The point of the R in the RC network i to ensure that the re of the transistor is not the only AC gain limiting part of the circuit.
I have simulated and build this configuration and it seems to work very well with no stability issues at all.
I have used it with a mosfet output stage with a current limited buffer between to ensure that protection does not overload the TIS/VAS stage and ensure more linear TIS loading.
Using the RC network, some sort of currentlimiter is a must, as the TIS is capable of delivering quite high peak currents.
Does anyone have a PDF of the "Feedback, Sensitivity and Stability of Audio Power Amplifiers"
Thanks
Jens
I can't understand what you are saying here at all. How about a full schematic?
Well,it's up to you of course, but my conclusion in Chapter 8 is that these sort of circuits positively bristle with problems.
This circuit has well-known problems with indeterminate circuit conditions. Remarkably, when I built it, it worked. (APAD6 p225) I hasten to add I do not claim this is repeatable.
Yes please
\\\Jens
Hi to all, didn't mean to tease you, just wanted to show it to Edmond first, to see if it's already known and then decide what to do.
Show it here, in Jan's Linear Audio or try to exploit it.
Not normally this reluctant to show a circuit but the clever CMCL implementation is a friend's idea and it simulates so nicely.
20 kHz distortion in the low PPM at full power, Slew rates of >200 V/us, noise around 1 nV/rtHz.
All so much better than needed that it would not be worth the trouble to complicate the circuit just to hit the numbers.
But they come naturally from a fairly simple circuit, combined with advanced compensation and a few twists.
Feedback stability looks fine and so does VAS current in the face of component tolerances and thermal effects.
But it's "only a simulation", as someone is sure to say, so I will build it first.
More soon!
Best wishes
David
I sincerely hope you did not take my post as a criticism.
On the contrary, I meant it as a courtesy, I did not wish to mention another author if you wanted to write on the subject. But after several days it looked like you weren't interested, so I felt a reference would not steal your thunder.
Best wishes
David
No, no offence taken. I wish I had time to explore fully every topic raised here, especially the compensation discussion, but life is busy.
Thank for these precisions.
On another note since the debate turn around compensations i ll add that in simulations TMC and TPC give about the same results for the reasons explained numerous time, that is, comparable loop gain around the OS but so far no two stage design using any compensation that i know can touch a 4 stages amplifier with 3 nested loops linearity wise, actually only such a design can yield ppm THD class amplifiers up to 20KHz without the need of an error correction a la Bob Cordell wich is indeed yet another gain stage in disguise.
Edit : More trivialy about the acronym dispute , it seems to me that TMC is the most analytical
expression proposed since it refers to the very principe of this compensation.
Last edited:
One more gain stage, i dont count the output power stage as a (voltage) gain stage.
Oestripper s amp is a good start i think, iirc i simulated it and it has good performances but there s no nested loops as it s stabilized thanks to a RC shunt at the output of the input stage as well as energic degeneration for all three stages and a resistive shunt at the output of the third stage.
The principle i m talking about has been proposed and published by Pr Cherry in the early 80s and has been much discussed in this forum but seems that discussions didnt yield practical implementations to my knowledge set apart the one proposed by Cherry himself.
In the link below you have an ETI article from Cherry, page 8 fig5 show the principle for a multistage amplifier, if we restrict the thing to what is drawn there s four gain stages that precede the output power stage with each stage having a nested feedback loop.
http://www.epanorama.net/sff/Audio/Circuits/Pre-Amps/60W NDFL Power Amplifier.pdf
I did simulations of such an amplifier using four stages and to my surprise it did yield very good results in respect of the transistor count and is notably superior to the blameless at high audio frequencies with single digit ppm THD up to 20kHz, i ll post a few graphs later as they re in another PC.
The idea is to reduce distorsion in the audio band but not at frequencies that are significantly above 20KHz, as mentioned by Cherry ultrasonic distorsion is inaudible anyway.
Last edited:
There is three nested loops whose return ratio are marked S tau F and a global feedback loop whose ratio is beta, the shorter loop enclose the output stage with what is called the VAS in a two stages design a la blameless.
I would like to thank everyone who has contributed to this discussion up to now.
However, I don't think anyone has commented so far on Chapter 7 of APAD6, the detailed analysis of VAS distortion, carried out in collaboration with Samuel Groner. I reckon that together we took the subject onto another level.
Do you agree? Any thoughts?
However, I don't think anyone has commented so far on Chapter 7 of APAD6, the detailed analysis of VAS distortion, carried out in collaboration with Samuel Groner. I reckon that together we took the subject onto another level.
Do you agree? Any thoughts?
Somewhat off-topic, but you might like to know that the Second Edition of Small Signal Audio Design has just been published. 50% longer than the 1st edition, no less.
http://www.amazon.com/Small-Signal-Audio-Design-Douglas/dp/0415709733/ref=sr_1_1?s=books&ie=UTF8&qid=1409729285&sr=1-1&keywords=small+signal+audio+design
http://www.amazon.com/Small-Signal-Audio-Design-Douglas/dp/0415709733/ref=sr_1_1?s=books&ie=UTF8&qid=1409729285&sr=1-1&keywords=small+signal+audio+design
Where to start. 3 new chapters on phono preamps, new chapter on headphone amps, much more on discrete opamps, equalisation, interfacing with microcontrollers, more on just about everything, really.
headphones are big - the small portable earbuds, iems, folders obviously sell a set with every DAP, smart phone - and the upgrades are there for good money
manufacturers of custom earmold iem for pro musicians now show at consumer audio events
high end, big circumaurals have had quite a bit of market growth over the past decade too
now there are many "audiophile" headphones for $k - Electrostats, Stax high end were always there but Sennheiser added a new model of dynamics that broke the $k, some other names in traditional studio and consumer headphones have followed seeing the market is there
Planar Orthodynamics have a new life with Neodymium Magnets several manu jumped direct to $k offerings from their recent founding
http://www.headphone.com/collections/over-ear-headphones?sort_by=price-descending
manufacturers of custom earmold iem for pro musicians now show at consumer audio events
high end, big circumaurals have had quite a bit of market growth over the past decade too
now there are many "audiophile" headphones for $k - Electrostats, Stax high end were always there but Sennheiser added a new model of dynamics that broke the $k, some other names in traditional studio and consumer headphones have followed seeing the market is there
Planar Orthodynamics have a new life with Neodymium Magnets several manu jumped direct to $k offerings from their recent founding
http://www.headphone.com/collections/over-ear-headphones?sort_by=price-descending
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.