INA226: Calibration problem with INA226 and INA238

Arkadiusz Mieczkowski

Part Number: INA226

I have the same problem as people mentioned in:

https://leap.tardate.com/electronics101/ina226/highsidemonitorbasics/ 
https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/903395/ina226-calibration-on-shunt-voltage  
https://github.com/RobTillaart/INA226/issues/2
https://www.eevblog.com/forum/beginners/playing-with-arduino-and-ina226-(current-and-power-monitoring)/

I wrote exactly the same values given in datasheet in Table 1. and tested it on laboratory equipment with Raspberry Pi 3B+. Source code from here: https://github.com/e71828/pi_ina226.The same problem with INA238. In any device I have to do correction to get expected values.

over 1 year ago

0 Guang Zhou over 1 year ago

TI__Mastermind 30910 points

Hi,

There could be a number of factors leading to the issue observed. As some of the comments pointed out, component/meter/source tolerance is certainly one of them.

To examine the accuracy of these devices, please measure the stable input (shunt or bus voltage) at the device pins, and then compare with register readings. There will still be some deviation due to device tolerance such as offset and gain error, but the readings should be very close to expected. Both these devices are highly accurate. More often than not, large discrepancy is caused by setup issues.

Regards, Guang

0 Arkadiusz Mieczkowski over 1 year ago in reply to Guang Zhou

Prodigy 10 points

sReading of shunt is much different...

0 Guang Zhou over 1 year ago in reply to Arkadiusz Mieczkowski

TI__Mastermind 30910 points

Hi,

Assuming the DVM shows actual shunt voltage and the screen shot shows what INA226 reports, the difference is huge. Bus current doesn’t look right either, and it is not consistent with shunt voltage and device setting.

Can you print out all register values in addition to voltage/current? This should help identifying the root cause.

Regards, Guang

0 Arkadiusz Mieczkowski over 1 year ago in reply to Guang Zhou

Prodigy 10 points

0 Guang Zhou over 1 year ago in reply to Arkadiusz Mieczkowski

TI__Mastermind 30910 points

Hi,

Thanks for the additional info. Looks like the shunt register reports 30mV while the DVM measured 20mV.

Please test with some other units if you have spares. Sometimes devices can get damaged during handling, the likelihood of multiple devices being out of spec to such extent should be rare.

Regards, Guang

0 Arkadiusz Mieczkowski over 1 year ago in reply to Guang Zhou

Prodigy 10 points

In other unit it is 30 mV with same configuration...

0 Guang Zhou over 1 year ago in reply to Arkadiusz Mieczkowski

TI__Mastermind 30910 points

Hi,

Here are all the paths forward I can see – (1) double check the hardware setup; (2) check the I2C waveform to make sure the register values align with the screen capture; (3) return these units for verification if you still suspect it is a device issue, although I doubt device issue is the root cause.

Regards, Guang

Amplifiers

Amplifiers forum

INA226: Calibration problem with INA226 and INA238