Hi Jan,
I may have not been on the same page as you. Having reread you post now it makes sense.
Cheers,
View attachment 358648
It is not desirable stability wise to have the local loop
encompassing three devices , stability will be also buffer
dependent.
Rather , why not keeping the cdom at its usual place ,
compensating the cascode and then by pass the buffer
at high frequencies such that we then have a classical
configuration with no buffer acting , the input stage current
being hefty it will supercede the buffer wich will be somewhat
relegated to provide the DC biaising current.
It is not desirable stability wise to have the local loop
encompassing three devices , stability will be also buffer
dependent.
Rather , why not keeping the cdom at its usual place ,
compensating the cascode and then by pass the buffer
at high frequencies such that we then have a classical
configuration with no buffer acting , the input stage current
being hefty it will supercede the buffer wich will be somewhat
relegated to provide the DC biaising current.
Attachments
Last edited:
I know this; I said exactly the same thing a while ago in this thread. So, what makes you think the circuit I posted has the Miller cap connected across three transistor?
Here is the post again:
http://www.diyaudio.com/forums/soli...lls-power-amplifier-book-390.html#post3552738
The Miller resistor makes the TIS local Zout more resistive rather than capacitive, so it acts like a phase lead, in addition to the zero. Which effect is critical depends on the circuit, but I have not seen many cases where a Miller resistor was helpful. The Rm, Xm, and complex current gain of the TIS all modify the effect of this resistor in their own special way so I wouldn't expect anyone to blurt out the whole idea in casual conversation, or to explain it with one simulation.
This is a great idea and I had already come up with the same basic schematic in my head after reading the notice Email. Nice to see some successful problem-solving here. Shall simulate...
This is a great idea and I had already come up with the same basic schematic in my head after reading the notice Email. Nice to see some successful problem-solving here. Shall simulate...
I know this; I said exactly the same thing a while ago in this thread. So, what makes you think the circuit I posted has the Miller cap connected across three transistor?
I would see no point in using schematics that have
inherent flaws so i only took your simulation exemple
at face value , literaly , are we expected to know all the
whereabouts of your private sims and schematics.?.
That said , whatever you pretend to have knowledge of ,
the point is to provide a better TIS:VAS stability , and in this
respect my point is not to ask you if you know whatever
god knows what but to propose a technical solution ,
if you are interested in commenting such tech datas , good ,
we can have a real debate , otherwise keep on your monologue ,
i guess that there s people who are far less patient than i am ,
that s why i keep posting in relation to you while others
simply , and wisely , decided to not interfere with erroneous
behaviour characters....
This date back from the 70s and can be found in Teac s
BX300/BX500 integrated amps , the power amp topology
is a fet buffered current mirrored Bjt differential that drive
an enhanced VAS , worth noting that the fets , differential
and current mirrors are all three in single dual devices packages.
Although the principle is just plain simple i must admit that
i did find the idea very elegant , at least simulator wise.
This idea was mooted by Samuel Groner in his commentary to D. Self:
http://www.sg-acoustics.ch/analogue_audio/power_amplifiers/pdf/audio_power_amp_design_comments.pdf
Last edited:
Actually, my post that you quoted had nothing to do with wrapping three transistors in the Miller loop. That was my point, although you now claim "...are we expected to know all thewhereabouts of your private sims and schematics?"
Last edited:
All the schematics I have posted so far are to illustrate specific points. Personally I would never design an amplifier in the fashion of those schematics.
This thread is something else. Good stuff from my point of view. While I don't contribute anything cause its way too advanced for my limited mind. It does, however, force me to re-evaluate my belief system
Thank you to kgrlee for the explanation of the feedback resistor bypass capacitor. It has helped in more than one way. The example asc file has also increased my understanding of LTspice. Never knew LTSpice could do a nyquist plot. Also, sent me back to the drawing board...
I can see that this TIS transistor emitter decoupling capacitor does involve adding a zero but its effect is totally dependant on the topology surrounding it. Don't ask me for any scientific reasoning for this "conclusion".
Thank you, all, for making this thread what it is.
Thank you to kgrlee for the explanation of the feedback resistor bypass capacitor. It has helped in more than one way. The example asc file has also increased my understanding of LTspice. Never knew LTSpice could do a nyquist plot. Also, sent me back to the drawing board...
I can see that this TIS transistor emitter decoupling capacitor does involve adding a zero but its effect is totally dependant on the topology surrounding it. Don't ask me for any scientific reasoning for this "conclusion".
Thank you, all, for making this thread what it is.
Well , if it s an exercise of style , why not....
The only one approaching randomly is fig 50 but the local loop ecompass
even the input stage , there s no efficient feedfoward path actualy.
This is well explained (and the cure) in the Lundberg paper ( available on the web) I gave the reference in a previous post ( also in Roberge but out of print)
WRONG argument. You claimed that a two stage miller amp is NOT an OTA.
I showed you that many many many people are use it as an OTA and gave you experts names in the field aswell as reasons why.
Gray and Mayer are calling it an opamp which is of course also correct. They do not say it is not an OTA as you assert.
As a matter of fact, they are never using the concept of OTA but it exists.
By the way, CA3080 is late sixties design as your own reference is explaining
Senior management as you have correctly pointed out is not a team it is people who want to , will ,and have climbed over other to get their job . Stopping those under them at all cost so as not to have them climb over the manager. Manager job is get promoted not make thing work and by all means not by any one under them . To say I agree with your view of Western management is an understatement.I started what I called Quality Meetings in our factory before I knew what Quality Circles were. I did it cos 'senior management' would argue & blame each other for problems to no useful purpose ... like our semantic/pedantic discussions.
Much later, when I first got a book on Quality Circles, I was so shocked that I went to the park and read it from cover to cover.
It described EXACTLY the problems & successes we achieved ... and also the opposition from 'senior management'. These dinosaurs where opposed to my Quality Meetings even when there was clear evidence that they resulted in substantial gains in productivity, quality & morale.
The SPC (which I didn't know was called SPC at the time) was part of this initiative. My Quality Meetings/Circles were teams. Senior Management weren't.
If this sounds like I'm prejudiced against Western Management .. that's cos I am.
Correct.
I did NOT assert that Grey, Hurst, Lewis and Meyer said that the two-stage Miller compensated amplifier is NOT an OTA. I merely stated that they do NOT at any stage call it an OTA, and neither did Solomon in his seminal paper. Now, why do you think that is?
Of course I am aware that Sansen and others call a two-stage Miller compensated amplifier an OTA, but as I've stated I don't believe this to be appropriate because shunt-shunt minor loop feedback about the second stage converts it into a transimpedance stage delivering a voltage output for a current input.
The shunt derived negative feedback courtesy of the Miller capacitor also has the effect of reducing the output impedance of the TIS pro rata with frequency. It is for these reasons that I do not accept that the two-stage Miller compensated amplifier is an OTA, as it does not even come close to an ideal current source at its output.
By the way, CA3080 is late sixties design as your own reference is explaining
This is academic; my point remains valid: the CA3080 and the other reference I provided prove that historically the term OTA did NOT refer to the topology of two-stage Miller compensated amplifier.
Last edited:
I think that paper is really excellent. I read it and took a copy a few years back but it deserves to be re-read so thanks for the reminder.
But I don't think it discusses the effect of the RHP zero on the minor loop?
Only the LHP zeros, as far as I remember. That is why I said Mike's comment was new to me.
Or perhaps I missed it, do you have a reference inside the paper, an equation or picture number?
Best wishes
David
'Positive' zeros usually have a bad effect on the circuit, and their effect on some people's temper is also not positive. So I call them 'RHP' now
Last edited:
Appendix V is the first time the RHP zero is even mentioned.
The zero cancellation resistor is discussed but no comment about the effect on the minor loop.
Anyone else have another reference?
Best wishes
David
Last edited:
If you introduce a resistor with the Miller capacitor, the expression for the zero becomes fz = 1/(2piCmiller(1/gm2-R))
If R > 1/gm2 the positive zero becomes a negative zero and can be used to cancel a pole as it is done in Lundberg's paper to cancel the pole introduced by the capacitive load; This as explained can degrade the stability of the minor loop requiring a // capacitor. a problem in audio if you change load or lenght of cable i believe