DIY Tube Tester: Design, Modifications, and Results

This article is about building a DIY tube tester and matcher, one of the most important tools for any tube electronics enthusiast.
My first tester was an EMC 203, and I still use it a lot for quick tube checks. It has a great range of heater voltages starting from 2V all the way up to 50V. I was happy with this tester until I realized that it is literally just a tester, not a meter. Being an emission-type tester, the EMC can give you an impression of how well the cathode releases electrons, but it gives you no clue about how well a tube actually amplifies a signal. 

Another drawback of this tester type is its relatively low test voltage, much lower than the voltage the tube will experience during operation. This becomes even worse when dealing with guitar amplifiers, as they usually run tubes close to or beyond their maximum ratings.

Looking elsewhere, I found out that I needed a tester capable of setting an operating point on a tube and measuring critical characteristics like plate current, Gm (mutual conductance), or the amplification factor. There are tons of great testers on the market, both vintage and modern, but the prices shocked me a bit. That is how I came across this DIY tester project and started building my own.

Planning the Build and Component Selection

When planning the build, I opted for the following components:

• High-Voltage Transformer: A separate Hammond 166G44 transformer (44VAC with a center tap). This allows for a slightly higher B+ voltage of up to 400V DC.
• Filament Transformer: A separate Hammond 166L12B transformer (12VAC with a center tap). This provides the two most popular heater voltages as well as the bias supply. Both transformers cost around $30 each on Digikey.
• Digital Meters: Very nice YB5135Di digital meters with isolated power were found on AliExpress, allowing me to feed them all from a single power supply. For the plate current, I selected a 0.2A meter. The bias voltage meter has a 200V maximum rating. Both feature a four-digit display, providing 0.1-step resolution.
• Connectors: I liked the original design's concept of using two rows of banana jacks, so I bought isolated jacks along with isolated male connectors.
• Capacitors: With less powerful high voltage transformer I opted for 330uf capacitors vs original 470uf. Nice ones from JCCON were found on AliExpress for about $20 for all

Custom Modifications and Modernizations

Compared to the original design, I made the following modifications:

• Dual-Range Bias: Two bias voltage ranges (0–20V and 0–60V) provide a broader range and better accuracy. I also used a precision multi-turn potentiometer for granular bias control.
• Fuse Protection: I installed fuses into the high-voltage circuits for additional protection against tube shorts and runaway currents.
• Neon Bulb Circuit: The neon short-test bulb was equipped with dropping resistors and a "Test/Power" switch, allowing the bulb to double as a pilot light once the short-test is complete.
• Dedicated Plate Voltmeter: Finally, I installed a third 500V meter to monitor the exact plate voltage on the tube under test. This required adding an extra diode to prevent the "short-test" voltage from showing up on the meter. I'm using the same meter type, with a 0-500V range this one has a 1V resolution, accurate enough to measure plate voltage. All three meters cost me around $40.

Enclosure and Woodworking

All components were installed into a custom wooden enclosure, with the controls and connectors mounted on an aluminum faceplate. For the enclosure, I bought a few poplar boards from a local hardware store. Two nice 6" x 12" aluminum sheets were found on Amazon for about $23. For the finish, I debated between Tried & True Wood Finish and Odie's Oil, ultimately opting for the former.

First Results and suggestions for version 2

Using the new tester, I measured my entire tube inventory and successfully matched several pairs. Overall, I am thrilled with the results, but intensive use revealed a few areas for improvement:

• Gas Test Sensitivity: With a 1MΩ resistor, some tubes arc during the gas test, destroying the tube and blowing the fuses. The 12AU7 family suffers from this particularly badly. Dropping this resistor to say 500K may make this a bit better, have not tried yet though.
• Fuse Position: To swap a fuse, I currently have to open the bottom lid, which not the most inconvenient thing. I wish I had mounted the fuse holders directly on the front panel.
• Double Triode Testing: The original design included a dedicated 12A*7 socket and a switch to route plate voltage to specific sections. Due to limited space, I opted out of this, meaning I have to re-plug the test leads every time I test a dual triode.
• DMM Resolution: Having only one decimal place gives me a 0.1 resolution step. This is insufficient for high-gain tubes like the 12AX7 or 6SQ7, which operate at very low bias voltages and low plate currents. While this still allows me to measure 1mA with roughly 10% accuracy, I would prefer more precise measurements. Other tubes are fine with current DMMs though
• Short test: The short test neon bulb starts dimming as soon as heater hits the tube. It is either too sensitive due to a high voltage supply or I am doing something wrong. Also, the test is designed to test plate shortage only, missing other pins on the tube. Still using my old EMC203 for the short tests.
• Lack of Regulated High Voltage: Some DIY designs include a Zener diode voltage reference and a MOSFET voltage regulator to set the plate voltage exactly to the tube's datasheet specifications. I chose to avoid that complexity. The drawback is that I now have to build a custom reference chart with operating points matched to the voltages this tester actually generates. Because of the 6-step multiplier, the high voltage sags depending on the load. For example, a 12AX7 will get 280V DC, whereas an EL84 will pull the circuit down to about 255V.

Some Upgrades

To overcome the dual-triode hassle, I built a simple external switch box to select between the two halves. Since all three of my tube sockets are wired in parallel, I can plug this switch into the octal socket, leaving the 9-pin socket free for the dual-triode under test. This works beautifully and speeds up my measurements. Another tweak I have in mind is to use the circuit as a high-voltage power supply for capacitor leakage testing.

I am very happy with the build. For around $200 I can match tubes and measure their critical parameters now. I even tested a few tubes for a local record studio and returned some investments :-)