Input Capacitance—common-mode?...differential?… huh?

RE: Input Capacitance—common-mode?...differential?… huh?

Jin Pan60 — Sun, 18 Feb 2018 19:00:34 GMT

Some how I feel the definition of common mode and differential input capacitance and resistance of an op amp to be confusing.

In TI's white paper Understanding Operational Amplifier Specifications (sloa011), Cp and Cn are in the position of Ccm of this article on the positive and negative input. The common mode input capacitance is defined as Cic = Cp // Cn; so I can also derive that the differential input capacitance is

Cid = Cd // (Cp in series with Cn).

Similarly for the input resistance, ric = Rp // Rn, rid = Rd // (Rp + Rn) where Rp and Rn are the resistance from each positive and negative input to ground.

It seems these definition are not consistent with the definition in this article. So I am not sure what definition the datasheets are using for common mode and differential input capacitance and resistance.

RE: Input Capacitance—common-mode?...differential?… huh?

Bruce Trump — Mon, 28 Jan 2013 13:04:20 GMT

Pawel--The common-mode capacitance value in the data sheet appears two places--one on each input to ground.-- Bruce

RE: Input Capacitance—common-mode?...differential?… huh?

Pawel Wierzba — Mon, 28 Jan 2013 09:22:15 GMT

I'm affraid I need some clarification regarding Figure 1. As far as I remember the common-mode capacitance is measured between the op-amp inputs connected together and the graound. If this definition is correct, the op-amp in Fig. 1 has 2*C_CM of common-mode capacitance. The question is what is listed in the datasheet, C_CM or 2*C_CM?

RE: Input Capacitance—common-mode?...differential?… huh?

Bruce Trump — Wed, 16 Jan 2013 01:51:45 GMT

You are correct, Ken, 1mV input is unnecessarily small to stay in the small-signal range of an op amp. It's convenient number for simulations but may be too small to conveniently view output on a scope. For lab measurements, anything up to 50mV input step should be okay. I suggest going no higher than 50mV, or so, as some input stages start to lose gain significantly beyond this point. -- Bruce

RE: Input Capacitance—common-mode?...differential?… huh?

Ken Dillinger — Wed, 16 Jan 2013 00:11:16 GMT

This is a good topic as it is a typical 'gotcha' when first designing with op-amps. I am a bit surprised that you are using a 1mV signal to test for ringing and overshoot. It makes sense for a noiseless simulator, but what about a real world lab test? I have always presumed (assumed??) a 25mVp to 100mVp step is ok to avoid slewing and is pretty darn close to 'small signal' analysis used in text books. I would love to see a real world lab test to determine stability by referring %overshoot to phase margin.

-Ken