|
Home
/ FAQ
/ News Classifieds / Events |
Audio Asylum Thread Printer |
Get a view of an entire thread on one page |
67.244.254.73
Dear technically-minded audiophiles,
Can you help me learn something about tube circuit topology and its consequences?
It seems to me that the early years of amplifier design are pretty well documented and can be traced through a relatively small number of designs. But after, what, the 1950s? things start to get murky. And by the time you get to today's amplifier's, I've totally lost track (beyond the most basic categories--single-ended, push/pull, etc.).
What I'd really like to be able to do--and cannot come close to right now--is place modern amplifiers into categories based on their basic design--as in, 'this is basically a Williamson push/pull with X modifications'--that sort of thing.
Ideally, I'd like to be able to associate each of the classic designs with a particular, subjective 'sound' so that I could interpret modern designs in that light.
- What are the most important designs?
- What do those designs sound like, relative to each other?
I'd like to keep it limited to tubes just to keep things (relatively) simple.
It's a lot to ask, I realize. But if you can clue me in to some good information resources--sources that concisely point out key design differences and their sonic consequences--I would be grateful.
Thanks much.
Jim
Hi Jim,
You are headed down a treacherous path.
The problems are two-fold. First and most important, there is not necessarily a strong link between circuit topology and sound quality. In tube amps especially, I would say that the power supply has *at least* as much influence on the sound as the actual circuit or even the parts quality.
For an example of this, try and dig up some information of the "Paoli mod" for the ARC SP-3a. The original preamp had an emitter follower (capacitance multiplier) to regulate the main rail. It was a zero-feedback design. Then a company out of Paoli, Pennsylvania offered a complete PCB that replaced the original regulator. It was a high-feedback design using a floating 3-pin regulator. The bass got *way* tighter and there was a lot more definition. (Remember, the slogan on the front panel is "High Definition".)
But after living with the thing for a month or two, out it came. Too fatiguing and non-musical.
The second problem is that there are simply too many variations to make a simple taxonomy that will cover them. But there are a couple of things you can do. One is go to a good research library and find the MIT series on tube designs written during the war. I believe Hillman was the principle author of the compilation and it featured the first known example of what came to be known as a "cascode" configuration. This was re-discovered in the '80s and used by a great many companies that are now out of business. It's even used by a few companies that are still in business.
The other thing you can do is download a bunch of the Audio Research schematics. There is an un-official website that has info on almost all of their products, including schematics of a great number of them. They were instrumental in bringing balanced circuitry to tube audio, and all of their power amps have always featured a thing that is sometimes called "partially cathode coupled", but in reality is just a local feedback loop around the output stage (as is the so-called "ultralinear" configuration).
Also, visit Allen Wright's, Vacuum State Electronics websites for some nice modern schematics.
Have fun, but don't expect to get anywhere close to where you thought you were going...
I'm convinced--but to reframe the issue and not to give up.
I'm now seeking, more modestly, a historically informed understanding of modern tube amplifier design.
I'll start with the Valley and Wallman book.
By the way, in searching for that book, I came across this page at Positive Feedback:
http://www.positive-feedback.com/pfbackissues/0603/Sfbiblio.6n3.html
I have no idea how well judged this bibliography is, but it looks useful.
Thanks very much.
Jim
Jim,
Looking at your original request, I do not believe any of these books will be of much help to you.
You are a reviewer, and I suspect your interest is in the relationship between circuit tpology and their resulting sound qualities. None of those books, including Valley & Wallamn will have even one word mentioning this.
In fact, I would say no book written by engineers will ever mention sound quality. They will talk about noise levels, and distortion levels, but never anything about recreating the magnificence of an orchestra in full flight, or in the instrospection/beauty of a solo jazz pianist like Keith Jarret's Koln Koncert.!
For that you have to get down in the dirt with us subjectivist designers - and then there will need to be a lot of confusion and BS filtering to gain much solid information advance.
But you could start with Page 2 of the below linked brochure of my dpa300B power amps. This gives a basic circuit analysis and sonic report on SE , Push pull and differential toplogies - of course from my experience and point of view.
Happy to answer any questions this page brings up for you or others.
Regards, Allen (Vacuum State)
the MIT book Charles mentions is better known as "Valley & Wallman" and it was Prof (?) Wallman that invented the cascode, or at least he was the first to put it in print.
All tube designers, professional or amateur, should/must have a copy of this book. It took my designs from copies of others 60's efforts to the 21st century.
Regards, Allen (Vacuum State)
There seem to be several editions, or at least printings, of this book--the original in '48, then one in '63, and even one in 2001. Are they identical? Any idea? Which is best?
Jim
Thanks Allen. I've ordered a copy. A few are available used on the Internet.
Jim
your message is asking to have the essence of tube amplifier lore summed up in some sort of well-laid-out table of some sort. Well, it doesn't quite work like that, I'm afraid.
The problem is that many companies over the decades have produced variances on a number of common circuit topologies and then embellished them with fat-worded marketing terms. Then along came the generations of tube brands, designs and the whole age-of-enlightenment opened up, so to speak. That is to say, throw in so many component and topology variables and you can't really sum up much of anything, sonically speaking, into a simple table of characteristics.
If so, there wouldn't be much to this hobby, would there?
Cheers,
David
Hey David. Thanks for your note. It seems to me that you said about the same thing I did in my message--that after a certain point (1950s?) we lost track. Yet I suspect that all those variations are still on the same basic set of themes--and that's what I'm trying to get at. What themes are today's variations on?
And, sure, by the time you mix all that up, there may not be much left of the characteristic sound of that circuit--too much else going on. Or, rather, the sonic characteristics smear together and overlap to the point where it's not a useful guide to the sound. I acknowledge that.
But the basic topology is the backbone, it seems to me. I guess my question can be summed up as--can today's tube amp topologies be readily understood in terms of the classic designs? If so, then which basic designs are still important? If, on top of that, I know a bit about what those prototypical designs sound, or sounded, like, then I start to feel like I'm getting a handle on the subject--even if knowing the circuit would NOT allow me to predict much about the sound (and it probably wouldn't).
So that's the basic question: If I look at, say, the top tube amps in Stereophile's RC list, could I classify their topologies with reference to important designs of the past? Or are today's designs new in meaningful ways--beyond the choice of components and a riff here and there?
Cheers,
Jim
Here are 2 aspects of our typical amp circuit topologies that you need to have studied:
- Amplifer class: A, AB, AB1, B, C, D, etc.
- Single Ended and Push-Pull
There are plenty of books on these subjects and even a bunch of stuff online. The proper way to study these, however, is to get some older college text books on circuit design and then analyze a number of the seminal topologies such as the Williamson. Also, the old tube books from Mullard and RCA had a lot of the analysis you're looking for.
As for making something like an Excel spreadsheet to try to simplify the whole thing, I wouldn't want to embark on such a mission to cover amps found today.
What I would suggest is that you take a look at the well respected amps of each generation. For instance, the 60s could be summed up with the HK Citation II, the McIntosh MC275/225, Fisher's 800 receiver and a Scott of similar stature. You'll find many, many commonalities in the circuit topologies but with variations like the Mac Unity Coupled Transformers, or the way feedback is used in the Citation II.
Since tubes died, for consumers in the 70s, you'll have to switch to SS topologies, of course. Still there's a similar trend found within SS gear where lower parts count drove the use of direct coupled amplifiers and then on into Integrated Circuits.
After the 70s, most tube amps didn't do much of any profound topology changes and just reissued amps with the classic topologies - despite what their marketing departments insisted had been done.
Cheers,
David
Thanks David -
> > After the 70s, most tube amps didn't do much of any profound topology
changes and just reissued amps with the classic topologies - despite what
their marketing departments insisted had been done. < <
That's what I assumed, more or less, and that, really, is where I want to get to. I want to be able to identify the classic topology behind a modern design. That's about it.
Jim
Disagree with doodlebug's assessment completely. For 20 years, from 1950 to 1970, the only two power amp topologies used were the Williamson (and slight variations) and the McIntosh. The Mac has yet to be copied, even though the patents have long since expired.
By using extra windings in the output transformer they were able to get enough local feedback to achieve unity gain. (May as well as used cathode followers instead!). Then they had to get extra gain to run the output stage. So they added an extra driver stage that had positive feedback ("bootstrapping"). In my book, the whole thing was kind of a big mess requiring a horrendously complex circuit *and* output transformer (which is probably why nobody else has touched it with a ten foot pole), but it worked, measured great, sounded pretty damned good, and was the foundation for a company that has truly and justifiably become a legend.
It was in the '70s and '80s that the real innovations started showing up. Between ARC's fully balanced designs and the resurrection of Valley and Wallman's "cascode" (thanks for the memory jog, Allen), a whole new generation of circuits was born (including the so-called "mu-follower" variation on the cascode).
Go look at a Vacuum State Schematic or any of the classic ARC power amps. No relationship to either Williamson nor McIntosh in sight.
...that the ARC (which one should I look at?) and the Vacuum State amps are, in your opinion, original and important modern (i.e., post-Williamson)topologies? Any others come to mind?
Thanks very much,
Jim
Nothing else springs to mind in terms of companies that were innovative AND are still in business.
You can spend weeks going through the ARC schematics and following their history step-by-step. They are unnecessarily complex in my opinion (at least the earlier ones) but they are certainly original.
Allen Wright's designs are also quite original, although he is not so prolific nor as open as ARC once was. (I don't think they still include the schematics in the owner's manuals as they once did.)
But like I said in the beginning, it's not going to lead you anywhere near where you want to go. It will, however, be entertaining and educational.
The early ARC efforts where quite complex, at least the input/driver stages, using a cathode follower between each and every stage. With a few exceptions, the designs have become far less complicated as the years have gone by.
You are correct ARC no longer includes the schematics, parts lists in the owner's manual. Don't know why. Most likely it saves a few $$$$ on printing and most owners don't care that they are no longer included.
ARC is more than willing to give the schematics, parts list for any product they make. Just give them a ring and it will be in the mail or on the fax machine in a jiffy at no charge!
...is a completely new topology never before used in audio.
The only hint of it ever being used before anywhere was from a friend who said he saw the same concept (but not exactly the same circuit) in some old schematics at a company who built control electronics for atomic power stations in Germany in the 60's!
I guess they needed serious precision as well...
Regards, Allen (Vacuum State)
...published "The VTL Book" which was about tube circuit topologies and their amplifiers.
You may want to contact Luke Manley at VTL to see if he still has any copies.
I'm not sure it's exactly what you're looking for, but it might help.
One of the problems I forsee in your approach is that while the individual designs may have a sonic signature, those will change with tube types, i.e. KT88 vs EL34. And even change again with different brands of tubes.
Post a Followup:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2009, Audio Asylum, all rights reserved ]
[ Go Green ]
[ Take5 ]