Affordable, higher-power SET monoblocks

Bob Reynolds · « **Reply #20 on:** 10 Jun 2006, 05:24 am »

Quote from: jeffreybehr

Bob, did it occur to you that we buy them to LISTEN to them, NOT to measure them? SETs have a well-deserved reputation of being probably the most-musical-sounding amps available. Connected to a reasonably sensitive (or even some NOT-reasonably-sensitive) speakers, SETs do a fabulous job of reproducing the MUSIC.

Of course you're welcome to your opinion, but I suggest you be a little more open-minded.

Well sure, but I prefer to listen to the source and not the distortion created by the playback system. I own both tubes (SET) and solid state amps.

It has nothing to do with being open-minded. A well designed SET will not appear to be broken on the test bench. Read the letter associated with the Sterophile review and you'll see that the original designer of the circuit topology was well aware of it's short comings.

The Stereophile archives have plenty of tube amps that measure quite well within their intended range of use. The debate between tube and solid state will go on forever, but today, well designed examples of each sound very similar.

It was not my intent to offend you or your choice/preference of amplifier. I was simply pointing out that that particular amp produces loads of distortion when producing more than 1 watt.

-- Bob

JoshK · « **Reply #21 on:** 10 Jun 2006, 06:58 pm »

Just to note, my opinion about said tubes and implementation is more from reading and learning rather than have tried all variations, but I do trust the knowledge learned by some who do have such experience. It also makes a lot of sense and correlates with why some tube amps sound a lot better than others. I just didn't want to misrepresent my own experience.

I also don't have a lot of experience on 9pins versus octals, only some. However, I tend not to like big warm euphonic sound. It sounds good with some music and makes a lot of wart ridden recordings sound good but it doesn't make all music sound good and is annoying on many well recorded albums, IMO. With DIY, I can build both very easily and not too expensively.

I still tend to gravitate to truth and low distortion and therefore lean toward's Bob's take on this subject. I also firmly believe that distortion profile matters as much as total distortion figures, if not more. This is why negative feedback can be a bad thing if lots of care isn't taken when using it.

PaulFolbrecht · « **Reply #22 on:** 10 Jun 2006, 07:10 pm »

I'm no DIY expert, but all of the good-sounding tube amps I've heard have been little or NO neg. feedback designs.

JoshK · « **Reply #23 on:** 10 Jun 2006, 07:30 pm »

There is a big reason for that, from what I've read, from those who seem to know what they are talking about. The idea is that many use neg feedback as a crutch to make a non-linear circuit, linear. But this sort of implementation tends to create a lot of higher harmonic garbage, which is far more offensive to our ears that even high low order harmonic distortion.

So, the idea behind using NFB is that the circuit should be able to stand on its own without it and should be very linear before NFB. Then using a *little bit* of NFB can be beneficial in some circumstances. Some technologies may be able to get away with using more NFB, but I think this requires a hell of a lot more knowledge than a lot of audio designers display, and requires really fast circuits, which most tube circuits aren't. In the end I think it is easier to design for not using any to get good results.

This puts a huge burden on the quality of PSU in most circumstances and that each load is more ideal and less reactive. This agrees with what Dmason said about power supplies. Remember in a SET amp, the PSU is IN the signal path, so any electrolytics in the PSU are electrolytics in the signal path. Guys like Thoersten work really hard to design without 'lytics in the power supplies of their amps.

In this way, "digital" amps are like SETs in that the PSU is of very high priority. In building the UcD amps, a huge amount of care in the PSU and especially grounding is essential to gaining the most from them.

Dmason · « **Reply #24 on:** 10 Jun 2006, 09:50 pm »

That's right. In a SET circuit, the output tube drives the output transformers. They are key. They also are the biggest obstacle, and by far the biggest expense, which is exactly what makes the T amps so sexy to me: they do away with all of that completely. I think the main reason the 2A3 is such a legendary favorite is the fact that it is also about the simplest circuit. Not much to muck things up. Take away more components by using 24awg magnet line, wired directly from the tx taps to your widerange driver contacts gets you even closer to the music. Use outstanding widerangers in proven loads gets you closer. This approach trumps the mondo boat anchor, crossover laden multi driver approach everytime, and wins converts. I believe this is exactly why there is currently a groundswell of interest in this approach. It is also far less expensive overall.

PaulFolbrecht · « **Reply #25 on:** 11 Jun 2006, 12:15 am »

Well, I'm not impressed with switching amps so far. Well, I am: they sound much better than class A/AB transister amps, especially for the money. But tubes still sound better to me. And I haven't listened seriously to SET amps yet.

Dmason · « **Reply #26 on:** 11 Jun 2006, 01:05 am »

Horses for courses. I prefer well implemented T amps to Push/Pull tubes, too much reassembly of sonics in PP for me, then PP, then sand, if forced to listen. Never liked the sound of transistor amps. Never. Using a good SET dual triode type pre amplifier with T amps is a truly stunning combo. One of the best compromises out there. Power AND glory.

PaulFolbrecht · « **Reply #27 on:** 11 Jun 2006, 02:22 pm »

Dmason,

I just finished auditioning a pair of NuForces with my PrimaLuna 3. It was pretty impressive, but I still liked my AES Six Pacs (PP, EL34, 0 feedback monoblocks) better. More air, better imaging, silkier mids. It was definitely *close*, though, and the NuForces were the best sand amps I've heard (they still use transistors, of course).

About SETs: you noted that in a SET the output tube(s) drive the output trannies directly. I knew that, but never thought about the fact that that's not the case in PP. I suppose that explains the generally greater impedence matching problems with SETs.

TheChairGuy · « **Reply #28 on:** 11 Jun 2006, 02:53 pm »

Damn, I'm actually following this semi-technical discussion. Thank you good Doctor D, Josh and Paul! Plenty of learning for me here in this topic.

Bob Reynolds · « **Reply #29 on:** 11 Jun 2006, 05:56 pm »

Regarding NFB... In an ideal world of reasonable speaker loads and never pushing the active power device out of its linear range, then NFB shouldn't be necessary. But, anybody and his brother can put drivers in a box and solder up a crossover, so the poor amp designer is caught. Also, especially with tubes, the linear range just does not produce enough power to be practical. It may be that an amp design uses no global negative feedback, but local negative feedback is almost always used. It's a tradeoff between building a practical amp that is stable into all sorts of crazy reactive loads and trying to satisfy the consumer that believes negative feedback is evil.

Regarding power supplies... It's easy to forget that the signal that is produced on the outputs of an amp is *NOT* the same signal at its inputs. The output signal is generated from the power supply and modulated by the input signal. In effect, an amp attempts to clone a larger version of the input signal. So, the power supply is critical in any type of amp.

Regarding class D amps... I may be way off base here, but it seems to me that they should appeal to those folks that like tube amps because they create tons of ultrasonic distortion. They can't be measured without the use of special filters because the noise swamps the testing equipment. Of course, all this noise is beyond the audio spectrum so it doesn't matter. Well, maybe so. Just like having super tweeters that range up to 40KHz should afford no benefits.

-- Bob

JoshK · « **Reply #30 on:** 11 Jun 2006, 06:04 pm »

I think most tube amps use lots of NFB to lower the output impedance (hence raise the dampening factor) so that the amp produces a flat frequency response into such a reactive load as Bob puts it. Many old skool triodes have reasonably low output impedance without it but virtually all pentodes have high output impedance so gNFB lowers the output impedance.

Its true as Bob said, that basically all amps, even ones stated as NO FEEDBACK are using local degenerative feedback.

tanchiro58 · « **Reply #31 on:** 14 Jun 2006, 10:19 pm »

Has anyone listened to the stereo 808 SET amp yet? The 808 output tubes have a midrange of 211 tubes and a bass of 845 tubes. The wattage is about 25W. The amp is custom built with a pair of interstage transformer, driver tubes are 8233s. This is the most powerful and musical SET amp I ever listened.

PaulFolbrecht · « **Reply #32 on:** 15 Jun 2006, 01:03 am »

What brand? I'm interested.