@atmasphere Yes, that's the issue with engineering options like triode vs UL. Based on all the OPT I've used, designed around, etc, a UL OPT will require a certain NFB characteristic for best performance. Using the same level of NFB when the UL taps are lifted may not be the optimal level for the reasons NFB is used. Marketing loves to have perceived needs to sell to.
I hate to say it, but now I think maybe I like my amp in ultralinear mode versus triode
It's a Cary V-12; it features a dozen EL34s and each pair has a switch in between them that configures that pair to either triode or ultralinear. In full triode Cary listed in the specs that it makes 50 wpc and in full ultralinear 100 wpc. For most of the twenty three years that I have owned this amp I have always felt that I preferred triode except for the occasions that I wanted to full out blast (it has literally been many years since I've felt the need to full out blast).
However, today I experimented with a couple of things in my system, and after listening to the same "Jazz Essentials" (compilation) red book CD a couple of times all the way through, the next thing I experimented with was switching to full ultralinear.
Maybe there was more "PRaT"? (Which is a term I am still not sure that I completely grasp.) Maybe . . . but what I do feel I noted for sure was that the imaging (particularly the imaging in the center) had more weight (meatier?) and was presented more forward, which I actually like.
I put a few more hours in (one more time with Jazz Essentials, Holly Cole/It Happened One Night, Dave's True Story/Sex Without Bodies, selected tracks from Rebecca Pigeon/The Raven and Once Blue/self titled and Norah Jones/Feels Like Home) after switching to ultralinear. (No booze during this session, just coffee.) The jury is still out on this, but I do have some CDs in mind that I want to listen to over the next few days as I continue to evaluate.
127 responses Add your response
@pickindoug To do it properly, the output transformer would have to have taps for triode operation since the optimal plate to plate load for a power pentode wired in triode is a lower impedance than that used for UL operation. That could make the transformer a lot more expensive and the switching a bit more complex. |
| Post removed |
| Post removed |
I wonder if the output transformer of amplifier with the triode/UL option is optimized equally for both modes, or if the default is to design for best operation in UL, then simply switch the screens to pair with the plates of the output tubes (usually via a resistor)? Theoretically, doubling power gets you 3 db of headroom, but if the linearity of the output transformer isn't consistent across both modes the perceived sonic difference may be greater. And to your point, the choice of music plays a role in which mode sounds better. I've got that same pleasant quandary with speakers and two systems in one room. |
| Post removed |
| Post removed |
Many years ago I had a tube amp w/8 EL34s that I had the mfr mod to allow each power tube to be triode/ultralinear switchable (Roger Mojeski's RM-9). The speakers were Vandersteen's then TOTL, the Vandersteen-4s. I greatly preferred triode mode with both TT and digital sources (especially with digital sources). Triode mode pushed the inherent musicality and realism of those big Vandy's as far it they went. I never felt the urge to use ultranlinear. What's more, I didn't need to: the internal subs of the speakers were powered by a separate amp, a 200 wpc SS amp. |
@atmasphere , that could be; this subject is over my head so I will not offer an opinion.. As far as the UL design, Mr. Had does state at least a couple of times in that manual how much he likes triode. He even refers to himself as "a triode crazed audiophile" in that manual and at one point goes on to say that this amp "sounds better" when operated in triode. |
@immatthewj I was commenting more in the fact that the manual was either poorly written or misleading. I doubt the 'triode only' thing since it is a UL design. The best interpretation is there's no global feedback. But there certainly is local feedback.
|
@atmasphere , I am not arguing with what anyone is saying; I have stated before that I am illiterate on these subjects. I was simply directly quoting from the owners manual that came with my amp when I bought it. Although he didn't write it that way, is it possible that what i quoted was meant to only be applied to triode operation? |
@immatthewj UL operation is feedback FWIW. To know how much you'd have to compare to pentode operation, which isn't an option... |
Ralph makes a very good point about the percentage of screen taps and their relationship to the effectiveness of UL operation. In an earlier post I referred to the original "Musician's Amplifier," the first iteration of the the British Williamson amplifier proposed by Sarser and Sprinkle. It employed tubes and transformers more readily available on the American market. In 1952, when Hafler and Keroes published the first "ultralinear" Williamson, Sarser and Sprinkle noticed that the output transformer they chose for the "Musician's Amplifier," the Peerless S-265-Q, had primary taps that could act as 50% loads for the screens. They urged hobbyists to convert their "Musician's Amplifiers" to "ultralinear" operation by some simple changes to the circuit. In fact, while the 50% taps of the Peerless transformer are not *ideal*, they actually brought the output tubes just a hair closer to true triode operation while doubling the power output. Having built this amp myself, I can tell you that the result, despite its flaws, is quite beautiful. |
As far as the CAD-805RS, what Cary lists on their site under specs for feedback is "0 to 10 dB continuously variable control." |
To @xenolith, agree, yes, you are fortunate to own those very unique and interesting big blue amplifiers from Gary and Charlie. I love one-off boutique custom builds like this with audio, cars, and motorcycles. Great story how they got finished. Very cool, and Thank You for sharing the info and photos. Happy listening. |
Re-checking my own prior Cary amps on this and there are several I can account for that followed Dennis’ vision for "absolute zero-feedback designs", and talk some here about "low- and zero-negative-feedback triode circuitry " and how it "tended to generate the sound he preferred—to his ears" unquote. Some of the older Steroephile articles in the early to mid 2000s challenge and mention this too; https://www.stereophile.com/tubepoweramps/1098cary/index.html On some search engines you can look for these words and hyphenated phrase +Cary +"zero-feedback" and different amps will pop up references this about some times.
|
@brianlucey , what sonic changes did you note when in 1/3 triode 2//3 UL and in vice versa? |
@dogearedaudio , coming back to the feedback, and again not arguing with you, but almost at the bottom of page 4, the page I believe should be just before the un-numbered specifications page, there is a header "OVERVIEW AND CLOSING THOUGHTS" and under that header Mr. Had had written that, "The V12 is in reality four single-ended amplifiers operating without any form of feedback." |
@knotscott , the owner’s manual says nothing about needing to adjust the bias when switching from triode to UL or vice versa, and I have checked the bias after I made this last switch and there was no noticeable swing in the bias. As far adjusting the volume when moving from triode to UL, I would have expected that myself, but it seems as if I am mostly leaving the knob on my SLP05 where it was. Maybe turning it just a hair down, but if so, only a hair or so. Not much at all. Mainly what I noticed was a change in "the body" of the sound. (or maybe the "shape of the sound" or maybe the "location of the sound." I am bad with terminology.) The gain on that preamp when using the balance ins and outs (and that’s all I have used so far, but I am thinking about experimenting with that also) seems pretty high to me, and with most CDs or SACDs it is about 9 oclock give or take a hair. |
@dogearedaudio , there should be a page labeled "SPECIFICATIONS" that interestingly is not numbered, but it looks like it would be page 5. (I say looks like, because the pages are starting to come loose in that particular manual.) Anyway, dimensions are listed. next weight is mentioned, next circuit type is mentioned, and then, under circuit type, it lists "FEED BACK: Zero". And I am not arguing with anything you say, as I truly have a limited understanding of most of this, but just telling you what the "CAD-280SA V12 OPERATING MANUAL" that came with my amp says. |
"Zero" feedback? Really? I’ve looked at the CAD-280SA V12 manual online and it doesn’t mention feedback, but perhaps that's a different model. Hard to imagine they use NO feedback. Maybe not global feedback, which encompasses the output transformer. Internal feedback can be used to good effect, but no feedback at all is hard to imagine. Oh sure, there are amps that will work well with different output tubes, and "non-optimized" can offer some interesting sound signatures. Like I said, nothing wrong with that. ;-) |
FWIW, Gary Dodd, shortly before he died in 2015, was contracted by a very rich person who heard his "Blue Monster" amps:
to build the "best sounding amplifier possible; even better than the Blue Monsters." What he came up with was monoblocks with 4 KT77 output tubes run in OPTIMIZED ultralinear mode with 2 6BL7 input tubes; custom would massive power supply and output trannies, huge inductors on both input and output stages, 700 volts on the plates and Dueland CAST coupling caps. He died before finishing the amps. 6 years later, his great friend and talented tube audio builder, Charlie Cocci, completed the two amps. Charlie, who has been building tube amps for at least 50 years, stated "These are not only the best soundfing amps I've ever built, they're the best sounding amps I've ever heard...and I've heard a lot of amps!" FWIW. Those amps are in my living room. The Dodd Audio Balanced Power Supply that is shown in the 2nd photo powering the Blue Monsters now powers the amps in my living room. I'm a very lucky person. Confirmation bias confession? I don't think so, I think I just agree with Charlie. FWIW. |
To your point, while you can run different output tubes in many of DH’s creations, stemming from his former life with Cary Audio, and recent times creating Inspire, like the early years of Cary V12, V12R, or SLA-70 @immatthewj is referencing periodically - his amps were/are designed and listened to with a very particular output tube in mind. in the case of the V12/R. It’s the classic EL34 tube, paired with particular transformers and input/driver tubes that make for an interesting blend an sound. Today, If you splurge and buy one of Dennis’ small and more recent 300B amps, you buy it with the new re-issues WE300B tubes, and you pay a bit extra for it and that’s how the designer likes it and wants it to be. Only comes with these tubes. Falling on my own sword, my existing mono amps were designed with specific power transformers and plate voltage (650v) and such to run TungSol KT150 output tubes in their optimum operating window. With my speakers, I enjoy both sets but also really like how KT120s sound in these amps. Go figure, lol. :) |
@dogearedaudio , I just took a look at the specs listed in the V12 owner’s manual and under feedback it says "zero." I do remember on another "switchable" amp I referred to, the Mesa Baron, on the back panel there were knobs to adjust feedback. I do not recall playing around with that particular adjustment during the two days that I auditioned that amp. As far as "tube roller’s dream", I should clarify that I dredged that up from my memory (which is USUALLY pretty good) but since I cannot find the blurb for whichever amp it was, I probably should not be providing a designer/manufacturer’s quotes that I cannot find to reference back to. |
I'll also add, in reference to "marketing," that IMHO a "tube-swappers dream" is definitely a marketing ploy, in that there is no way an amp like that can be truly optimized for each tube--as Ralph cogently points out. I don't believe there's any harm being done here, but anyone with a technical understanding of tube amplifiers (and I'm barely on the periphery of that circle) will shudder at the idea of a "one-size-fits-all" amplifier. ;-) |
When Ralph and others talk about "optimizing" the amp for UL or triode mode, one thing they’re talking about is maintaining appropriate levels of feedback for each topology (assuming the amp uses global negative feedback). Switching from triode to UL increases the gain of the output stage. Switching the other way, from UL to triode mode, will decrease the gain. This affects the level of negative feedback the input stage will see. Let’s say an amp is optimized for UL mode. If you switch to triode mode without changing anything else in the circuit, you *decrease* the amount of feedback. I think this is often why people hear an "improvement" when switching to triode mode. You’ve reduced the amount of feedback, resulting in a sense of "clarity." The problem is, you’ve also reduced the output impedance of the amp and thus reduced the amount of control the amp will exert on the speaker. So the other comment you get is that bass sounds "looser." Conversely, then, if an amp is optimized for triode mode, switching to UL mode will *increase* the feedback, and this may detrementally affect the stability of the amp as well as the sound. Some "switchable" amps will include a circuit that changes the feedback network for each mode, thus "optimizing" the feedback circuit either way. Whether the Cary amp does this or not is a question for Cary. But more to the OPs point, I have built amps both ways and have always preferred UL arrangements, sonically speaking. For those who want to understand a bit more about these different topologies, there is an excellent article geared for the layman here: https://oestex.com/tubes/ul.html And if you want to better understand the issues of "optimizing" an amplifier for various modes of operation, there’s a slightly more technical but superb overview offered by Dave Gillespie over at Audiokarma: https://audiokarma.org/forums/index.php?threads/regilding-the-gilded-lily-heaths-w-2m.767851/ Even if you stop after the short history of feedback amplifiers Dave provides, you’ll have a better grasp on how your own tube amplifiers work! The "Gilded Lily" Dave refers to is actually the amplifier I build for myself, using a modern copy of the original Peerless output transformer. This is where, for all intents and purpose, the "ultralinear" configuration is in every way preferable to triode operation--to my ears, anyway. ;-) |
@immatthewj I have a highly modified version of that amplifier and you can actually run combinations so for example. UL T UL
or
T UL T
(I used the amp for mastering for like four years and it's been sitting around so if someone is interested in a very nice example, I'm happy to sell it for a low cost deal) |
| Post removed |
Yes, @viridian , that was the Andy I was referring to. I enjoyed talking to him. Somehow he got on the subject of amps and the people who designed those amps that he didn’t care for. I won’t mention the names, but as we were talking about the amp, I said (and I do remember this word for word), "well, with the ____ I guess you get a lot of bang for the buck," and he replied with this kind of crazy laugh, "Hoo hoo hoo, or some times just a lot of bang!" And I remember him laughing some more. I don’t know why my memory for certain detail that doesn’t really benefit me in any way is so good, there are subjects that if I could have remembered them as well actually would have done me some good. |
@viridian , going on about 25 years ago I was buying some 12AU7s and 6922s for a preamp and amp I had, from Andy from Michigan. I probably made two or three phone calls to get the deal done, but in the course of those calls and conversations, Andy got off the topic of the tubes he was selling me and got to talking about a push pull amp he had recently restored or built (I wish I could still remember his exact words) but it was just a massive amount of class A wpc and when he said something about ". . . liquid fire coming out of the . . ." his tone of voice was almost orgasmic. Your post made me think of that. |
Thank you, @mulveling ! I have printed that as well. You asked for a diagram, @emergingsoul , it appears as if your wish has been granted a couple of times over. |
Regardless of what we each think of the term "UltraLinear", and its alignment on the good-versus-evil spectrum, I found the following diagram most succinct and easy to digest. It’s from a web search that hits el34world dot com. I assume the exact % of the UL taps’ positioning between outer taps and center tap is what’s at issue for "optimal" UL loading to a given output tube type - as circuit gurus like Ralph have pointed out. In pentode mode, both tube screens are tied to the center tap. In triode mode, each screen is tied to its own tube's anode plate. |
Yes, that is good. I am happy for you. But do not thank me; thank @atmasphere , as he posted the link a few posts before you said you would like to see a diagram. |
@emergingsoul wrote:
@atmasphere had previously written:
Again:
|
@immatthewj to answer your questions, after I recapped with those specific caps, and tried different designs of silver-over copper and more pure copper interconnects between source/preamp, preamp/amp, in my case, and for my particular speakers it was a dead-on match for my taste and ears. Not to over think this too much - I simply enjoyed those particular EL34s at the time. I figured out who was making them for Ruby lable brand [at that time] and the sound is where I wanted it to be, for me. So, I did not change to different tubes, although I came close to trying the reissue Genelex KT77s regardless of what anyone says about them, and the only reason I did not do that is I wanted to try separate tube mono blocks with a dedicated ultralinear circuit, double the transformers and power caps instead - with my speakers. Someone really wanted to buy my V12R at the time, and I struggled with selling it. I sold it only to try the monos as an experiment to run with my speakers [trying different amps with my own custom speakers I build/built], to install the exact same caps, and even better Nichicon power caps, and compared from there. Everything was how I needed it on the V12R. Had I kept it, I might have tried KT77s, and once again, keeping bias down to 200-210 per side, max. Sound find, tubes last longer. btw, I’m trying new input tubes this weekend, on the monos to try and model one input tube set I had on the Cary V12R, trying to bring the sound of the new amps closer to what I had with the V12R prior. Just testing, trying, listening. Oh, and No, I did not want to mess any more with the prior amp or the switchable circuit or trying to un-switch it, left it as is, it sounded great as-is, and decided to move to dedicated ultralinear mono amps. This has been yet another chapter of learning, upgrading, listening and has paid off nicely too fwiw. |
Thank you for that, @atmasphere , that's a nice nutshell that even a layman (at best) like me can just about grasp. |
@decooney , well, okay, but you did tell me that the only tube you wound up using were the Chinese Ruby branded EL34s, right? At the time I needed to retube that amp, Jim had a bunch of tubes listed in his inventory, including the GLs, which was part of the reason I contacted him for that retubing. (The last I read he said he was retiring and prior to that wasn't going to sell any tubes made in Russia anymore.) I was quite happy with those re-issued Tung Sols, and I will say that he gave me what seemed to be better personalized service than any other tube seller I've bought from (with the possible exception some NOS small tubes from of Andy from Michigan). Jim would send me an email every day or so telling me how the matching process was going, and as I recall, it took a few days or maybe even a week until he emailed me telling me he had two matched sex-tets ready to ship. A couple of years prior to that I had bought a dozen reissued Mullard EL34s from Viva, and it was wham, bam, thank you ma-am, and I had them, but they didn't run long at all (about two years) before the bias(es) started doing goofy things. Anyway, as another aside question, did you ever experiment with running your V12 configured in anything but FULL triode or FULL ultralinear? I mean like 1/3 + 2/3s? I know that the owner's manual says only run FULL one way or the other, and after the amp arrived to me I talked to Kirk Owens about it, and he didn't sound too crazy about the idea. . . . I know that the Stereophile review said it was okay, but I always felt better about defaulting to what the manufacturer said than what Stereophile said. I think I remember that I may have tried it for a real real short time, but I do not remember how I felt about it, meaning it was nothing at all close to as dramatic as the Mesa Baron was. |
@immatthewj You don't have to dig deep on this one. The Dynaco ST70 is arguably the most famous ultra-linear amplifier made although all the Dynaco amps were ultra-linear. |
