Theory, Practice and Sexlife of the "Reflected Plate Amplifier" Hybrid, low Voltage Tube circuit

[emrr]

Certainly a lot of ancient tube gear lacking NFB loops amplify and sound remarkably similar at a very low voltage and their stated voltage, ignoring headroom. I’ve listened to many after restoration with a variable supply coming up from 0V.

 

[gyraf]

Relatively low voltage tube circuits are sonically valid, in any context, as they do not run "starved anode"

Aah, cynical. I honestly don't consider me being in that category

As I have been trying to say a couple of times already in this thread - just listen to it, for pete's sake! And if you really like what it does for you, then all power to you - you then have a simple, cheap and safe way of getting what you want - what's not to like about that?

..remember, "starved anode" is just a label we came up with to describe the way our ears don't really seem to like the results from these circuits. IMO this whole family of circuits has some common sonic qualities that are not exactly the part of "tube sound" that often makes us prefer tubed circuits. It's not that it sounds bad, it just dosen't do what you'd expect from a genuine tube circuit.

Could very well be my ears that are acting difficult or just plain wrong, but I have put quite a lot of effort into pursuing architectures like these - and never been able to come up with anything even slightly "better-sounding" in adding tube when comparing these transistor-loaded circuits to all-transistor/fet circuits of the same type. I honestly think there's good reason why these hybrid setups went out of fashion in pro audio already before the turn of the century - if it did what it says on the tin, this would be all-over.

Dragging in V76 is poor argumentation

/Jakob E.

 

[thor.zmt]

As I have been trying to say a couple of times already in this thread - just listen to it, for pete's sake! And if you really like what it does for you, then all power to you - you then have a simple, cheap and safe way of getting what you want - what's not to like about that?

You tell me, as you seem extremely negative.

..remember, "starved anode" is just a label we came up with to describe the way our ears don't really seem to like the results from these circuits.

Which circuit are you referring to?

IMO this whole family of circuits has some common sonic qualities that are not exactly the part of "tube sound" that often makes us prefer tubed circuits. It's not that it sounds bad, it just dosen't do what you'd expect from a genuine tube circuit.

This is NOT my experience. And I certainly did way, way more high voltage tube circuits than low voltage. Maybe you did your circuits wrong?

I do know a fair numbers of not just bedroom Studio's that used the iFi tube box in mastering, I had to deal with the tech support queries "what is the best way to connect this balanced - to which my common answer was "buy two, use XLR +/- split to LR RCA.

I even know a few guitarists who like to use this directly from their guitar.

Could very well be my ears that are acting difficult or just plain wrong, but I have put quite a lot of effort into pursuing architectures like these - and never been able to come up with anything even slightly "better-sounding" in adding tube when comparing these transistor-loaded circuits to all-transistor/fet circuits of the same type.

I made some like this where the input device was switchable between tube & J-Fet on the fly. Generally speaking, tube was preferred.

I honestly think there's good reason why these hybrid setups went out of fashion in pro audio already before the turn of the century - if it did what it says on the tin, this would be all-over.

Out of fashion? Look here:

Tube Preamps | Sweetwater

How many do you think are "pure tube" running on high voltages?

Dragging in V76 is poor argumentation

Why? My point was that you cannot make a genuinely sonically identical copy of 1950's gear today. So why not build something else?

It is worth reviewing how this thread came about. Someone who wanted ways to get more "colour" into his recordings was scared of playing with tubes because of high voltages.

Using the right tubes and circuits it is possible to get "tube sound" without lethal voltages. This makes it much less scary for budding DIY'ers to play around, get their feet wet.

And you really think this is a bad thing you must dump on? Go on at length how bad it must sound (which it does not)? Must be your way or the high way?

Thor

 

[JohnRoberts]

I recall a slew of low voltage 1u rack mount tube skus back in the 90s addressing the bedroom recording business. I refused to do such a gimmicky SKU while I was in charge of all product management inside Peavey, my replacement was less steadfast.

Of course the customer was always right, even when not.

JR

 

[Mr. Moscode]

Aah, the scope of this discussion is much clearer to me now - it's intended / aimed for a hifi-audience?

Note that they hold beliefs and aesthetics quite different to those of pro audio: There is some overlap, but not at all as much as you'd think..

Yup. Too bad.

And yes, I agree that this way of low-voltage-using real tubes is a great idea for the hifi segment, and listening to what it does - in that context - really makes sense now

/Jakob E.

In the hi end hi fi I don't think anyone cares if it is low voltage. The sound is the thing.

 

[Mr. Moscode]

No, it can be applied just as much in pro audio. The fun is, that by adjusting operation conditions for the tube and open loop gain a wide range of sound characters can be obtained, adjustable/switchable even.

Now we're talking audio filters, not transparency. It's a compromise.

I think your view is excessively cynical, Re "hifi".

Relatively low voltage tube circuits are sonically valid, in any context, as they do not run "starved anode", are excellent for DIY, due to the lack of dangerous voltages, more so for beginners.

As long as 48V are present why not a DIY API 500 rack tube mic pre, or gain make-up Amp for a passive summing mixer?

Tube sound does not come from using high voltages.

WTF is "tube sound" on this forum, anyway? In many posts it's a prejorative. Strip away the transformers, keep it simple, low NFB can sound very clean and not lacking anything in my experience.

So I'd be happy to (for example) discuss a mic pre with classic Sziklai differential Frontend as "electronic transformer" and after 26dB of gain there, two stages of "hybrid tube" gain.

Will it sound like a TAB V76? Not really, though it could be tuned if desired to sound s lot like a TAB V76. But if you really want a TAB V76, buy the original, no DIY copy in 2023 will be accurate enough.

Thor

Exactly.

 

[Mr. Moscode]

Aah, cynical. I honestly don't consider me being in that category

As I have been trying to say a couple of times already in this thread - just listen to it, for pete's sake! And if you really like what it does for you, then all power to you - you then have a simple, cheap and safe way of getting what you want - what's not to like about that?

I agree. It either gives you pleasureable listening or or leaves you wanting something . . .

..remember, "starved anode" is just a label we came up with to describe the way our ears don't really seem to like the results from these circuits. IMO this whole family of circuits has some common sonic qualities that are not exactly the part of "tube sound" that often makes us prefer tubed circuits. It's not that it sounds bad, it just dosen't do what you'd expect from a genuine tube circuit.

Starved anode pertains usually to current starved pentodes which result in very high gain in 1 stage. For instance the 6EJ7 input tube of a Julius Futterman OTL amplifier has a 1.3 meg plate resistor but 60db of gain but decling gain with F, more like an op amp. However as part of that circuit it the amp is incredibly revealing and transparent. IMHO one of the best sounds on the planet.

Could very well be my ears that are acting difficult or just plain wrong, but I have put quite a lot of effort into pursuing architectures like these - and never been able to come up with anything even slightly "better-sounding" in adding tube when comparing these transistor-loaded circuits to all-transistor/fet circuits of the same type. I honestly think there's good reason why these hybrid setups went out of fashion in pro audio already before the turn of the century - if it did what it says on the tin, this would be all-over.

Dragging in V76 is poor argumentation

/Jakob E.

Pro and semi-pro audio went all op amp for all the advantages that come with it. Low heat noise cost size. Tubes make much more heat and cost more to keep a studio cool. A piece of specialty tube gear that does something that nothing else can do earns a spot in the rack.

I have yet to hear a solid state amp that keeps me interested. Just sayin'

 

[thor.zmt]

Now we're talking audio filters, not transparency. It's a compromise.

Nothing is objectively or subjectively transparent. Indeed, too transparent in specific areas sometimes exposes problems that have no fix.

I always like to compare old Leica (Leiz) lenses, like the 1930's pre war example I have on my Sony Digicam. Objectively the lens is not great, but pictures taken with this glass have a quality, colour rendition, everything seems RIGHT. Use much better moden glass, it lacks this quality.

Thor

 

[gyraf]

+1 for the subjective quality of the Leitz Elmar - I know exactly what you mean here...

 

[Mr. Moscode]

Nothing is objectively or subjectively transparent. Indeed, too transparent in specific areas sometimes exposes problems that have no fix.

I always like to compare old Leica (Leiz) lenses, like the 1930's pre war example I have on my Sony Digicam. Objectively the lens is not great, but pictures taken with this glass have a quality, colour rendition, everything seems RIGHT. Use much better moden glass, it lacks this quality.

Thor

I get it. In my biz transparency is key. There are a lot of things that can eat away at that. Interconnects, output chokes, unsavory caps, speaker cables, oxidation in wires and contacts, too much nfb. I'm talking mostly power amps but in low level signal preamps it can be worse - switches, jacks, pots. There isn't that much signal to begin with.

Nice chat. BTW - the Aphex circuit with some beefy output mosfets and power supply would make a cool power amp. The only thing is, IMVHO, open loop gain should be no more than 40 db if it is low distortion to begin with, say <1% open loop. W/10 db nfb I'm happy at .1% THD at full output. You would need a low gain tube. The 6S4 is a hi Perv tube. It might work, but I doubt it's cheap or plentiful. Also, experience has shown me that a bootstrapped tube follower sounds better driving mosfets. Would the other work? Probably, but would it be interesting to listen too? Meh?, if past is prologue. . . .

 

[thor.zmt]

I get it. In my biz transparency is key.

I have been in the same biz (remember, multiple Stereophile Class A/B products minus paid advertising...)



I went Frankie on this...

There are a lot of things that can eat away at that.

True. But sometimes a little soft focus and extrema colour saturation can turn a girl with bad porus skin in a terrible place into a wet dream.

This is why I consider.Audio, I mean really good audio, if I a say so, a kind of magic.

Nice chat. BTW - the Aphex circuit with some beefy output mosfets and power supply would make a cool power amp.

I might use "Bimos" with Lateral MOS drivers at 120mA IQ and a bunch of the china copies of the old RHOM (IIRC) Ring Emitter Transistors as actual output stage, 120mA Iq per pair of outputs.

But yes. And I did make something like that as a labrat.

The only thing is, IMVHO, open loop gain should be no more than 40 db if it is low distortion to begin with, say <1% open loop. W/10 db nfb I'm happy at .1% THD at full output. You would need a low gain tube.

No, I would use the inner loop to have output stage and VAS in one loop, leaving just the right amount of NFB around our tube, to get the correct sound.

The 6S4 is a hi Perv tube. It might work, but I doubt it's cheap or plentiful.

I'd go with 6N3PE. I like this tube (NOS Russian), though a GE 5670 ***** is better.

Also, experience has shown me that a bootstrapped tube follower sounds better driving mosfets.

Mosfet's store not tubes. Actually often an emitter followed is great. At LF the Mosfet input impedance is essentially infinity, so the emitter followed input impedance will be beta times infinity.

Once (parasitic) capacitances become dominant, using a decent dlfast super beta BJT as follower, voila, the VAS load is C/Beta...

My favorite ever hybrid was like this:

6072A SRPP input (ground referenced - 200V VA)

5687WB VAS referenced to -200V with bootstrapped load from +200V

5687WB cathode follower

Bimos Output on +/- 100V with 8A lateral MOS as drivers and 6 pairs of RET BJT's (90A peak I out), 720mA Iq and 8W/8R Class A.

Feedback loop closed to input tube cathode.

500W+ @ 8R. 2 Ohm load capable. Killerwatt level into 2R.

Thor

 

[Mr. Moscode]

I think we're on the same wavelength. Magic is what we sell. I have a customer who owns 20 amps, a collector. He just got one of mine and so far likes it the best. So he says. But collectors are always looking for another.

I'm always asking myself how I can make it better. I can't help it. That's why people pay for my upgrades. No more manufacturing. I'm done with that.

"

The only thing is, IMVHO, open loop gain should be no more than 40 db if it is low distortion to begin with, say <1% open loop. W/10 db nfb I'm happy at .1% THD at full output. You would need a low gain tube.

No, I would use the inner loop to have output stage and VAS in one loop, leaving just the right amount of NFB around our tube, to get the correct sound."

One man's "correct" is another's Sansui . .

My amp has an open loop triode input tube. That ouput sums with the NFB at the input grid of the 2nd triode > bootstrapped K follower > mosfet out, with a bootstrapped bias network. So NFB only around 1/2 of the amp. Very nice top to bottom, still short of Futterman OTL but in the ballpark.

 

[thor.zmt]

I think we're on the same wavelength. Magic is what we sell.

Precisely, "Magic" with a "C" as in "just an illusion":

I'm always asking myself how I can make it better. I can't help it. That's why people pay for my upgrades. No more manufacturing. I'm done with that.

Better is such a imprecise measure. I am looking instead how I can achieve the results I desire with whatever technology is currently available (not always successfully, mind you).

One man's "correct" is another's Sansui . .

Sansui made many excellent Amplifiers, funnily enough.

But what I call "musical correctness" or "rightness" is a quality that is immediately obvious and is obvious as much with acoustic Jazz and Classical as it is with death metal or EDM.

For example, a system that sounds great with "Gurl & Geetur" but falls to pieces with Rush is not "right", it may be pleasing with audiophool drivel, but in reality it's wrong... Like classic LS3/5A driven by 300B SE Hong Kong systems that sound great with Choi Kam and turn to total mud with 1812...

My amp has an open loop triode input tube.

I have done this on many Amplifiers. It depends on context. As I used a low gain (high current) middle stage it was better to loop the feedback to the input cathode, to actually have some loop gain. With a 6072A SRPP first stage gain is 22 and the 5687WB has a gain of 16, so total OLG is only 50dB and with ~ 30dB gain closed loop gain only 20dB NFB is used.

That ouput sums with the NFB at the input grid of the 2nd triode > bootstrapped K follower > mosfet out, with a bootstrapped bias network. So NFB only around 1/2 of the amp. Very nice top to bottom, still short of Futterman OTL but in the ballpark.

For fun, try my style one day. Not hard. And yes, you can use 6N3PEV and 6N6PEV russkie tubes, but that does not sound like NOS 6072A/5687WB. I use direct coupling to the output stage, so there is just one coupling cap in the whole circuit. Naturally, there is a servo.

Thor

 

[JohnRoberts]

Precisely, "Magic" with a "C" as in "just an illusion":

yup, magic is not real AFAIK

Better is such a imprecise measure. I am looking instead how I can achieve the results I desire with whatever technology is currently available (not always successfully, mind you).

AKA design engineering

Sansui made many excellent Amplifiers, funnily enough.

But what I call "musical correctness" or "rightness" is a quality that is immediately obvious and is obvious as much with acoustic Jazz and Classical as it is with death metal or EDM.

I had the good fortune of knowing and working with Rudy Bozak (RIP), the well known loudspeaker manufacturer (as a consultant on a couple of consumer electronics products). Rudy's big speakers (like his Concert Grand) were all about "sounding" right when reproducing classical music. I never got him to explain what that means objectively. When I was with him I mostly listened.

For example, a system that sounds great with "Gurl & Geetur" but falls to pieces with Rush is not "right", it may be pleasing with audiophool drivel, but in reality it's wrong... Like classic LS3/5A driven by 300B SE Hong Kong systems that sound great with Choi Kam and turn to total mud with 1812...





I have done this on many Amplifiers. It depends on context. As I used a low gain (high current) middle stage it was better to loop the feedback to the input cathode, to actually have some loop gain. With a 6072A SRPP first stage gain is 22 and the 5687WB has a gain of 16, so total OLG is only 50dB and with ~ 30dB gain closed loop gain only 20dB NFB is used.

My last phono preamp design used an input JFET gain stage running open loop. The 75 uSec RIAA pole was implemented passively at the drain of that input JFET. NF was used from there, the rest of the way to the final output. In my experience NF makes good circuits better. I ran this phono input stage open loop so the preamp input could accept crazy high edge rates and just passively roll them off.

For fun, try my style one day. Not hard. And yes, you can use 6N3PEV and 6N6PEV russkie tubes, but that does not sound like NOS 6072A/5687WB. I use direct coupling to the output stage, so there is just one coupling cap in the whole circuit. Naturally, there is a servo.

Thor

I am not a tube guy, not even for fun... but you guys enjoy, I'll just watch.

JR

 

[thor.zmt]

Rudy's big speakers (like his Concert Grand) were all about "sounding" right when reproducing classical music.

They do great with rock and metal too, an example of "right".

My last phono preamp design used an input JFET gain stage running open loop.

All my various phonostages that get awards left, right etc. Use open loop first stage, sometimes open loop after.

I am not a tube guy, not even for fun... but you guys enjoy, I'll just watch.

Don't you know, electrons that travel in Vaccuum are faster and bigger? ; )

For the best sound, 833 Transmitter tubes in Class A1 with 5kV anode voltage on exposed anode cap's. Honestly I tried. Until it fried the office cat. We decided not to commercialise the design at that point.

Thor

 

[Mr. Moscode]

yup, magic is not real AFAIK

AKA design engineering

I had the good fortune of knowing and working with Rudy Bozak (RIP), the well known loudspeaker manufacturer (as a consultant on a couple of consumer electronics products). Rudy's big speakers (like his Concert Grand) were all about "sounding" right when reproducing classical music. I never got him to explain what that means objectively. When I was with him I mostly listened.

My last phono preamp design used an input JFET gain stage running open loop. The 75 uSec RIAA pole was implemented passively at the drain of that input JFET. NF was used from there, the rest of the way to the final output. In my experience NF makes good circuits better. I ran this phono input stage open loop so the preamp input could accept crazy high edge rates and just passively roll them off.

I am not a tube guy, not even for fun... but you guys enjoy, I'll just watch.

JR

Thanks, JR. You're pretty humble. A hybrid RIAA net. Someone else did that in some hi end phono pre but I don't remember who. Maybe it was you.

I'm not against NFB, In my stuff I use it in moderation. My experience as a musician who has performed jazz almost all my life, covid excepted, is what tunes my ear. Cymbals, shakers, gourds, percussion are revealing of transparency in the high end. I have found with adequate power, if that sounds right heavy metal sounds right. I don't think my audience are metal heads, not my generation.

 

[Mr. Moscode]

Precisely, "Magic" with a "C" as in "just an illusion":





Better is such a imprecise measure. I am looking instead how I can achieve the results I desire with whatever technology is currently available (not always successfully, mind you).

Yup. Better, by my thinking, is transparency that reveals low level detail-room tone, reverb tails, inner harmarnic parts with clarity, piano chords where you can plainly hear each note. Better is loud ffff passages that maintain the same soundstage imaging and don't collapse under the demands of power and maintain the same harmonic balance as pppp. Better is eliminating the feel of artificiality the best I can. Better is natural sounding voices without overtones that don't belong there.

Sansui made many excellent Amplifiers, funnily enough.

Yes they did but 70s receiver sound is what I mean. I designed an all tube amp for Sansui.

But what I call "musical correctness" or "rightness" is a quality that is immediately obvious and is obvious as much with acoustic Jazz and Classical as it is with death metal or EDM.

Exactly.

For example, a system that sounds great with "Gurl & Geetur" but falls to pieces with Rush is not "right", it may be pleasing with audiophool drivel, but in reality it's wrong... Like classic LS3/5A driven by 300B SE Hong Kong systems that sound great with Choi Kam and turn to total mud with 1812...

That's expecting a Prius to haul a tractor trailor. But a tractor trailor can haul a Prius. - Tortured metaphor, I know.

I have done this on many Amplifiers. It depends on context. As I used a low gain (high current) middle stage it was better to loop the feedback to the input cathode, to actually have some loop gain. With a 6072A SRPP first stage gain is 22 and the 5687WB has a gain of 16, so total OLG is only 50dB and with ~ 30dB gain closed loop gain only 20dB NFB is used.

Class A triode is as basic as it gets. My early Moscodes did exactly that - NFB to the K of the input stage w/20db NFB. I decided to go NFB to the 2nd stage on out with only 10 db. Of course that leaves overall gain up to the 1st stage, but no one complained.

The amp has a 6/12 fil auto select for tube rolling. I.e. 6DJ8 or 12AU7 works.

For fun, try my style one day. Not hard. And yes, you can use 6N3PEV and 6N6PEV russkie tubes, but that does not sound like NOS 6072A/5687WB. I use direct coupling to the output stage, so there is just one coupling cap in the whole circuit. Naturally, there is a servo.

Thor

I have 2 caps and a servo around the output. So far so good.

Did you ever look at Modjer Modjeski's Counterpoint SA4 OTL? 1 servo from back to front using zener level shifters run at very low current bypassed with coupling caps. They sound very good but not quite the magic of a Futterman OTL with Quad 63s.

As Harvey would say, "The Search for Musical Ecstasy".

 

[Tubetec]

5KV , jeeeeezus ,
nothing like the smell of hot ***** first thing in the morning eh?

 

[thor.zmt]

My experience as a musician who has performed jazz almost all my life, covid excepted, is what tunes my ear. Cymbals, shakers, gourds, percussion are revealing of transparency in the high end.

Funny decades as sound engineer and drummers always ***** on a recording their expensive Zildjan/Paiste/et al Cymbals sound like cheap sh!t. Thy rarely b!tch with me, even live with an all digital system and Class D amps, shows it is not what you do, but how.

Thor

 

[thor.zmt]

5KV , jeeeeezus ,
nothing like the smell of hot ***** first thing in the morning eh?

555555

That's Thai for Ha ( = 5)

Thor