Rude remarks removed. Please be polite.
Hi Pano,
I wonder whether you have read this post: http://www.diyaudio.com/forums/analogue-source/154210-mpp-204.html#post2392615
let's hope the same won't happen with this thread.
Cheers,
E.
</p>
Hi Edmond, that is incorrect. Set aside not all second order systems are overshooting and exhibit a phase dip (but only the overdamped systems, or otherwise said in the frequency domain, with close pole/zero pairs), the Bode integrals theory shows clearly that for a constant unity loop gain frequency, and while keeping the closed loop system stable, you can't get more loop gain than what Miller compensation (as a first order system) can provide. Or otherwise said again, for a first order system, the gain times bandwidth product is a constant.
Obviously TMC provides more loop gain than Miller. That extra loop gain has no other way to appear other than from a pole-pole-zero distribution. That is, roll the gain steep and late (in frequency) and bring the roll off to 20dB/decade around the unity loop gain frequency, to keep the closed system stable.
What you can say is that TMC, as a second order system, can provide the same loop gain as TPC (also a second order system) while not requiring the system to be overdamped. This property is indeed an advantage of TMC over TPC, and is one reason why I was looking after an analysis on if and how TMC implements pole splitting.
I would love to further discuss these things, they are indeed fascinating, but it's holydays and I need to get a life
Hi YWN,
There is a bit of a second-order TPC-like nature to TMC if you recognize that the voltage gain of the output stage is less than unity. With TPC, the T resistor is connected to ground and it has no signal on it (and no bootstrapping). With TMC and a nominal output stage gain of unity (but allowing for injection of distortion components into it), the T resistor is completely bypassed and you merely have a largely pure transistion of Miller take-off point from VAS to output.
If you have a nominal output stage gain of 0.9, you have partial bootstrapping of the T resistor, and this probably gives rise to some TPC-like behavior - the resistor has 10% of the signal across it and it looks similar to a resistor of 10X the value connected to ground.
Just thinking out loud here.
Cheers,
Bob
YWN,
on the other hand, under dynamic conditions each stage effects the Loop, each stage is effected by the Loop and vital parameters, like Loop gain, bandwidth, the poles and zeros of the transfer function constantly vary with the load. The Loop is more or less characterized by the formula z = z² + c
(L)oops!
on the other hand, under dynamic conditions each stage effects the Loop, each stage is effected by the Loop and vital parameters, like Loop gain, bandwidth, the poles and zeros of the transfer function constantly vary with the load. The Loop is more or less characterized by the formula z = z² + c
(L)oops!
Last edited:
<
Set aside not all second order systems are overshooting and exhibit a phase dip (but only the overdamped systems, or otherwise said in the frequency domain, with close pole/zero pairs),
So overdamped systems are the ones that show overshoots ?...
Yet another completely non sensical sentence..
One has to wonder if it s not pure rethoric...
YWN,
.........
The Loop is more or less characterized by the formula z = z² + c
(L)oops!
My Loops are characterized by the formula z = mc2
Sorry, that would be underdamped. My fault, but then I suspect you understand what I meant. But for you, as for your buddy Edmond, it's just another opportunity to spew some venom.Happy holidays!
Incorrect. TMC behaves as an overdamped (or critically damped) second order system.
</p>
Yes. Actually by varying the TMC resistor, you can get an over- under- or critically damped second order system.
Maybe Ophelia told him this (they know each other).
Yes, we exchanged a couple of emails. You don't want to know what I was told in terms of avoiding this place and in particular certain members. But I don't take anybody's opinion as granted.
I would appreciate if you could stay on topic and make your technical points. I have personally zero interest in yours (and any others) dirty laundry.
Book Supplementary Material Posted
Hi Folks,
Finally I have posted quite a bit of supplemental material for the book on my web site at CordellAudio.com - Home. There will be some more, but this is a pretty big chunk.
The material includes virtually all of the semiconductor models used in the book and a few more. VDMOS models for the MOSFETs are included, both for vertical and lateral MOSFETs. EKV models for the MOSFETs will follow.
Many simulations (about 40) for the tutorial examples in Chapter 19 are included, and are ready to run with model files and plot files included in the simulation directory.
There is also a convenient download link for LTspice and a simple circuit to verify operation of the download. One can be up and running a supplied simulation in about 10 minutes, even without LTspice experience.
There may be a few errors and bugs, so let me know if you have any problems.
Cheers,
Bob
Hi Folks,
Finally I have posted quite a bit of supplemental material for the book on my web site at CordellAudio.com - Home. There will be some more, but this is a pretty big chunk.
The material includes virtually all of the semiconductor models used in the book and a few more. VDMOS models for the MOSFETs are included, both for vertical and lateral MOSFETs. EKV models for the MOSFETs will follow.
Many simulations (about 40) for the tutorial examples in Chapter 19 are included, and are ready to run with model files and plot files included in the simulation directory.
There is also a convenient download link for LTspice and a simple circuit to verify operation of the download. One can be up and running a supplied simulation in about 10 minutes, even without LTspice experience.
There may be a few errors and bugs, so let me know if you have any problems.
Cheers,
Bob
Hi YWN,
Wherever did you get the idea that TMC gives phase linearity which TPC cant provide? This is completely untrue.
TPC gives closed loop phase linearity that is ruler flat within the pass band just as with ordinary single pole miller compensation.
I have demonstrated here
http://www.diyaudio.com/forums/soli...terview-negative-feedback-55.html#post1160809
that TMC is merely TPC applied only to the second and output stages living the input stage to a single pole roll off loop gain response.
Nobody here has persuaded me with hard evidence that limiting TPC to the output stage (TMC) instead of the whole amplifier (TPC) is preferable.
Last edited:
To be precise TMC generates virtually the same total loop gain about the output stage as the equivalent TPC network.
Indeed, TMC applies the TPC loop gain characteristic about the output stage alone while TPC provides the same double-pole loop gain characteristic for the WHOLE amplifier.
TPC is therefore vastly superior to TMC.
Hey syn09,
Wrong! TMC neither behaves as an over-damped system, nor as a critically damped second order system. Apart from minor side effects, TMC behaves as a first order system. BTW, perhaps you don't know it, but a critically damped second order system is not, I repeat is NOT equivalent to a first order system. I also like to remind you that the absence of overshoot does NOT automatically mean that we are dealing here a with critically or under-damped 2nd order system.
Again, totally wrong!
Last edited:
Hi michael, I would not go that far. I think it's fair to call it "a matter of taste"
For audio purposes, the discussion TMC vs. TPC is to me ultimately irrelevant and pointless. As a student, I am actually curious if TMC can be applied in linear ICs, but understanding that needs some in depth theoretical analysis regarding the limitations, side effects, etc... That's what I am looking for, without much success here
People seem to be more concerned about scoring rather than discussing technical matters.</p>
Hi michael, I was talking about the loop gain phase, that always dips for an effective TPC compensation. Correspondingly, the closed loop TPC compensated amp always overshoots.
A correctly implemented TMC does not phase dip and overshoot in the closed loop.
care to choose which of the Yellow traces show a "1st order" loop gain system?
the Vout is ~= output Z due to the AC 1 added to each output Q bias CCS
plotting 1/Vout gives the more familiar looking ~=gain curve presentation of output conductance
(I chose this "odd" approach to ward off arguments over where you should place a simple Middlebrook V test source to properly measure the loop gain - clearly this is a measurement of an external amplifier property and not subject to choice/debate)
the .step statement switches between TPC and TMC and both steps are plotted in yellow
green is the Cdom only output 1/Z
if you want to play you can plot each "step" with V(out)@1, V(out)@2 syntax
its really hard to see why anyone would want to get hung up on the semantics - if you want to call the TMC output step response "almost 1st order" I suppose you could - but I can tell the difference by eyeball in this sim and no ther aspect of TMC can be profitably talked about as a "1st order" system
the Vout is ~= output Z due to the AC 1 added to each output Q bias CCS
plotting 1/Vout gives the more familiar looking ~=gain curve presentation of output conductance
(I chose this "odd" approach to ward off arguments over where you should place a simple Middlebrook V test source to properly measure the loop gain - clearly this is a measurement of an external amplifier property and not subject to choice/debate)
the .step statement switches between TPC and TMC and both steps are plotted in yellow
green is the Cdom only output 1/Z
if you want to play you can plot each "step" with V(out)@1, V(out)@2 syntax
its really hard to see why anyone would want to get hung up on the semantics - if you want to call the TMC output step response "almost 1st order" I suppose you could - but I can tell the difference by eyeball in this sim and no ther aspect of TMC can be profitably talked about as a "1st order" system
Attachments
Last edited: