One thing about MJ's book (at least the 2nd ed. I have) is that he seems to gravitate to the 6DJ8 and friends just about wherever possible. It always struck me as a little odd because I would think he'd want to expose the reader to other possibilities. Was this tube in style at the time (late 90s) or is it just a matter of it being his personal favorite tube? I ask here because I seem to recall that one or two folks here seem know him personally.
What brought it up was that I finally found my "Bevois Valley" amp chassis. It never worked due to stability issues, but I know a great deal more now about tubes than I did a decade ago and so I want to fix it. I'm think of converting it over to use 12AT7s instead of 6DJ8s as I've had more experience with the former. In the past year or so that I've been on this forum, I really haven't seen anyone use the 6DJ8 in any new design. Any particular reason for that?
The amp is on a compact chassis meant for a bookshelf and our office needs a real amp, but that is another thread....
What brought it up was that I finally found my "Bevois Valley" amp chassis. It never worked due to stability issues, but I know a great deal more now about tubes than I did a decade ago and so I want to fix it. I'm think of converting it over to use 12AT7s instead of 6DJ8s as I've had more experience with the former. In the past year or so that I've been on this forum, I really haven't seen anyone use the 6DJ8 in any new design. Any particular reason for that?
The amp is on a compact chassis meant for a bookshelf and our office needs a real amp, but that is another thread....
The ECC88 just about fits for every application with good performance. It is also one of my favourite types. I get great results from it in whatever configuration and it's variants are plentiful and fair priced.
This tube was produced en mass in the 1950's for UHF TV tuners and early FM and finds itself ideal for amp front ends.
richy
This tube was produced en mass in the 1950's for UHF TV tuners and early FM and finds itself ideal for amp front ends.
richy
The reason it's popular with MG and commercial designs is obviously the availability. I wouldn't normally use it simply because so many other tubes sound better. 12AT7 is not one of them IMO.
He talks about several other tubes in the 3rd edition (worthwhile to get!). But the ECC88 is indeed very easy to get, performs well, and is pretty inexpensive. An example of a 12AT7 used in a circuit with the same ancestry as the Bevois can be found in my Red Light District article. I haven't heard the two amps side-by-side, but I'd suspect that the differences between them at levels lower than clipping would be pretty minor. The RLD would, I believe, have better clipping and overload recovery behavior.
Since I am about to embark on an amp with a 12AT7 driver, what would you suggest as a similarly parametered alternative that sounds better? My application is a direct input 12AT7 LPT with CCS tail driving PP EL34s.
OK, well obviously that is a matter of personal taste.
As another example of where I am coming from, the overall consensus here and elsewhere is that the 12AU7 is not a good tube to use in a gain stage because of its non-linearity. I built a preamp back in the same era using one...mainly because I had a pile of them but also because they look like a good candidate on paper. I used it in a "mu-follower" config (using MJ's design process) and didn't find the linearity to be all that horrible. Maybe because there is some local feedback in that design?
I also used it as a cathode follower which worked very well, but obviously that has no gain to speak of. My other "Thomas" amp uses one as a phase splitter, but again...no gain.
So yeah...I was just trying to make sure I wasn't missing something. 🙂
I have successfully used E280F and D3a triode wired in a similar application. Obviously two tubes will be needed to replace one 12AT7 which means more space, heating power and expense.
I wouldn't use such a strong term to desribe any tube. After all the ARC SP10 is by many considered the best preamp ARC ever made and it's chock full of ECC88s. And Jadis use 12AU7 in top flight amps. So, yes, it's largely a personal preference.
Last edited:
One thing to keep in mind about Morgan Jones is that he is very opinionated. This shines through quite clearly in his book. For power amps it's 6SN7 and friends, push-pull, global feedback or go home. None of this crappy, inefficient, good-for-nothing, single-ended directly heated triode sh....tuff. That said, I do find that I frequently return to the 3rd edition of his book simply because it's such an easy read and covers some approaches that aren't strictly tube-only (SS CCS, LED bias, etc). The 3rd edition of the book has a fairly comprehensive comparison of a bunch of different tube types with respect to THD, and I find his reasoning behind picking a particular tube over another to be quite reasonable.
Reading Morgan Jones reminds me a lot of reading Douglas Self. With opposite sign, of course (Self hates tubes with a passion).
~Tom
Reading Morgan Jones reminds me a lot of reading Douglas Self. With opposite sign, of course (Self hates tubes with a passion).
~Tom
Most of the applications of 12AU7s has been in RF circuits: oscillator/buffers, frequency multipliers, PP, Class C power drivers, or quasi-digital applications like astable multivibrators, one-shots, SR latches. Not much in audio applications.
12AU7s are said to work especially well in SRPPs and Mu-stages. Active plate loading also just might improve the linearity enough to make these workable in audio designs. This might also apply in LTPs, if the bulk of the distortion is h2, balanced topologies would cancel most of it. Cascode operation is another possibility that just might improve the linearity. (I got outstanding performance from 6BQ7s that have no audio pedigree whatsoever in a cascode LTP.)
Cathode follower operation isn't quite so critical. I considered using 12AU7s as CF grid drivers, but the durn things aren't all that much less expensive than 6SN7s or the nine pin mini version: 6FQ7, so why bother?
Thanks, a 6gk5 was also suggested. I will start with the 12at7 as I have a chassis with too few holes to modify up to two separate triodes, but maybe will play with some of these others. Why the heck not.
I have some E180Fs, but the might not be enough for this application.
Indeed, it shines through in the 2nd ed. pretty well too. At the time, I took a lot of his advise a bit to close to heart and spent a lot of effort finding resistors with non-magnetic caps and so forth to rebuild an amp that (still) has completely inadequate OPTs. The law of diminishing returns became somewhat lost in all the details. Not saying what he talks about with component technology isn't interesting or accurate, but some of those details translate into big bux quickly with little benefit.
Tubelab, this forum, and the passage of time ignoring that other forum helped bring me back from the dark side. 😉
Yes, Morgan does not lack for opinions. 😀 However, I'll point out that his pride and joy, the Crystal Palace, does NOT use global feedback.
The Crystal Palace...really need to pick up that 3rd ed. Not cheap though...if you want both books. 2nd ed. is only one book. 🙂
The book is cheaper than a set of Chinese EL34s.
(OK, Morgan, you owe me a pint of Summer Lightning)
(OK, Morgan, you owe me a pint of Summer Lightning)
Maybe "hate" is too strong of a word. "Strongly dislikes" may be better. In the first chapter of his book, where he outlines his design philosophy and distances himself from the subjectivists, Self states quite clearly that he doesn't see a point of tube circuits in modern audio amplifiers as tube amps produce more distortion than solid state amps. He goes on to further state that IF (emphasis mine) tubes indeed do produce a nice characteristic tube sound, amplifiers should be fitted with a "niceness knob" to allow the user to dial in the desired amount of "niceness" - and have the ability to turn it off.
The rants may be entertaining but they also cause subjects to be rejected merely because the author is dead set against them. It becomes "subject A isn't worth spending time with" rather than "subject A is beyond the scope of this text for these reasons, but if you want to know more here are some references".
I prefer statements of the pros and cons of possible solutions rather than a prescription of which opinions I should have on which solutions.
I'd still recommend Mr. Jones' book. You just have to keep in mind that he's set in his ways and quite opinionated.
~Tom
- Status
- Not open for further replies.