Question about TMC

[jcx]

The deterioration of phase margin of TMC and TPC are different.

not by enough to easily see - you just have to look at the gain and phase margin by cutting both of the TMC feedback paths around the output stage and compare

its easy to get TPC loop gain measure to lie on top of the TMC curve within a decade of the gain intercept frequency

as I've simmed - pointed to in links to the discussion in other threads


L is avoided as much as possible in precision design as generally inductor's parasitics are usually much more significant than good quality R, C
open solenoids make good mag field pickups too
and ferrite/ferromagnetic hysteresis is often faulted as a particularly bad source of distortion with a annoying dependence on saturation history

 

[jxdking]

Hi, jcx
Thank you for replying. About the simulation on same chart, maybe you can post the link here. I am traveling in China and searching on the mobile phone with VPN is painful(I tried).
What I want to point out is that TMC feedback is some kinds of short cut. It partially skips the input stage, directly feed into VAS. If the simulation is not too simplified. The miller effect on input transistor will take some place, and you will see the difference on bode plot.

 

[jpc2001]

Both TMC and TPC are using RC network to switch feedback path. Did anybody try to use LC instead of RC network? As my first calculation, the required inductor values is about within tens of uH. That is completely do-able!

Nice idea. There are at least two possible variants of this:

1. Replace TMC resistor with an inductor, say 100uH.
2. That, and also replace the TMC capacitor (facing the VAS collector) with a resistor, say 470 or 820.

Spice suggests that both topologies can run stably.

(2) has worse distortion performance, relative to TMC. I guess it's really a nice feature of the TMC capacitor that it almost vanishes from the circuit at audio frequencies, and we lose that by replacing the cap with a resistor. I suspect the resistor is loading the VAS too heavily and contributing to this distortion; not sure.

(1) has equally good distortion performance as plain old TMC. I had to include a small resistor in the network also, to improve stability and avoid ringing. 100 ohms can work, in series with either the inductor or with the TMC cap.