JimL

Although it is true that tube amps tend to have higher measured distortion than solid state, IMHO a properly designed tube amp or preamp has low enough distortion that it is inaudible. As with all class A amps, distortion generally goes down as levels decrease, so a typical listening levels which is usually at least 20-30 dB below clipping for electrostatic headphone amps, measured distortion tends to be buried by noise. I agree that tubes sound different from solid state, I just don't buy that tube sound is due to "pleasant" distortion.

So just to point out, the SRX Plus, based on the Stax SRX DIY design, has two differential amplification stages (first is cascode) and only one cap between input and output stages. The Plus refers to the use of MOSFET cascode constant current sources (CCS) replacing resistors to optimize circuit operation. The CCS set the circuit operating parameters but the tubes do all the signal amplification.

As I recall when Walt Jung was doing his series on current regulators in AudioXpress he found some noise problems with the 317 regulator, don't know if that applies to the LT1117.

Which is interesting because they claim it's a single-ended Class A, which is normally 25% efficiency at the most - should mean lots of heat. Supposedly the case is the heat sink, switching power supply (ugh - yeah, I know the SRM252 uses a switching PS but it's not in a $5000 system), "linear regulators" to the audio circuits, but I see no spec on their website about much power it draws from the AC line, which would allow a guess on how much voltage and current the amp draws. Stax solid state amps run about 30-55 watts from the wall (except for the 252) and they are push-pull so theoretically more efficient than single-ended. I doubt like hell that the Sonoma draws even that much or the case would be uncomfortably hot. Oh, yeah, and the max output is 145 VRMS, so a single ended supply of 0 and 450 volts should handle that. Plus MOSFET outputs, so easy to drive.

I believe he is referring to Tyll putting the Sonomas on his Wall of Fame.

Re: tube rolling, understood. For another tube flavor, you could try the SRX Plus. Can be built point-to-point, tubes handle all the signal chores, but with cascode MOSFET current sources, which greatly enhances the efficiency of the output tubes. Could probably drive dynamic headphones through a transformer, but I haven't really tried that.

Couple interesting things from Tyll's review. First, Sonoma says they DO use DSP for distortion correction on the single-ended driver. Second, the distortion measurements show approx 1% 2nd harmonic at 95 dB, 500 Hz, and 3% 2nd harmonic distortion at 105 dB. Also note the THD levels at 100 dB are in the 1-5% range from 200 Hz to 3 kHz - curiously they are lower below 200 Hz, where you would expect higher distortion due to increased diaphragm excursion - could that be where the DSP is being applied? By comparison the THD levels for the SR007 and SR009 are in the 0.1-0.5% range at 100 dB throughout the entire audio range.

One issue with replacing tubes for different music is that sooner or later, those tube sockets will loosen, causing noise problems. Tube sockets were not designed for repeatedly changing tubes.

Can you be more specific? Better in what way? Thanks in advance.

Yeah, my guess was a 6DJ8 diff amp into the rest of the SRM727. There is only a few volts across the DN2540 (the 10M90S/DN2540 cascode basically runs the DN2540 at constant voltage), so I wouldn't push the combination past +/-450, and would want to have it somewhat under to allow for over voltage on the AC line. Anyway, the difference between, say +/-425 volts and +/-500 volts is less than 1.5 dB in ultimate output.

Gee, and that took less than the 3 years of development time that Stax says they used for the T8000. Now the next step is to change to GG EL-34 outputs, and you'll have a modernized, simplified T2.

I was actually thinking of a used Bifrost Multibit. Pretty much all my digital is ripped off CDs.

Yeah, that was my concern too, not to mention if it blows a gasket you have to send it back.

Thanks. Any in particular? I'm looking to upgrade over a Schiit Modi that I got as part of a deal - mostly to get a Stax LNS.

Does that listening experience include the DACs or just the amps? Tried goggling computeraudio and all that pops up is computer audiophile, which I assume is a completely different website. Is there a link you could post?

Well, Heinlein's version is definitely before Buckley, but I believe Buckley said it before Hanlon, as he was referring to the John Birch Society and that ilk, back in the 60s. Not sure I can find a citation for Buckley, though.

Cool! I'll have to remember that.

Actually, Tyll addresses that in his post, he thinks it was just lucky or unlucky that he got that pair, as it was in a group that got sent out to different reviewers. Personally, I'll go with that unless/until there is more evidence to the contrary. As the late William F. Buckley used to say with regard to the John Birch conspiracy types, never underestimate the power of stupidity. But to raise another issue, on SBAF they discussed Focal's warranty policy, and apparently Focal do not carry over warranties to a second buyer. That could/should impact on the resale value for those who try to sell off after reading Tyll's post.

Well, Tyll posted earlier today the results of measuring seven pairs of Utopia headphones, and it turns out that they all measure pretty close - except for his review pair, which was about 5 dB lower from about 2.5-10 kHz. http://www.innerfidelity.com/content/seven-focal-utopias-measured#PYM2m50L5BvTDLGH.97 I listened briefly to a pair driven by a HeadAmp Gilmore Lite at a meet in Albuquerque, and it sounded tonally closer to a SR009 than an SR007 MkII.

Excellent!

I think we're going around in circles on this. Here's my argument. We can roughly model the roll-off at higher frequencies as a capacitor. So let's say the BJT current source is equivalent to a 300 kilohm resistor and a parallel 2 pf cap, whereas the MOSFET current source is like a 15 megohm resistor and a parallel 2 pf cap. The MOSFET impedance will start to bend over earlier in the audio spectrum because its DC resistance is higher to begin with, and the BJT will look flatter over the audio spectrum. At the highest frequencies they will both look like a 2 pf cap so they will eventually meet. BUT, within the audio spectrum, the BJT will burn up more signal current because it has a lower resistance, so the output device will have to work harder with the BJT than the MOSFET, and therefore will cause more distortion due to the output device working harder. Again, if we model the headphone as a resistor and capacitor in parallel (the resistor being there to model the fact that we're using energy to make sound), then the output device sees the headphone load (resistor and capacitor) and the current source (resistor and capacitor) all in parallel. The amount of signal current it has to generate depends on the the resistors in parallel and the capacitors in parallel. The higher the signal current generated the higher the distortion. If the current source "resistance" is higher it wastes less signal current in it. If the current source "capacitance" is small, the amount of signal current it uses to swing the capacitance doesn't really differ much from that needed for the headphone alone. The cascoding of the MOSFETs ensures that this extra capacitance is on the order of a few pf, much less than that of the headphones alone, so it really doesn't make much difference. The ideal current source has infinite impedance, and wastes no signal current, so it should be sonically invisible. My gut instinct is that the less signal current it uses, the less sonic imprint the current source should have. A flatter impedance may result in a more consistent sonic imprint, but a higher impedance should result in a lesser sonic imprint, so my argument is, higher is better.

So basically very little advantage over the cascode current source alone. My measurement of the cascode using a 10M90S upper and DN2540 lower gave somewhat better results that a dual DN2540 cascode, I wonder if the simulation would also show better results.

So the T2 current source calculates out to an effective impedance of around 280 kilohms, the cascoded DN2540 to around 14 megohms. Wonder how good the performance would be if you replaced the stacked pnp/resistor combo of the T2 with the cascode MOSFET current source feeding D3/D4?

Photo please!! Sorry for the duplicates!

One correction. I thought that they would be running more power since the consumption was listed as 95 watts, but I see that that spec includes the "optional slot", without it the power consumption is listed as 58 watts. The 727 consumes 46 watts, figure 2-3 watts for the 6922 tube heaters, so another couple watts per side for the output stage.