When experimenting with vacuum tube circuits on the workbench, one of the most useful tools to have is a reliable and adjustable anode power supply. Unfortunately many people end up improvising with makeshift supplies, old transformers, dropping resistors and other temporary solutions that may work, but are neither practical nor safe. After years of doing tests this way, I decided to build a small bench power supply specifically dedicated to vacuum tube experimentation.

The idea was to obtain a simple but robust instrument capable of delivering an adjustable anode voltage up to 350V, with overload protection and stability sufficient for use during circuit development and testing. At the same time I did not want a complicated design full of rare or difficult to source components. On the contrary, the goal was to build something that could use common parts, easily recovered even from old equipment.

The project presented here was born exactly with this philosophy: a tube regulated power supply intended as a laboratory tool for hobbyists and for anyone who enjoys experimenting with vacuum tube electronics. It is not meant to be elegant or minimalist, but rather practical, robust and tolerant of the inevitable mistakes that happen when testing circuits still under development.

Useful tool for experimentation

As I mentioned in my previous article about the transistor curve tracer, I owned an old Chinaglia oscilloscope with a CRT that was already exhausted and not worth repairing. Since I am the kind of person who never throws anything away, I immediately dismantled it to recover some components, mainly tubes and tube sockets. Then, while I was throwing the remaining parts into the box for disposal, I had the idea of salvaging the entire chassis and turning it into a small anode power supply that could be useful when experimenting with circuits on the bench, also replacing an old bulky power supply that I had built many years ago when I was still a beginner.

I started sketching a circuit that could work well with simple and inexpensive components. As the regulating tube I chose a 6GE5 (a very robust compactron TV horizontal output tube), and as the error amplifier a very common EF80. I also included, partly for the pleasure of using them, three 85A2 gas regulators: one stabilizes the screen voltage of the 6GE5 while the other two, connected in series, provide the two negative voltages required for the EF80 biasing. There is also a comparator circuit built around a TL081 opamp which monitors the current drawn from the power supply output and, in case of overload (adjustable), disconnects the screen grid supply of the 6GE5 forcing the tube into cutoff. A push button allows the normal operation of the power supply to be restored. As the title suggests, the output voltage can be freely adjusted from 0 up to 350V, while another potentiometer allows adjustment of the protection circuit sensitivity within a range between 5 and 50mA. I also added a 6.3V 1A AC output, protected simply by a fuse.

Someone might ask why I did not build it using solid state devices instead of tubes. The answer is simple: it is actually easier this way. Solid state designs require extremely fast protection circuits because transistors can fail in a fraction of a millisecond, while good old vacuum tubes are much more tolerant. If for a brief moment they deliver a bit more current than expected, they usually do not care at all. The second reason is that with tubes there is no need to search for and mount large heat sinks. And finally, the last reason is that I simply like tubes and I enjoy using them. Now let’s take a look at the schematic of the power supply, click to enlarge:

–> Download the schematic by clicking here alimentatore-6GE5.zip <–

The following photos show my construction…

In operation

Since the oscilloscope cabinet is almost completely sealed, in order to prevent the inside from turning into a small oven I drilled a hole on the side and installed a cooling fan (I will also need to add a grille so nobody can accidentally stick their fingers inside) and drilled several ventilation holes on the top above the power tube. In this way the fan pushes fresh air into the cabinet while the hot air escapes from the top.

What is the difference between a simply adjustable power supply and a stabilized one? On the internet you can find many schematics for anode power supplies, both tube and transistor based, but a large portion of them are actually only adjustable supplies. In these circuits there is a potentiometer that allows you to vary the output voltage, but the selected value does not remain truly stable. The voltage changes with the load and is also affected by the normal fluctuations of the mains supply. In practice this type of regulation is rather crude.

In a stabilized power supply the situation is different, because a circuit called an error amplifier is used. In my project this function is performed by the EF80. The role of the error amplifier is to continuously compare the output voltage with a reference voltage and correct it in real time so that the selected value remains constant regardless of variations in the mains supply or in the current drawn by the powered circuit.

However it is important to clarify a rather common misunderstanding: the presence of zener diodes or gas regulators does not automatically make a power supply stabilized. I have seen several DIY amplifier builds where, for example, a 6080 tube was used together with a gas regulator between grid and ground to supply a 300B. In theory it should behave as a regulator, but in practice the 6080 ends up providing only a “more or less” controlled voltage that continuously drifts. When you analyze the circuit behavior carefully you often realize that the entire stage could be replaced by a simple dropping resistor with practically the same result, sometimes even better.