[Engineer2you] built a nixie tube clock and claims it is the simplest design. We felt like that was a challenge. In this design, the tubes are set up as a matrix with optoisolators on each row and column. With 60 segments, the matrix allows you to control it all with 16 bits. There are six columns, each corresponding to a digit. That means each row has 10 lines.
The Arduino code reads the clock and produces the output to the tubes fast enough that your eye perceives each digit as being always on, even though it isn’t.
It may be semantics, but part of what makes the design simple isn’t that it is simple on its own, but that it does use a small number of dense modules. For example, the clock is a DS3231, and there is a DC step up board to generate 390V for the tubes. So instead of minimizing part count, this design really minimizes how many parts you have to connect by employing modules, including the Arduino. That’s still something, though.
It looks as though the nixie tubes used are of Soviet origin. They need no more than 170V to ignite and at least 120V to stay lit. Not a problem with a simple DC to DC converter since the current is very low — on the order of 2.5 mA or so.
We suppose one day the stock of nixie tubes will be gone. But there are still people making them. Or you can do a modern version with light pipes.
What an obnoxious video, with subscribe interludes? Closed immediately
Oh be nice…it’s not that bad, there’s no horrible techno, or sped up footage where he builds it. And asking for subscribers is normal now, it least he only did it for 15s instead of a minute and a half commercial.
obnoxious ? your nuts, and many people do the subscribe thing. I think you just forgot to take your meds today.
Let’s say. It took me almost 1 month to finish the project to gain your view and subscribe. How it can be called “obnoxious”. Before joining to Youtube, i also have same thinking with you, but now i feel it’s normal. Thanks
Optos? Incidentally its feasible to just use LDRs but these are made of unobtainium these days.
Some clever folks made a rotary version using a gapped disk punched with holes.
Bonus is that it can display custom figures simply by having unused lines.
LDR are very very slow, so might not work too well for driving multiplexing displays.
http://lednique.com/opto-isolators-2/light-dependent-resistor-ldr/
>Response time of 2 – 50 ms. (Much slower than photo-transistors.)
If the switching time is large enough faction, then it will affect the brightness of individual digits.
FYI: optoisolators can switch around 1-10us, regular transistors around tens of ns.
Using an ESP32 instead of the Arduino would be simpler, because you could use NTP for the clock instead of depending on a separate RTC board. But i do appreciate the write-up.
How’s it simpler to have at least wifi, network, and a server (which will have a real time clock) than just using a real time clock?
Well it would be one less part…considering that it’s on YouTube, probably a safe bet that he has access to the he internet, as would anyone seeing the project…i don’t think “you” need a server, you’re asking an existing one for the time.
ah, i assumed we were talking simplest hardware design, i.e. minimal components. Anyway, wifi is ubiquitous and there are more than 4000 public NTP servers available (check out the NTP Pool project). NTP also provides for a very accurate clock, since it always keeps it in sync within a couple of milliseconds.
NTC is very accurate, but in between queries you are relying on providing the timing locally subjecting to drifts and tolerances of the crystal.
Or you can do it with a firmware only RTC on a $0.30 microcontroller in less than 4kB. I implemented a 24-bit NCO in 3 lines of code to offset the tolerance of a cheap crystal and so far achieved +/-1 sec per month accuracy. I calibrated it without any fancy equipment – just by eyeballing it with atomic clock website.
Project documented here: https://hackaday.io/project/169509-a-very-accurate-led-clock
“They need no more than 170V to ignite”
Soooo, we’re featuring DIY time bombs now?
Sorry, I’ve been feeling super pedantic all morning and just had to get that out.
Here in Germany Nixies and other glow discharge devices actually do “ignite” in our language.
“Ignite” is commonly used in english too. Piecutter is easily triggered. Or ignited, perhaps.
3……2……1…….Ignition!
Pedantic explorer has achieved liftoff!
Well, I suppose if you were to trust that Merriam Webster fellow. It just seems a bit counter intuitive of a term for what is considered a cold cathode device where nothing is burning, no fuel is consumed, just warm glowy plasma. By the same definition do we ignite an incandescent filament?
Modern language is so full of nonsensical derivative gibberish from a thousand dead languages! We need some standards, man! And by standards, I mean one!
I vote that we stick to the original word used by our primitive ancestors when fire was first discovered;
Ngaaah!
As opposed to the word:
Oooonnh
used for things like Aurora, glow worms and such.
There. Nice and tidy. Don’t you think?
Deutsch ist meine zweite Sprache, ich Frage bloß. Anzünden? Nixies zünden an? Auch Glühbirnen? Und LEDs?
Sie “zünden” einfach nur. Ohne “an”. Genauso wie ein Thyristor, eine Funkenstrecke (spark gap) oder eine Leuchtstoffröhre (fluorescent tube).
“Zünden” means just “trigger”.
I hope that helps.
“They need no more than 170V to ignite”
Actually, in English I think “strike” is usually used in place of “ignite” here. 170V is the “Striking Voltage”, or more formally the “Breakdown Voltage” of the gas mixture within the nixie tube. The breakdown voltage is determined by Paschen’s Law which gives the breakdown voltage needed to start a discharge arc between two electrodes in a gas. The breakdown voltage is a function of the gas mixture, gas pressure, and gap length. See more here:
https://en.wikipedia.org/wiki/Breakdown_voltage
https://en.wikipedia.org/wiki/Paschen%27s_law
You’ll find many examples of different usages here, but it’s crucial to realize the distinction between igniting a glow discharge in a tube and striking an arc in the tube: an arc is definitely not desired in a neon or Nixie tube, where it will cause damage. Contrast this with the arc that happens in (say) a xenon flashtube when it is triggered. They are very different processes, and the terms ‘ignite’ vs. ‘strike’ are useful to distinguish them.
It is possoble to see glow in the non-lit numbers. There is no mechanism to keep the off-digits with a voltage bigger than the sustainable voltage. (See EEVblog about the subject) … So, the “simple” construction could be designed in other ways.. nevertheless, good ideia!
Only “zünden”. And only bulbs that have a gas discharge. So Nixies yes, filament bulbs (Glühbirnen) no, LEDs no.
Danke! Gut zu wissen.
Vielleicht, wenn wir die Spannung verdoppeln oder verdreifachen. Dann wird alles den grammatikalischen Spezifikationen entsprechen.
Habe ich mehrmals mit LEDs gemacht. Übrigens benutze ich kaum mehr mein Deutsch. Die Übung ist mir gut.
Only “zünden”. And only bulbs that have a gas discharge. So Nixies yes, filament bulbs (Glühbirnen) no, LEDs no.
390 Volts? Say that again?!
No, it just need 150VDC to light up. 390VDC is maximum value of DC step-up module
I know. I built several things with Nixie tubes. It’s just the text “and there is a DC step up board to generate 390V for the tubes” that bugs me as hell …