I respect Self (indeed all of his Selves) but I haven't acquired that edition, and don't recall which one I have in storage (it didn't make the cut for books in the apartment, although his Small Signal Audio Design did), but Deboo is good for eliminating an extra cap---albeit entailing closely matched resistors. Cordell's suggestion of the use of a standard inverting integrator followed by an inverter loses the precision resistor requirement. But perhaps as important, it sidesteps the common-mode distortion attendant on designs like the differential integrator and other designs (including Deboo) with input common-mode swing.
Now one can make some very high CMR input stages, and of course there are some voltage-feedback op amp ICs with exemplary performance thus. Not a lot of JFET ones afaik.
As I haven't read his discounting of capacitor mismatch issues in the differential integrator that "everyone" uses, I can't critique it. I can say though that a 5% worst-case cap mismatch results in a very different response to a step excitation in the diff integrator, and a very different frequency domain appearance. If you care about it you can measure the caps and adjust the input resistors for identical time constants.
Another alternative: a noninverting current conveyor output-loaded with a capacitor and buffered. Unfortunately integrated CCs are scarce afaik.
Last edited:
Are we supposed to make our choices under the authority of Douglas Self ?
Or to think by ourselves ?
Last edited:
I don't think Doug wants anyone to take him as written. He wants people to think and innovate. He just gives his views. He does hate subjectivists. I found a 1984 article by Lipshitz which has some comments about the (then new) war on subjectivism.
Christophe, over the lasr few years, D. Self had acquired a tremendous following in the DIY crowd. One of the reasons is that far too few people have even heard of Bob Cordell's book. On a personal level, I much prefer Bob Cordell's work. The reason why D. Self has become so popular is that in Europe there is a wide spread belief that the Brits are THE experts on audio in general and the fact that D. Self is reputed to have taken part in the design of Cambridge Audio gear.
Looking at it like that, personally I am underwhelmed by Cambridge Audio audio gear in general, although of course some models came out better than others, just like with everybody else's lineups. And while I do not deny obvious knowledge D. Self has shown in his book, I happen to have different opinions. I do not completely agree with Bob Cordell either, mostly regarding NFB, but I feel his book is easily the best one I have ever seen on the topic which was written for common people rather than hard core engineers..
Nevertheless, D. Self has acquired a status of a demi god among the DIY crowd, even if sratching below the surface shows that many don't really understand all of it, they think him infallible. That's all there is to it.
Last edited:
And that differs from building your own electronic how ? Just asking . That syndrome manifests it self in so many interesting way does it not ?
Exactly! Same thing.
At the time a friend and I were developing my speakers, we agreed not to trut our own selves because it was already too personal for us, so we got quite a lot of people into several panels to judge for themselves. People from audiophiles to professors on te academy of music, from all walks of life. Only when they said it was a great loudspeaker did it become a true, real good speaker for us as well.
Not as good as the graph shown for that JBL model, which as far as I could see was doing like 30-20.000 Hz +/- 0.75 dB or do, a hell of an achievement I can only salute in silence, we got it 40-18.000 Hz +/- 1.5 dB, but then we didn't have even 10% of what JBL had in developing those and I would wager ours don't cost nearly as much as that JBL, even if I don't know its price.
Stan Curtis was the man behind Cambridge audio. Even primary grade google skills would have shown you that. Doug has been actively publishing DIY articles for over 30 years (almost as long as a Bob and a Mr J.M.Didden) , so at least in UK he is part of the audio furniture. Pre-internet of course DS was mainly known in UK and BC was mainly known in USA. Nothing more nothing less.
Personally for hobbyists its great that there are multiple and detailed analyses of the problem. No one views him as some sort of design deity though, just a good reference point.
Brad, I recall reading a post from you someplace in response to a question from Samuel Groener regarding driving the substrate of a 2sk389. I have tried looking for that post but cannot find it here. But if memory serves correctly, you suggested that we could bootstrap the substrate (and by inference reduce the distortion due to the non -linear input capacitance when the jfet is driven by a hi-z source).
The issue was covered by articles from Walt Jung (http://www.analog.com/media/en/technical-documentation/application-notes/742022599AN232.pdf) and more recently by another contributor to these forums, John Caldwell (http://www.ti.com/lit/an/slyt595/slyt595.pdf).
I tried (what i recall of your suggestion) of connecting the substrate pin to the follower output but was rewarded with gross distortion for my trouble.
I feel that there is something in this but am unsure what it is i'm doing wrong. Any other suggestions ?
The issue was covered by articles from Walt Jung (http://www.analog.com/media/en/technical-documentation/application-notes/742022599AN232.pdf) and more recently by another contributor to these forums, John Caldwell (http://www.ti.com/lit/an/slyt595/slyt595.pdf).
I tried (what i recall of your suggestion) of connecting the substrate pin to the follower output but was rewarded with gross distortion for my trouble.
I feel that there is something in this but am unsure what it is i'm doing wrong. Any other suggestions ?
Bill, I am not even suggesting, let alone stating, that it was D. Self's fault in any way. The fault is in the head of those who misused his book. As yiou surely know, modern kids have trouble reading anything beyond three short sentences, so most in fact never actually read the book cover to cover. So they ask questions answered in the book several pages on. It's us old folks who are used to books that any such book makes any real sense because we usually read it first and dicuss it later. Then we read what other smart folks said about it, like for example Samuel Groner, and we listen to what you here have to say about it.
BUT, the inevitable "but", anyone who writes and publishes a book does it for a reason, he/she wants to be heard and wants his views to be taken up by others as well because they believe in what they have written. Anyone denying that is a liar. I am not theorizing here, I published a book on word processing on PCs in 1990. First edition sold out in less than 2 months, second edition went in 4 months or so, and the local wars prevented the third edition from being printed. And that's easy writing, you simply follow the programs as you go and explain what might be unclear. Self's and Cordell's books are way more complicated to write. After publshing over 200 pieces on PCs in local magazines, I feel qualified to say that.
All book authors love to be accepted, read and quoted. When you are aacepted (as an authority), read and quoted, you can't help getting a bit of water in your ears. It's only human. If not checked, you soon become intolerant of other people's views which differ from yours. You feel your authority is being questioned. Some deal with it better than others. Bob Cordell seems to read postings here, and on a number of occasions he could read that I beg to differ from his views on GNFB and OL bandwith. I differ with a reserve, because whila I like very wide open loops bandwidths and relatively low GNFB factors of around 20 dB or so, Ialso say that achieving them is still no guarantee the amp will sound good, just a better shot at being good. So much is going on in any amp no single or even dual conditions being met will guarantee anything, at best they will make it more PROBABLE the amp might sound good, assuming you got everything else right. Bob never argued the point, I would rememer that. He obviously has his ego under good control, and that's what makes him even more credible. It helps me as a reader to understand and respect his views on technical merit, not on hurt ego.
BUT, the inevitable "but", anyone who writes and publishes a book does it for a reason, he/she wants to be heard and wants his views to be taken up by others as well because they believe in what they have written. Anyone denying that is a liar. I am not theorizing here, I published a book on word processing on PCs in 1990. First edition sold out in less than 2 months, second edition went in 4 months or so, and the local wars prevented the third edition from being printed. And that's easy writing, you simply follow the programs as you go and explain what might be unclear. Self's and Cordell's books are way more complicated to write. After publshing over 200 pieces on PCs in local magazines, I feel qualified to say that.
All book authors love to be accepted, read and quoted. When you are aacepted (as an authority), read and quoted, you can't help getting a bit of water in your ears. It's only human. If not checked, you soon become intolerant of other people's views which differ from yours. You feel your authority is being questioned. Some deal with it better than others. Bob Cordell seems to read postings here, and on a number of occasions he could read that I beg to differ from his views on GNFB and OL bandwith. I differ with a reserve, because whila I like very wide open loops bandwidths and relatively low GNFB factors of around 20 dB or so, Ialso say that achieving them is still no guarantee the amp will sound good, just a better shot at being good. So much is going on in any amp no single or even dual conditions being met will guarantee anything, at best they will make it more PROBABLE the amp might sound good, assuming you got everything else right. Bob never argued the point, I would rememer that. He obviously has his ego under good control, and that's what makes him even more credible. It helps me as a reader to understand and respect his views on technical merit, not on hurt ego.
Fine, I stand corrected. And BTW, I have so much to do I have little little time or inclination to hand out on Google and such like. I make very sure that my PC stays what it was bought for, to be a tool, not a black hole sucking up my time. That I leave to those with more free time on their hands.
BTW, Stan Curtis is well known and needs no introduction. I studied his pure class A 60WPC amp very carefully.
I remain underwhelmed by Cambridge Audio gear. And I was taken by Stan Curtis' work in form of the Italian made Monrio amps. I cannot judge why, whether it's the design or some changes made different in serial models from the basic design. All I know such changes do happen, I have seen a lot of original architectural plans which were way better than what was actually made because of "cost saving" changes made by the builders. On occasion, they managed to completely ruin the original design beyond recognition.
It is not a good position to be a believer and refer to some books of vulgarization instead of trying to understand thinks by ourselves, using a minimal scientific knowledge.
I have not read any of those books, but i had a strange answer from this D.Self, while he was asking for opinions about his future book. I said-him that it would be nice to add a chapter on CFA (my favorite topology). His answer was "i will not", and under my insistence, "you can not ignore the subject ?" his response was: "Yes i can".
For all the other aspect of this personage , i found his manners very arrogant, and many of his position so definitive and with such a lack of nuance, that i don't have a great sympathy. Most of this, i was surprised to not agree at all with some of his definitive statements.
While, everything i found under Cordell signature looks both perfectly in accordance with my own little knowledge and experience.
We have a sentence, in French from a great grammarian: "What is well conceived is stated in a clear way...with easy words;-) Bob Cordell is always clear.
Audio electronic is not complicated. It requires 70% of correct engineering, 10% of experience (about components and schematic evils), 10% of feelings, 10% of personal imagination.
You have not read any of those books, but you don't like how D. Self answered to you. I don't see that as good engineering from your side, I have received some "arrogant" answers from D. Self, but still I have learned many things about audio amps from his WW articles and books.
Don't be arrogant yourself.
Can you link to those posts so we can see the context. CFA is an emotive topic. I am sure the exchange was not quite that blunt, although a precursory search suggest you are very much a fan of cfb and feel it audibly superior? This does appear to have ignited some heated discussions amongst some.
Those of a subjectivist bent might argue with your 20% there. Some feel it is 100% engineering+experience. vive la difference.
About the God Self, i refer to a very long thread, long enough to get an idea of the character ;-)
I don't think i'am arrogant, despite my poor English don't allow-me to always express humor and nuances.
On the contrary, i'm unsure of thinks and don't take as a scientific rule some of my positions, even if they were correlated with many different experiences.
And i love to change of position IE discover new things.... and consider things upside/down in the hope to have new ideas.
ex: http://www.diyaudio.com/forums/loun...torch-preamplifier-part-ii-5.html#post4274770
Definitively, i apply to myself a rule that i taught to young electronicians all my life: "Verify everything BY YOURSELF". Analyses (understand), Thesis (experiment), antithesis (other's opinions), Conclusion....Demonstration...
Well, i don't feel as a necessity for me to read those books (they don't existed when i entered in this professional activity) , it is just a pleasure when you find somebody witch analyse things the exact same manner than you: you feel less lonely ;-)
- Status
- Not open for further replies.