Triodes and Tetrodes and SET, Oh My!

[Audiocrack]

Thanks to each of you for contributing to this thread!

I, for one, have learned from this thread a lot about a lot of amplifiers -- without actually hearing anything new myself. (Of course we have to listen for ourselves. But, I believe that when you understand someone’s sonic preferences, and you have calibrated your preferences to their preferences on equipment you both have heard, then it is possible to triangulate meaningfully on, and understand, what they mean what they discuss components you have not heard.)

I have had my VTL MB-750s for something like 18 years and I intend to keep them forever. I know well that Luke and Bea and many owners of Siegfried IIs consider the older generations of big VTLs dated sonically. For my purposes and preferences I do not necessarily agree.

I am sure the new generation of VTLs, with bigger and better power supplies and transformers and capacitors, exhibit flatter frequency response and better control over woofers and more extended high frequencies. But I like a bit of high frequency softness and roll-off. In any event it never occurred to me that I would ever consider a different amplifier.

Then I began auditioning ribbon-based panel speakers (Genesis, Gryphon Pendragon, re-built Apogees). And then I began listening to horn speakers driven by SET amplifiers. I now understand the attractions of SET amplifiers.

So I wanted to investigate whether I might prefer the sound of the midrange and treble frequencies of a ribbon-based speaker driven by a high-power SET amplifier to the sound produced by my VTLs in triode mode.

The NAT amplifiers obviously are beloved by many Europeans who have heard them or own them. From numerous reports the KR amplifiers, too, are extremely highly regarded. The NATs and the KRs seem to represent a close and fascinating contrast.

On the one hand I like that the NAT Transmitter and Magma are all-tube designs. On the other hand I like that KR manufactures their own triode tubes designed specifically for audio, rather than pressing NOS transmitter tubes into audio service as NAT does.

From the posts on this thread I believe there is a reasonable consensus that the KR -- with its “reverse hybrid” transistor input and driver stages -- exhibits slightly tighter grip in the low frequencies, and may have overall slightly greater “slam” and dynamics, among other desirable characteristics, compared to the all-tube NAT. But there also seems to be a consensus here that the NAT will sound a smidegeon warmer than the KR.

My application of an SET amplifier to power only the midrange and tweeter drivers of a ribbon panel speaker would mitigate the low-frequency advantage of the KR over the NAT. Plus, purely philosophically, if I can stay with an all-tube design then that is my preference.

So who says audiophiles cannot agree on anything? I think there is a statistically significant consensus on this thread that all-tube NAT is a bit warmer than the KR SET which may be a bit warmer than VTL push-pull in triode mode which (according to Michael Fremer) is a bit warmer than the big darTZeel. So there is some agreement on at least part of a tight spectrum of amplifiers at a very high level of sound quality and equipment which straddles the tube/solid-state sonic line.

Since I would be looking for a new amp to power only the midrange and treble drivers of ribbon speakers above 100 hz (Genesis) or 200 hz (Gryphon Pendragon) the low frequency advantage of the KR over the NAT would not help me in my particular application.

If I were to try any new amplifier for this specific purpose it would be the NAT Magma New.

PS1: Where exactly Jadis amplifiers fall in this group of amplifiers remains a mystery to me. But since I would want to try an SET I would focus on NAT and not on the push-pull Jadis.

PS2: If we were to expand this tight, high-power, SET-based discussion beyond NAT and KR to include Lamm and Ypsilon and Aesthetix Atlas, I fear any consensus about sonics would collapse quickly!

Interestingly nobody mentiones the Kondo sets. To me the Gakuohs set amps of around 22 watts deliver enough power in my Tidal LA set up in my view but if I remember correctly Ron you would have preferred (somewhat) more power. Have you ever considered the 50 watts Kondo Kagura amps? Apparently the Kagura's are able to deliver enormous power.

 

[Steve williams]

Interestingly nobody mentiones the Kondo sets. To me the Gakuohs set amps of around 22 watts deliver enough power in my Tidal LA set up in my view but if I remember correctly Ron you would have preferred (somewhat) more power. Have you ever considered the 50 watts Kondo Kagura amps? Apparently the Kagura's are able to deliver enormous power.

50 watts are 50 watts are 50 watts,no

 

[Audiocrack]

50 watts are 50 watts are 50 watts,no

In my experience 50 (class a) tube watts are more powerful than 50 solid state watts.

 

[Ron Resnick]

Interestingly nobody mentiones the Kondo sets. To me the Gakuohs set amps of around 22 watts deliver enough power in my Tidal LA set up in my view but if I remember correctly Ron you would have preferred (somewhat) more power. Have you ever considered the 50 watts Kondo Kagura amps? Apparently the Kagura's are able to deliver enormous power.

I think they have not been mentioned in this particular discussion perhaps only because the discussion has focused on higher-power SETs: KR Konzilla DX (100w), NAT Transmitter (120w), NAT Magma New (170w).

I honestly do not remember thinking your Tidal LA needed more power. I was astounded at the transparency and dynamics I heard with you! (Of course, now that you tell me you were feeding them only 22 watts I definitely want them louder! )

Plus -- I see the Gakuohs are listed at US$150,000?!?! For that price (compared to Kronzilla DX or Nat Magma New) they better come with a Japanese foot masseuse for my wife and me while we listen to the system!

Not for any good or specific reason but I feel that I would want more than 50 watts on Genesis panels or Apogee panels or Pendragon panels. Flemming drives the Pendragon panels (specified in Gryphon literature as 6 ohms at 89 dB) with a Gryphon Mephisto. When I heard the Infinity IRS V Michael Kay was driving the midrange/tweeter panels with Jadis JA-200s.

 

[Believe High Fidelity]

No OTL's as an option Ron?

 

[Steve williams]

In my experience 50 (class a) tube watts are more powerful than 50 solid state watts.

Really??

 

[Ron Resnick]

No OTL's as an option Ron?

I have always admired the design of Ralph's Atma-Sphere amplifiers, but I have never heard them.

I know OTLs are reputed to be generally great on electrostatic panels, but do they do as well on ribbon drivers?

 

[JackD201]

Hi Ron

I spent a lot of time bi-amping and tri-amping with differing amplifiers. It can be done but really I just thought getting the coherence to match multi-amping with like amps can leave you with nagging feelings up until you get it right which could take a long time. In the end I ended up with just one pair of amps I really liked, two identical sets with just differing input tubes and one that allows for internal selection of Bi-AMP, MONO (single rail dual supply) or Bridged Mono Solid State. I would hesitate to use a SET only for the highs because IMO their true strength is in the midrange. Sweeter softer highs are just a few tube rolls away.

Miss ya bud!

Jack

 

[Ron Resnick]

Hi Ron

I spent a lot of time bi-amping and tri-amping with differing amplifiers. It can be done but really I just thought getting the coherence to match multi-amping with like amps can leave you with nagging feelings up until you get it right which could take a long time. In the end I ended up with just one pair of amps I really liked, two identical sets with just differing input tubes and one that allows for internal selection of Bi-AMP, MONO (single rail dual supply) or Bridged Mono Solid State. I would hesitate to use a SET only for the highs because IMO their true strength is in the midrange. Sweeter softer highs are just a few tube rolls away.

Miss ya bud!

Jack

Thank you, Jack, for describing your experiences with bi-amping and tri-amping with different amplifiers. I, too, would be hesitant to use a SET only for the highs.

In the scenario I am envisioning the SET would cover all frequencies above 100 Hz (in the case of Genesis) or above 200 Hz (in the case of Gryphon Pendragon).

I miss you too!

 

[Believe High Fidelity]

I have always admired the design of Ralph's Atma-Sphere amplifiers, but I have never heard them.

I know OTLs are reputed to be generally great on electrostatic panels, but do they do as well on ribbon drivers?

They are speaker dependent, but when they can be paired most that go that route never reconsider going any other route. Just want to make sure you don't sell yourself short not matter which direction you end up taking.

Sounds like you are getting closer though. :thumbs up:

 

[Mike Lavigne]

I have always admired the design of Ralph's Atma-Sphere amplifiers, but I have never heard them.

I know OTLs are reputed to be generally great on electrostatic panels, but do they do as well on ribbon drivers?

the advantage of an OTL is neutrality and linearity at the frequency extremes. the rolled off bottom and top is minimized compared to an SET with a traditional transformer. the take away is less ability to control the speaker driver......so a more even impedance curve is preferred....and more heat and fragility.

Ralph's big Atmasphere's are able to deal with tougher loads that most any other OTL. and it's quite a neutral sounding tube amplifier.

don't personally have any experience with an OTL on a ribbon driver. but for 3 years (2001-2004) I did own the Tenor 75 watt OTL's.....when they did not explode they sounded outstanding. still on my short list of finest sounding tube amps. I'm told that the new current version of that amp is now rock solid.

 

[microstrip]

(...) I know OTLs are reputed to be generally great on electrostatic panels, but do they do as well on ribbon drivers?

Unless the ribbons are of the ultra low impedance type, I would expect OTLs to drive ribbons better than electrostatic - they are mainly a resistive load and electrostatics are capacitive. But I can imagine that matching a tube OTL with an SS amplifier can be a nightmare!

 

[Ron Resnick]

the advantage of an OTL is neutrality and linearity at the frequency extremes. the rolled off bottom and top is minimized compared to an SET with a traditional transformer. the take away is less ability to control the speaker driver......so a more even impedance curve is preferred....and more heat and fragility.

Ralph's big Atmasphere's are able to deal with tougher loads that most any other OTL. and it's quite a neutral sounding tube amplifier.

don't personally have any experience with an OTL on a ribbon driver. but for 3 years (2001-2004) I did own the Tenor 75 watt OTL's.....when they did not explode they sounded outstanding. still on my short list of finest sounding tube amps. I'm told that the new current version of that amp is now rock solid.

Thank you, Mike, for writing about your experience with OTLs.

If an advantage of OTLs is linearity at the frequency extremes then that suggests to me that, for my particular equation which does not require linearity at low frequencies (built-in bass amps will handle that), combined with the fact that I do not value extreme high frequency extension, SET is likely to be my best solution.

 

[Mike Lavigne]

Thank you, Mike, for writing about your experience with OTLs.

If an advantage of OTLs is linearity at the frequency extremes then that suggests to me that, for my particular equation which does not require linearity at low frequencies (built-in bass amps will handle that), combined with the fact that I do not value extreme high frequency extension, SET is likely to be my best solution.

I understand. everyone has their own sonic compass. being in love with the mid range of a top level SET is completely understandable. you are not alone. and I know its not that simple.

every approach has it's attraction.

 

[microstrip]

(...) If an advantage of OTLs is linearity at the frequency extremes then that suggests to me that, for my particular equation which does not require linearity at low frequencies (built-in bass amps will handle that), combined with the fact that I do not value extreme high frequency extension, SET is likely to be my best solution.

Although I can be misunderstanding what you mean by linearity, if both amplifiers do not overlap significantly in sound characteristics by a large bandwidth around the crossover zone the result can be very poor. For example, we can quote Andy Payor on the subject. "The notion that one can use a 15W single-ended triode for the main amplifier (because it doesn’t have to contend with making bass) coupled to a 1000W class-D amplifier for the bass may sound like the best of both worlds, but in practice it doesn’t work very well. This approach has been responsible for robbing the true performance potential of many a biamped system, as have poor setup and integration of levels within the system. " http://www.soundstageultra.com/index.php/features-menu/general-interest-interviews-menu/343-active-loudspeaker-systems-on-the-rise-peter-roth-talks-with-andy-payor-laurence-dickie-and-richard-vandersteen.

Blending the low bass frequencies with the rest of the spectrum is critical - when designers address this aspect of matching the sound of both zones we find they have strong opinions on it. IMHO this expertise is what separates great bass from just more bass.

 

[Diapason]

The OTLs that I owned for years (GRAAFs, pictured in my avatar) could sound incredibly transparent, fast and "alive", but they didn't often attain the relaxed, liquid "fullness" of a SET at its best. That said, it may have been a speaker issue more than anything, as they mated beautifully with Quad ESLs when I owned them too. In any case, I'd strongly recommend hearing some good OTLs (on the right speakers) as they provide a kind of sound I've never really heard elsewhere.

 

[Ron Resnick]

Although I can be misunderstanding what you mean by linearity, if both amplifiers do not overlap significantly in sound characteristics by a large bandwidth around the crossover zone the result can be very poor. For example, we can quote Andy Payor on the subject. "The notion that one can use a 15W single-ended triode for the main amplifier (because it doesn’t have to contend with making bass) coupled to a 1000W class-D amplifier for the bass may sound like the best of both worlds, but in practice it doesn’t work very well. This approach has been responsible for robbing the true performance potential of many a biamped system, as have poor setup and integration of levels within the system. " http://www.soundstageultra.com/index.php/features-menu/general-interest-interviews-menu/343-active-loudspeaker-systems-on-the-rise-peter-roth-talks-with-andy-payor-laurence-dickie-and-richard-vandersteen.

Blending the low bass frequencies with the rest of the spectrum is critical - when designers address this aspect of matching the sound of both zones we find they have strong opinions on it. IMHO this expertise is what separates great bass from just more bass.

I hear ya’ Francisco!

I read the full article in the link you posted. It was very interesting, and the speaker designers generally agreed with the theoretical benefits of actively bi-amped or built-in bass amplifiers designed specifically to match the speaker in question. This view certainly supports the designs of Evolution Acoustics, Genesis, Gryphon and Von Schweikert Audio, all of which supply built-in bass amps and custom crossovers for their dynamic driver woofers (and the active crossover Rockport Arrakis to which you add your own bass amp).

I would not hesitate to twist any argument I might proffer into a pretzel to avoid disagreeing with Andy Payor. But I think, as is often the case, that the success or failure lies in the implementation.

I certainly understand the theoretical objection to bi-amping with amplifiers of radically different circuit topologies (e.g., high power class D and low power SET, as Andy used in his example). But there also are many examples of successful mixing of different amp technologies.

Gary Koh would know best, but I believe there are many examples of audiophiles using tubes, and even SETs, on Genesis ribbon panels. Mike Kay at Lyric Hi-Fi in Manhattan used Jadis JA-200s on his Infinity IRS Vs.

It seems pretty common for Apogee owners to employ solid-state amplification on the woofer panel with tubes on the midrange and tweeter ribbons.

All of the MartinLogan speakers with actively-powered, built-in, class D woofer amps (e.g., Renaissance series, Summits) obviously require some additional amplifier for the electrostatic panels. I personally would never use anything but high-power tubes on a ML electrostatic panel.

I am pretty confident I would be very happy with the sound of tubes, whether push-pull or SET, on Pendragon or Genesis ribbon panels.

Maybe the MB-750s in triode mode would be a closer match to the built-in bass amps than the SET NAT Magma New would be. I cannot be certain in advance. I think that would make for a very interesting comparison test!

 

[Ron Resnick]

The OTLs that I owned for years (GRAAFs, pictured in my avatar) could sound incredibly transparent, fast and "alive", but they didn't often attain the relaxed, liquid "fullness" of a SET at its best. That said, it may have been a speaker issue more than anything, as they mated beautifully with Quad ESLs when I owned them too. In any case, I'd strongly recommend hearing some good OTLs (on the right speakers) as they provide a kind of sound I've never really heard elsewhere.

So it seems that for me, for this particular mission, I do not have to investigate OTLs. A relaxed, liquid, full sound is what I want on the ribbon panels. I am sure I get that from the MB-750s in triode mode. The question is would I get even more of that, with the unique SET transparency and liveness, from NAT Magma New. I suspect that I would.

 

[morricab]

IMHO we can not separate the Jadis sound from their preamplifiers. I have now a JPL + Defy7 mkIII in my room, and have owned the great JP80 + the JA80's power amplifiers. The JLP is a very dynamic and alive preamplifier, an excellent example of positive tube coloration and matches the Defy7 perfectly. The JP80 was much more refined, with delicate but great dynamics, and less colored. It was a great match with several SS amplifiers. Unfortunately I never tried the JP80 with the Defy7. The JA80 sounded lush, nice sounding, but lacked dynamics in my system. Probably with more efficient speakers my opinion would be different.

BTW, do you have any details about the silver cabling that was used in the Defy III?

Would be cool to try it with the Defy 7. What do you think is it possibly better than the JA80?

 

[morricab]

IMHO we can not separate the Jadis sound from their preamplifiers. I have now a JPL + Defy7 mkIII in my room, and have owned the great JP80 + the JA80's power amplifiers. The JLP is a very dynamic and alive preamplifier, an excellent example of positive tube coloration and matches the Defy7 perfectly. The JP80 was much more refined, with delicate but great dynamics, and less colored. It was a great match with several SS amplifiers. Unfortunately I never tried the JP80 with the Defy7. The JA80 sounded lush, nice sounding, but lacked dynamics in my system. Probably with more efficient speakers my opinion would be different.

BTW, do you have any details about the silver cabling that was used in the Defy III?

I think it was Siltech cabling but not with gold.