Arduino Based Low Voltage Vacuum Tube Tester

In this episode we take a stab at building a vacuum tube tester that gives us some proper, usable data! Granted, it’s all at super low voltages, just 24V in this case, but that actually works out perfectly for the types of circuits that we commonly build. Most importantly though, this was all done for extremely cheap – in total, there is less then $30 worth of parts involved in the entire build!

Schematic, gerber and code files here:

If you want to support the channel please hop over to Patreon:

Also, we now have some epic shirts for sale!

Timelapse Music:
Artist: C7, Jamphibious
Title: Mega Man 3 "Shadow's Lounge"

Intro Music adapted from:
Artist: The Runaway Five
Title: The Shinra Shuffle

Come join us on Discord and Twitter!

Thanks for watching!

Chapters
0:00 Intro
2:47 Using the tester
4:08 Testing the “bad” tubes
6:43 Tester construction
9:33 Arduino code
12:32 Building the PCB
14:46 Testing the finished version
18:04 Outro
19:17 Bunny!

[anandawijesinghe]

I really enjoy your unusual approach to tube electronics and your circuit construction techniques!

I have one suggestion: Tube data books usually give the Anode Characteristic curve, which is the anode current Ia vs the Anode to Cathode voltage Vak for different fixed Grid to Cathode voltages Vgk, and the Transfer Characteristic curve, which is the Anode current Iak vs the Grid to Cathode voltage Vgk for different fixed Anode to Cathode voltages Vak. Since these curves use voltages directly applied to the tube pins, they are independent of the choice of an Anode load resistor (your choice is 33k).

The condition of a tube can be checked directly against published tube data if your system enabled the standard Anode Characterics and the Transfer Charactristics to be plotted, instead of the Anode to Cathode voltage Vak against the Grid to Cathode voltage Vgk, which depends on the value of the anode load resistor.

To determine the Anode current all you would need to do would be to use a precision resistor for the anode load, and measure the voltage drop across it.

Please keep publishing your highly enjoyable videos on tube electronics ! 🙏!

[paulawillaminachandler-ren]

That is one sophisticated tester as is, but if you are planning on building one to test at higher voltages. That would be beyond amazing! You really know your stuff that's for sure.

[algorithminc.]

Really glad I stumbled across your channel. I engineer modern electronics and algorithms, but love playing in the old ways, designing tube circuits. There is a lot to learn from the old ways - those old ARRL Handbooks and the like. It's always great fun to move from paper and pen to the soldering iron in some old tube radio. Thank you - great channel - and I look forward to watching the computer building series (reminds me a bit of that 90's series called "The Machine that Changed the World"). Nice Usagi ... we have a few wild that run around the backyard. Cheers

[jorgeferreira]

Hi.
First of all, thank you for such wonderful videos. Being an EE that got into the Uni in 1980 and went all the way to the digital world, for me the tubes were mainly a curiosity from older days. And your videos are filling a gap in my endless curiosity.

About your tester gadget.
Those massive resistor values used for sampling the plate voltage are most probably causing erroneous readings of the ADC (AnalogRead). The ADC inputs for micro controllers have very specific source impedance requirements. Common values are in the range of 2, 5 ~ 10K values. Not complying to it means that the capacitor of the sample & hold front end of the ADC has not enough time to charge up to the voltage level being sampled.
I would suggest an extra OPAMP wired as a buffer (unity gain) between the resistor divider and the Arduino analog input.

[gordonwelcher]

It is good the way you put the video into chapters.
The old tube tester you have is an Emissions tester. It just connects the plate and grids together, applies a DC voltage and measures the current.
The better type is called a Mutual Conductance tester. It tests the tube in an actual circuit. It places an AC signal and bias on the grids and measures the AC output from a plate resistor.
I would use a higher voltage for my circuits. The performance is better and you can use the curves and specifications from the manufacturer.
When I was 14 years old I built amplifiers using 2000 volts and I am still alive.

[akkudakkupl]

Nice project! But I have some ideas for improvement ;-)

I propose that you read the analog voltage 16 times and accumulate the results (add), then divide by 4 - you just gained 2 bits of resolution for 'free' (a bit of time). Also go for faster serial speeds, there is no reason to sit at 9600 baud, you can safely do 115200 baud.

You can also use internal pullup of the arduino for the switch with pinmode(start, INPUT_PULLUP) - the input will stay at high state untill you press the switch, then it will go low - you would need to invert the logic in first state of your state machine if (digitalRead(start) == LOW && go == 0), but you need 1 resistor less and simplify the layout.

With the 555 you could have configured it differently to also save some parts - resistor from output to trig and treshold and capacitor from trig and treshold to ground - this way you will get 50% duty cycle with just 2 timing components and simpler layout.

I'd also go for a TL072 for the opamp - it has way lower drift and better input impedance so won't load the PWM voltage much, u741 is super outdated. You can also use the second half of 072 for an active filter to lower ripple of the produced voltage. Also dividing the voltage at the pasive PWM filter as you are doing seems like it increases ripple of the produced voltage, would be better to lower gain at the opamp.

[lindoran]

Amazing! Your videos have been such an inspiration to me. I'm currently working on a 1 bit ALU using only open loop lm324n quad op amps I don't know if I'll make a whole computer like you are doing but it's always in the back of my mind. I love that you are making your own tools and pcbs. Just awesome!

[tomhague]

I arrived at your channel quite by accident. I was intrigued by your tube tester and will be experimenting with a prototype of my own. When will you have the time to investigate that cheap tube curve tracer. It may prove beneficial to me. Vacuum tubes were still a part of my education when I was at college and the transistor was up and coming. My first exposure to computers was a PDP11-05. I still like the vacuum tube. Isn't technology fascinating? The future is built on the past. i am looking forward to your design. Keep up the good work.

[rafaelabrahaosantos]

Awesome project!

Improvement idea: I was reading the diagram and, in my opinion, it would improve your testings if you made use of a buffer of some kind at the valve output (anode) terminal, before the voltage divider.

[vladnickul]

If you can finish that project, you will become a God, to all audio tube loving guys :)

[jannepeltonen]

And at the end, we have a bunny for scale :)

[vdvdtsp]

Great project!!! It's the easy way to get the diagram in special connections!

[kilovolt]

It would be so nice if there is a tube tester that recognizes tubes where the text has disappeared. I have hundreds of tubes where paint is unreadable.
And also whether the values are still good.

[justovision]

Those milled PCBs are beautiful. Do you have a video on your hardware and workflow?

[winstonsmith]

Fantastic! I eagerly await your more capable V2.0.

[acoustic]

Very nice! I'd love to see a versatile, low cost vacuum tube tester project which is adaptable for many different types of power, preamp and rectifier tubes. Something that would be useful for a technician. There are a couple of good testers out there such as Roe Tester. However, that's a very large project. Thanks.

[ReallifeBambiDeerattheFarm]

Always love seeing the cute bunny at the end! Now that's a cool way of using today's technology for yesterday's electronics. Perhaps I might build something like that.

[stefandumoulin]

This is amazing! And just in time before ordering my first arduino. I guess i'll have to order two now ;)

Do you think this could be adapted to higher voltage? For example scaling up the anode voltage by 10 and then changing the voltage divider accordingly? And i guess maybe also the grid voltage driver.

Thank you so much for your amazing content

[andypughtube]

1950s time traveller "So, you are using 5 brazillian tubes to measure the performance of one tube, which costs more than all the others together? Which works worse?"
Nice work, but it's fun to re-watch this in the mindset of someone from the vacuum tube era.

[RectifiedMetals]

I am looking for exactly what you described for a general tube tester. Not only for most dual triodes, but one specific that has a higher current and filament voltage.