TMC
Hi Bob,
Investing time and energy in someone who refuses to listen is not my favorite hobby.
Besides, I'm not equipped to deal with such a monumental task. 😉
edit: BTW, did you notice this post?
This is the last thing I have to add to the TMC vs TPC debate.
(TMC: 1.25ppm vs TPC:1.58ppm)
Cheers,
E.
Hi Bob,
Investing time and energy in someone who refuses to listen is not my favorite hobby.
Besides, I'm not equipped to deal with such a monumental task. 😉
edit: BTW, did you notice this post?
This is the last thing I have to add to the TMC vs TPC debate.
(TMC: 1.25ppm vs TPC:1.58ppm)
Cheers,
E.
Last edited:
It depends on the quality of he models. But even if the models are not very accurate, at least a simulation shows in which direction to go. IOW, it's unlikely that a predicted decrease will turn out in an increase in real life.
Hi Andrew,
Simulators are great, especially when models are good, but I wouldn't thread the needle with one.
Cheers,
Bob
NNFB amplifiers
Hello Bob,
Reading one of the last chapters about NNFB amplifier I saw your exemples of IPS stages that can be used. You also show a graphic between such design and a classic LTP. The latter has 10 times more Thd at high input voltages.
I know that GFBA throw less stress on the IPS because only the error signal (between input and feedback input) is applied to the IPS whereas in a NNFB the full input signal is applied.
The comparison you make, I suppose, is with the classic LTP used in an NNFB situation?
Anyhow, would it be usefull to use the more complex NNFB IPS even in complementary mode in conjunction with a GFBA?
Is this not the case at all?
Is this theoretically possible but other problems will arise swamping the gain?
Is this possible but the complexity, cost etc. is not worth at all the quasi insignificant gain.
Or is it just a good thing to try...if one accepts a bit more complexity?
Thnx
Olivier
Hello Bob,
Reading one of the last chapters about NNFB amplifier I saw your exemples of IPS stages that can be used. You also show a graphic between such design and a classic LTP. The latter has 10 times more Thd at high input voltages.
I know that GFBA throw less stress on the IPS because only the error signal (between input and feedback input) is applied to the IPS whereas in a NNFB the full input signal is applied.
The comparison you make, I suppose, is with the classic LTP used in an NNFB situation?
Anyhow, would it be usefull to use the more complex NNFB IPS even in complementary mode in conjunction with a GFBA?
Is this not the case at all?
Is this theoretically possible but other problems will arise swamping the gain?
Is this possible but the complexity, cost etc. is not worth at all the quasi insignificant gain.
Or is it just a good thing to try...if one accepts a bit more complexity?
Thnx
Olivier
Hello Bob,
Reading one of the last chapters about NNFB amplifier I saw your exemples of IPS stages that can be used. You also show a graphic between such design and a classic LTP. The latter has 10 times more Thd at high input voltages.
I know that GFBA throw less stress on the IPS because only the error signal (between input and feedback input) is applied to the IPS whereas in a NNFB the full input signal is applied.
The comparison you make, I suppose, is with the classic LTP used in an NNFB situation?
Anyhow, would it be usefull to use the more complex NNFB IPS even in complementary mode in conjunction with a GFBA?
Is this not the case at all?
Is this theoretically possible but other problems will arise swamping the gain?
Is this possible but the complexity, cost etc. is not worth at all the quasi insignificant gain.
Or is it just a good thing to try...if one accepts a bit more complexity?
Thnx
Olivier
Hi Olivier,
You raise some good questions. I like input stages that provide good dynamic range and large input signal handling capability. I often think that a well-executed NNFB amplifier can be a good exercise because it forces the desgner to do the right thing in many instances, since there is no NFB to clean up open-loop shortcomings. The input stage required for an NNFB amplifier is a good example. Indeed, if early NFB amplifiers had been designed with such input stages, TIM and slew rate limiting issues would have been barely on the radar. We give up a little bit of noise in such input stages, but noise is not as big a deal in audio power amplifiers anyway. I don't see any fundamental reason why these input stages could not be applied to NFB amplifiers.
Cheers,
Bob
Nothing to do with ego.
I just hate to see people delude themselves with a mish mash of misleading anecdotal simulations which completely ignore the underlying theory.
Well at least jcx and a few others got the point.
And another interesting question; what will be different in the 2nd edition?
Cheers
S
Hi Mike,
That is a good question. The simple answer is that it is up to McGraw-Hill.
The first thing that will happen is the second printing. At that time I'll be able to correct spelling errors, perhaps some word or sentence errors, and errors in the drawings. I don't know exactly what the limit is on degree of correction, but I'm guessing anything that alters pagination will be off limits or highly discouraged. I've been keeping a file on all of the feedback that I get on errors and the like, and encourage such feedback.
The timing of the second edition is largely a judgment call on the part of MGH and me. When enough of the subject matter has changed, or when it is desirable to add additional subject matter to keep the book fresh and competitive, then the book may be ripe for a second edition. I've heard that a rule of thumb is that about 25% of the text is revised or new in a new edition. I'm new to this, so don't take it as gospel. I'm guessing the second edition would come out 2-4 years after the first. Doug Self is on his Fifth edition, so if one divides the total age of the book at the time of the 5th edition by 4 one might arrive at an average interval between editions.
On the other hand there are books whose second edition is much longer in coming. One example I can think of is the late Randy Slone's amplifier book. I think his book was out quite a few years and, sadly, his second edition was unfinished at the time of his passing.
It is also unclear by how much the size of the book can grow with new editions.
I'm keeping a file on input from others and ideas that I have for revisions and additions to the second edition, and I continue to encourage such input. Some of the discussions on this thread have been extremely helpful in pointing to areas that are candidates for revision. Another obvious area that may deserve updating and expansion is the section on class D amplifiers. I've even been asked by some to add some material on tube amplifiers. As you can imagine, a second edition is a substantial undertaking.
Cheers,
Bob
I would discourage adding anything on tube amplifiers; this is the 21st centuary after all.
I would recommend leaving such to specialist texts already in existence.
I'll wait for the second edition.
Maybe only tube as preamp and hybrid design. Tube with output transformers are too Steampunk.
Hi Stinius,
Well, I would hope fewer errors, better explanations, coverage of things I may have overlooked, and updates that reflect technology advances.
Let me know what you would like to see 🙂.
Cheers,
Bob
You mean you didn't get the first edition? Think of all you will have missed over the next few years 🙂.
Cheers,
Bob
See my tube amp article in the latest Linear Audio. I love the "warmth" of the sound of those big fat KT88's.
Cheers,
Bob
"VAS"
Bob,
One small detail for the second edition:
The second stage of a power amp. is not a "VAS" (voltage amplifier stage) because this suggests it takes a voltage as its input and delivers a voltage as its output.
Infact, the second stage is a TIS (transimpedance stage) because it takes a current as its input and delivers a voltage at its output.
Bob,
One small detail for the second edition:
The second stage of a power amp. is not a "VAS" (voltage amplifier stage) because this suggests it takes a voltage as its input and delivers a voltage as its output.
Infact, the second stage is a TIS (transimpedance stage) because it takes a current as its input and delivers a voltage at its output.
Always pondered that ??
I suppose the first and second stages together would truly be a "voltage amplification stage".
Looking at what comes out of the first stage seems to be a voltage (.1-1v+p-p) ,resulting in the much larger voltage swing that the output drivers use.
With a BJT amp , the whole circuit is current controlled from input to output . From the LTP to the "VAS" 😕 this current just happens to result in a large increase in AC voltage. I would guess a FET based circuit would be the only true voltage stage.
OS
I suppose the first and second stages together would truly be a "voltage amplification stage".
Looking at what comes out of the first stage seems to be a voltage (.1-1v+p-p) ,resulting in the much larger voltage swing that the output drivers use.
With a BJT amp , the whole circuit is current controlled from input to output . From the LTP to the "VAS" 😕 this current just happens to result in a large increase in AC voltage. I would guess a FET based circuit would be the only true voltage stage.
OS
Hi Mike,
I think you are technically correct, at least for the middle stage of an amplifier that employs conventional Miller compensation. There may be other arrangements that might not fit neatly into the transimpedance description.
Technicalities aside, VAS is the accepted term for the middle stage, and it would be confusing to deviate from that. Sometimes we have to be a little flexible in our semantics as long as most reasonable people understand what we are saying.
Cheers,
Bob