pentode – wauwatosa tube factory

One of my first tube projects was a TubeLab SSE. In fact, I still use this amp as a reference whenever I build something new.

George, of TubeLab, is notorious for torturing tubes and generally just blowing crap up in his experiments. He is also a fervent supporter of DIY and frequent poster on diyaudio.com.

A couple of my recent posts have looked at ways of applying local feedback to pentodes in order to force them into more triode-like behavior. It’s funny how experiments and research in the DIY hobby community converge. Here’s yet one more example.

Yesterday George posted some exciting but cryptic experiences with a new design here. The challenge, in his own words:

[With triode strapping] the pentode takes on triode like qualities with the associated triode disadvantages, most notably the inability to pull its plate down near the cathode voltage thus limiting the available power output. Another issue that needs to be overcome is the screen grid voltage limitations of most TV sweep tubes. Wire them as a triode, and most will eventually blow up when left alone idling which is worse case for a class A amp (maximum dissipation).

This is a great summary of the limits of single-ended triode amplification. Power is limited to single-digits by tube perveance, voltage maximums, and the ability to dissipate heat. Pentodes, able to swing outputs much closer to zero and operated with fixed screen voltages, go a long way towards solving the conundrum. But the trade-off is linearity and output impedance (which is why SETs are popular in the first place).

George goes on to tease his new design:

I arrived at a new topology that I can’t find anywhere in recorded vacuum tube history…..yes, there are several close similarities, but this is truly unique…I called this topology the Composite Electron Device for lack of a better name, since it is a composite of a vacuum tube pentode, a mosfet, and a hand full of discrete parts to create triode like curves.

We don’t get any schematics (yet), but he finally gets to the measurements pudding:

THD was 0.197% at 100 mW, rising to 0.235% at 1 watt. 5 watts brings 0.662%, 10 watts 1.61%, with 2.48% at 15 watts and 4.04% at 20 watts clipping sets in at 20.8 watts where the THD hits 5%.

So what is it, this single-ended not a tridoe? We know there’s some local feedback going on, but George says this is actually something new under the sun. That doesn’t happen everyday in tube land, so I’m following this one with a lot of interest.

By the way, if you want to build a traditional SET and prefer a PCB, take a look at George’s TSEII or original SSE designs!

Oops, I bought another pair of output transformers. This iron is 60W rated push-pull with a 6k6 primary, so the natural pairing would be 6L6GC.

I’ve used the 6L6GC in the past (see the Luciernaga). Because it is common in guitar amplifiers, it’s an easy to find tube both new and vintage. While it’s no power-house in triode mode, the 6L6GC is quite capable in pentode or ultralinear.

Operating the output tubes as pentodes means higher output impedance and distortion. The Quality Amplifier from a couple weeks ago avoids the issue entirely by operating the outputs as triodes. The other approach, which is actually more common today, is to sacrifice some of the circuit’s gain to bring output impedance and distortion back down. That’s called (negative) feedback.

The question is how do we want to apply it?

This post won’t get into the nitty gritty of feedback (that deserves its own page on the website), but there are generally two prevailing approaches. Global feedback is what we see most often; this takes a signal from the output of the amp and wraps it all the way back to the input. The Williamson amplifier is the quintessential example of this.

The second approach to feedback is to use local loops. These affect a circuit just like global negative feedback, but are isolated to just a stage or two. Local feedback, because it involves fewer stages and phase shift, is more stable than global negative feedback. That means they’re generally simpler to employ.

The circuit above is a variation of what seems to be unofficially referred to as an E-Linear stage. Feedback from the output transformer primary is applied via the ultralinear taps directly to the load resistor for the input stage. This local feedback drastically reduces the output impedance of the 6L6GC.

The input stage is commonly a pentode because the high plate resistance is a benefit here to applying feedback. In this case, the input stage is a hybrid cascode, which still has a high “plate resistance” owning to the MOSFET upper device. That also gives us more options for the lower triode tube.

I like the simplicity of this circuit quite a lot. In Class AB, it looks like a good 25W should also be available with a pretty modest B+ (or a little less in Class A). Seeing as how I’ve got the iron on hand, I hope to give this one a test at some point!