[philipp]

I've been thinking about what Frank wrote a while back about tubes handling overload much better than transistors and its audible effect on the reproduction of music (Frank, feel free to correct my memory if it's faulty).

I was wondering if the benefits of the Ultra hybrid approach are most significant at the amplifier-speaker point or the source-preamp point. Or, to put the question more bluntly: which would move the speakers with a higher quality recreation of the source, an Omegastar pre/Ultra amp or an Ultra pre/Omegastar amp?

I'd also love for Frank or someone else at AVA to elaborate on this topic of feedback and how overload occurs at different points in the component chain. I'm asking mostly on an objective level, since I've read plenty about the subjective impression of tubes vs. solid state.

Thanks!

[Feanor]

I've been thinking about what Frank wrote a while back about tubes handling overload much better than transistors and its audible effect on the reproduction of music ...

I was wondering if the benefits of the Ultra hybrid approach are most significant at the amplifier-speaker point or the source-preamp point.  ...

On the face of it, I'd guess you'd need tubes where ever there is a feed back loop, but what do I know?  Frank, enlighten us.

[WEEZ]

I'll take a stab at this..

From a purely subjective stand point, I prefer the use of tubes closer to the source (in the pre-amp). YMMV.

I think if you ask Frank which Ultra component to buy first..it would be the pre-amp; but I could be wrong.  

WEEZ

[Bill Baker]

In my opinion, the more glowing glass the better. If I had to choose in steps, I would have to agree with WEEZ and have at least the preamp tubes in the beginning.
 Depending on synergy and obviously personal preferences, a solid state preamp with a tube amp can provide exceptional results.

[pacifico]

funny..Iwas just thinking about a prima luna pre amp for my NAD C372 and wondering if it was woth it. this will likely be my only purchase for the year in terms of audio equipment. Next year I will replace my speakers, $500 paradigm monitors. think gallo satellite speakers until I get a biger place. not sure if I should do speakers first or preamp. Spacewise the preamp would be easy and I really havent settled on  a speaker choice yet either. Still waiting for that next great affordable product with WAF (like the gallo stuff in general).

[avahifi]

My choice would be for tubes in the preamplifier first as that is where any harshness would be generated, if, and only if the tube preamp or hybrid preamp has been designed to eliminate some of tube circuits usual major flaws, so that the tube unit does not actually make things worse.

The basic engineering purpose of using the tube is that the summing node of input and feedback on a reasonable tube circuit will have in excess of 200 volts of headroom before overload.  A typical solid state device circuit will have 0.2 volts or much less headroom before the feedback loop clips and fails, all other things being equal.

Of course all things are not equal.  The tube has weakness in having a high output impedance, low drive current, and significant load sensativity depending up the circuit design.  That means tubes don't like to drive real world loads unless you give them lots of help.  So, all other things being equal a tube circuit with high overload capability but bad drive capability might sound better or worse than a solid state circuilt with bad feedback loop overload capability but good drive capability.  Actually, we suspect that most of the purple prose about tube sound and solid state sound over the years is simply jillions of efforts to actually described these effects as they relate to the listener musically, without the listener really knowing what is causing the effects he is hearing.  Much like the six blind men and the elephant, multiplied times ???

Anyway, we think we are getting a pretty good handle on how to use a tube for its superior overload capability but without the tube getting into drive current and impedance issues that would detract from its performance.  We do this with our Ultra circuits with a patented transimpedance amplifier design, using a small signal power mos-fet as part of the vacuum tube, providing a two piece device (along with some passive parts) that has high feedback overload immunity, low output impedance, high output drive current, and no load sensativity at all.

To make the new Transcendence Eight all tube preamp work as amazingly as it does, the approach was to provide each half of each tube with its own isolated high voltage high speed analog power supply regulator so that the supply impedance is a small fraction of an ohm broadband and there is no interaction between sections through the power supply at all. It appears to have given the tubes "a big shot of steroids", and we think the T8 will hit home runs for you even driving difficult solid state amplifier input impedances.

Both of our approaches seem to work very well, but of course the T8 still has a bit of that "tube like" character in its sound. Is it right or wrong, I just don't know, but it sure sounds lovely.

Frank Van Alstine

[philipp]

I appreciate your reply, Frank. There's a lot of information and opinion out there on the web but it's difficult to put it all into perspective. I'm only making the effort because I'm saving up for a stereo system that's going to cost me considerable money and stay with me for a very long time. Listening sessions are informative but so is knowledge of the issues surrounding audio reproduction.

Along with your earlier comments on the topic, Stan White's article about his Powertron amplifier inspired me to ask you for an elaboration on this topic. I'll quote something from Stan's article:

Knowledge of audio tube amplifiers has had a similar history.  What were seen as facts early on were subsequently discarded as flawed.  When transistor amplifiers came on board, tube amplifiers were discarded as outdated.  There was some basis for this.  When tested with resistive loads and short leads between the load and amplifier, transistor amps tested superior in distortion, frequency response and transient response.  The testing was flawed.  The assumption that tests performed with resisters were applicable to loudspeakers was an error.  Loudspeakers do not present a resistive load to an amplifier.  Loudspeakers are very complex impedance loads.  Tube amplifiers handle such loads better than transistor amplifiers.  It took kids, listening to live music to figure out that the engineers were wrong---tubes sound better than transistors. With that discovery, tube amplifiers were gained a new life.

Now, this may or may not be accurate but it got me wondering about a common thread I see in conventional audiophile wisdom that a tube pre and a solid state amp are a dandy combo. Stan talks extensively about the imperative need for excellent transient response in order to recreate the sound of real music. He also insists that only tubes are up to the task. Does this jibe with your research and experience? It seems your comments about tubes' superior headroom and overload capability would bear that out but I'd like to know.

[WEEZ]

With all due respect to Frank, I think there's more to it than than j-u-s-t a feedback loop...

(here's a theory):

Vacuum tubes in a pre-amplifier drive a steady load impedence (the amplifier input z) and (unless you use goofy wires) a mild capacitance.

Power amplifiers drive a much more difficult load (the loudspeaker) where the impedence varies with frequency (sometimes wildly! ) and also exhibits a more reactive load (wire inductance; wire capacitance; crossover components which have usually an inductor and a capacitor, etc., etc.).

The argument as to whether one topology sounds better than another will likely go on forever. I've owned shitty sounding amps and pre-amps that were tubed. And shitty sounding components that were solid state. (Fortunately, I've owned good sounding equipment of both types too   )

Tubes are said to be more linear than transistors. Don't know if that's really true- but I think it is. The curcuits in tubed equipment are usually much simpler and some use little or no feedback at all. The expense in tubed equipment seems to be output transformers (in power amps)..and the general low volume demand for tubed equipment vs. solid state.

The bottom line is that excellent equipment exists of both types. A 'perfect' tubed amp would sound the same as a 'perfect' solid state amp.

(if you find one for cheap, let me know   )

What about it , Frank?

WEEZ

[JoshK]

We do this with our Ultra circuits with a patented transimpedance amplifier design, using a small signal power mos-fet as part of the vacuum tube, providing a two piece device (along with some passive parts) that has high feedback overload immunity, low output impedance, high output drive current, and no load sensativity at all.

I don't know what your patented circuit does but can't you do all that with cascoding a tube with a mosfet, so the tube is "standing on the shoulders" of the mosfet?  Gives you relatively high drive current, low output impedance and the tube provides your gain (transconductance) which I am assuming you are referring to as "high feedback overload immunity" since you have a lot of gain to work with with most tubes.