Settling Time

RE: Settling Time

Tim Sobering — Sun, 30 Jun 2013 14:06:42 GMT

Bruce,

I'm glad you mention what can be a significant difference between inverting and non-inverting amplifiers...the presence of a changing common mode voltage at the inputs which can potentially use a significant portion of the CM range. While the non-inverting configuration has the advantage of a higher input impedance, it is not without a cost.

"...leave it to my successors" -- are you retiring?

Tim

RE: Settling Time

Bruce Trump — Mon, 24 Jun 2013 17:55:53 GMT

Bonnie—There would be quite a dramatic difference in settling behavior between gain of G=+1 and G=-3 as the noise gain is 1 versus 4, respectively. The final settling would be considerably slower due to reduction in closed-loop bandwidth.

I believe you may have intended to compare G=+2 and G=-1 because they have equal noise gains of 2. For a given output step size, these two gains will have similar settling behavior through the transition to small signals. In spite of the equal noise gains, there can be significant differences in behavior with a real op amp. The G=+2 case applies a changing common-mode signal to the op amp that can reveal different behaviors in the input stage. In comparison, the common-mode voltage of the inverting amplifier is unchanged before the step and after settling. This effect is on my list of possible topics but I’ll probably leave it to my successors. — Bruce

RE: Settling Time

Bonnie Baker - WEBENCH Design Center — Mon, 24 Jun 2013 06:19:45 GMT

Bruce,

This is a very good article. Thank you for covering this topic. Is the settling time for an amplifier different if you have the amplifier in a positive gain (i.e. +1V/V) versus a negative gain (- 2V/V)?

RE: Settling Time

Hugo Coolens — Sat, 15 Jun 2013 12:50:31 GMT

Thanks for the link and the modelling info

regards,

hugo

RE: Settling Time

Bruce Trump — Wed, 12 Jun 2013 20:23:09 GMT

Hugo-- I hastily prepared a design note to show how I simulated the two op amps as I cannot attach schematics in this comment. I thought others might be interested in the liberties I took in creation of the graphs in this blog. The link is below. -- Regards, Bruce.

e2e.ti.com/.../2446.an-op-amp-with-35-degrees-phase-margin.aspx

RE: Settling Time

Hugo Coolens — Wed, 12 Jun 2013 09:57:10 GMT

Hi Bruce,

Can you show us how you modelled that 35° in your circuit?

regards,

hugo