Tube for a shunt voltage regulator

[Wavebourn]

...or take one FET, one Zener, and one resistor. No FET and resistor needed if the current is in the range of specs for a zener. Or take a P-channel FET and mount it directly on a chassis to eliminate heatsink.

[Geek]

Originally posted by SY


As long as the voltage is below 250-275V. Otherwise, it won't work.

Right. In that case, tie the screen to something lower. Anode can go anywhere (almost).

[JohnAtwood]

I've been using tube-type shunt regulators for several years, and have had good sonic results with them. The basic circuit is a high-transconductance pentode (typically a video amplifier tube) using its own gain to regulate. A filtered negative reference voltage is used. The regulation is not high, but if it is only used for one stage, never shared between channels, the results are good. With a little tweaking, the impedance vs frequency can be made essentially flat.

I've built this circuit with 6CL6, 6AG7, PL83, and the pentode section of a 6CX8. It would have better regulation with super-high transconductance tubes such as the 12GN7 or 7788, but I am somewhat allergic to frame-grid tubes in my circuits.

Here is a somewhat simplified version of the 2A3 amplifier circuit that I wrote about in Vacuum Tube Valley issue number 15. It is the one with RF-powered filaments. It is also one of the best-sounding amps I've built.

Simplified 2A3 amp

The top part of the schematic is a basic three-stage SE amplifier using a 5879, 2E24, and 2A3 (all pentodes triode connected). The 2.5V filament supply for the 2A3 is returned to ground through a current-sensing resistor. The 12AU7 is a cathode follower that is used to filter any hum or noise out of the -150V source. It also helps isolate this reference voltage between the two channels of the stereo amp. The 2.2K 10W resistor feeding the shunt regulator sets the gain of the regulator. If a higher voltage is available, a larger dropping resistor can be used, thus increasing the gain. I've tried using a choke here, but have run into stability problems. The 0.22uF cap across the 365K resistor allows the regulator to have full gain across the audio band. The 6CL6 screen is bypassed by a big capacitor. If you want even better stability at DC, a VR-tube could be used here.

- John Atwood

[gingertube]

Jarthel,
I finish a 4 week stint in Cairns as support engineer on the Navy's Laser Airborne Depth Sounder today and fly home to Adelaide tomorrow. As far as I know I'm about the only guy in Adelaide who has the full Vaccuum Tube Valley collection so if you want to reference John's article give me a buzz Saturday or after (82693539).
I also recall that there was a series of articles on the TubeCad webpages about suitabilty of various tubes in voltage regulator circuits. They may have been talking about series regulators rather than shunt (don't recall) BUT any tube suitable as a series regulator device is also suitable as a shunt regulator device. Do a search on that web site.
Cheers,
Ian

[ray_moth]

Here is the article on pass tubes