Bob Cordell's Power amplifier book

[YWN]

Originally Posted by wahab

I did provide a valuable explanation to back my sayings.

One more time, the VAS has inherently more gain in a
TMC configuration than in an equivalent TPC one,
due to heavier VAS loading of the latter..

You don t agree with that?...Surprising...
Seems that you didn t catch the curves meaning..

Edit : I repost the sims.

The red curve is the OLG of an uncompensated amp.
Green is OLG of the amp with influence of TMC vas loading.
Blue is the same amp OLG with influence of TPC VAS loading..

These curves talk by themselves, so please, no more "wrong"
as you re only making fool of yourself with such unsubstancied claims...

Hi wahab, interesting curves, could you explain how you got them (Schematic, method, loop breaking point, etc...)? A LF loop gain of about 132dB and a unity loop gain frequency of 40MHz makes me believe there's a possible mis-alignment in the analysis methodology.

I myself don't believe TMC is creating extra loop gain, but rather using the same loop gain available in TPC to (mostly) linearize the output stage and (less) to linearize the VAS. Ingenious, but the end result, from a linearizing perspective, is exactly the same. Actually I think TMC could be worse than TPC, if the VAS and the input stage are incompetently designed.

I am wondering if TMC provides any pole splitting effect; I guess so, but to a lesser degree, because of the lower loop gain available around the VAS. I wish I had a quantitative evaluation of all these...

[wahab]

Originally Posted by YWN

Hi wahab, interesting curves, could you explain how you got them (Schematic, method, loop breaking point, etc...)? A LF loop gain of about 132dB and a unity loop gain frequency of 40MHz makes me believe there's a possible mis-alignment in the analysis methodology.

I myself don't believe TMC is creating extra loop gain, but rather using the same loop gain available in TPC to (mostly) linearize the output stage and (less) to linearize the VAS. Ingenious, but the end result, from a linearizing perspective, is exactly the same. Actually I think TMC could be worse than TPC, if the VAS and the input stage are incompetently designed.

I am wondering if TMC provides any pole splitting effect; I guess so, but to a lesser degree, because of the lower loop gain available around the VAS. I wish I had a quantitative evaluation of all these...

Hi, YWN

True that i was laconic about the sims protocols.

The red curve is obtained by breaking the NFB loop
with a high value inductance and removing the compensation
network, and as, we have the "brut" OLG of the amp.
No wonder in these condition that the unity gain bandwith
extend to 40mhz or so.

Green and blue curves are obtained by connecting the TMC/TPC networks
as they should be, safe that the cap that feed the VAS input
is connected to ground rather than to the vas input.

Of course, in normal operation, seen from the star point,
this cap is connected to a -180° source rather than to earth,
increasing the vas loading to an equivalent doubled capacitance,
but the effect of connecting to earth is little and will not provide
a false advantage when TMC is compared to the same TPC network,
since for the test of this one, the same cap is also connected to earth,
so the eventual errors are only in absolute value and are of the same
order with the two compensation networks, leading to a right relative
comparison.
As you surely noticed, we then obtain the OLG of the amp BEFORE
compensation is implemented but with the VAS loaded by its respective
compensation networks.
As can be seen, the difference between TMC and TPC is due only
to the latters heavier VAS loading.

What remains of the OLG, after the VAS loading is counted can
be used in many way, wether splitted between TMC local NFB
and the amp global NFB, or used almost entirely as global NFB by
implementing TPC, but this latter will forcibly have a slightly lower OLG
to start with , on the order of 15db in the curves i posted,
and at frequencies that are largely in the audio band.

[stinius]

Originally Posted by YWN

Hi wahab, interesting curves, could you explain how you got them (Schematic, method, loop breaking point, etc...)? A LF loop gain of about 132dB and a unity loop gain frequency of 40MHz makes me believe there's a possible mis-alignment in the analysis methodology.

I myself don't believe TMC is creating extra loop gain, but rather using the same loop gain available in TPC to (mostly) linearize the output stage and (less) to linearize the VAS. Ingenious, but the end result, from a linearizing perspective, is exactly the same. Actually I think TMC could be worse than TPC, if the VAS and the input stage are incompetently designed.

I am wondering if TMC provides any pole splitting effect; I guess so, but to a lesser degree, because of the lower loop gain available around the VAS. I wish I had a quantitative evaluation of all these...

Hi YWN

Syn08 has done a very good analysis of TMC and TPC here

Read from post #63

Too bad that he don’t post here anymore, but I see that you are quite up to date YWN so maybe you could ask him if he allows you to post his analysis with the attachments at this forum?

Cheers
S

[wahab]

Originally Posted by stinius

Hi YWN

Syn08 has done a very good analysis of TMC and TPC here

Read from post #63

Too bad that he don’t post here anymore, but I see that you are quite up to date YWN so maybe you could ask him if he allows you to post his analysis with the attachments at this forum?

Cheers
S

A very good analysis, indeed.
Thanks for bringing theses clues.
Though, i don t agree with a little part of his conclusion,
when he states this :

"Bottom line, I think this analysis shows the ultimate TPC and TMC equivalence; TMC has, from a loop gain perspective, absolutely no advantage over TPC; it just a simple and clever way to split the loop gain across the VAS and OPS. The equivalent loop gain is identical"

As discussed in this thread, it appears that they are not
equivalent, in respect of LG ,at least in my understanding..

[Edmond Stuart]

Originally Posted by stinius

Hi YWN

Syn08 has done a very good analysis of TMC and TPC here
............
Cheers
S

Originally Posted by wahab

A very good analysis, indeed.
.............

Hi Stein & Wahab,

Her analysis contains an error, see: Bob Cordell Interview: Negative Feedback , Mystifying TMC.

Cheers,
E.

[michaelkiwanuka]

Originally Posted by Bob Cordell

Hi Mike,

Where is Edmond when we need him to defend TMC? :-)

Anyway, it probably depends on your definition of what consitutes TMC that is equivalent to TPC. Maybe you can tell us and also show some simulation that supports your assertion. My inclination is that it is wrong to say that TMC is pointless relative to TPC.

Cheers,
Bob

Same component values for both is what i mean.

Bob, see below:
Bob Cordell Interview: Negative Feedback

[michaelkiwanuka]

Bob, try the following sims:


Bob Cordell Interview: Negative Feedback

Bob Cordell Interview: Negative Feedback

[michaelkiwanuka]

Originally Posted by wahab

"Bottom line, I think this analysis shows the ultimate TPC and TMC equivalence; TMC has, from a loop gain perspective, absolutely no advantage over TPC; it just a simple and clever way to split the loop gain across the VAS and OPS. The equivalent loop gain is identical"

That conclusion is 100% correct.

[wahab]

Originally Posted by michaelkiwanuka

That conclusion is 100% correct.

It would be correct if the VAS loading of the relevant compensation
networks was not existent or equivalent, but since it exist and is not
equivalent, and is easily measurable, this conclusion is wrong and
you are wrong as well.

I will mention that i used the same components values
for TPC and TMC sims, so that only aggravate your lack of clues
about what is discussed..

Unless trolling is the only thing you re interested in ?...
Who knows...

[YWN]

Originally Posted by wahab

The red curve is obtained by breaking the NFB loop
with a high value inductance and removing the compensation
network, and as, we have the "brut" OLG of the amp.
No wonder in these condition that the unity gain bandwith
extend to 40mhz or so.

Hi wahab, I am still very uncomfortable with 130dB of loop gain and 40MHz of unity loop gain frequency, these values are out of any scale for an audio amp, at least according to the definition of "loop gain" I was taught. Could you please post a schematic with the relevant component values and the loop gain probe in place? Perhaps you are testing the Miller loop only? But even so, 40MHz unity loop gain frequency would be outlandish.