• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 496 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

INA236: Current measurement change depending on conversion time

patrice sarrasin
Prodigy 30 points

Part Number: INA236

Tool/software:

Hi,

we are using the INA236 to measure the current of an output. We would like to measure up to 20A, our bus voltage is around 12V.

Our current configuration is the following:

VS = 3.3V

RShunt: 0.0005

ADC range: 20-48mV

CurrentLSB: 20 /2^15 = 0.00061

Continuous conversion mode for the bus and shunt voltage

ShuntCal = 0.00512/(0.00061*0.0005) / 4= 4194

Current measure on the multimeter: 1.32A

Changing the average doesn't seems to have a significant impact on the current value, but changing the conversion time does. As we can see below, the higher the conversion time, the closer we are to the measured value.

Average        Conversion Time        Current Value
1                   1100                            1.77
1                   2116                            1.56
1                   4156                            1.45
1                   8244                            1.398
1024             8244                            1.40
64                 332                              2.4
16                 1100                            1.76

If we change the conversion mode to continuous shunt voltage, the conversion time has no more impact and the read value almost equal the measured value.

Average        Conversion Time        Current Value

1                    4156                           1.34

1                    332                             1.34

64                  332                             1.35

64                  4156                           1.35

Is it because the INA236 need more settling time when changing from the shunt voltage to the bus voltage conversion?

Any way to fix this?

Regards

  • 0
    TI__Mastermind 36640 points

    Hi,

    Are the two sets of data collected from the same board? The two modes of measurement shouldn’t affect accuracy.

    Can you perform a similar experiment but instead of looking at current, compare shunt voltage instead. Use DVM to measure across IN+ and IN- pins as reference. Use INA236 to measure Vshunt in the two modes, and loop a few times to observe if the values are stable.

    Regards

    Guang

  • 0 in reply to Guang Zhou
    Prodigy 30 points

    Hi Guang!

    The two data sets were taken on the same board, same chip, same output, same load...

    I compute the shunt voltage with the following formula:  Register value * 0.000000625 = shunt voltage

    For now, the results seems to follow previous tests, but we can see that the value of IN+ and IN- seems to change depending on the mode. Is it normal?

    Continuous conversion for bus and shunt voltage

    Average        Conversion Time        Register Value          Voltage           measure shunt voltage

    1                    332                             1932                          0.00121          0.0012

    1                    4156                           1172                          0.000732        0.0012                    

    64                  332                             1936                          0.00121          0.0012

    64                  4156                            1173                         0.000733        0.0012

    Continuous conversion for shunt voltage

    Average        Conversion Time        Register Value          Voltage         measure shunt voltage

    1                    332                              1079                          0.00068       0.0006

    1                    4156                            1088                          0.00068       0.0006

    64                  332                              1090                          0.00068       0.0006

    64                  4156                            1087                          0.00068       0.0006

    Regards

  • 0 in reply to patrice sarrasin
    TI__Mastermind 36640 points

    Hi,

    The input pins of INA236 will not cause the input voltage to fluctuate. It has to be external. If input is noisy, you could consider an input filter, which would be a good idea.

    Another experiment is to remove the shunt resistor and apply a known voltage source to the input. This way you remove one uncertainty. Once convinced, then switch back to current load and move forward.

    Regards

    Guang

  • 0 in reply to Guang Zhou
    Prodigy 30 points

    Hi,

    Ok, thanks will make some check on the HW.

    Best regards