If you built this in lab and starting measuring signals, you'd divide what you'd get for a non-common mode signal (i.e., just drive one input with the other grounded) by what you get with both inputs driven from the same source. For increased accuracy, you'd probably increase the input level in the CM case to avoid excessive noise (in that situation, you have adjust the calculation for the increased input, in other words, if you increased the input level by a factor of 100, your CMRR is 100x larger than the raw calculation would indicate).
So... CMRR is made much smaller by introducing a current mirror to the Diff-tail. The Av has dropped to next to nothing. Do we care? We can amplify it back up in the next stage while minimising internal noise/distortion. Would it be fair to say that the purpose of the diff-amp is not to amplify but to 'condition/clean' a signal? Thanks again, these are the videos that I've been building up to.
I think you've got this backward. Using the current mirror has increased the CMRR. That is, there is greater rejection of the (undesired) common mode signal. The gain for differential signals (the desired signal) is not changed.
@@ElectronicswithProfessorFiore Yes, sorry. I meant to say (apologies for a bad question), As CMMR increases, as a result of the current mirror, Av decreases. The way you talk about this makes me think that the diff-amp is really used at the front (incoming) end of a larger circuit as a method of reducing noise/distortions introduced in the circuit itself by means of feedback to the 'other' input. (Confused knowledge in my brain here).
@@simonyoungglostog There are two gains to consider here. The first is the normal signal gain. That doesn't change. The second is the common mode mode. The current mirror reduces that gain. Ideally, you'd like the CM gain to be zero. Yes, the diff amp is used at the front end of a system. The differential configuration reduces noise that is induced into the incoming signal wires (particularly a problem when the signal voltages are very small, as they would be with something like a microphone). Diff amps can also be used with negative feedback, as you pointed out. NF can reduce distortions by sizable amounts but does not impact noise performance.
That's TINA-TI. It's free. You can download it from the Texas Instruments site (ti.com). It's a special version of the TINA simulator they made for TI. Runs on Windows. If you don't have Windows, you can get an online student version of TINA relatively cheap. I've used it. It's pretty nice. I've created a bunch of videos on how to use TINA-TI. They are in the Simulators playlist.
This is by far the most comprehensive explanation of the Differential Amplifier circuit I have found.
EXCELLENT EXPLANATION!
Is only the absolute value of CM gain considered in CMRR?
Yes, it's magnitude only.
How would you go about calculating CMRR from this circuit?
If you built this in lab and starting measuring signals, you'd divide what you'd get for a non-common mode signal (i.e., just drive one input with the other grounded) by what you get with both inputs driven from the same source. For increased accuracy, you'd probably increase the input level in the CM case to avoid excessive noise (in that situation, you have adjust the calculation for the increased input, in other words, if you increased the input level by a factor of 100, your CMRR is 100x larger than the raw calculation would indicate).
So... CMRR is made much smaller by introducing a current mirror to the Diff-tail. The Av has dropped to next to nothing. Do we care? We can amplify it back up in the next stage while minimising internal noise/distortion. Would it be fair to say that the purpose of the diff-amp is not to amplify but to 'condition/clean' a signal?
Thanks again, these are the videos that I've been building up to.
I think you've got this backward. Using the current mirror has increased the CMRR. That is, there is greater rejection of the (undesired) common mode signal. The gain for differential signals (the desired signal) is not changed.
@@ElectronicswithProfessorFiore Yes, sorry. I meant to say (apologies for a bad question), As CMMR increases, as a result of the current mirror, Av decreases. The way you talk about this makes me think that the diff-amp is really used at the front (incoming) end of a larger circuit as a method of reducing noise/distortions introduced in the circuit itself by means of feedback to the 'other' input. (Confused knowledge in my brain here).
@@simonyoungglostog There are two gains to consider here. The first is the normal signal gain. That doesn't change. The second is the common mode mode. The current mirror reduces that gain. Ideally, you'd like the CM gain to be zero. Yes, the diff amp is used at the front end of a system. The differential configuration reduces noise that is induced into the incoming signal wires (particularly a problem when the signal voltages are very small, as they would be with something like a microphone).
Diff amps can also be used with negative feedback, as you pointed out. NF can reduce distortions by sizable amounts but does not impact noise performance.
Prof what simulator are you using? thank you
That's TINA-TI. It's free. You can download it from the Texas Instruments site (ti.com). It's a special version of the TINA simulator they made for TI. Runs on Windows. If you don't have Windows, you can get an online student version of TINA relatively cheap. I've used it. It's pretty nice.
I've created a bunch of videos on how to use TINA-TI. They are in the Simulators playlist.