Hello Bob
I'm not sure I can see that many measurements, but I wholly agree there are some, eg Fig 13-2, 13-4, 13-6. It would be helpful if you could point out where some of the others are. I think you have done the book a disservice in not making it absolutely clear what is a measurement and what is a simulation. I try to make it clear in every diagram caption.
I cannot agree. The simulation is one (at least) step away from reality; the approximations in SPICE are well-known. (Early effect being a prime example) Your results do not agree that closely with mine. I think using simulation obscures matters rather than otherwise.
If nothing else, AP screenshots are as close to reality as you can get- they have not even been through .wmf or other vector graphic conversions. I would never replot data because, apart from anything else, I am not confident of my ability to never make a mistake.
We obviously have different styles, different strengths, different weaknesses, and sometimes vastly different opinions. My readers seem to like that.
Simulation and measurement are both very important, and the balance sometimes varies depending on what point you want to make.
I can only say that in the second edition I will do my very best to help my readers in designing and understanding power amplifiers of many different topologies and technologies. That will undoubtedly involve additional measurements, but there is a lot more to it than that.
Cheers,
Bob
Hi Mr. Cordell
I have both books AND I studied both, Your book is one of the BEST book I ever read. Believe me, I studied a lot of books including Grey & Meyers, EM theory by Cheng, Kraus, Griffiths, Jackson, Ulaby, RF circuits by Polzar, Ludwig....Antennas by Balanis, Kraus........... I studied a lot of book to make this comment
Your's is one of the absolute best. I went through a lot of your formulas in the book to verify, It is one of the LEAST mistake book I've seen and this is very refreshing. AND this is only the 1st edition. I actually wrote to Roland Best and William Egan on their mistakes in their Phase Lock Loop test books and, to my big surprise, they actually care enough to reply!!! Best even offer the new manuscript.
I might be quite new in audiophile electronics and have a lot to learn. But I have been around the block enough times. I spent 7 years studying all the books after I retired, 3+hours a day of hard studying. I know books. AND yes, I went through the pain of deriving most of the formulas in the book. If I found inconsistency, you'll know about it like we talking about the current issue of the complementary VAS and offset. I don't say what I said about your book lightly.
I am not going to compare books, I made plenty of comments here about other book and pointed out inconsistencies, error, typos and all in a few threads here.
Last edited:
Bob,
I know many people really like your writing style, it isn't easy writing a book that involves the reader and makes it hard to put the book down. The fact that you don't have pages of scope analysis is fine, we can read that type of thing in most any text book on the subject. Keep your style, just add to what you have already done. I for one would gladly purchase a new edition with expanded sections.
I know many people really like your writing style, it isn't easy writing a book that involves the reader and makes it hard to put the book down. The fact that you don't have pages of scope analysis is fine, we can read that type of thing in most any text book on the subject. Keep your style, just add to what you have already done. I for one would gladly purchase a new edition with expanded sections.
I really like the step by step explanation from p53 to p71. How you start with the basic LTP, explanation why degeneration linearized the LTP, increase the input range, why it can increase slew rate. Then move onto darlington VAS, and CCS load and why it reduces distortion. They cascoding VAS to keep collector voltage of VAS constant to improve linearity. Then the Oliver's optimization and the reason behind it.
You don't need scope picture. We are not having the same circuit or designing the same way. The key is to arm people with the knowledge so they can design on their own.
We don't need a book to show us the solution, we want a book to inspire us to using our imagination with the knowledge armed by the book. I don't need the result bar graphs to show the exact number and jump on the author if it happens to be off a little( just an example, there is no reason for me to think the graph is not all but accurate). It's the concept, the way to approach the design that is valuable.
I actually reference Grey and Meyer all the time when I was reading this book the first time, it is SPOT ON. I read a lot of chapters like 5 times, still get new idea out of it.
You don't need scope picture. We are not having the same circuit or designing the same way. The key is to arm people with the knowledge so they can design on their own.
We don't need a book to show us the solution, we want a book to inspire us to using our imagination with the knowledge armed by the book. I don't need the result bar graphs to show the exact number and jump on the author if it happens to be off a little( just an example, there is no reason for me to think the graph is not all but accurate). It's the concept, the way to approach the design that is valuable.
I actually reference Grey and Meyer all the time when I was reading this book the first time, it is SPOT ON. I read a lot of chapters like 5 times, still get new idea out of it.
Last edited:
Hi Alan,
Thank you for your very, very kind words. This kind of encouragement makes me want to work even harder on the second edition. Thanks for going through my book so thoroughly and I always look forward to helpful feedback.
Cheers,
Bob
Thanks! I'm glad you like my writing style - I probably could not change it if I wanted to -
.One of the biggest challenges was prioritizing what material to fit between a given page limit in a given amount of time. McGraw-Hill was kind to me in both respects: the original page target was 500 and the original publication target was about 6 months earlier.
Cheers,
Bob
Bob,
I know it isn't easy to write this type of book without making it very dry reading, you don't have that problem. I was once asked to write a technical book and that was the comment they made to me after giving them some examples of my writing. I never could find the time to write it, a very different subject. I have given seminars to engineers in my field and the post comments were always just that, I made it a conversation rather than a lecture, That was exactly what the participants liked, I let the engineers ask questions and this made the conversation come alive, it could actually address their questions. I've sat through seminars like that where no questions were allowed until the speech was done, it was like sitting in a boring class room. Don't even try to change your writing style, it is spot on.
I know it isn't easy to write this type of book without making it very dry reading, you don't have that problem. I was once asked to write a technical book and that was the comment they made to me after giving them some examples of my writing. I never could find the time to write it, a very different subject. I have given seminars to engineers in my field and the post comments were always just that, I made it a conversation rather than a lecture, That was exactly what the participants liked, I let the engineers ask questions and this made the conversation come alive, it could actually address their questions. I've sat through seminars like that where no questions were allowed until the speech was done, it was like sitting in a boring class room. Don't even try to change your writing style, it is spot on.
Just curious, how on earth can you compare an EM theory book with a book targeted the DIY audio community?
Anyway, your broad engineering experience and the sheer number books you managed to study are duly noted. I guess everybody here is deeply impressed
.Just curious, how on earth can you compare an EM theory book with a book targeted the DIY audio community?
Anyway, your broad engineering experience and the sheer number books you managed to study are duly noted. I guess everybody here is deeply impressed
I am new in here and I am still learning. For me to back up my post in praising Mr. Cordell, I have to establish that I did read a lot of books in great detail.....It's not like the kids said to the mother "You are the best mother in the world!!
" as they only have one mother!!! I just want to make sure people don't look at me in that way. Particularly I read Self's book and I had enough posts to challenge his book, picked out all the typos in the 6th edition. Typos in a lot of the pages I read in the 6th edition??!!! Assertions that doesn't not make sense. I am just too lazy to dig up the posts as it's been a few months. I stop reading his book after studied over 100 pages. I stopped after reading Chapter 18 on the so called Class XD. That just does not add up. You don't reduce any power dissipation from running as Class A. You still have to pull as much idle current and burn just as much heat. When the bottom half is off, it's nothing more than a NPN EF using a CCS load!!! How is that an improvement? On top, you have to put in extra current sink. A chapter on that? That's when I drop the book cold. I did the calculation at the time, I posted in some thread here a few months ago in more detail. I don't remember that clearly anymore.
I want to note that I don't give praise easy, and I really meant what I said. I don't know Mr. Cordell in person, it's all about the book.
Last edited:
Hi Mr. Cordell
I am just giving a fare assessment of your book. May be because I did design bjt IC, we have something in common, may be the book particularly resonate with me.
I encourage people to reference Grey and Meyer for more details where the formulas come from. Your book is a continuation of that book, explaining in more detail in distortion and things that matter for hifi electronics.
Don't change a bit in what you have in the book, just add. I would like to see you cover the OPS used in Haraga, one that has higher output impedance and how does that affecting the sound. It's a different approach from EF OPS that stress on low output impedance.
I would like to see more in part of the chapter on CFA and how the distortion analysis if possible.
I really think people should use the text book to learn how to design, how is every stage affect the distortion, noise and slew rate. A text book should not try to feed readers a particular design, just follow. Then it becomes a "cook book", not a book to really understand the theory and trayoff. I got spoiled by your book, how clear it explain the circuit. I don't follow your designs in the book, particularly not the values. You arm me with the knowledge to make my own mind, change the circuit to fit what I want. To me, that's the essence of a good text book.
You feed me, I'll go hungry tomorrow, you teach me the tools of survival, I can feed myself.
Just finished my work by the end of Fin. Year deadline so I would like to catch up a few comments.
So do I, and I would appreciate the benefit of your experience.
Pete looks to have done a professional and competent job, and my compliments and respect to him.
But a detailed examination of the linked results shows it's not so simple.
140dB down noise from the soundcard with inputs shorted, excellent.
but only
100dB down for 2nd harmonic distortion from the HP 8903 oscillator alone at 20 kHz.
Sounds plausible for that era, equals .001%. But there's more traps
80dB down hash at about 48 kHz, probably from some internal clock.
and
80dB 20 kHz 2nd harmonic once the interface is introduced i.e. .01%
even worse, only
72dB down for loopback thru the interface. Not sure about the impact of this.
As I wrote, this is no disrespect to Pete Millet, it just shows how hard it is to do low level distortion measurements.
And that is with what looks to be a pretty decent soundcard.
So at least 40dB short of what is needed to accurately measure sub PPM. The data could probably be post processed to improve the results but 40dB is a lot to ask.
I hope I can do this myself, one of these days.
Any comments?
Best wishes to all
David
So do I, and I would appreciate the benefit of your experience.
Do you have a preference or comments on how they compare?
I have considered this as a low cost option. What problems did you find?
Pete looks to have done a professional and competent job, and my compliments and respect to him.
But a detailed examination of the linked results shows it's not so simple.
140dB down noise from the soundcard with inputs shorted, excellent.
but only
100dB down for 2nd harmonic distortion from the HP 8903 oscillator alone at 20 kHz.
Sounds plausible for that era, equals .001%. But there's more traps
80dB down hash at about 48 kHz, probably from some internal clock.
and
80dB 20 kHz 2nd harmonic once the interface is introduced i.e. .01%
even worse, only
72dB down for loopback thru the interface. Not sure about the impact of this.
As I wrote, this is no disrespect to Pete Millet, it just shows how hard it is to do low level distortion measurements.
And that is with what looks to be a pretty decent soundcard.
So at least 40dB short of what is needed to accurately measure sub PPM. The data could probably be post processed to improve the results but 40dB is a lot to ask.
I hope I can do this myself, one of these days.
Any comments?
Best wishes to all
David
Last edited:
Just a postscript to my previous post, as I catch up.
I am aware Bob's DistortionM box would undoubtedly help, it even multiplies the distortion by 40dB, how convenient.
So thanks to Bob for the idea.
Also like to add that in some cases I found that his use of a simple simulation was actually more educational than a real measurement with real but conceptually irrelevant (at that level) clutter.
Obviously need both, I have both books and appreciate the different perspectives.
Best wishes
David
I am aware Bob's DistortionM box would undoubtedly help, it even multiplies the distortion by 40dB, how convenient.
So thanks to Bob for the idea.
Also like to add that in some cases I found that his use of a simple simulation was actually more educational than a real measurement with real but conceptually irrelevant (at that level) clutter.
Obviously need both, I have both books and appreciate the different perspectives.
Best wishes
David
I am new in here and I am still learning. For me to back up my post in praising Mr. Cordell, I have to establish that I did read a lot of books in great detail.....It's not like the kids said to the mother "You are the best mother in the world!!
Particularly I read Self's book and I had enough posts to challenge his book, picked out all the typos in the 6th edition. Typos in a lot of the pages I read in the 6th edition??!!! Assertions that doesn't not make sense. I am just too lazy to dig up the posts as it's been a few months. I stop reading his book after studied over 100 pages. I stopped after reading Chapter 18 on the so called Class XD. That just does not add up. You don't reduce any power dissipation from running as Class A. You still have to pull as much idle current and burn just as much heat. When the bottom half is off, it's nothing more than a NPN EF using a CCS load!!! How is that an improvement? On top, you have to put in extra current sink. A chapter on that? That's when I drop the book cold. I did the calculation at the time, I posted in some thread here a few months ago in more detail. I don't remember that clearly anymore.
I want to note that I don't give praise easy, and I really meant what I said. I don't know Mr. Cordell in person, it's all about the book.
Hi Alan,
Thanks again.
I do largely agree with your comments about "class XD", which Doug did patent. It is largely the same as the old trick of putting a pull-down resistor or current source on the output of an op amp to force it into class A. That works fine for a small-signal situation driving relatively high load impedance, but one can pay dearly for using it to force the class B power output stage into a limited class A region that displaces the crossover away from zero current by a few watts. The result, of course, is an asymmetrical output stage to which many here would likely have a knee-jerk negative reaction to.
Cheers,
Bob
as a patch on a chip you can't otherwise do much about internally then XD may be an option
but the "First Watt" principle should still probably apply - so most of your listening is "in Class A"
even with discrete amps that you can modify Class AB still has nonlinear transitions as you leave the Class A region which Bob treats fairly well
but there is only passing mention of "non switching" output bias - even the LT1166 few pages don't really go deeply into the non-switching concept, ends with casting doubt on its utility with Bob's preferred Mosfet outputs
but the "First Watt" principle should still probably apply - so most of your listening is "in Class A"
even with discrete amps that you can modify Class AB still has nonlinear transitions as you leave the Class A region which Bob treats fairly well
but there is only passing mention of "non switching" output bias - even the LT1166 few pages don't really go deeply into the non-switching concept, ends with casting doubt on its utility with Bob's preferred Mosfet outputs
A patent on this? Must have a good lawyer!!!
There is no advantage moving the crossover away from the center. In my design, I run 1.5A idle so I get 18W of Class A before it crossover to Class AB. This is also moving the crossover away from the center, the amp work in Class A until the drive current exceeds 3A. so the crossover is away from the center!!! difference is I don't have an extra current source that needs a heat sink, more circuit, more heat.
I did the tricks for years using a pull down resistor on opamps to avoid the crossover distortion. BUT those are low current, no consequence. You are talking about few amps here, there is a steep price to pay.
That's like the owner of Mesa Boogie guitar amps. He got all sort of patents and if you look at them, they are laughable. I own patents and I even wrote the last one on my own. I know how you can twist words and have the right "bus word" in the Claims to get a patent.
That was the last straw before I just drop that book. There are other things I don't agree like his definition of Class B and others. But that's the most glaring one. Also, the typos!!!! One word of advice to you when you write the revision.....Make sure you double check your writings to refer to the correct Figures and tables. Nothing is more frustrating than to read the book that refer to a wrong diagram and there are times it's not obvious and can get very confusing. It can happens when you add more diagrams and tables in the revision and forgot to change the wordings.
Last edited:
So, to summarize, you studied books from Gray & Meyer, to EM theory, to microwave antenna design, designed bipolar ICs, broad EE engineering experience, patents owned and written recently, Mr. Cordell must be really proud of getting your attention here.
Given you level of expertize, on behalf of Mr. Self, I'm extending apologies for not being up to your expectations, and deeply disappointing you with his books.
