Tubes vs. SS

[mikebse2a3]

How is saying that one prefers solid state equipment over tubes bashing anyone? Why so defensive?

Read my post carefully Don because I never said that nor do I believe that..!

Then who hijacked your keyboard and wrote this?

"What I will never understand is why some people think that because they heard one or even a few examples of either technology (Tube or SS) in systems and rooms that are unique unto themselves believe that their limited experience entitles them to bash another persons opinions of the value of their chosen technology."

So what is it? Have you forgotten what you wrote on Dec 13 already?

Whats funny to me is how because I have pointed out IMHO that Tube Technology has much to offer and shouldn't be dismissed because of it's age as some here have insinuated is seen by you as being defensive.

What is funny to me is how I said that I prefer solid state equipment, and even though a person is listening through tube equipment the recording was likely made with solid state recording equipment, could be construed as me being defensive.

Don I know exactly what I've posted and your taking things the wrong way and have even assumed evidently that it was directed at you so insults aside Don I hope you reread my post later when your calm and then maybe you will understand what my points have been.

By the way I didn't say you were being defensive you asked that of me and I answered you.

miketn

[Don Richard]

What's funny to me is how, because I have pointed out IMHO, that Tube Technology has much to offer and shouldn't be dismissed because of it's age as some here have insinuated, is seen by you as being defensive.

Punctuation corrected. Easier to understand now.

[pzannucci]

SS would seem to have less interaction between components and frequency variance during the recording process. Now depending on what part of the recording process and what instruments / vocals you are recording, perhaps tube preamps or front end might be good for that particular instance. Some rounding and smoothing many times occur.

The SS vs Tube debate hits home harder when you talk about playback. The variance and rounding of the sound with tubes are a lot of times more what people want. It sounds more analog vs digital (kind of). You can associate the proper words with the proper device - cold, sterile, lean, lifeless or euphonic.

These are usually words that going in order describe Class D -> typical SS -> Tubes.

Boxx hit it with a similar analogy where you record a CD to tape and the tape sounds better. I used to do this with a very good cassette deck. Apparently my ears liked some of the distortions introduced in that chain because we all know what is broken in the first recording, can only get worse when re-recorded. The recent magazine articles point out similar results with digital and tape. Digital sounded more like the mic feed but the tape sounded better - huh?? We can't explain what we like in each but what can stay truer to the original intent during recording until we play it back is best. WE individually decide what we like is the best.

Same with SS vs. tube. No clear answer just enjoy you distortions.

[ClaudeJ1]

Same with SS vs. tube. No clear answer just enjoy you distortions.

My conclusion to this entire thread would be a modification of the above quote:

"Enjoy your PLEASANT distortions." (2nd harmonics from tubes, and a heavily "dithered" noise floor from tape).

[DrWho]

The myth that tubes have 2nd order distortion and SS has 3rd order is rather misleading and distracts from the true science of the differences in my opinionopinion. The circuit topology dominates the sonic characteristic followed by the load curve of the active device. Then you have the sonic characteristic of the output transformer. In other words, both active devices are capable of having any distortion characteristic and you almost always have a little of everything. Somewhat related, I find it kind of ironic that McIntosh uses an output transformer on their solid state designs...

[Chris A]

I like this article on part of that subject (from Mr. Pass):

http://forums.klipsch.com/forums/storage/4/1746312/Audio%20distortion%20and%20feedback%20-%20Nelson%20Pass.pdf

[Ski Bum]

I find that I personally like the extreme ends of the spectrum: clean, linear, unclipped ss power on the one hand, and full blown tube processing of SETs on the other. PP is not tubey enough for me, I guess. The strange thing to me is that the approach with no redeeming technical merit achieves more subjectively pleasing results. Count me in with the distortion lovers, I guess.

[Chris A]

I like this article on part of that subject (from Mr. Pass): http://forums.klipsch.com/forums/storage/4/1746312/Audio%20distortion%20and%20feedback%20-%20Nelson%20Pass.pdf

I do understand why some folks like SETs from reading the Nelson Pass article linked above: they are getting higher overall distortion, but it is all second harmonic distortion. If this harmonic distortion level isn't greater than ~1%, the SETs will probably reproduce string orchestras and violins fairly well (but not as well as amplifiers with lower harmonic distortion without feedback being used).

The issue still remains with an output transformer being used to lower the amplifier's output impedance. If the amp's resulting output impedance is still greater than ~1/10th the lowest input impedance of your speakers, then that "tube sound" it is coming at a high price--via interactions with the speakers' drivers to produce non-flat output (among other issues). Klipsch Heritage speakers, in particular, have widely varying input impedances over the audible frequency range, and therefore interact complexly to high output impedance amplifiers.

One of those other issues is that Klipsch speakers are also so efficient that they also feedback the room's early reflections into the amplifier's output terminals (i.e., the speakers are acting as good microphones), which produces audible reverberation effects NOT found on the original source material--if the amplifier's output impedance is high.

Bottom line: I recommend using the 4 ohm tap setting...

Chris

[Ski Bum]

Chris, I totally agree with the points you're making. I would love to try one of those First Watt jobs you've been recommending. Ironically, it was a Pass amp that sent me on an inquisitive trip to the SET zone in the first place. I'll be back to class A ss eventually, as I would like the low level goodness without the SETbacks. In the meantime, I still find SETs undeniably enjoyable, warts and all. They are among the most intriguing sound processors I've ever messed around with.

[DrWho]

Pass says this in his article:

If you see a curve with distortion level climbing as the output goes down, it implies crossover distortion caused by the gap between the two push-pull gain elements, and this implies high order harmonics.

My experience has been that the curve rising as you go lower in amplitude is because the noise floor of the amplifier starts to dominate the distortion measurement. If the two amps he's comparing in that example have the same noise floor voltage, then I might agree about the distortion claims....but the THD (total harmonic distortion AND NOISE) results he's showing sound about right for the noise floor of a class B.

Another thing to keep in mind about the masking of low order harmonic signals is that it assumes "western music" or music where 2nd and 3rd harmonic tones are common to the typical instrumentation and tonal structures. Shift to a different kind of instrument or musical genre and you may not experience the same level of masking...it's always going to come back to the source material in this regard.

[Chris A]

The 300B SET monoblocks that I used to drive my TADs (which are nominally 16 ohm drivers) had their output transformer settings on 16 ohms. The sound was terrible - so bad, that I put those two units into storage. Then I tried the 4 ohm settings: better but still they had issues like the ones that I detailed above.

I understand about the First Watt amps - the price tag for Nelson Pass-built units is a little high. But I've got to tell you that it's like the difference between the living and the dead--and I'm comparing it to the Crown D75A that was driving them. No contest.

So I'm not as hard over as one might think on SETs. but the issue that I had was that the user directions that came with the SETs were...well...not good. I recommend using the 4 ohm settings on SETs as a default (i.e., the lowest impedance setting available) in order to minimize their output impedance, thus minimizing the issues that I talk about above.

I believe that I do "get it" with regard to the second harmonic-only thing: this is a really big deal. I'd actually recommend a SET amplifier for discerning ears if the budget is inflexible for FET amps of the Nelson Pass variety, and I'd recommend even higher quality tube-type amplifiers with lower output impedance, but with minimal amplifier stages, but the down side is maintenance costs/time, and warm-up times.

Chris

[DrWho]

Bottom line: I recommend using the 4 ohm tap setting...

Ditto that....the only time to move up a tap setting is when you can't get enough power into a higher impedance load.

Although I haven't wrapped my head around the non-linearities inherent to an audio transformer (both distortion and frequency response). You will move the low frequency or high frequency corner around as you change the tap settings - and you'll also change how much distortion the core is adding. It is almost impossible to get a passive device to behave well at "high power" levels - even with perfect components (basically anything over 100mW). Transformers are about one to two orders of magnitude away from being a perfect engineering solution....which is annoying as heck as an audio designer (transformers have a crap ton of benefits to them).

Btw, I just wanted to mention that I really highly regard Nelson Pass's work - that dude is an analog genious.

[Daddy Dee]

The myth that tubes have 2nd order distortion and SS has 3rd order is rather misleading and distracts from the true science of the differences in my opinionopinion. The circuit topology dominates the sonic characteristic followed by the load curve of the active device. Then you have the sonic characteristic of the output transformer. In other words, both active devices are capable of having any distortion characteristic and you almost always have a little of everything. Somewhat related, I find it kind of ironic that McIntosh uses an output transformer on their solid state designs...

DrWho,

Could you say more about the Mac design with an OPT in SS designs? Are there any other SS amps which follow this design? Also, do you think there is a substantial difference in sound based on this strategy?

[Chris A]

It is almost impossible to get a passive device to behave well at "high power" levels - even with perfect components (basically anything over 100mW). Transformers are about one to two orders of magnitude away from being a perfect engineering solution....which is annoying as heck as an audio designer (transformers have a crap ton of benefits to them).

This is also the problem that I have with transformers and passive crossovers. Pass also made the same comments when he decided to build higher quality volume controls.

...that dude is an analog genius...

As you can probably tell, his article on amplifier feedback really opened my eyes/gave me answers to questions that I've had for a long, long time...and he didn't even talk about psychoacoustics []

Chris

[tube fanatic]

To add to this fun discussion, take careful note of what Bob Carver sought to accomplish in the design of this amplifier:

http://www.theabsolutesound.com/articles/carver-black-beauty-305-power-amplifier/?utm_campaign=Newsletter&utm_medium=email&utm_source=email-52

[Schu]

< .5% distortion?

[Chris A]

Just some idle points based on first reactions:

I don't believe that I'd use Leonard Cohen's voice as a standard for auditioning anything (YMMV). And that's a lot of power for guys on this forum - too much, in fact.

And I don't believe that I'd ever go back to using the passive crossovers in loudspeakers (i.e., I bi-amp and tri-amp) - so mentioning amplifier bass performance traded against midrange and hf doesn't have a lot of relevance.

What bothers me about this guy doing the reviewing are his speakers. Here's a clue: they're not horn-loaded and they're not very efficient--meaning that that have will tons of modulation distortion in their bass and midrange drivers...at least.

Here's another clue: four pairs of his speakers are planars--which have serious in-room placement issues that require me to ask for a picture of his listening room. He doesn't describe the room that they're in, and we all here have learned that this is a pretty important point, i.e., artto's pictures of a couple of magazine critics' listening rooms showing that they really don't know what they're doing for a living.

I owned Magnepan MG-IIIa's in the 80s and I drove them with a Carver M1.5t (i.e., a Mark Levinson ML-2 double). You can make all the negative comments you want, but I was suitably impressed. I've always been impressed by Carver's approach: I think that he's been extremely audacious in an industry that is very set in its ways (i.e., the industry tends to look backward much more than forward). Carver's discussion of amplifier reverb is where I found the clue to the question: "why do so many people like SETs?". The answer is: high output impedance. One thing that I found is that Carver isn't/wasn't sensitive to speaker modulation distortion issues which is something that set PWK apart from the crowd--a long time ago.

Chris

[DrWho]

The myth that tubes have 2nd order distortion and SS has 3rd order is rather misleading and distracts from the true science of the differences in my opinionopinion. The circuit topology dominates the sonic characteristic followed by the load curve of the active device. Then you have the sonic characteristic of the output transformer. In other words, both active devices are capable of having any distortion characteristic and you almost always have a little of everything. Somewhat related, I find it kind of ironic that McIntosh uses an output transformer on their solid state designs...

DrWho,

Could you say more about the Mac design with an OPT in SS designs? Are there any other SS amps which follow this design? Also, do you think there is a substantial difference in sound based on this strategy?

They actually use an autoformer (which is a device that should be familiar to the Klipsch crowd). I think the main reason they do this is to maintain a similar sonic signature between their SS and Tube designs. In other words, the output autoformer is definitely dominating the sonic signature of their well designed amplifiers. It's a necessity for the Tube amps, and I think they have a strong business strategy where they're reusing as many parts as possible by making a high voltage SS design that can plug into the same power supply architecture. McIntosh has also gone to great lengths to maximize the performance of their autoformers as well - although I seriously dislike that they choose to mystify a process that is well known in other industries (random wiring of the autoformer instead of fixed wiring so as to reduce the parasitic effects). They also aren't using litz wire which is really the cats meow in these types of applications - albeit way more expensive (though I gotta wonder since they're hand winding).

Btw, the power supply shouldn't be ignored in these amplifier comparisons - and McIntosh could make a strong argument that capacitors behave better when the less current is flowing through them. Running a really large voltage rail reduces the effects of non-linear capacitor impedance because it reduces the current in your output stage. That said, capacitor non-linearities can be reduced by simply having more of them - which at audio frequencies doesn't create any extra concern that one might otherwise have with high speed digital designs. They're also cheaper and smaller to implement than the autoformer, which is why I think it's a business strategy and voicing concern for McIntosh to feed a large voltage SS stage into the same autoformers. I would not use this approach doing a ground up analog design from scratch, but then I don't have a sonic signature and company image to maintain either...

I'm not sure of any other amplifier manufacturers using an output autoformer, nor anyone else doing a solid state into any form of a transformer. I'm sure it exists though.

[Daddy Dee]

Thanks Mike. Interesting.

[tube fanatic]

< .5% distortion?

You missed the point! I was referring to this extract from the article:

http://www.theabsolutesound.com/articles/carver-black-beauty-305-power-amplifier/?utm_campaign=Newsletter&utm_medium=email&utm_source=email-52

Obviously, this is not the ideal amp for a “transparency to sources”
listener like me, in that it makes virtually everything, no matter how
recorded, sound some shade of dark and ravishing. And yet...and yet. The
Black Beauty has the kind of midrange resolution and presence that make
voices and instruments sound not merely beautiful but quite real and
quite “there.”

In the user manual, Carver says unequivocally that dark and gorgeous was precisely the kind of performance he wanted from
his Black Beauties. He was looking to provide “a warm, rich sound with a
sumptuous soundstage, great front-to-back depth of field, and very
tight, pin-point imaging within that larger acoustic,” and in doing this
he was looking to please the “serious music lover.”

So, even at this price point, Carver intentionally chose to create a "type" of sound which is not necessarily true to the original recording.

Maynard