Why Tube Amps Sound Different (and better) Than SS Amps
- Thread starter mep
- Start date
- Status
Good point. I need a new CD transport, with my old one dying, and I had been considering a used Esoteric P03. There are several reasons why I decided not to go with it, and weight was one of them. The 71 pounds were simply too impractical in every way (it's just a transport, for cryin' out loud).
Well now you have jumped to a couple of conclusions not warranted by what I have written.
Firstly, yes, I still would say a 1/3 improvement, and yes I consider that a substantial and large difference. I nor my acquaintances thought the Spectral was even a serious competitor to the VTL. It however wasn't boring, it wasn't dead, you are letting your stereotypes of SS clinical sound lead you to a wrong conclusion. The Spectral was a fine, and enjoyable amp, but once you heard it up against the VTL it was pretty much no contest. My experiment took awhile to process and re-orient my opinions about such things. But I didn't ditch the VTL's and keep the Spectral.
I experimented some more with the Spectral, loaned it around to get the handle on it in different systems, and sold it a few months later to purchase a Wadia DAC. I had the VTL 3 or 4 more years. Other circumstances caused changes in my system or I might still have it. On occasion I still wonder if I shouldn't have kept it as a pre-conditioner for larger amps. I eventually started doing Room Correction and found the benefits to that to allow much of the same musical enjoyment and more accuracy. So I wasn't crazy enough to sell something I enjoyed vastly more and keep something I didn't like as much just on a principle of accuracy. Others might make that choice, but not for music in my home thank you.
Now in my mind the ideal amp is straight wire with gain. However, I don't mind things that accidentally depart from accuracy if they trick you into better enjoyment. I still wouldn't call this inaccurate though lovely portrayal better as an amp. It suits my preference, and not all have the same preference. The accurate amp is better. Tubes might be preferred, just remember they aren't better in a technical sense. I have seen so much conjecture, discussion, ink spilled etc. about how tube amps measure worse and are superior. Attempts to make SS amps as good as tubes are. The idea we are measuring the wrong stuff because we don't measure what makes tubes superior etc. etc. All of that is a wild goose chase. The good SS amps are simply more transparent and accurate. The other preferences really should be done at pre-amp level or as signal pre-conditioning. Something like building a miniature push-pull triode as a pre-amp that you can load appropriately for the same effect as one possible example. Or DSP to convolve the sound of tubes painlessly. Tube amps simply have a smaller performance envelope than good SS. That is really all there is to it.
Further, at least back when I did this experiment of connecting amplifiers in series, not all SS amps were fully transparent. I did this with a handful of other amps. Some were clear, and some were an amalgamation of two sounds. You could hear the tube sound, but it was partly overlaid with other characteristics. Maybe someone with a nice current tube amp with a characteristic sound should try it with some good class D units and see what happens.
Not sure what baby you are talking about. I said the VTL wasn't transparent. I never said it was more than what it was which is these two amps in this particular case. Though I did try it with a few more SS amps at the time I haven't described those.
Both amps were loaded with an 8 ohm wire-wound power resistor. And yes that isn't quite the same as the complex impedance of an ESL. That should have been no issue with the Spectral, and the VTL had it windings connected to be optimum at 7.5 ohms. The resistors for dividing the voltage to feed to the next amp were large enough not to effect anything or reduce response in the high frequencies. If my memory is good I used 1.1 kohm Vishay and a smaller value to ground to create a unity gain output. So the loading by the dividing network was more than 1100 ohms, and the output impedance was less than 100 ohms. The interconnects were low capacitance and a bit short of two feet long.
same reason many men cant stay married to one woman. I like to play the field too, the safe (cheap?) way is with audio gear
. that said, when the synergy is strong with tubes and their matching speaker its special indeed. I just need to open the wallet wider, CLSs need lottsa tube watts to make 'em sing well.Mark, im looking forward to reading your review. keep us posted.
As briefly as I can, and I'll touch on the major points, as I'm waiting for a customer to arrive...
1. a good amplifier should not have a 'sound' --it should be a straight wire with gain. No particular signature, neutral.
2. vacuum tube sounds different for the following reasons, and depends also on speaker type:
a: Much lower damping factor. The back EMF isn't shunted by the amp's source impedance nearly to the degree that a SS amp will shunt. Bass reflex speakers will sound 'fattier' on vacuum tube amp than SS.
b: Back EMF generated by coasting voice coil isn't shunted, and, in the case of feedback network in the amp, this EMF voltage gets injected into the feedback network and reamplified a few microseconds later.
c. most vacuum tube amps are transformer coupled. Most modern transformers have fairly high distortion and limited bandwidth. With exception of a few transformer made from 1955 to 1966, most every 'boutique' transformer I've tested has been rather poor, not only at the low frequency extreme, but at high frequencies, more surprisingly. Take an old Peerless model 16432 output transformer, which goes from 37Hz to 40KHz at near full power, and far beyond those limits at reduced power, and compare it to transformers wound today and it's a different world. The art of transformer winding has been lost with the collective knowledge of UTC, Thordarson, etc.
Note: bass reflex is more dependent on amplifier electrical damping, as it lacks mechanical damping at resonance and at frequencies below the tuning frequency.
Back EMF is the electrical current generated by the coasting cone/voice coil in the magnetic field. It is an electric generator during this return to resting position phase.
That's a few of the major points off the top of my head. Got to run.
1. a good amplifier should not have a 'sound' --it should be a straight wire with gain. No particular signature, neutral.
2. vacuum tube sounds different for the following reasons, and depends also on speaker type:
a: Much lower damping factor. The back EMF isn't shunted by the amp's source impedance nearly to the degree that a SS amp will shunt. Bass reflex speakers will sound 'fattier' on vacuum tube amp than SS.
b: Back EMF generated by coasting voice coil isn't shunted, and, in the case of feedback network in the amp, this EMF voltage gets injected into the feedback network and reamplified a few microseconds later.
c. most vacuum tube amps are transformer coupled. Most modern transformers have fairly high distortion and limited bandwidth. With exception of a few transformer made from 1955 to 1966, most every 'boutique' transformer I've tested has been rather poor, not only at the low frequency extreme, but at high frequencies, more surprisingly. Take an old Peerless model 16432 output transformer, which goes from 37Hz to 40KHz at near full power, and far beyond those limits at reduced power, and compare it to transformers wound today and it's a different world. The art of transformer winding has been lost with the collective knowledge of UTC, Thordarson, etc.
Note: bass reflex is more dependent on amplifier electrical damping, as it lacks mechanical damping at resonance and at frequencies below the tuning frequency.
Back EMF is the electrical current generated by the coasting cone/voice coil in the magnetic field. It is an electric generator during this return to resting position phase.
That's a few of the major points off the top of my head. Got to run.
Accurate reproduction of the source is certainly relative, but it is not theoretical.
Tim
It's very different. "Perceptual coding" in digital is controllable, adaptable, reversible. It can, and most often is, used to correct effects that have taken the sound at the listener's ears further from reproduction of the recording. A deliberate alteration of the input, built into an amplifier which offers no ability to bi-pass or adjust that distortion, is exactly the opposite, in both intent and result.
Tim
I can't speak for the Spectral DMA-50, but the Spectral DMA-260 sounded just as non-clinical and exciting in my system as my own tube amps. Why? Because it sounded very similar, see my post # 22.
So how do you explain my findings about a Spectral SS amp sounding in my system very similar to my tube amp in post #22 (page 3 of thread)?
Because the measured and well-understood conclusions are not something that some tube lovers are willing to accept.
Tim
We've gone through the topic of this read a hundred times, and yet here we are. If not for rewind and restart, most of the board would vanish. Why do you wish to bar this particular topic from further discussion?
Tim
We should note that tube preamplifiers share similar sound characteristics as power amplifiers. And curiously they do not share most of the problems that have been pointed out before in power amplifiers. So, IMHO they should be much more adequate to debate the "SS" versus "tubes".
I have often said that such think as "tube sound" does no exist. We have sound characteristics of the implementation, and surely the preferred topologies are different with different type of devices. Each type has intrinsic problems when used to amplify signals, and designers must overcome them. However, good tube designs show an absence of "solid state flavor", characteristic of many early designs using this technology. The following statement is attributed to Nelson Pass "Pass said. "I seek to apply the sonic lessons taught by amplifiers which don't necessarily measure spectacularly well but sound very good. Triodes just happen to fall into that category. I would not see to duplicate triodes per se, except possibly as an exercise."
IMHO the variation of sound characteristics in each type does not allow a generalization of "tube sound" versus "solid state sound", unless we consider that the bass of a Krell amplifier is typical of the whole SS family, and the airiness and holography of an Atmasphere OTL is characteristic of all tubes. We all have a general feeling about tube or SS sound qualities, due to past experience, but not systematic enough to make science. And as the level of our debates is too basic, and does not exceed damping, THD figures or frequency response, the debates will be just a sharing of opinions and endless. I having followed similar debates in the DIYaudio forum years ago and they were much more technically oriented - although not much more conclusive ...
High-end has has changed a lot along three decades - it is not easy to make generalizations that can cover it all.
I have often said that such think as "tube sound" does no exist. We have sound characteristics of the implementation, and surely the preferred topologies are different with different type of devices. Each type has intrinsic problems when used to amplify signals, and designers must overcome them. However, good tube designs show an absence of "solid state flavor", characteristic of many early designs using this technology. The following statement is attributed to Nelson Pass "Pass said. "I seek to apply the sonic lessons taught by amplifiers which don't necessarily measure spectacularly well but sound very good. Triodes just happen to fall into that category. I would not see to duplicate triodes per se, except possibly as an exercise."
IMHO the variation of sound characteristics in each type does not allow a generalization of "tube sound" versus "solid state sound", unless we consider that the bass of a Krell amplifier is typical of the whole SS family, and the airiness and holography of an Atmasphere OTL is characteristic of all tubes. We all have a general feeling about tube or SS sound qualities, due to past experience, but not systematic enough to make science. And as the level of our debates is too basic, and does not exceed damping, THD figures or frequency response, the debates will be just a sharing of opinions and endless. I having followed similar debates in the DIYaudio forum years ago and they were much more technically oriented - although not much more conclusive ...
High-end has has changed a lot along three decades - it is not easy to make generalizations that can cover it all.
The main problem is that tubes and solid state have become stereotyped over the decades. Tubes are smooth but rolled off on top and flabby below and SS is grainy, edgy but has great bass. It is exactly the same stereotype as analog vs digital. The same stereotyping is even applied to brands, feedback, name it. How many of each topology has to go the opposite way before the stereotyping ends? Welcome to the twilight zone.
Now if we all read the actual OP carefully, the title was used only to get attention and the post itself was one person's personal theory.
Now if we all read the actual OP carefully, the title was used only to get attention and the post itself was one person's personal theory.
You really can't blame the manufacturer, though. Putting a 1 lb transport in a 69 lb box can raise the margins by 1000%.
Tim
What do you know about TEAC/Esoteric's VRDS-NEO and what the engineering goals of that device are Tim?
- Status
Similar threads
