to Mark Johnson:
judging from the first name on the author list I judge that it is fictional and you must be a fabrication of Russian hackers attempting to lead us to a false hope in science to cure the ills of music recording. Furthur evidence of this is your inclusion of the icon of the Evil Empire in an icon of the Great American genius, the cpu.
I love the gun totter blowing up the design with his jingoism.
Cheers
Alan
judging from the first name on the author list I judge that it is fictional and you must be a fabrication of Russian hackers attempting to lead us to a false hope in science to cure the ills of music recording. Furthur evidence of this is your inclusion of the icon of the Evil Empire in an icon of the Great American genius, the cpu.
I love the gun totter blowing up the design with his jingoism.
Cheers
Alan
It seems to me you could make a pretty low distortion output stage if it were
The schematic below shows the general idea. Simply replace Q28 and Q30 by Sziklai pairs and you've got it. (Circles with a "V" inside (e.g. item 43) are ideal voltage sources, to generate the cascode biases).
Feel free to substitute MOSFETs for selected BJTs, if you wish.
_
- Constant Current
- Constant Voltage
- Class A
The schematic below shows the general idea. Simply replace Q28 and Q30 by Sziklai pairs and you've got it. (Circles with a "V" inside (e.g. item 43) are ideal voltage sources, to generate the cascode biases).
Feel free to substitute MOSFETs for selected BJTs, if you wish.
_
Attachments
Here are the 19 IEEE publications authored by Don Draper so far:
IEEE Xplore Search Results
Please observe that many of them were published before the "Mad Men" TV series came out (July 2007).
Last edited:
WTF?
And IIM will make having designed such an
amplifier irrelevant. Except as a doorstop or boat anchor.
And IIM will make having designed such an
amplifier irrelevant. Except as a doorstop or boat anchor.
Exactly!
There was a DIY design for an amp that did something a lot like that, back in the late 80's - early 90's. The active output followers were power FETs, with power bipolars bootstrapping them for constant Vds. The mounting flange on the heatsinks got quite crowded, but as racing engine designers say, "There's no substitute for cubic inches". Low distortion, and since the output current was divided among so many output devices, it was bulletproof. I built one for a friend of mine running massive, hideously capacitive electrostats, and tested if by driving a full-power squarewave into a HUGE 4 uF cap, preceded by an equally huge custom non-inductive 0.25 Ohm power resistor that we christened "Frankenstein's Zobel Network". Basically, a homebrew Bryston for design masochists who 'just gotta do it themselves'.
There was a DIY design for an amp that did something a lot like that, back in the late 80's - early 90's. The active output followers were power FETs, with power bipolars bootstrapping them for constant Vds. The mounting flange on the heatsinks got quite crowded, but as racing engine designers say, "There's no substitute for cubic inches". Low distortion, and since the output current was divided among so many output devices, it was bulletproof. I built one for a friend of mine running massive, hideously capacitive electrostats, and tested if by driving a full-power squarewave into a HUGE 4 uF cap, preceded by an equally huge custom non-inductive 0.25 Ohm power resistor that we christened "Frankenstein's Zobel Network". Basically, a homebrew Bryston for design masochists who 'just gotta do it themselves'.
And IIM will make having designed such an
Maybe this one is not so
and even could play a music.http://www.diyaudio.com/forums/solid-state/260308-gainwire-ngnfb-classb-poweramp-12.html#post4084025
The much-appreciated Sony TA-N7 did the opposite: bipolar outputs bootstrapped with VFETs which were directly driven from the output (iirc 12V Vgs). 3 pairs of each outputting 200W/8R. Still regret having sold it, but my friend needed it to drive his difficult electrostatics.
Sony TA-N7B on thevintageknob.org
Jan
Last edited:
Comments on design philosophies and a cautionary tale from a geezer
That one looks very well designed. If you don't mind, I'll probably even build one.
I wasn't referring to any amp in particular with my comment; rather to the (apparent) expressed blanket attitude that TIM was irrelevant, and IIM so unimportant that its mention wasn't even replied to. In my opinion (and I'm not alone in this) a designer shouldn't ignore any known distortion mechanism, and its well-known architectural causes, at the outset.
From looking at the schematic, it looks like you truly look like you know your stuff. And it wasn't a slam on John, either. His track record speaks for itself. He's brilliant, and if he doesn't mind me characterizing him this way, he's one of those 'intuitive' designers, who thinks outside the box. I've followed his design career for years, and he has designed products for various companies that are well-known and well-respected (the products themselves, as well as the companies). I doubt he would remember having met me - at the time he was
"John Curl" (and still is, just like another legendary designer I respect, who will always be "Nelson Pass"). And I was just some computer designer nobody knew about (and deliberately so - the companies that hired me had strict "no talking about your project" rules).
When I started seriously designing analog, and digital, audio circuitry, I was suddenly walking around the edges of a very small pond, already containing many large fish. John, Nelson, Matti Otala...etc, and other guys swimming their way in - (Jan Didden, Malcolm Hawksford among them) were starting to get attention for their genius and getting hired by audio companies. Competition was fierce for the few positions as 'chief designer' available.
The PC had just started getting broad market penetration, and the CD standard had just started the digital revolution in home audio, and there were serious flaws in the early implementations, which fueled a lot of research into why, which ended up being a good thing.
A lot of people criticize one designer or researcher or another for various reasons. Everybody's got a viewpoint, and design philosophies can change and evolve. That's the point. It's not called a 'learning curve' for no reason. And one of my favorite quotes is "Research is what I'm doing when I don't know what I'm doing." That one is from a physicist whose name escapes me at the moment. Einstein was also one of those "intuitive" guys, and some people love to point out he flunked his first Algebra course. When someone told him that he found relativity hard to understand, he replied "I wouldn't let that concern you. Now that the mathematicians have gotten hold of it, even I find it hard to understand."
That design you pointed to was how many years ago? Having looked at schematic and the scope photos, I don't doubt that it works extremely well, and no doubt has been already built by quite a few people who find it a wonderful-sounding addition to their sound system.
But as I started to say earlier in this post, making blanket statements that certain design criteria or distortion measurements are irrelevant, without backing it up with references or explanations, is, in my opinion, something one shouldn't do. Other people who don't understand why might just accept that as Gospel, and run around repeating it and defending it. And if it turns out to be mistaken, it's hard to undo - and the damage is done.
Let me preempt what some might interject at this juncture, pointing to my criticism of Leach. He designed that amp back in 1973 or '74. Mark Alexander designed his in 1992. It would be MORE surprising if Mark's WAS NOT better-performing. That doesn't mean Leach's amp isn't worth owning, or doesn't 'sound good'. And as far as his being arrogant...well, he was a young, very smart, up-and-coming designer with design ideas to defend in an arena of many others with the same agenda. As I've stated elsewhere, it's pretty much a given that there's going to be more than enough ego to sometimes spill over, and rub people the wrong way. It's obvious I have done so.
But the event that actually prompted my comment, I will never forget.
My roommate, who was taking one of Leach's classes and was really enthusiastic about Leach's amp, suggested I stand outside the door of the class, and he would introduce me. When in the course of this, i mentioned Jim Bongiorno's AmpZilla design, he did nothing but criticize it for its flaws, which were easily compensated for (a bit of a pun there - the amp was what is known as "conditionally stable" and certain loads presented problems for it, but all you had to do was adjust the values in the compensation networks accordingly), while ignoring what a breakthrough it was: This was the first time (in a while, at least) that, in a very competitive era of consumer audio that an amp that powerful and sonically capable had its design, schematics and PCBs offered to the public in a nationally-published electronics 'hobby' magazine, not in a research journal. And Leach was trying to compete with it.
Apparently he found it stressful, and was reacting badly....or maybe he was just having a bad day.
My college roommate, and the other students there, listening with great attention to what started as a technical discussion, but rose quickly to what could only be described as a "tirade" were actually shocked into stunned silence. I was just asking technical questions. My roommate ended up abandoning building Leach's amp, and his speaker designs as well, and started his own design from scratch and bought a pair of Magneplanars. He and I became roommates during our co-op work quarters at NASA as well at school, and I helped him debug his design. Driving the Magneplanars, it had a tendency to oscillate at 75 Megahertz, amplitude-modulated at 10 Mehahertz.
Easily fixable by unsoldering the output FETs, putting ferrite beads on the source leads, adding small series resistors in the gate connection points and soldering them back in. A case of having TOO much gain outside the desired bandwidth of operation, but since the oscillation was occurring locally, in the output stage itself, he was unable to solve it with the compensation techniques he'd learned in Leach's class. Leach's amp, and the AmpZilla, used bipolars in the output stage, and this problem - basically an RF Colpitts oscillator - wasn't one you had to look out for.
It changed my life too. I went to see assigned faculty advisor, who was the dean of the EE department, and asked what I should do. Leach was known to be very influential, so any continued study of analog design was out for me.
I would not find slots open in the better professors' classes, or would get a frosty reception if I did. And Leach's work was garnering attention, for him as well as Tech. And there were other things going on as well, and the dean did not want any unwanted attention drawn to the area, which an official complaint of unprofessional conduct would bring. I considered changing my major to Computer Science, but was talked out of it, and a track in computer hardware design was laid out. I still studied analog design outside my core curriculum, and designed some analog synthesizer circuits. Bob Moog was really doing interesting things with Walter (now Wendy) Carlos at the time and publishing widely (no, he wasn't at Tech). But I wouldn't receive any academic credits for the work. No problem - when I left Tech and went to work for Data General, who later opened a position for me at their Semiconductor Division in Sunnyvale to assist the team designing their new 32 bit minicomputer, synthesis was going digital, and Stanford was at the forefront - so I just did graduate level research there, and consulted on the side in the rising digital audio wave cresting at the time. So, as I've said elsewhere, it all actually worked out better for me.
Maybe I should thank him. Maybe he's got his temper under better control.
It's been over 40 years. Who knows. It would be MORE surprising if he hadn't.
We all mellow with age.
Think I'm insufferable now, at 60?
Just imagine what I was like back then.
I used to tell bedtime stories like this to my niece, in the hopes that hearing about problems I'd run into might give her perspectives from which to view her own. She unfortunately ran into a scenario of 'death by misadventure' late last year. I've done some rather dangerous things myself, almost managing to oops my way out of this life at 21, 23, 26, 36, 37 and most recently, 58. But I successfully retired from doing all that, and if all I have to worry about is verbal shots from guys on forum posts....well, I won't lose any sleep over it. It's circuit design, math and physics that I let keep me awake.
Maybe this one is not so
http://www.diyaudio.com/forums/solid-state/260308-gainwire-ngnfb-classb-poweramp-12.html#post4084025
That one looks very well designed. If you don't mind, I'll probably even build one.
I wasn't referring to any amp in particular with my
comment; rather to the (apparent) expressed blanket attitude that TIM was irrelevant, and IIM so unimportant that its mention wasn't even replied to. In my opinion (and I'm not alone in this) a designer shouldn't ignore any known distortion mechanism, and its well-known architectural causes, at the outset.From looking at the schematic, it looks like you truly look like you know your stuff. And it wasn't a slam on John, either. His track record speaks for itself. He's brilliant, and if he doesn't mind me characterizing him this way, he's one of those 'intuitive' designers, who thinks outside the box. I've followed his design career for years, and he has designed products for various companies that are well-known and well-respected (the products themselves, as well as the companies). I doubt he would remember having met me - at the time he was
"John Curl" (and still is, just like another legendary designer I respect, who will always be "Nelson Pass"). And I was just some computer designer nobody knew about (and deliberately so - the companies that hired me had strict "no talking about your project" rules).
When I started seriously designing analog, and digital, audio circuitry, I was suddenly walking around the edges of a very small pond, already containing many large fish. John, Nelson, Matti Otala...etc, and other guys swimming their way in - (Jan Didden, Malcolm Hawksford among them) were starting to get attention for their genius and getting hired by audio companies. Competition was fierce for the few positions as 'chief designer' available.
The PC had just started getting broad market penetration, and the CD standard had just started the digital revolution in home audio, and there were serious flaws in the early implementations, which fueled a lot of research into why, which ended up being a good thing.
A lot of people criticize one designer or researcher or another for various reasons. Everybody's got a viewpoint, and design philosophies can change and evolve. That's the point. It's not called a 'learning curve' for no reason. And one of my favorite quotes is "Research is what I'm doing when I don't know what I'm doing." That one is from a physicist whose name escapes me at the moment. Einstein was also one of those "intuitive" guys, and some people love to point out he flunked his first Algebra course. When someone told him that he found relativity hard to understand, he replied "I wouldn't let that concern you. Now that the mathematicians have gotten hold of it, even I find it hard to understand."
That design you pointed to was how many years ago? Having looked at schematic and the scope photos, I don't doubt that it works extremely well, and no doubt has been already built by quite a few people who find it a wonderful-sounding addition to their sound system.
But as I started to say earlier in this post, making blanket statements that certain design criteria or distortion measurements are irrelevant, without backing it up with references or explanations, is, in my opinion, something one shouldn't do. Other people who don't understand why might just accept that as Gospel, and run around repeating it and defending it. And if it turns out to be mistaken, it's hard to undo - and the damage is done.
Let me preempt what some might interject at this juncture, pointing to my criticism of Leach. He designed that amp back in 1973 or '74. Mark Alexander designed his in 1992. It would be MORE surprising if Mark's WAS NOT better-performing. That doesn't mean Leach's amp isn't worth owning, or doesn't 'sound good'. And as far as his being arrogant...well, he was a young, very smart, up-and-coming designer with design ideas to defend in an arena of many others with the same agenda. As I've stated elsewhere, it's pretty much a given that there's going to be more than enough ego to sometimes spill over, and rub people the wrong way. It's obvious I have done so.
But the event that actually prompted my comment, I will never forget.
My roommate, who was taking one of Leach's classes and was really enthusiastic about Leach's amp, suggested I stand outside the door of the class, and he would introduce me. When in the course of this, i mentioned Jim Bongiorno's AmpZilla design, he did nothing but criticize it for its flaws, which were easily compensated for (a bit of a pun there - the amp was what is known as "conditionally stable" and certain loads presented problems for it, but all you had to do was adjust the values in the compensation networks accordingly), while ignoring what a breakthrough it was: This was the first time (in a while, at least) that, in a very competitive era of consumer audio that an amp that powerful and sonically capable had its design, schematics and PCBs offered to the public in a nationally-published electronics 'hobby' magazine, not in a research journal. And Leach was trying to compete with it.
Apparently he found it stressful, and was reacting badly....or maybe he was just having a bad day.
My college roommate, and the other students there, listening with great attention to what started as a technical discussion, but rose quickly to what could only be described as a "tirade" were actually shocked into stunned silence. I was just asking technical questions. My roommate ended up abandoning building Leach's amp, and his speaker designs as well, and started his own design from scratch and bought a pair of Magneplanars. He and I became roommates during our co-op work quarters at NASA as well at school, and I helped him debug his design. Driving the Magneplanars, it had a tendency to oscillate at 75 Megahertz, amplitude-modulated at 10 Mehahertz.
Easily fixable by unsoldering the output FETs, putting ferrite beads on the source leads, adding small series resistors in the gate connection points and soldering them back in. A case of having TOO much gain outside the desired bandwidth of operation, but since the oscillation was occurring locally, in the output stage itself, he was unable to solve it with the compensation techniques he'd learned in Leach's class. Leach's amp, and the AmpZilla, used bipolars in the output stage, and this problem - basically an RF Colpitts oscillator - wasn't one you had to look out for.
It changed my life too. I went to see assigned faculty advisor, who was the dean of the EE department, and asked what I should do. Leach was known to be very influential, so any continued study of analog design was out for me.
I would not find slots open in the better professors' classes, or would get a frosty reception if I did. And Leach's work was garnering attention, for him as well as Tech. And there were other things going on as well, and the dean did not want any unwanted attention drawn to the area, which an official complaint of unprofessional conduct would bring. I considered changing my major to Computer Science, but was talked out of it, and a track in computer hardware design was laid out. I still studied analog design outside my core curriculum, and designed some analog synthesizer circuits. Bob Moog was really doing interesting things with Walter (now Wendy) Carlos at the time and publishing widely (no, he wasn't at Tech). But I wouldn't receive any academic credits for the work. No problem - when I left Tech and went to work for Data General, who later opened a position for me at their Semiconductor Division in Sunnyvale to assist the team designing their new 32 bit minicomputer, synthesis was going digital, and Stanford was at the forefront - so I just did graduate level research there, and consulted on the side in the rising digital audio wave cresting at the time. So, as I've said elsewhere, it all actually worked out better for me.
Maybe I should thank him. Maybe he's got his temper under better control.
It's been over 40 years. Who knows. It would be MORE surprising if he hadn't.
We all mellow with age.
Think I'm insufferable now, at 60?
Just imagine what I was like back then.I used to tell bedtime stories like this to my niece, in the hopes that hearing about problems I'd run into might give her perspectives from which to view her own. She unfortunately ran into a scenario of 'death by misadventure' late last year. I've done some rather dangerous things myself, almost managing to oops my way out of this life at 21, 23, 26, 36, 37 and most recently, 58. But I successfully retired from doing all that, and if all I have to worry about is verbal shots from guys on forum posts....well, I won't lose any sleep over it. It's circuit design, math and physics that I let keep me awake.
- Status
- Not open for further replies.