Never heard of Kleeman, but I've seen a Maybach 62S at the Toronto Auto Show and it's stunning. Costs 0.5M, a true auto Blowtorch, but worth every penny (if you can aford it
).
Last edited:
Anatech, please understand that I did use 2 uH output coils for decades, from 1970-1990, fully understanding the NEED of the 2uH coil at my output.
Do you remember that I was one of the first designers of the complementary differential transistor input stage and most probably, the very first designer of the complementary differential jfet input stage? I also designed with complementary jfet input stages as well over 37 years ago. What is the most popular design today, Doug Self's design? Bob Cordell's? Well, I fit in there, somewhere, even though I only helped to introduce the comp-diff bipolar or fet input stage. Some people find it valuable, even today.
Actually, the use of the output coil is more a function of feedback and circuit stability, not anybodies 'popular' topology. For highest fidelity, it is better to not have a coil than to use one, so long as stability under difficult loads can be maintained.
Do you remember that I was one of the first designers of the complementary differential transistor input stage and most probably, the very first designer of the complementary differential jfet input stage? I also designed with complementary jfet input stages as well over 37 years ago. What is the most popular design today, Doug Self's design? Bob Cordell's? Well, I fit in there, somewhere, even though I only helped to introduce the comp-diff bipolar or fet input stage. Some people find it valuable, even today.
Actually, the use of the output coil is more a function of feedback and circuit stability, not anybodies 'popular' topology. For highest fidelity, it is better to not have a coil than to use one, so long as stability under difficult loads can be maintained.
Don't step in it
Magura
BTW, the offer on carbon is still on John.
JC.
I too have reasonably extensive experience with thye use of carbon in audio construction. I really would take Magura's offer seriously and accept. It will not take very much time - or effort - to evaluate it's potential. (Re read your signature)
Besides the fact that you said one should never use MORE than a 2 uH coil , I see inconsistencies in your opinion(s).
I did scope a DIY amp (my amp) with a piezo speaker array , saw more ringing w/ sqaurewaves (200 watts synthesizer OP) on a scope. This did not destabilize the amp , but the coil partially corrected this. On the stupid commercial chip amp , which just survived 15 years of nightclub duty , I also see the 2uH coil (attachment).
I do not take the "chip amp"(stk-4050V) seriously , it can not hold it's own against modern discretes , but the coil has given it decades of "life"
,even bridged into 4R.Yes , myself and others have run without the coil , it falls into the same "leap of faith" category as running without overcurrent protection. We have amps that will do it , but the circuit might be called upon to drive a load like this ( the piezo array or electrostatics).
On my new triple , I noticed not much of anything with the capacitive load , but , like a "thick old man" , still used a coil. Using an inductance meter on the speaker cable, I saw 1-2 uH to start with , so maybe that is why I could run without the coil. ??
OS
Attachments
Ostripper, you have now shown your own colors. I tried to be polite about using a 2uH coil. Sometimes people use them, I used to, but they are audible with the best equipment. To quibble with me, over my statement is just petty criticism, in an attempt to try to find something wrong with my general comments.
Never heard of Kleeman, but I've seen a Maybach 62S at the Toronto Auto Show and it's stunning. Costs 0.5M, a true auto Blowtorch, but worth every penny (if you can aford it
I was drooling looking at Brabus M-470 in 1999. So what? In 2005 I bought Nissan Armada that outperformed it in terms of specs significant to me, and I paid for it many times less. Armada worth more than every penny, in comparison with that Blowtorch made from a stock Mercedes that was greatly overpriced, right for "NO COMPROMISE" guys.
The point is, John is wrong regarding myself as a competitor. I would never ever compete for NO COMPROMISE customers; they are spoiled and demanding, and don't know what they really want. I prefer professionals who knows what for and which gear they are buying. They don't need advertisements on a shiny paper; you may regard absence of advertising on a shiny paper as a compromise to cut costs down.
Hi John,
I do recall your claim to inventing the complimentary diff pair, and I am extremely familiar with how it works. I was wondering when you figured that circuit arrangement out. Industry has a habit of inventing and utilizing useful circuit arrangements.
The complimentary diff pair may work extremely well, if and only if (IFF) all the NPN and PNP (or N channel and P channel) transistors are closely matched together. I know this from working on several examples of this design from several different brands of audio components. Matched by hand, they very audibly out perform the exact same circuit the way factories generally put them out the door. In other words, the performance of a single differential pair exceeds that of a complimentary diff pair in normal manufacturing. This means that using a complimentary diff pair is not a good idea for any line of amplifiers with higher production numbers than custom, low volume brands. It simply takes far too much care (=time + labour) to put out high numbers using the complimentary diff pair.
I think one of the earlier attractions of the complimentary diff pair was that it enabled the designer to run the input stages at far higher current levels, more easily than could be done with a single differential pair. Of course, by making the DC resistance seen by each base equal, the DC offset issue would not be an issue as long as the single pair was matched. Of course, the single pair was far easier to match and they even came out with matched pairs in a single package (like 2SA798 for an inexpensive solution). Once the J-FET came into widespread use, the problems with current vs. DC offset pretty much evaporated, again as long as they were matched (uPA68H as an example).
I still see the complimentary diff pair design as a liability rather than a plus for most manufacturers.
Other interesting input stages are the old single, which I really don't care for. It's an ancient design as far as a time line is concerned. They already had tube differential pairs, so why did the single transistor input stage appear? Later on, a more interesting input stage appeared using complimentary devices - one of each. These could be BJTs or J-FETs. Rotel had one example, and there must be more. These are sufficiently weird enough to throw most technicians off.
The last class of input stage that is gaining in popularity are the op amp types. For manufacturing, these probably represent the easiest way to get very good performance at low cost (compared to the matching process). Then I guess we must include the amp-on-a-chip, or chip amps. I have never heard one I like, although the mean is improving. Compare these to transistor amplifiers of the 60s and 70s. For chuckles and grins, I restored an old, old unit. It was popular in the early 70s. Let me just say that it sounded truly horrid. It's only good point was that it didn't have enough power to hurt itself. Looking at the low end of audio, it's amazing how far the bottom end has advanced in both quality and reliability.
Wow, I didn't think I'd have that much to say on the types of input stages!
Now, as far as cars are concerned, give me a Corvette any day over a 928 or whatever Porshe you come up with. They are also very good on gas as long as you don't floor it. For me, a BMW 750 or 745 sounds really nice. I enjoy driving those. Still, for a day to day driver, nothing wrong with a 328 or similar either. It's just that the fuel economy isn't the greatest. On the low end, it's hard to beat a KIA for the money. I used to joke that KIA stood for Killed In Action, until mine was (Sedona company vehicle). It did save my life and it still was drive-able afterwards even though it was a write-off. The best car I had ever owned was a '92 Buick Roadmaster wagon, 5.7 L. Many toys and wonderful fuel economy, I was actually getting close to 30 mpg (Canadian gallons).
I think if I could choose any car to run around in, I'd probably go with a BMW 750 type. I've got nothing to prove anymore. They are very comfortable and handle well. They are also easy on passengers, and this is important if anyone is still married.
John, I've seen that you've used GM quality as an insult type comparison. You're too kind. If you really want to insult someone, make a comparison to a Ford. That'll bite down hard! When AMC was around, that was a worthy low point. Just to warn you, I come from a car family. I get to drive almost anything I want to try. Not recently as I can't even make it down to my brother's lots anymore. Too far to drive for me. Suffice to say that I have driven most cars out there. Even a Jensen Interceptor (years ago), from a LeCar (terrible thing), original rustang and Hyundai Pony and on up to 'Vettes, Caddilacs, Ford (what's a four letter word beginning with "F") on u to BMW, Mercedes, SAAB (really nice also), Jaguar (very unreliable, and "mushy") - on and on, including full trucks (straights - one chassis, box on rear rated as 5 or 10 ton normally). Most Porshe models are decidedly uncomfortable. I can't figure out why anyone in California wou7ld own one given all the restrictions. The Bentley I can understand - almost. Too expensive to insure. It does a great job at making a statement though, and they can be pretty. You couldn't go to a mall in it, too much parking lot damage.
Any further comments on the complementary differential pair? What was the driving force, the issue you were trying to solve when you designed this circuit John? The audio landscape was pretty grim back then as I recall.
-Chris
I do recall your claim to inventing the complimentary diff pair, and I am extremely familiar with how it works. I was wondering when you figured that circuit arrangement out. Industry has a habit of inventing and utilizing useful circuit arrangements.
So, we both understand and agree that we both know the whys and whats of the output inductor and it's relationship to the feedback network used. That's comforting then. I must admit that I have not followed your career so closely as to what designs you favored and when. Note that I didn't say you had never designed this way, just that it is not your current style.
Being ignorant of each designer's preference, I can only reply to what I see "out there". The most popular input stage is the single differential pair, and for a good reason.
The complimentary diff pair may work extremely well, if and only if (IFF) all the NPN and PNP (or N channel and P channel) transistors are closely matched together. I know this from working on several examples of this design from several different brands of audio components. Matched by hand, they very audibly out perform the exact same circuit the way factories generally put them out the door. In other words, the performance of a single differential pair exceeds that of a complimentary diff pair in normal manufacturing. This means that using a complimentary diff pair is not a good idea for any line of amplifiers with higher production numbers than custom, low volume brands. It simply takes far too much care (=time + labour) to put out high numbers using the complimentary diff pair.
I think one of the earlier attractions of the complimentary diff pair was that it enabled the designer to run the input stages at far higher current levels, more easily than could be done with a single differential pair. Of course, by making the DC resistance seen by each base equal, the DC offset issue would not be an issue as long as the single pair was matched. Of course, the single pair was far easier to match and they even came out with matched pairs in a single package (like 2SA798 for an inexpensive solution). Once the J-FET came into widespread use, the problems with current vs. DC offset pretty much evaporated, again as long as they were matched (uPA68H as an example).
I still see the complimentary diff pair design as a liability rather than a plus for most manufacturers.
Other interesting input stages are the old single, which I really don't care for. It's an ancient design as far as a time line is concerned. They already had tube differential pairs, so why did the single transistor input stage appear? Later on, a more interesting input stage appeared using complimentary devices - one of each. These could be BJTs or J-FETs. Rotel had one example, and there must be more. These are sufficiently weird enough to throw most technicians off.
The last class of input stage that is gaining in popularity are the op amp types. For manufacturing, these probably represent the easiest way to get very good performance at low cost (compared to the matching process). Then I guess we must include the amp-on-a-chip, or chip amps. I have never heard one I like, although the mean is improving. Compare these to transistor amplifiers of the 60s and 70s. For chuckles and grins, I restored an old, old unit. It was popular in the early 70s. Let me just say that it sounded truly horrid. It's only good point was that it didn't have enough power to hurt itself. Looking at the low end of audio, it's amazing how far the bottom end has advanced in both quality and reliability.
Wow, I didn't think I'd have that much to say on the types of input stages!
That is already agreed upon but each of us. My guess is that you are talking about a terminology thing now. The basic layout and design of a circuit is what I call topology. This encompasses how the feedback loops are designed as part of that definition. I can't see how you could retain the same topology and change the feedback. That changes the fundamental way the circuit will work. The term "popular" was used to indicate the popularity of global negative feedback, it is the most popular way to implement feedback. Any nested or local loops are simplified to one global network for the sake of simplifying for this discussion, otherwise we could consider all kinds of designs that differ slightly.
That idea is gaining ground and I'll accept it, although I have not been able to prove this is true when looking at the amp as a system. In the world of trade-offs, sometimes the choice comes down to complete performance and that may favor a good old global negative feedback with output choke. It depends on the trade-offs made to take the feedback earlier on in the circuit. Think here of Nelson's Stassis type amplifiers. I am using the Nakamichi version in my bedroom, and one trade-off is that it "likes" higher load impedances.
Read earlier where Morten posted this the first time, or ask nicely if he would tell you what he was thinking back then. All I can say is that it is difficult to machine.
Now, as far as cars are concerned, give me a Corvette any day over a 928 or whatever Porshe you come up with. They are also very good on gas as long as you don't floor it. For me, a BMW 750 or 745 sounds really nice. I enjoy driving those. Still, for a day to day driver, nothing wrong with a 328 or similar either. It's just that the fuel economy isn't the greatest. On the low end, it's hard to beat a KIA for the money. I used to joke that KIA stood for Killed In Action, until mine was (Sedona company vehicle). It did save my life and it still was drive-able afterwards even though it was a write-off. The best car I had ever owned was a '92 Buick Roadmaster wagon, 5.7 L. Many toys and wonderful fuel economy, I was actually getting close to 30 mpg (Canadian gallons).
I think if I could choose any car to run around in, I'd probably go with a BMW 750 type. I've got nothing to prove anymore. They are very comfortable and handle well. They are also easy on passengers, and this is important if anyone is still married.
John, I've seen that you've used GM quality as an insult type comparison. You're too kind. If you really want to insult someone, make a comparison to a Ford. That'll bite down hard! When AMC was around, that was a worthy low point. Just to warn you, I come from a car family. I get to drive almost anything I want to try. Not recently as I can't even make it down to my brother's lots anymore. Too far to drive for me. Suffice to say that I have driven most cars out there. Even a Jensen Interceptor (years ago), from a LeCar (terrible thing), original rustang and Hyundai Pony and on up to 'Vettes, Caddilacs, Ford (what's a four letter word beginning with "F") on u to BMW, Mercedes, SAAB (really nice also), Jaguar (very unreliable, and "mushy") - on and on, including full trucks (straights - one chassis, box on rear rated as 5 or 10 ton normally). Most Porshe models are decidedly uncomfortable. I can't figure out why anyone in California wou7ld own one given all the restrictions. The Bentley I can understand - almost. Too expensive to insure. It does a great job at making a statement though, and they can be pretty. You couldn't go to a mall in it, too much parking lot damage.
Any further comments on the complementary differential pair? What was the driving force, the issue you were trying to solve when you designed this circuit John? The audio landscape was pretty grim back then as I recall.
-Chris
Thank you John; but I think people who design for NO COMPROMISE customers need more luck than myself. I have more freedom.
I am actually a moderate , neither for, nor against anybody. My "color" is white , BTW (all colors , not a racial epitaph ).
The only reason for my previous comment was to clarify the statements presented , not as a personal attack. Since you DID use them , and now your designs have evolved beyond them , they are a valid route to stability.
I have seen designs that do not use them (class A stealth) , but instead use a ferrite bead in the feedback loop of the current amp to ensure stability.
Not mine , maybe Mark Levinson's work ?? Self and Cordell are very much in the "loop" , but the audio world is a much bigger playing field.
OS
Hi Wavebourn, going to Denver to drum up business. I am being sent by one of my business partners. Good place to go, I hear. Never went there, myself, before. We will be showing our new amp, but my new business partner has lots of new stuff to show. How about all digital from phono cartridge to loudspeaker? That company is Behold. It will be interesting to compare it to other stuff.
Last edited:
I had my first complementary differential power amp running 41 years ago in 1968. I brought it to The Burning Amp meeting, last year and it was shown to Wavebourn, Scott Wurcer, and Nelson Pass, if you need any proof.
John Iverson developed it separately, about the same time.
Bell Labs had something, one year earlier, but it didn't perform as an amp or a typical preamp, so don't know how to classify it. I changed to complementary differential jfet designs in 1971-72, and have not used the bipolar version, since.
John Iverson developed it separately, about the same time.
Bell Labs had something, one year earlier, but it didn't perform as an amp or a typical preamp, so don't know how to classify it. I changed to complementary differential jfet designs in 1971-72, and have not used the bipolar version, since.
Last edited:
Good luck John!
I am going to offer something to potential Behringer, Mackie, Bose, Peavey, and similar, customers. There is an enormously huge niche for High-end for professionals, if I offer them affordable gear that sounds much better than a standard mass production.
That I concede , I should of said his "design team". I became interested because his "design team" must have done the genesis stealth. (same style)
The products I have studied are the model N (degree)20 - 53 amps. Ah, hah .. I see the JC- 1-3 in the "fan site" for M - L.
I noticed exactly the same "style" of the levinson N (degree) 27.5 voltage amp stage (7 diode strings / same jfets / same CM 's as in the genesis stealth , so I became interested in all the Levinson designs.
You were associated with this group ??
OS
- Status
- Not open for further replies.