Building a single ended amplifier using the 211 tube is not, for me, just a “more powerful variant” of the usual triode projects. It is a different path altogether, with precise rules, real compromises, and very high stakes. The 211 (VT-4C) is a historic tube, fascinating and demanding, capable of enormous sonic scale and dynamics that, if handled correctly, can challenge many more modern solutions. But precisely because it is a “large” tube, it demands to be treated with respect: high voltages, proper insulation and safety margins, serious drive capability, transformers up to the task, and a power supply that does not collapse when things get difficult.

This article documents an old SB-LAB project dating back to 2017, completed after a long gestation period of almost ten months, including calculations, prototyping, transformer design, and mechanical construction. The photos you see therefore belong to that specific period and to that exact version. If I were to rebuild these monoblocks today, I would not make them identical. The core concept would remain the same, but several details would be revised and improved to reflect everything I have learned over the years, both electrically and mechanically, including internal layout and optimization of grounding paths and wiring. This is exactly the beauty of “important” projects: they are never truly finished, they are milestones that teach you something, and over time they show you what you would do better today.

After almost ten months of development, I finally brought these two single ended monoblocks with 211 tubes to life. The project was anything but simple. I wanted to use the 211 properly, and by “properly” I mean that a tube of this size cannot be reduced to delivering 10 watts, as many do simply because they are unable to design a proper driver stage. According to the datasheet, the 211 requires and expects mixed A1+A2 operation, with both negative and positive grid operation. If you disagree with this, my suggestion is to use an 845 instead. For this reason, weak driver stages and interstage transformer coupling were ruled out from the start, as they tend to smear the sound, limit bandwidth, and prevent proper control when the grid swings positive.

When discussing the 211, many people focus immediately on anode voltage and output power, but the truly critical point, the one that separates a “clever” design from a well executed one, is the driver stage. The reason is simple: the 211 is not a tube that is satisfied with a modest voltage swing. If you want it to operate in its natural territory, you must provide wide excursions and, above all, maintain linearity when the load becomes brutal, that is when the grid starts drawing current. That is where many circuits collapse, compress, distort, or become unstable. The goal of this project was very clear: to control the 211 even when it enters A2 operation, without losing grip, without smearing the upper range, and without letting the low end fall apart when current demand becomes real.

The project started on paper, and I had to think carefully about the schematic because a very large voltage swing was required, one that remained linear into a strongly non linear load represented by the G1 grid of the 211. This grid draws virtually 0 mA over the entire negative swing, then increases up to a maximum of 160 mA during positive swing. The behavior of my circuit is shown in the graph below, where the signal waveform is shown in blue and remains linear, while the grid current of the 211 is shown in green and is clearly non linear. This circuit delivers 25 watts RMS with 850 volts of anode voltage into a 9k load. It would be possible to increase output power to about 35 watts RMS by raising the supply to 1000 or 1200 volts, but I chose not to, because this would have required series connected capacitors in the power supply, which in my opinion could have compromised sound quality. The current supply uses a voltage doubler followed by an LC filter with a 200 uF, 1000 volt polypropylene capacitor.

Here lies one of the key points. As long as the grid of the 211 remains negative, it is almost “invisible” to the driver stage, and many amplifiers deliberately stop there to avoid complications. But if you truly want to use the 211 with authority, you must accept that in A2 operation the grid becomes a current drawing load, and a highly non linear one. This means the driver must not only supply voltage, but also current, and it must do so without being pulled out of linear operation. In other words, it is not enough to simply reach the required voltage, it must reach it properly, with margin and control, and without letting distortion run away.

I chose 850 volts and a 9k load because this represented a well reasoned balance between power, electrical stress, insulation, long term reliability, and power supply quality. Yes, increasing the anode voltage to 1000 or 1200 volts would increase power, but at that point many things change: clearances, component availability, transient management, and above all you enter a territory where the power supply must be even more monolithic to avoid betraying the design. At that stage, I preferred to stop where I could guarantee a truly solid, high quality supply, without compromises such as series electrolytics which, although manageable, were not what I wanted in a project of this kind.

The graph visually summarizes the goal: keeping the signal waveform clean even while grid current becomes a dominant factor. In listening terms, this translates into the absence of compression on peaks, better control of dynamic swings, and that sense of authority which, in well designed single ended amplifiers, does not depend on nominal power but on how the circuit behaves when the music becomes demanding.

After finalizing the circuit from a theoretical standpoint, I built the transformers and moved on to the CAD design of the amplifiers.

In a project like this, the magnetic components are not an “accessory”, they are an integral part of the result. Output transformer, filter chokes, power transformer, and the management of parasitic capacitances and leakage all determine bandwidth, stability, noise floor, and transient behavior. CAD design in this context is not just about fitting everything inside a chassis. It is about deciding signal paths, distances, shielding, core orientation, and management of stray fields. These are all aspects that, if underestimated, will come back to haunt you later, when you are already at the end of the project and chasing hum or unwanted coupling.

And then came the physical construction. Mechanical execution, especially with such high voltages, is a chapter of its own. This is not only about aesthetics. Wiring, insulation, choice of grommets, distances between points at different potentials, mounting methods, ventilation, and accessibility for maintenance are all critical. A 211 amplifier must not only look good and sound good, it must also be built in a way that avoids unpleasant surprises over time, especially considering the energy involved and the transients during power up and power down.

The following images, as mentioned earlier, belong to the original project. They document a real journey, with the solutions of that specific moment in time. If I were to revisit this design today, some choices would certainly be refined, not because they were “wrong”, but because accumulated experience inevitably leads you to optimize details that may not have been priorities back then. This applies to internal layout, certain grounding paths, management of parasitic capacitances, serviceability, and also conservative choices in components and wiring, in order to make the project even more robust without altering its core character.

Instrumental Measurements
Power 25 watts RMS
Bandwidth at 1 watt: 15 Hz – 50 kHz -3 dB
Bandwidth at 25 watts: 30 Hz – 22 kHz -3 dB
THD at 1 watt: 0.3%
THD at 6 watts: 1.7%
THD at 15 watts: 3.8%
Damping factor: 5
Rout: 1.6 ohm

Instrumental measurements must be interpreted in the proper context. In a single ended triode amplifier of this type, nominal power does not tell the whole story. What matters is how distortion rises with power, how bandwidth behaves near the limits, and above all stability and transient response. The damping factor here is not particularly high, and it does not have to be, but it is sufficient to provide credible control with many loudspeakers.

Tubes used 7
1x 12AT7 / ECC81: input stage, excluded from the global NFB loop
2x 6J5: voltage gain
1x EL34 triode connected: follower, driver, bias control, DC coupled to the output tube
1x 6080: voltage regulator for the EL34 anode supply
1x PC900: error amplifier driving the 6080
1x 211 / VT-4C: output tube

The tube chain was chosen with a very specific goal: to reach the 211 with a driver stage capable of wide voltage swing and current delivery, while maintaining a controllable and stable structure. Excluding the input stage from the global feedback loop helps preserve its behavior and sonic character, preventing feedback from indiscriminately affecting everything. The pair of 6J5 tubes provides voltage gain in a clean and predictable way, while the triode connected EL34 does the heavy lifting: following, driving, and managing bias via DC coupling to the output tube. In a design that must enter A2 operation, this is not a minor detail. Coupling and operating point stability must remain solid even when the 211 grid starts drawing significant current.

The stabilization section using the 6080 and PC900 was designed to keep the driver supply consistent, because driver linearity and stability also depend on how much its high voltage moves with load variations. In practice, it is not enough to design a good driver, you must also prevent the power supply from turning it into something else when you turn up the volume.

Spectrum at 1 watt

Spectrum at 6 watts

Spectrum at 15 watts

At high power levels the spectrum shows how well the system remains under control near saturation and how credibly A2 operation is handled. This is where the value of a serious driver becomes clear. When the circuit is designed for this region, there is no sudden change in character, no nervous compression, and the music retains a natural quality that is often missing in designs that avoid A2 out of fear of complexity.

Square wave at 100 Hz

Square wave at 1 kHz

Square wave at 10 kHz

At 10 kHz the square wave is a severe test, because it highlights real bandwidth limits, parasitic capacitances, leakage, and compensation. A clean result here is not just nice to look at. It is often correlated with more believable harmonic decay, sharper transients, and a clearer soundstage, without resorting to an over analytical sound that later leads to listening fatigue.

How does it sound. The first impression was that these were truly big amplifiers, and after long term listening they absolutely are. In my controlled environment, after many tests and recordings, I can only say they are even better than they initially appeared. They deliver an imposing yet always refined mid and high range, and a bass that is solid and statuesque. They are melodic and rich in detail, and with certain recordings you can hear the pianist breathe, the guitarist resting their hands on the instrument. I have no doubt placing them among the best amplifiers I have ever had the chance to build and listen to. It is hard to explain, but when you encounter amplifications like this, you realize that everything you once thought was immense was, in reality, only big. The effort required for such a demanding build was absolutely worth it.

With hindsight, I would add one important observation. The sonic character of a 211 single ended amplifier is not just about power and presence. It is about authority, continuity, and naturalness when the music moves from pianissimo to fortissimo without changing character. With these monoblocks, the sensation is one of reserve, not so much in watts, but in control. The mid high range is wide and luminous, yet never sharp, and when the recording is good, a microdynamic quality emerges that lets you perceive gestures, breaths, nuances, and space in a very physical way. The bass, when the loudspeaker allows it, is not only extended, it is firm, articulated, and readable. This kind of solidity does not come from emphasis or excess, but from a circuit that does not lose its grip when the demands become serious.

If I were to revisit this design today, I would work even more on certain construction and optimization details, because that is where the real progress is made. But one thing would not change: the core idea of treating the 211 with the respect it deserves, with a driver worthy of A2 operation, transformers designed for this mission, and a power supply that is not an accessory, but an integral part of the sound.