INA225: Non linearity in gain

Thomas CN1

Part Number: INA225

The INA225 shows non linearity in gain when the current is low. We have set the gain to 100 by design. But seeing the gain as low when the current is lower. Please find the test data below.

R sense = 20milli Ohms

Vout is the INA225 output voltage seen

It can be seen that the gain is only 39 when the sensed current is 241mA and the gain reaches close to 100 only when the sensed current is 624mA. What could be the reason for this behavior?

Test data

Vout (V)	I sense (A)	R sense	Calculated Gain
0.187	0.241	0.02	39
0.378	0.333	0.02	57
0.597	0.437	0.02	68
0.907	0.533	0.02	85
1.233	0.624	0.02	99
1.276	0.651	0.02	98
1.297	0.661	0.02	98
1.292	0.661	0.02	98

Schematic

over 2 years ago

0 Angel Rodriguez de San Miguel Bermudez over 2 years ago

TI__Expert 8130 points

Hello Thomas,

I will take a closer look and get back to you by tomorrow.

Best Regards,

Angel

0 Thomas CN1 over 2 years ago in reply to Angel Rodriguez de San Miguel Bermudez

Intellectual 300 points

Hi Angel,

Is there any further update? Did you get a chance to look into it? Expecting the update soon.

Regards,

Thomas CN

0 Angel Rodriguez de San Miguel Bermudez over 2 years ago in reply to Thomas CN1

TI__Expert 8130 points

Hi Thomas

My first thought was that the error could be caused by input offset since this tends to be the dominating source of error at lower currents. However, at your lowest current shown of .241A, the voltage drop across the shunt resistor is around 4.82mV. At the worst offset voltage scenario of 150uV, it could be causing around 3% of the error seen. Although potentially significant, this is much smaller than the error you are measuring.

It is likely that this large error is being caused by an imbalance at the common-mode input filter caps C55 and C56. Any imbalance in the common mode input low pass filter will degrade common mode rejection. This seems to be causing a significantly larger error at lower currents.

A similar issue is discussed in detail in this E2E thread:

(+) INA280: any issue if the input common mode cap value is mismatched? - Amplifiers forum - Amplifiers - TI E2E support forums

I would suggest replacing with lower cap values and tighter tolerances or removing these capacitors to see if there are improvements in the measurements.

Best Regards,

Angel

0 Thomas CN1 over 2 years ago in reply to Angel Rodriguez de San Miguel Bermudez

Intellectual 300 points

Hi Angel,

Sorry for the late response and Thanks for your previous detailed reply.

However we tried removing those capacitors completely. But there is no change seen in the behavior. We are seeing this behavior on all our boards.

In order to confirm if it is due to any of our design issue, will you be able to try reproducing it on any of the evaluation board at your end and see the results are matching with our observations or not?

Thanks in advance for your extended support and efforts

Regards,

Thomas CN

0 Angel Rodriguez de San Miguel Bermudez over 2 years ago in reply to Thomas CN1

TI__Expert 8130 points

Hi Thomas,

I will procure a board within the next couple of days and will update you on any findings.

Best Regards,

Angel

0 Thomas CN1 over 2 years ago in reply to Angel Rodriguez de San Miguel Bermudez

Intellectual 300 points

Hi Angel,

Can you please update if you got a chance to check the issue at your end?

Regards,

Thomas CN

0 Angel Rodriguez de San Miguel Bermudez over 2 years ago in reply to Thomas CN1

TI__Expert 8130 points

Hi Thomas,

I tested your conditions with our EVM which has the following schematic.

In all of my tests I could confirm a gain of 100V/V at the output for different currents/Vshunt values through Rsense.

It was done with constant DC currents through Rsense to replicate the different Vshunts and no load at the output.

How are you measuring your I sense (A) data and know what the current through Rsense is?

Are you measuring your Vshunt?

Are you measuring simple DC currents or are there transients involved?

Your inquiry seems to be related to this previous E2E post: (+) INA225: wrong gain - Amplifiers forum - Amplifiers - TI E2E support forums, in which case it is likely that you are having the same problem observed there.

There are stability issues due to the device's output impedance, where the device cannot drive the capacitive load at the output.

I ran a TINA simulation where I switch between 0 to 250mA of current through Rsense at a frequency of 2.98kHz.

You can see how the output voltage is unstable for your conditions (R85=100Ω, C58=0.1uF), and I plotted the response for different R85 resistor values

Unstable system due to capacitive load is likely the cause for the behavior you are seeing.

I would suggest increasing R85 to at least 300 ohms or higher (lower the value of C58 if you wanted to keep the same RC constant) at the output, and see if that helps with the stability issues.

Best Regards,

Angel