Semiconductor Free Nixie Clock Pt1

[@videolabguy]

That is absolutely the most amazing thing I have seen today! Thank you for making my morning coffee so much better. You have restored much of my faith in the next generation. Best of luck with all of your projects and your obviously brilliant future career.

[@jdflyback]

Thank you!

[@garydejaen6938]

Incredible! Perfect timing on the video, I was just thinking about JD!

[@jdflyback]

Thank you so much Gary!

[@jdflyback]

0:00 Overview 3:40 Explanation 11:00 Detail about Hours Reset 15:40 More of the Clock 20:00 Details about construction 20:58 Detail about aging the Tubes

[@mikegLXIVMM]

Fascinating project, thanks for posting!

[@bigjd2k]

I remember taking apart an old calculator which used trigger tubes when I was young!

[@jskratnyarlathotep8411]

aha, that explains why neon logic didn't expanded more and why we don't see almost any of it today left

[@robinsutcliffe-video_art]

Your projects make me laugh purely because they are so fantastic! Did you get it back together again? Can you make a video synth from tubes please?

[@quantumlab9130]

That’s really awesome, I never knew about trigger tubes. I’ll have to experiment with them a little bit.

[@twirlywhirly555]

Very very Nice , Not seen the MTX-90 used before like this ^^ , I had a working build with the Z700U but never quite got the hours reset working

[@jdflyback]

That is awesome, I hope to build one with the Z700U someday. I go in to the design of the hours reset at about 11:00, it was hard to get it to work.

[@frankowalker4662]

OMG ! You are insane. LOL. This is just mind blowing. Never change.

[@chanheosican6636]

I had once a 200 in 1 radio shack project kit. It used NAND gates and flip flop jk to pulse leds off and on.

[@skungpid]

I can tell by your voice you are younger than me (slight jealousy). Still, you have chosen the long path of Non-LEGO circuitry. You, my boy are a full JEDI. I salute you. You have the understanding of Were there more like you who really understood non-cookie-cutter circuits! Again, I salute you - Dogeza (土下座). Besides a full on JEDI, you are a master of the Gopnik, par excellance. Kurwa mać! But, my son.....don't forget to get a life.

[@alexkart9239]

I'm not sure exactly, but it looks like these lamps (MTX-90 thyratrons) are intended more for indication, and not for control functions. They are cheap and unstable because they are not intended to perform any critical functions.

[@nikomaukkonen4994]

That's an awesome clock! I've had something similar on my to do list, but never got around to do it. Maybe i should give it a shot Did you have to "burn in" the thyratrons to reduce the differences in firing voltages or was it not necessary with these kinds of tubes? Also was it a lot of work to get the ring counters to work properly? I remember building a five tube counter from "TH4B" trigger tubes and it was a lot of work to get it running in a stable manner.

[@jdflyback]

Yes, I ran in the tubes for about 24- 48 hr at about 4mA each. They were useless without doing so. It took me a few weeks just to make a single working row of 5 but after I figured that out The rest were fairly straight forward. After burn in the tubes will function but they are not matched so I had to use two potentiometers per tube to set the working point and bias for the next tube. Now that it is working I am trying to replace the pots with fixed resistors but the tubes are so finicky I might just keep them.

[@legobuildingsrewiew7538]

youre still alive wow

[@KeritechElectronics]

Still talking when there's science to do :)

[@therealchayd]

That is some serious tubage - great work! have you considered dekatrons to minimise the component count and make a more portable clock and minimize device mismatch issues?

[@q9a]

You are use 16 = 10+6 elements to devide 10*6 = 60Hz to 1 Hz. 60 = 2² * 3 * 5. Because of 2*2=2+2 the Optimal predevider is 3 * 4 * 5 = 60Hz with need 3+4+5 = 12 Elements. This reduces the Hardware for Predeviding by 25%

[@KeritechElectronics]

Oh damn, this is absolutely awesome!

[@flatfingertuning727]

How precisely controllable are the thresholds? Could one handle the 1Hz signal generation by setting up three relaxation oscillators whose periods would be about 44ms, 60ns, and 76ms when enabled, but which when triggered would be held on until the end of an AC line cycle, and generate a 1Hz pulse when all three outputs are active simultaneously? I was also wondering if it would be possible to generate a multiplex-drive clock by having a 2kHz or so master clock divide a 16-way ring, whose first ten outputs would be for unit digits and whose other six would be for tens digit, and then for each digit use a relaxation oscillator with a period that would normally be just under 16 master-clock cycles, but whose trigger would be delayed when the previous digit's oscillator output was "active", and which if triggered would remain active until the next master clock pulse for its "side" of the divide-by-sixteen ring. Sorry if my description isn't clear, but the idea would be to have each digit be represented by which of sixteen-phases a relaxation oscillator was in. For units digits, ten of the sixteen states would be valid "holding" states, and for tens digits the other six states would be valid holding states. Advancing a counter would cause the digit to "slip" a state, but if the digit slipped off the last valid state its oscillator would stay triggered during the valid states for the next digit, thus causing that digit to slip by a state. All of the digit wires for units digits would be energized sequentially, followed by the digit wires for tens digits, all controlled by the master ring, but each digit's anode would only be energized when its oscillator output was triggered. One would need a bunch of trigger tubes for the divide-by-sixteen master ring, but addiitional digits would only need one tube each for counting.

[@getcartercarpark.]

May I ask if the trigger points of the tubes are affected by ambient light? I know neons can be temperamental and not light at all if there is no ambient light external to them. Thank you for another interesting video! K.

[@eDoc2020]

I don't know if you're still working on this but watching the hours counter during the reset process I think the issue might be a simple race condition. When running in fast set mode the time between hour pulses seems like it's about the same as the time needed to complete the reset on 6Hz.

[@sneugler]

Nice job! Have you made any progress on this protect since then? Having a wall-mounted nixie clock where you're able to observe all of the MTX-90s switching would have a great look to it.

[@rampart1234]

Spectacular!

[@chanheosican6636]

A counter with no ic cool.

[@robot797]

this is awsome is there any way for us to get your circuits in digital form?

[@jdflyback]

I mainly referenced other clock schematics in my design, I added links to all of them to the description. I may create my own schematic in the future.

[@robot797]

@@jdflyback I knew of them but yours is using the same tubes as I have and so it would be a lot more reliable to use your design (I want to build a no ic clock with crt display)

[@jdflyback]

I will see if I have time to make a proper schematic, All of the counters use the same component values to the six stage one I show at 8:56 just with more stages added on. The only different stage is the tens of hours at 11:00 which has different taps for the trigger electrodes. The schematic is copied from the www.sgitheach.org.uk/nixie3.html schematic except the cathode resistors are replaced with the two pots, and the anode resistor replaced with a 220K pot. He also uses a 12 stage counter for the hours avoiding the complicated hours reset I use.

[@jstro-hobbytech]

That is cool. I'm a clock junkie

[@legobuildingsrewiew7538]

Can you give me the components to your 2kw flyback driver please

[@jdflyback]

Hello, I used this Schematic markobakula.wordpress.com/power-electronics/500w-royer-induction-heater/ with 3 .47uf caps in parallel, and IRFP260N mosfets. The diodes are uf1004s but any fast high voltage diode should work. The flyback winding was 8 turns in place of the heating coil shown in the article.