I have been looking around for the Common mode rejection ratio value on the datasheets for comparators but there don't have one. Why is this?

This ratio is crucial in terms of choosing a particular comparator. So why don't the datasheets one, i.e. LM311, LM 393, LM 2903, etc.

If I am understanding this concept wrong, what other specifications in the datasheet would I need to consider in order to determine which comparator would be more suitable?

  • 2
    \$\begingroup\$ How can you define it? Comparator by definition does Vout = infinity * (V+ - V-). So it's pure differential... \$\endgroup\$ – Gregory Kornblum Sep 29 '18 at 20:31
  • \$\begingroup\$ I believe we could search for something like"input offset voltage dependence from common mode voltage" ;) \$\endgroup\$ – carloc Sep 30 '18 at 8:11

The CMRR is limited by your source impedance for each input. e.g. the RLC equiv cct since the comparator is a nonlinear switching device, no IC specs apply. Often hysteresis is added with % positive feedback.

CMRR issues can arise from long unbalanced single ended cables and improved by balanced shielded cables with a CM choke like ethernet ports. Normally the shield is grounded at Rx end only to avoid AC ground loops but may also be terminated to earth ground with 1 to 10nF caps.

An EMC assessment is necessary. (Electromagnetic Compatibility ingress<< egress)

Only Linear aspects of IC's have CMRR specs in the datasheet, not comparators.

  • \$\begingroup\$ Um, it looks like you have a duplicated answer. \$\endgroup\$ – Long Pham Sep 30 '18 at 11:10
  • \$\begingroup\$ Looks like a bug on SE. \$\endgroup\$ – Tony Stewart EE75 Sep 30 '18 at 11:13

A comparator's data sheet (like an opamp) will specify the worst case input offset voltage. That offset voltage might be +/- 5 mV and that means that if your inputs are within 5 mV of each other, you cannot predict whether the comparator output is high or low. That rule defines how a comparator works irrespective of whether both inputs are at several volts negative or several volts positive.

Sure, that input offset voltage might change a tad as the common mode input voltage changes between a negative value to a positive value but, the maximum limits of Vos covers that small change.

Because an opamp has a linear output, the common mode voltage effect will be an unwanted small change in the output level and so it is important for an opamp data sheet to specify CMRR and unimportant for a comparator.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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