INA212 seems to be dead.

Hi Nikita,

I am sorry to hear you are experiencing difficulties with the circuit.  Based upon your description of the problem I think there must be a high voltage applied to the INA212 and we will need to check it.  Before I make my observations to you I will answer your question on input voltage range...the INA212 is specified for an input range of -0.3V to 26V independent of the power supply voltage...so if you had a 3V power supply at V+ pin you could have up to 26V on the IN- and IN+ pins and have no damage.  That is how the part is supposed to work.

In your circuit I do not exactly recognize the symbol on the left hand side but I am guessing it is a photovoltaic voltage source.  I also see a Zener diode connected in series with this source.  My concern is the voltage generated by the photovoltaic source...on your schematic you show that it is regulated to 3.501V on the right hand side of the diode...since the diode will be forward biased by the current flow that you indicated on the schematic this means that the photovoltaic voltage source will be about 0.250V higher than 3.501, so this is about 3.751V...is this the case?  Also if there is 1mV across the 0.1Ohm shunt this means the amount of current flowing is 10mA...where does this current flow?  Is terminal UB connected to a load of some sort?  You described the output of the INA212 as going higher than the V+ pin...when the device is functioning properly this is not possible...so please double check the voltage at the V+ pin and ground pin right at the INA212 to confirm it really is the 3V that is indicated on the schematic.  With 1mV of input the output should be 1V...since you are seeing higher than 1V, please confirm the voltage between the inputs IN+ and IN-...it might be possible that the differential input voltage is greater than 1mV.  By taking voltage readings at the following pins I think we can further trouble shoot your circuit:

Please measure the voltage at the pins of the INA212 between V+ pin and the ground pin.

Please measure the voltage at the pins of the INA212 between IN+ and IN-.

Please measure the voltage at the IN+ pin and the ground pin.

If you make the above measurements on a new unit and since your circuit works OK at time zero and seems to fail over time you may need to make continuous voltage measurements (logging the data) to see how these voltage will change with time.

Alternatively you can make the voltage measurements on a failed unit, perhaps we can find the problem from just those measurements without the need for the two hour measurements and logging the data.

One more note is I do not see a power supply bypass capacitor on the V+ pin...I do not think this has anything to do with your issue, but it is always recommended to place a capacitor at the V+ pin to ground.  Please reply with the measured data and we will see what the next steps will be.

Yes, on the left is Solar Cell, and diode is Schottky 1n5158. So 3.5Volts is because of Li-Ion cell, connected directly to UB (UBattery).

this is plot of measurement, when IC died. As you can see at 3:30 AM Current could grow, but it still was dark. Later, when i have disconnected solar cell, voltage on OUT was 3.5, what is slightly higher than ADC's measurable (2.5Volts Inner reference). The blue plot is battery voltage, and as shown above it started to fall rapidly and at about 4:33Am because of low power, transmissions stopped. So some kind of short circulit was possible too.

Please note, that measurements could no be very precise, due to project developing stage.
 http://i.imgur.com/pUq7xlt.jpg

I have already destroyed bad INA IC's. But i have few new with (x200 multiply rate) and i will check that again.

Best regards, Nikita.

Thanks for the additional information...what is the output voltage of the solar cell under full illumination?

Voltage on disconnected solar panel is about 5-6 volts. But it's connected to Li-ion , so there is more dependent from battery's voltage instead of solar panel.