I have been researching triode emulation. It is the dream of many experimenters and developers to design a solid-state device that exactly replicates the characteristics of a vacuum tube triode. Run a search on “triode emulation” and you will get 53,700 results.
Undaunted, I am trying various circuits that claim to emulate the triode. I’m just curious.
On the surface, a vacuum tube triode diagram looks an awful lot like a transistor.
There are three elements in each: A Plate (Anode), Grid, and Cathode for the tube, and a Drain, Gate, and Source for the transistor. What happens when each of these devices amplifies a signal is where there are differences.
An audio signal is nicely amplified by both devices, if the gain is low and there is no distortion of the signal. The way in which they distort, however, is very different.
One of the last bastions of the tube triode is the guitar amplifier. The reason has everything to do with the way tubes distort when they are overdriven. The distortion produced is pleasing to the ear; full of harmonics and overtones. The solid-state device? Not so much.
The relationship of plate voltage of a tube triode to the grid is defined by the three-halves power law, which produces both even and odd harmonics. As the signal level increases, the 2nd harmonic rises, followed by the 3rd, and then higher orders.
On the silicon side of things, the JFET drain current is related to the gate-to-source voltage by a square law. This means that only 2nd harmonic distortion is produced.
For many decades, people have been trying to make solid-state amplifiers sound like vacuum tube amplifiers, and while there have been considerable gains made (no pun intended), the net result is that the characteristic sound of a tube triode is still easily differentiated from a solid-state device.
The first real commercial solid-state replacement for a tube triode was the Fetron.
Fetrons were manufactured by Teledyne Semiconductor starting in 1967, and they were a plug-in replacement for 9-pin tubes like the 12AX7. They were used in a variety of applications, but for our purposes, the Fetron’s key application was in the Mesa Boogie Mark series of guitar amplifiers. Mesa used the Fetron in the V1 input stage; there was a switch on the amp to select either Fetron or 12AX7 operation. The reason Mesa used the Fetron was due to microphonics problems with 12AX7s in the high-gain circuit. Mesa discontinued the use of the Fetron when production 12AX7 tube quality improved and microphonics were no longer an issue.
An important paper was written by Dimitri Danyuk, titled “Triode Emulator” (from the 116th Audio Engineering Society Convention, May 2004 in Berlin, Germany). Basically, it said that a JFET could approximate the 3/2 law of the triode by precise calculation of the source resistor and elimination of the bypass capacitor.
The drawback is that the Danyuk emulation only occurs at a single, narrow operating point of the JFET.
Jump ahead to 2017, and we have AMT Electronics 12AX7WS, which is a “solid-state analogue” of the vacuum tube triode. Also, Roberts Audio Technologies “Retrovalve”, which is a 12AX7 emulator, 100% analog, and is available in three different gains.
I suspect, although I don’t know, that these devices have their start with the Danyuk paper. The Roberts patents indicate that the Retrovalve incorporates an analog-to-digital (ADC) circuit, a digital signal processor (DSP) and a digital-to-analog converter (DAC). The DSP output provides real-time signal processing. So, they are centered around a JFET, with a lot of signal processing going on.
Reviews of the Retrovalve are, as you would expect, mixed. There are those who really like the sound produced, and those who say that they work well but don’t sound like a vacuum tube. So, the jury is still out, and the effort to emulate a tube triode continues…
