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.
Tool/software:
We have a 0.02Ohm current sense resistor in use with the LMP8640-F, which gives a current sense gain of 1V per A.
Second, we have a 0.05Ohm current sense resistor with the LMP8640-T, which also gives a current sense gain of 1V per A.
Now you should theoretically expect the combination of 0.05Ohm and LMP8640-T with only a gain of 20V/V to give a more accurate measurement, Like the other combination, since the larger measuring resistor produces a larger basic signal / voltage at the same current, which then only needs to be amplified even more weakly. Due to the weaker gain, offset error and other errors of the current measurement op would have to have a lower effect and the entire measurement, as just described, should be more accurate, especially for small currents by a factor of about 2.5.
But you can't really observe this in real measurements here.
In the case of the INA197 with 0.02Ohm, i.e. again 1V per A and in the case of the INA196 with 0.05Ohm, the combination of INA196 with 0.05Ohm can be measured about a factor of 2 more precisely than the combination with INA197, which is theoretically also logical.
Is the LMP8640 differently built, which are both good combinations, or why can’t you see any improvement here with the 0.05Ohm and LMP8640-T over the combination with LMP8640-F?
Or would you have to see an improvement in measurement accuracy here too?
Hi,
Your understanding is correct however only in theory. The analysis assumes there is no part variation, and the intrinsic nonidealities are the same between two gain versions (eg between LMP8640-T and LMP8640-F). In reality, none of the assumptions is true.
If you however construct the amplifiers one at a time with the same core components, then you should be able to observe the roughly 2.5x reduction in error, which is mainly caused by input offset.
Regards
Guang