INA226 detecting wrong shunt voltge

Hi Vivek,

What is the raw register value that INA226 reports back in both cases? Do you see large variation when multiple measurements are taken? If the measurements are stable and repeatable, have you measured the differential input at the INA226 to see if it is what you expect and further if it correlates with what INA226 reports back?

Regards, Guang

I measured differential input at INA226 it is for -12v 0.1mV and for -24v it is 0.2 mV using 24 ohm load. And Raw values received in first case is 112 and second it is 218. not seen large variations when multiple measurements taken

Hi Vivek,

The measurements are too much off. Do you mind posting relevant schematic here?

Regards, Guang

Hello Guang,

Thank you for response and I am very sorry for giving wrong values previously as they are obtained correct connections.

But after debugging that issue and correcting read write into INA226 I took  following reading on INA 226

Config Reg value : 0x9FF

Rshunt = 0.0002 ohms so Maximum expected current will be 409.6

Current_LSB = 0.0125

Calibration reg = 2048

please verify these setting also attached schematic for this design

test condition 1 : 12v and 24 ohm load expecting 500 mA  current through Rshunt and 0.1mV drop across it

Actual reading from meter are 482 mA  0.122mV drop across Rshunt

Raw values read from INA226  

Shunt reg = 0x38

Bus voltage reg = 0x981 (3.041 refer schematic for this value)Current Sensor.pdf

current reg = 0x38

final current value according to this will be 0x38 * Current_LSB(0.0125) = 0.7 A  which is not required current value.

test condition 2 : 24v and 24 ohm load expecting 1A  current through Rshunt and 0.2mV drop across it

Actual reading from meter are 915 mA  0.213mV drop across Rshunt

Raw values read from INA226  

Shunt reg = 0x6C

Bus voltage reg = 0x123E (5.81375 refer schematic for this value)

current reg = 0x6C

final current value according to this will be 0x6C * Current_LSB(0.0125) = 1.35 A  which is not required current value.

So please help me with this process to get correct values from sensor.

Thanks and Regards 

Vivek Borse

Hi Vivek,

Not a problem. Can you please label on the schematic (hand drawing is OK):

(1)   How the 12V (or the 24V) source is connected

(2)   How the 24Ohm load is connected

(3)   How the GND_ISO is connected and what is the relationship with GND_INA

Based on the metered shunt voltage, the INA226 reported current values are somewhat close within about 10%, it is better, although not great.

Regards, Guang

Hello Guang Zhou,

please find in attachments connection diagram of my current sense setup

GND_ISO and GND_INA are isolated 

Hi Vivek,

Looks like the attachments didn’t come through, can you try again.

Regards, Guang

this is the wiring diagram

Hi Vivek,

How is GND_ISO connected? I understand you said it is not connected to GND_INA, but is it connected to earth ground or left floating? How is the INA226 supply generated? Where is it referenced to?

I think you should get correct readings if you leave GND_ISO floating. Since you didn’t say, I assume it is, but then the VBUS measurement doesn’t make sense at all. It should be almost 0V, instead of several volts.

I strongly encourage you make some hand measurements with respect to GND_INA, including on the IN+ and VBUS pins.

Regards, Guang

Hello Guang Zhou,

Did you look my schematic and my wiring diagram then you can see that I generated INA226 supply using Z1 and R5. which power INA226 and ISO1541.

I am connecting GND_ISO connecting to earth (wiring diagram) ,if it is not connected to earth ground then INA226 is not power up not responds to I2C communication.

I did measure with respect to GND_INA on IN+  and VBUS when GND_ISO is floating no voltage is observed on these points.

When I connected GND_ISO to earth with respect GND_INA and Vbus is 3v and GND_ISO and IN+ is 0.106mV .

Hi Vivek,

You said before that the measurements are repeatable, but please be sure to take a few more and see what the variation is. Based on your Feb 9^th post, the measurements are not too far off. Remember that INA226 measures a voltage; it has not idea of a current. You’ll have to use the hand-measured shunt voltage as the bench mark, not the current meter reading. 

Also you can try increasing the load current to see if the accuracy improves. Generally speaking it should.

Regards, Guang

Guang Zhou

I mentioned in that post the shunt voltages measured with meter for both conditions which are not matching with INA226 readings.

for test condition 1 : shunt voltage is  0.122mV using voltmeter and reading from INA226 of shunt register is  0x38 

for test condition 1 : shunt voltage is  0.213mV using voltmeter and reading from INA226 of shunt register is  0xC6 

Hello Vivek,

It's been several days since your last post, so I assume your issue has been resolved. I will close this thread for now. If you need more assistance, you can reply to re-open the thread or start a new one with a new question.

Best regards,

Ian Williams
Applications Manager
Current & Magnetic Sensing