,

    I think the B and C versions should be sufficient. However these parts are for ESD events and handle fast slew rates better. If under normal system operation the device inputs still see transients that exceed the common mode max for a relatively long duration. Then you will still want to add the zener diodes. One way to determine if you need it would be to place oscilloscope probes on your inputs and have the trigger set near or at the absolute max, if the signal never crosses your trigger point then I would say you can save some cost and board space and forego the additional external transient protection circuitry.
  • Hi Patrick,

    Great! I am going to buy B or C version in this case.

    I run the boost converter without the INA199A2 and looked at the signal which is going be connected to my sensors. It won't exceed 26V so I should be able to run the system without additional protection.

    Many thanks for your help! I really appreciate that.

    Best Regards,

    Bartosz

  • Glad to be of service. Good luck with the rest of your system!