Dear Johan,
My yes, very bad indeed. You and I are in totally different stages in the audio world.
Respectfully, here is precisely how I feel.
If a person is into tubes, they should use at least a 94 dB per Watt per meter sensitivity of a speaker system, and much preferably, about a 97 dB or better speaker. In the later case, one can design with the very best sounding amplifying devices ever made, which is a triode, maybe even a directly heated triode.
So, if you wanna be in tubes, be into low powered tube amps, where one can "have it all" if desired.
If you are designing a 90 Watt P-P tube amplifier, it has no resemblence what so ever to my types of preferred tube designs. I need mine to perform in the first one tenth to five tenths of a Watt, where my efficient speakers operate 90% of the time. Truth is, you are lost in audio - lost building amps inappropriate for your speaker load. You should just go out and BUY a solid state amp from Bolder or Spectral, as it will far likely outperform your 90 Watt DIY tube design on your lower efficiency speakers.
One other thing, you seem to miss. A two stage direct coupled tube amp is the best possible design IMHO overall, because it will have a transparency, that NO three stage amp will ever even hope to have, by the virtue of not passing the signal trough an un necessary middle (3rd) stage.
Another point, if you can't see the difference between passing a signal between two stages using a two inch length of Siltech ( silver with a little bit of gold in it, wire ) versus ANY series connected capacitor at any price you wanna name, well, I will have to give up !!!
If you are unfamiliar with the difference between a well made triode tube amp with zero negative feedback loop, and any feedback amplifier, I suggest you explore both designs and learn for yourself what sounds most like music. Myself, and others, already have - for decades now. Ohh, darn, you can't DO that, your inefficient speakers force you to compromise and use ultra non-linear or pentode operation. Sorry, what a shame. You never ever experienced it, or much of what else I was writing about . Sorry indeed.
Like I said in my previous post's early paragraph, " quality not quantity ."
Also, and maybe not relevant, I have over sixty years of practical experience listening to these things, experience I rely upon. I am afraid the two of us could never agree on much of anything. Thanks for your post. 'Tried to reply civily to you as I possibly could. Have fun in your audio trip !!
Jeff Medwin
My yes, very bad indeed. You and I are in totally different stages in the audio world.
Respectfully, here is precisely how I feel.
If a person is into tubes, they should use at least a 94 dB per Watt per meter sensitivity of a speaker system, and much preferably, about a 97 dB or better speaker. In the later case, one can design with the very best sounding amplifying devices ever made, which is a triode, maybe even a directly heated triode.
So, if you wanna be in tubes, be into low powered tube amps, where one can "have it all" if desired.
If you are designing a 90 Watt P-P tube amplifier, it has no resemblence what so ever to my types of preferred tube designs. I need mine to perform in the first one tenth to five tenths of a Watt, where my efficient speakers operate 90% of the time. Truth is, you are lost in audio - lost building amps inappropriate for your speaker load. You should just go out and BUY a solid state amp from Bolder or Spectral, as it will far likely outperform your 90 Watt DIY tube design on your lower efficiency speakers.
One other thing, you seem to miss. A two stage direct coupled tube amp is the best possible design IMHO overall, because it will have a transparency, that NO three stage amp will ever even hope to have, by the virtue of not passing the signal trough an un necessary middle (3rd) stage.
Another point, if you can't see the difference between passing a signal between two stages using a two inch length of Siltech ( silver with a little bit of gold in it, wire ) versus ANY series connected capacitor at any price you wanna name, well, I will have to give up !!!
If you are unfamiliar with the difference between a well made triode tube amp with zero negative feedback loop, and any feedback amplifier, I suggest you explore both designs and learn for yourself what sounds most like music. Myself, and others, already have - for decades now. Ohh, darn, you can't DO that, your inefficient speakers force you to compromise and use ultra non-linear or pentode operation. Sorry, what a shame. You never ever experienced it, or much of what else I was writing about . Sorry indeed.
Like I said in my previous post's early paragraph, " quality not quantity ."
Also, and maybe not relevant, I have over sixty years of practical experience listening to these things, experience I rely upon. I am afraid the two of us could never agree on much of anything. Thanks for your post. 'Tried to reply civily to you as I possibly could. Have fun in your audio trip !!
Jeff Medwin
DC-coupled amps,I like that
Some time ago after seeing a design with even bigger cathode resistors and losses(heat),I said to myself can't that not be done better.
This is what I came up with,for those who like to experiment.I am to old to do that
Mona
Hi Ketje / Mona,
Yes, very good. You have the basic topology, a differential front end as I suggested, direct coupled to the grids of the Finals. KISS rules.
Too old BALONEY, I am 69 !!!
All else in your schematic needs FAR more thought as it needs simplification to make it better and more elegant.
(1) Get rid of 100% of the solid state devices, except for " something" on the common cathode tail of the 12AX7.... one LM334 ?? Tube rectifiers sound nicer. 5V3A or maybe a 5U4GB.
(2) Get rid of EL-84s.
(3) Get rid of feedback loop, global. UGH, UNNECESSARY.
(4) Get rid of any non-triode operation.
(5) Address heater to cathode ratings on the Finals tube, by biasing the Finals' Ef supply.
FORGET about "heat" from large value Cathode resistors. It isn't a problem at all. A NON-ISSUE !! Two Mills MRA-12 resistors will sound great and become bullet proof as Rks. The problem is dissipating large amounts of power to ground through the Finals tube itself, stressing it. !!!
(6) Run Finals conservatively at Golden Ratio plate power dissipation, ( 62% ) reducing mostly current, not VDC across it P-K.
(7) Run 12AX7 at about .0006 A. per triode section, and over 150 VDC Ea.
(8) Apply an elegant one resistor ( Mills MRA-12) shunt regulator to each 12AX7 tube, also called a draw down resistor by some old timers, so Finals swinging current does NOT possibly upset the front end B+.
All this I have HEARD and bits and pieces of it have been done over the years, by different advanced designers. No ones "put it all together yet" as I have above, and applied to an ACRO 20-20, or 20As, as I have offered to do .
Jeff Medwin
Hmm,
Jeff I probably philosophically fall into your camp, I have very high efficiency speakers (>100dB) and a pair of big SE GM70 power amplifiers, but..
Johan has probably been in this game as long as you, and simply has chosen a different path, it works for him, maybe not for you, or me..
Ketje, again has been around the block, and has lots of useful insights..
I think there are a number of paths to a goal.. I am not a fan of big PP amplifiers, but I have heard some awfully good ones lately.. A 94dB speaker system is a lot more tube friendly than some of the systems I have encountered over the years.
These are amongst the nicest and most tolerant people you are going to find on this forum. I found your posts responding to them a little less than respectful even if I find myself in agreement with many of your comments.
Jeff I probably philosophically fall into your camp, I have very high efficiency speakers (>100dB) and a pair of big SE GM70 power amplifiers, but..
Johan has probably been in this game as long as you, and simply has chosen a different path, it works for him, maybe not for you, or me..
Ketje, again has been around the block, and has lots of useful insights..
I think there are a number of paths to a goal.. I am not a fan of big PP amplifiers, but I have heard some awfully good ones lately.. A 94dB speaker system is a lot more tube friendly than some of the systems I have encountered over the years.
These are amongst the nicest and most tolerant people you are going to find on this forum. I found your posts responding to them a little less than respectful even if I find myself in agreement with many of your comments.
Hello Kevin,
I like your work and I made friends with Dave Davenport two years ago at RMAF. Dave is a sweet person.
I agree, I was very rough on Johan, but only because he was seemingly on opposite sides of the field as I am in audio. I don't sell anything in audio to make my living, so I just speak my piece, and people can take it or leave it.
I disagree with you. I do not think there are many ways to Nirvana using TUBES. I think high efficiency RULES the roost if you do tube amps, and that such a speaker is mandatory. Otherwise, if you have the resources, buy a new or second hand Spectral. Or spend $500,000 on a MBL Omni and their amps, it wipes out a whole lotta audio stuff I hear at shows.
I could sense from what and HOW they said it, that BOTH these guys are top notch, so I offer each of them an apology if I was a bit too direct and offensive. Thats "me", no real harm intended.
So Kevin, what the heck you doing with a clunky GM70 tube on 100+ dB speakers? ( LMAO, Just pushing you).
I would like to leave a tiny legacy amongst the existing ACRO 20-20 and 20A owners, to gratis give them a worked-out fully documented modification, using ALL the things I have found to be of value, so as to improve the circuit and the amp's long term reliability.
I have studied the DC amp over the last year or so, and simply await someone to send me a clunker or even their working amp for a re-do. Put a LOT of good thought, and good choices into what to do, and what not to do.
I have just eBay sold a 1970's DC P-P amp, made by Art Loesch himself, of what JC Morrison wrote up in the FI Primer as a 6CK4 DC P-P amplifier. It used 6AH4GTs as Finals, and it sounded good on my 97 dB EVs, but not equal to a SE DC two stage amp I normally refer to. An extra 3 or 6 dB of speaker efficiency can "tell you" a LOT about how well a tube amp is sounding, and how to design same.
Hey Kevin, I am going to an ALTEC A-7 system in 2014, so I TOO will have 100 dB plus. Got the 515Bs on my kitchen table, and the 802s and horns on the counter top, awaiting copious doses of my love and money. Run a 2 stage DC amp on them, chokes all under 7 Ohms DCR. and under two pounds each.
Happy New Year. Nice meeting you on line.
Jeff Medwin
Hello Sy,
I have happened to personally own both of those British speakers, and heard them with numerous high quality tube amps. They were not long term satisfactory for me, YMMV, fine !! I don't know what "great success" really means to you .
I will stick with my original statement.
I sold the speakers you named after messing with them, but my hands will stay on my two stage DC amp, until it turns cold and they have to pry them away from my non breathing body. 100 dB efficient speakers and really good two stage SE DC amps are, I find, THE amp-speaker match made in heaven.
Hi Jeff,
Yes, you have certainly made YOUR point, not even deigning shedding hints of disrespect/refined disdain left and right for those daring to disagree with you. The former I can respect, the latter .....
I will try to be as succinct without the personal slants of what one person thinks/imagines the other knows. In my posts here I have only mentioned matters sustained by science and subsequently experienced by scores of folks from the days of Williamson, through Walker, Baxandall, Keroes, up to Andy Grove and other recent gurus. The background for these can be found in e.g. RDH 4 et al. They have been supported by more folks I can recall, since you seem to value weight of personal experience and count of personalities.
Your own choices on the other hand, mostly seem to be limited your own personal preferences, with reluctance (or that is the impression you leave) to give more than cursory attention to alternatives - no reasons why or such, which leaves one with information quite devoid of technical substantiation.....
I think purely on that I rest my case as the cliché goes. With less judgmentalism than you showed for me, personal preferences never translates into dogma without proof/technical substantiation.
As such you really took a trip into second-guessing me. Most of my tube amplifiers I built in the past were in the 10W - 20W class. One of my most successful amplifiers was a SE EL84 plus EF86 4,5W jobbie as a student (if of any meaning to anybody, parts price cost £7-10, sold for £15 at the time). It was quite successful almost as an icon; never mind the numbers sold.....
Other frequently-built ones were the Willamson and Leak topologies, etc. My mentioned 100W example was built on request, not because of any prefernce on my side; simply quoted technical figures from that because those are at hand now.
So you will excuse me before I bore everybody; you sadly must go down in my experience as a person cherishing personal choices at the sad lack of technical substantiation at all. I respect you choices - but in science that does not automatically translare into dogma per se. (How nice it would have been had one had that luxury .....)
I apologise to readers for my rebuttal having gone personal here; Jeff opened the door. I will not comment on this thread further, but wish you a peaceful Christmas and an exciting 2014.
Regards,
Johan
Yes, you have certainly made YOUR point, not even deigning shedding hints of disrespect/refined disdain left and right for those daring to disagree with you. The former I can respect, the latter .....
I will try to be as succinct without the personal slants of what one person thinks/imagines the other knows. In my posts here I have only mentioned matters sustained by science and subsequently experienced by scores of folks from the days of Williamson, through Walker, Baxandall, Keroes, up to Andy Grove and other recent gurus. The background for these can be found in e.g. RDH 4 et al. They have been supported by more folks I can recall, since you seem to value weight of personal experience and count of personalities.
Your own choices on the other hand, mostly seem to be limited your own personal preferences, with reluctance (or that is the impression you leave) to give more than cursory attention to alternatives - no reasons why or such, which leaves one with information quite devoid of technical substantiation.....
I think purely on that I rest my case as the cliché goes. With less judgmentalism than you showed for me, personal preferences never translates into dogma without proof/technical substantiation.
As such you really took a trip into second-guessing me. Most of my tube amplifiers I built in the past were in the 10W - 20W class. One of my most successful amplifiers was a SE EL84 plus EF86 4,5W jobbie as a student (if of any meaning to anybody, parts price cost £7-10, sold for £15 at the time). It was quite successful almost as an icon; never mind the numbers sold.....
Other frequently-built ones were the Willamson and Leak topologies, etc. My mentioned 100W example was built on request, not because of any prefernce on my side; simply quoted technical figures from that because those are at hand now.
So you will excuse me before I bore everybody; you sadly must go down in my experience as a person cherishing personal choices at the sad lack of technical substantiation at all. I respect you choices - but in science that does not automatically translare into dogma per se. (How nice it would have been had one had that luxury .....)
I apologise to readers for my rebuttal having gone personal here; Jeff opened the door. I will not comment on this thread further, but wish you a peaceful Christmas and an exciting 2014.
Regards,
Johan
That the vast majority of people who have done (including me, and in the case of the Quads, the designer) have done so with great results; these are classic combinations. In the 45 years or so that I've been doing audio, you are the very first person I have ever known to state that Quad and LS3/5A are unsuitable for tube amps. I suppose each person needs to find their niche, and in a distribution, all curves have a tail.
Thanks for your reaction 
Simply filling some reactions in too
Simply filling some reactions in too
MonaHi Ketje / Mona,
Yes, very good. You have the basic topology, a differential front end as I suggested, direct coupled to the grids of the Finals. KISS rules.
Too old BALONEY, I am 69 !!!
Perhaps,not for me.I am 72 but disabled,only one hand more or less usable.
All else in your schematic needs FAR more thought as it needs simplification to make it better and more elegant.
(1) Get rid of 100% of the solid state devices, except for " something" on the common cathode tail of the 12AX7.... one LM334 ?? Tube rectifiers sound nicer. 5V3A or maybe a 5U4GB.
As long it's a reliable current source,the bias depends on it.
What do you have against sand?The tubes are made of it too (glass
(2) Get rid of EL-84s. Why? good tubes
(3) Get rid of feedback loop, global. UGH, UNNECESSARY.
In that case,connect the grid to ground and put some cathode resistors to limit the gain.
(4) Get rid of any non-triode operation. g2 to a
(5) Address heater to cathode ratings on the Finals tube, by biasing the Finals' Ef supply.
Is the case,separate heater windings on appropriate powersupply points.There are two ground symbols,one for the inputsection and one for the poweramp a +150V.
FORGET about "heat" from large value Cathode resistors. It isn't a problem at all. A NON-ISSUE !! Two Mills MRA-12 resistors will sound great and become bullet proof as Rks. The problem is dissipating large amounts of power to ground through the Finals tube itself, stressing it. !!!
Those big resistors must be decoupled with chemicals
(6) Run Finals conservatively at Golden Ratio plate power dissipation, ( 62% ) reducing mostly current, not VDC across it P-K.
Adjust current source !
(7) Run 12AX7 at about .0006 A. per triode section, and over 150 VDC Ea.
They are now .001A and 140V,not that far off
(8) Apply an elegant one resistor ( Mills MRA-12) shunt regulator to each 12AX7 tube, also called a draw down resistor by some old timers, so Finals swinging current does NOT possibly upset the front end B+.
I don't see the use of that.The +300V is referenced to the cathode of the finals with a zener.It's allways 150V above those cathodes and the drop on the anode resistors is fixed by the current source giving a well determind bias voltage.
All this I have HEARD and bits and pieces of it have been done over the years, by different advanced designers. No ones "put it all together yet" as I have above, and applied to an ACRO 20-20, or 20As, as I have offered to do .
Jeff Medwin
Last edited:
One of the best combinations I've ever heard was a pair of Quad ESL-63s driven by a fully regulated (and heavily) modified McIntosh 240. It still is one of my "mental" benchmark systems, and I've heard plenty of Altec based horn systems done right. Most of the triode amps I've heard - especially the low or "no" feedback ones - sound colored to the point where the differences between recordings become less apparent. Sure, they add a golden "hue" to the sound (hello, special FX box!), but it's not accuracy.
There are many ways to get to good sound: The Altec 604, for example, can be greatly improved through time-alignment and equalization via a digital system + bi-amping.
My own speakers, UREI 813As, aren't particularly efficient, but I still haven't heard a system that can match their scale and dynamics, all while being (relatively) uncolored. A good 15-20W amp gets them going, while Eico HF-60 power is even better.
Good to meet you online.
I've never been that good at being an absolutist.. Having said that I am generally heading in the same direction as you having gone to high efficiency speakers well over a decade ago. After much saving and scheming my father in law and I built the cabinets for my Onkens in early 2006 - much has evolved in the ensuing years, first JBL 2420 and 2402 drivers, then 2440 and 075. Various horns on the 2440 were evaluated until I end up with the current narrow dispersion set up that works well in my room. The X-O are 2nd order Linkwitz - Riley which work a whole lot better than the Butterworth based design they replaced.I'd been using 300B since 2004 as I found the 2A3 did not really get the job done, I suspect complex load impedance plays a role here, efficiency is certainly high enough.. The performance of the GM70 was a surprise, the amps are two stage with fixed bias in both stages and IT coupling. The driver is a D3A running at 20mA in triode connection. Basically in terms of overall resolution, and bass control the GM70 clobbered my old 300B amplifier which I eventually sold. The 300B was no slouch either.
I always thought the 2020 was an interesting amplifier, had one on loan for a short while and liked it, the owner did warn me that it was electrically fragile and to be careful with it.
I never got the fascination with the Rogers LS3/5A even with the dipole subwoofers. Dynamically constrained even with big amps and the subwoofers.. I've heard several systems with them and big ARC (and other) amps - left me luke warm. I'd quickly take a Vandersteen 2x or similar over these.. lol
I've heard some great sound out of Quad ESLs - in particular the ESL-57 driven by 17W trioded Dyna ST-70(with now mediocre KTA driver board), and oddly enough a 300B SE amp based on one of my older designs.
I've heard a lot of colored tube amps of both SE and PP persuasion, the end result depends heavily on the design goals, preferences and experience of the designer/builder. I've also heard some very neutral sounding amps of both types, often without gfb employed.
I've heard some great sound out of Quad ESLs - in particular the ESL-57 driven by 17W trioded Dyna ST-70(with now mediocre KTA driver board), and oddly enough a 300B SE amp based on one of my older designs.
I've heard a lot of colored tube amps of both SE and PP persuasion, the end result depends heavily on the design goals, preferences and experience of the designer/builder. I've also heard some very neutral sounding amps of both types, often without gfb employed.
Right room, right music, they can be wonderful. They are not made for heavy metal. Best I ever heard them sound was in my old efficiency apartment in Salt Lake City, driven by 60 watt monoblocks, and in a small living room in England, driven by the Crystal Palace.
Subwoofing is tricky and must be done actively to compensate for the high Q bass alignment.
Best I ever heard them sound was in my old efficiency apartment in Salt Lake City, driven by 60 watt monoblocks, and in a small living room in England, driven by the Crystal Palace.Subwoofing is tricky and must be done actively to compensate for the high Q bass alignment.
Both interesting examples.
I am not sure about all of those damper tubes in the 300B design, I'd be inclined to tinker some of the other circuit constants to get rid of them. Two stages would be more practical IMO with this approach using something like the C3G or D3A as drivers - triode connected of course.
The Mullard 3-3 looks intriguing and simple enough to try, again the EF86 is a wimp, and good ones are expensive. I would be inclined to try a 6Z9P instead with the requisite redesign.. No idea whether the end result would justify the effort..
I am not sure about all of those damper tubes in the 300B design, I'd be inclined to tinker some of the other circuit constants to get rid of them. Two stages would be more practical IMO with this approach using something like the C3G or D3A as drivers - triode connected of course.
The Mullard 3-3 looks intriguing and simple enough to try, again the EF86 is a wimp, and good ones are expensive. I would be inclined to try a 6Z9P instead with the requisite redesign.. No idea whether the end result would justify the effort..
Hi Bear,
Several questions in my mind.
Yes, I have seen these schematics before, and have visited Dennis Boyle, a good person, and his web site.
(1) Is this a real product??
(2) Other than one, maybe two prototype amps a score of years ago (thanks Abe), does and has anyone in the public domain own one of the AXIOMS?? How many are out there??
(3) It seems clear to me that this three stage DC topology will be easily beaten sonically, by a two stage DC amp with a high mu driver tube. The two stage amp will be more transparent sounding, due to one less stage in the path, so ... why be concerned with ever-so-complex second-best topologies?
It IS an interesting schematic to view and contemplate, but.... we gotta listen to amps ultimately with our ears and hearts, not our eyes and brains.
After twenty years of mildly wondering about it, can anyone out there in audio-land answer my point (2) above ??
Thanks for posting that again Bear. it was existing under my Radar screen for a long time.
Cheers,
Jeff
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.