300b lovers

Ralph, you’re the acknowledged OTL expert. How does an OTL amp work without feedback? I’d like to know. The last I checked, tubes designed for series regulator use like the 6080, 6AS7, or the Russian 6C33C have a Zout on the cathode side somewhere around 100 ohms. I’m not an expert on this class of tubes, but I wouldn’t expect any tube to have a Zout in the 8 ohm or less range.

(For the reader following along: a Zout of 8 ohms gives a damping factor of 1, a Zout of 2 ohms a damping factor of 4, and so on. Zout in the 0.1 ohm range, typical of transistor amps, gives a damping factor of 80. It should be mentioned that damping factors much greater than this are kind of pointless, since the speaker cable, which is in series with the DCR of the lowpass inductor in the crossover, will typically have a DC resistance of 0.1 ohm or so. I don’t expect speaker cables to use superconductors any time soon, so we’re stuck with copper or silver at room temperatures.)

I’ll admit this is kind of an idle curiosity since I will never design an OTL amplifier. The amps I’m interested in use transformers to solve various circuit-design problems. But I’m always curious how things work, whether solid-state, vacuum tube, or Class D, or hybrids of all three (like ZOTL’s).

@lynn_olson You might want to study the Wiggins Circlotron patent for the answer, or look at the setup/spec sheet of one of ElectroVoice's amps that used the Wiggins patent. Here is an example. 

I recently restored one of these (A20C model) that I had inherited about 20 years ago. Like you, I was very curious about it since EV and the patent claimed no zero crossing artifact if the amp is biased class B. I wanted to see if that was true. It is! With a sine wave input even at 1 milliWatt the output is excellent. Its first Watt performance (and sound) is really quite good.  

In the Features paragraph at the link you'll see that the primary impedance of the output transformer is 1/4 that of conventional PP amplifier circuits. That applies to an OTL in the same manner. Since the Circlotron allows for a fully differential/balanced Voltage amplifier and drive circuit, its symmetry means that distortion will be lower. So you can run it without feedback (you can do that with a totem pole circuit as well but its not nearly as easy).

We were the first to manufacture a Circlotron OTL and as such they were also the first really reliable OTLs as well because they were unconditionally stable with any output or load condition. I think that is due to the symmetry of the design and the lack of feedback, which can be a destabilizing factor especially if the amp operates near its phase margin limits (which can change with the load).

The Futterman OTL circuit employed positive feedback to equalize the drive to the top and bottom tubes of the totem pole output circuit, so that design seems to need negative feedback to really function right.

We've run feedback as well but do it with balanced feedback loops mixed with the audio in a manner similar to an opamp. This is because the cathode circuit is unavailable (there's a CCS circuit there after all); after doing that it took a while but I finally realized there's an advantage doing it that way because the feedback signal is not distorted by the tube as it would if the feedback were received in the cathode. In this manner there is less higher ordered harmonic and intermodulation generation caused by the feedback (meaning it sounds more relaxed).

I am convinced feedback to the cathode of the input tube is a reason why feedback has gained a bad rap in high end audio. Funny how traditions like that hang on for so long.

That’s a subtle aspect of feedback theory that is often overlooked. The summing node must be distortionless, and also free of overshoot or slewing artifacts. Applying input signal to a grid, and feedback to the cathode, impresses tube distortion (of the input tube) on the entire feedback loop.

Of course, we can get in trouble with a differential circuit as well, since the summing node is spread across two grid/cathode circuits, not one. Assuming perfect match, the nonlinearities of the pair should cancel. In practice, we should expect 2~3% gain mismatch in the differential pair, so there will be a residue of mismatch and associated nonlinearity, but not much.

If I understand your previous post correctly, the Zout of a Circlotron amplifier will be high, maybe in the 10 ohm or higher range. Or there is local feedback somewhere in there, reducing it to "normal" levels of an ohm or less.

As you can see, Don and I are taking a brute-force approach to distortion reduction.  Drivers that run at 40 mA per tube, and balanced deep Class A operation. The even-order distortion of the driver section is cancelled in the primary of the interstage transformer, reducing driver distortion even further. This presents a highly symmetric drive to the paired 300B grids.

It may strike some readers as weird that Don and I are taking a minimalist approach to a PP amplifier, more like a SET than a typical PP. The signal path is simple:

Optional SE/Balanced input transformer -> Balanced 6SN7 -> Interstage #1 -> Balanced KT88’s in triode in Class A mode -> Interstage #2 -> Balanced 300B’s in Class A mode -> Monolith Output Transformer -> Loudspeaker.

Similar to a fancy Japanese-style SET done twice. The hard part are the interstage transformers, which are not easy to find off-the-shelf, and ideally should be designed for the specific tubes in the circuit.

Don and I tried many variants to get rid of Interstage #1, since that operates at the highest impedances (thanks to the 6SN7) and is hardest (almost impossible) to design. The first version was simple RC coupling to the driver stage, with 6V6’s as driver running at 24 mA each. It sounded pretty decent and measured quite well, as you would expect from RC coupling. But Don wanted more ... so we tried paired dynamic loads for the 6SN7, which reduced its distortion about three times and was noticeably clearer sounding. But ... and there’s always a but, isn’t there ... there was just a faint trace of solid-state coloration from the current sources. Not much, but there. This was a version the folks at Spatial really liked, and we built some of the early "shoebox" format amplifiers in this format.

My grumble was the insanely long "burn-in" time for the super-deluxe coupling caps between the input and driver tubes. 50~100 hours. I am wary of burn-in times this long, since I suspect the part might be chemically unstable and never actually settle down, always changing its sonic presentation over time.

We tried another version, replacing the current sources with 100 Hy custom inductors. Was it any better? I’d say different, with deeper tone colors, no solid-state coloration at all, but losing a bit of snap and attack compared to the current sources. All expected ... no transistor sound, but unwanted stray capacitance in the high-value inductors, and question marks about linearity in the bass region.

Our transformer designer saw the high plate impedance of a balanced 6SN7 as a personal challenge, and insisted he could design an interstage transformer just for us. I was skeptical it could be done ... I only knew of one other transformer that could do that, from Tribute in Europe, and that thing was quite large and a one-off project done for the Karna amplifier. It might still be available from Tribute, for all I know. Tribute transformers are pretty special and the equal of any Japanese confection.

But ... a couple months went by, and Don got a special care package from our transformer designer. It was a pair of special custom Interstage #1 transformers, made just for us. Very simple wiring, compared to all the other variants we had tried, just six wires from the primary and secondary, as simple as it gets. In it went.

And that was the winner. Burn-in time was much faster, an hour or two, no super-exotic caps with their temperamental burn-in times, the circuit was way simpler, and the depth of tone colors was much deeper than any of the other versions. Obviously better than the inductor or current-source loads for the 6SN7, and no stinking coupling cap.

In fact, the IT coupling revealed the pretty obvious "cap coloration" of all the many coupling caps we had tried. The brutal fact is all caps have a sound, even though they measure extremely well. Worse, there is zero subjective correlation between sonics and the standard Df and Da measurements, not to mention they really do change sonics quite a lot over the first hundred hours. And nothing about that shift over the first hundred hours is measurable ... at all!

You hear the cap coloration by its absence. Go to true direct coupling, or IT coupling, and it is gone. After you’ve evaluated the sonics of twenty or more different brands and types of caps, you realize they all share a common coloration, with some much worse than others, but they all have something going on that is veiling the sound. And whatever it is, it can’t be measured with the instruments we have now.

I have no idea what it is. And I am taking about a coloration much more obvious than a cable swap between components. I’ll go out on a limb and say caps are the dominant sound of most tube electronics, whether preamps or power amps. Swap the coupling caps, and you have a brand-new amplifier, with brand-new colorations of its own. It makes you realize why there is a whole industry of guitar-amp tuning ... the tone coloration combinations are limitless.

But ... get rid of the caps, all of them in the signal path, and you are in a different sonic world. Do transformers have a sound? Well, if they are cheap transformers, yes they do. Murky and dull. The best, though, are very clear and have no cap sound at all.

That snap and clarity is what solid-state enthusiasts crave. Hey, I get it! The best solid state is very good and is free of cap coloration, as it should be. But ... solid state has its own sound, too. Unfortunately. Traces of it intrude on tube circuits if they have dynamic loads, or regulators that are not well designed. Don and I have gone to some trouble to use B+ regulators that are well behaved and have very high noise rejection (130 dB) and very low stray capacitance (using cascoded MOSFETs).

So I’ll go out on a limb (again) and say most audiophiles have never heard tube electronics without coupling caps. Ever. They think tube electronics just "sound that way" and tolerate burn-in times of hundreds of hours, with the sound changing hour by hour, sometimes better, sometimes worse, and sometimes just weird.

This plagues exhibitors at hifi shows, too, because the transport process results in some of the caps needing a re-set at the show, so the sound on Friday and Saturday can be remarkably different. And the culprit? Not the cables. Not the speakers. The coupling caps buried deep in the circuit, temperature cycling up and down, and going through unknown electrochemical changes deep inside.

And guess what? Solid-state electronics have caps too, not in the direct signal path, but as filter caps, typically large-value electrolytics. And you can bet they have a sound, too.

One variant we did not try was a center-tapped inductor to load the 6SN7 plates, and direct coupling between the input and driver tubes. That requires a high B+ voltage for the driver, but that’s no problem when the input + driver have their own power supply that’s fully isolated from the output section power supply.

It’s a good question if this is better or worse than a special interstage transformer. The center-tapped inductor shares many of the design challenges of an interstage transformer without necessarily having any advantages. This falls into the category of "build it and see". A minor challenge with a 450 to 500 volt regulated B+ supply, but nothing impossible.

Another similar option, instead of a center-tapped inductor, are paired current sources and direct coupling between input and driver tubes. This has an ugly disadvantage that a small DC imbalance in the input tube turns into a 5 to 10 volt offset for the driver tubes, which is intolerable because the driver section is then grossly imbalanced with different operating points for each driver tube.

To prevent this, it would require something like an opto-isolated DC servo circuit to balance the plate voltages of the input tube. Doable but kind of nasty, adding a lot of pointless complexity that adds nothing directly to the sound ... basically, another point of failure. The alternative would be manual adjustments (with limited range) on each current source and a meter so the user could hand-adjust DC balance.

Alternatively, the center-tapped inductor, because it has a moderate DCR of a few hundred ohms, limits the amount of plate-to-plate DC imbalance to a volt or so. However ... small as that is, that’s more than an interstage transformer, where the output going to the driver grids is always perfectly balanced, with zero DC offset from grid-to-grid.

What the interstage transformer does is offload circuit complexity to the ingenuity of the transformer designer. The circuit schematic looks simple, but what’s going on inside that transformer is very complex, requiring sophisticated modeling tools to fully understand.

I am still puzzled why capacitors measure so well, yet are so audibly colored. And why on Earth do they require hundreds of hours of active operation, not just polarized but signal going through them, to finally stabilize sonically? What’s going on inside them? My only guess is a (very) slow electrochemical process that subtly alters the dielectric properties of the plastic film.