1
\$\begingroup\$

I'm working on an application to measure current in a buck converter using a current sense amplifier. According to this video from Texas instriment, the equation for common mode error is same as eqn 1 below, Where Vcm_pds is the common mode voltage used when during device testing and Vcm_sys is the actual common mode voltage in the application.The formula just doesn't look right. Looks to me like all you have to do to eliminate the common mode error is to make Vcm_pds=Vcm_sys ecmrr(%)=0 . It cant be that simple to eliminate common mode error can it?enter image description here

\$\endgroup\$

1 Answer 1

Reset to default
1
\$\begingroup\$

The common mode error appears as a delta in the offset voltage of the amplifier so even when you eliminate the common mode component of the error some offset will remain.

In short if you avoid any changes in the common mode voltage and match the design CM voltage, there will be no error contributed by the changes in the common mode voltage. There will still be other error sources.

In general, in some, not all, cases, the common mode voltage of the amplifier does change significantly so we need to consider the CMRR.

For example, if you have a high-side shunt and you are measuring the voltage across the shunt, when the output voltage changes you will have a change in common mode voltage.

\$\endgroup\$
2
  • \$\begingroup\$ My confusion is, say for example the Vcm during testing the device cmrr (Vcm_pds) is 12V and the Vcm in my circuit is 6V, the e_Vcm at 6V will be much less than if my circuit had a Vcm of 0V ie ABS(6-12) will give less error than ABS(0-12). Seems counter intuitive that a 0Vcm would yield greater error that 6Vcm. Is that the case? \$\endgroup\$
    – J Leo
    Commented Apr 17, 2017 at 7:57
  • 1
    \$\begingroup\$ Yes, it's the case. The further you are from the test condition CM voltage, the larger the magnitude of the error (due to CMRR) is, potentially. Example from AD8676.. more typically for a dual supply op-amp the offset (say +/-100uV maximum) is specified with Vcm = 0 and you have a CMRR of min (say) -105dB for -3 < Vcm < 3, so offset will be +/-100uV at Vcm = 0 and (worst case) +/- 117uV for -3 < Vcm < 3V. \$\endgroup\$
    – Spehro 'speff' Pefhany
    Commented Apr 17, 2017 at 9:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.