Hi Jay! Been looking over Doug's shoulder at your fascinating work on the NIXIE tube clock. I'm a total layman on electronics, but love learning and creativity. The way you envision and develop a concept like this is inspiring. Grateful that I get this glimpse into your talent and mastery! You, Sir, are fun to follow! All the best, Chuck Heinrich
Awesome project man, Ive been searching for an authentic looking nixie-style led clock for over a year now to no avail and this project is the closest ive seen, far better than the edge-illuminated laser engraved acrylic ones you see everywhere. I might actually have to give this a try myself! I think the floating digits looked very snazzy though, maybe I will give them a try and see how far I can get
When nixie tubes are very rare like chicken teeth. And when you do find they are expensive and in america and im in south africa! I have fallen in love with nixie clocks. But due to problems mentioned. Im also looking to do something similar as in video. Although a bit weard and crazy maybe. It would probably let people ask Why?.
Question: I got those LED filaments here, where is the anode where the cathode part ? One end has a small hole in it, so it is definitely A/C but hardly to know which. Dont want to waste any filements, so any hint is great
the end with the small hole is the positive (anode) side. They are sold as "3 Volt" filaments but I found they work good at 30mA which was about 2.72 Volts.
@@jayhamlin9991 That does the trick. Working (technically) now, however some faults, maybe you got an idea: seconds 01256789 working , but 3 is without the lower line, it's on when 4. Would assume something with the logic. Or do you got an idea on how to troubleshoot the display ?
actually, i noticed that the lower line (_) is reversed in anode/cathode, e.g. your old design is slightly different to the use. Question would be: How can i easy test out everthing (like console i type in a number and it gets displayed ? would help others as well.
@@JanCzmok I think I know the answer but I have not had time to confirm, this is what I think it is. If you are using the Arduino charlie7 code, I had an early prototype with the lower line LED diode reversed. I think the code in gitHub is for this prototype. Sorry. We need to swap the '1' and '0' for the lower line. I can look at it tomorrow .
Yes, what you are saying is true - with 4 pins you could turn on any 1 LED segment. The 6 pins are also mechanical, holding the ends of each segment. Also, with 6 pins, it can make any number digit without multiplexing.
@@jayhamlin9991 0:58 I can see how each segment can be turned on at a time, but how would you make the digit "8" without multiplexing? If pin 5 is high and pin 2 is low, do you mean make pin 5 high, make pin 2 low and the rest high impedance? I think that might work, though current would be split down some paths. It looks like you still multiplexed each segment, but maybe you designed for the possibility of doing it differently.
Ok, you are right, you got me. Enough time passed that I forgot how it was done. Only 1 segment at a time is turned on and the digits are multiplexed. You can check out the code on GitHub and see how it's done. @@Enigma758
@@jayhamlin9991 Thanks for the responses and sorry for all the questions (sometimes I think too much ☺) . I think the fact that you used 6 pins made it easer to multiplex since you could simply break out each intersection. That way you wouldn't have to hard wire the additional connections required for 4 pin charlieplexing. And considering that was probably decided before you added the backing PCB, it seems like a reasonable decision. But now that I think about it again, the digits are driven by the shift registers and not the microcontroller I/O pins so you really can't do charlieplexing since you can't set an individual shift register pin to high Z. So I believe you are actually doing "traditional" multiplexing at the segment level. At any rate, you did an amazing job on the clock and it inspires me to do something similar!
Wow, those After Dark PCB's look so freaking amazing. Good job on the clock!
I really like OSHPark. I didn't document the soldering but I used a stencil and reflowed the board in a Black & Decker toaster oven.
Absolutely brilliant idea, actually far more interesting than a conventional NIXIE clock
Very nice project, it looks awesome.
Really cool project, I love the idea of using these LEDs ! Good job
Great idea to use those LED filaments for it and the clock came together beautifully. Well done!
This looks pretty nice! Good work!
I'm back watching your video again. Absolutely stunning result :)
Those are a thing of beauty. Really well made and I love the idea. Thanks for sharing.
Thank you very much!
Amaizing project! Looks very nice! 😁
Hi Jay! Been looking over Doug's shoulder at your fascinating work on the NIXIE tube clock. I'm a total layman on electronics, but love learning and creativity. The way you envision and develop a concept like this is inspiring. Grateful that I get this glimpse into your talent and mastery! You, Sir, are fun to follow! All the best, Chuck Heinrich
Very cool!
Awesome project man, Ive been searching for an authentic looking nixie-style led clock for over a year now to no avail and this project is the closest ive seen, far better than the edge-illuminated laser engraved acrylic ones you see everywhere. I might actually have to give this a try myself! I think the floating digits looked very snazzy though, maybe I will give them a try and see how far I can get
I can't wait to see what you come up with.
Great idea! But what they're lacking is a grid in the front face, like in the Nixie nubes
Amazing, thank you! :)
Wonder what it would look like with a nixie-like mesh screen in front of the digits inside the tubes to diffuse the light
they actually look more like numitron tube's
When nixie tubes are very rare like chicken teeth. And when you do find they are expensive and in america and im in south africa! I have fallen in love with nixie clocks. But due to problems mentioned. Im also looking to do something similar as in video. Although a bit weard and crazy maybe. It would probably let people ask Why?.
Question: I got those LED filaments here, where is the anode where the cathode part ? One end has a small hole in it, so it is definitely A/C but hardly to know which. Dont want to waste any filements, so any hint is great
the end with the small hole is the positive (anode) side. They are sold as "3 Volt" filaments but I found they work good at 30mA which was about 2.72 Volts.
@@jayhamlin9991 That does the trick. Working (technically) now, however some faults, maybe you got an idea: seconds 01256789 working , but 3 is without the lower line, it's on when 4. Would assume something with the logic. Or do you got an idea on how to troubleshoot the display ?
actually, i noticed that the lower line (_) is reversed in anode/cathode, e.g. your old design is slightly different to the use. Question would be: How can i easy test out everthing (like console i type in a number and it gets displayed ? would help others as well.
@@JanCzmok I think I know the answer but I have not had time to confirm, this is what I think it is. If you are using the Arduino charlie7 code, I had an early prototype with the lower line LED diode reversed. I think the code in gitHub is for this prototype. Sorry. We need to swap the '1' and '0' for the lower line. I can look at it tomorrow .
I did not write any test code other than the counting 0-9 in the code. It is only Arduino, you could easily add a console function.
Why 6 pins for each digit, charlieplexing should reduce that to 4. (n^2 - n)
Yes, what you are saying is true - with 4 pins you could turn on any 1 LED segment. The 6 pins are also mechanical, holding the ends of each segment. Also, with 6 pins, it can make any number digit without multiplexing.
@@jayhamlin9991 0:58 I can see how each segment can be turned on at a time, but how would you make the digit "8" without multiplexing? If pin 5 is high and pin 2 is low, do you mean make pin 5 high, make pin 2 low and the rest high impedance? I think that might work, though current would be split down some paths. It looks like you still multiplexed each segment, but maybe you designed for the possibility of doing it differently.
Ok, you are right, you got me. Enough time passed that I forgot how it was done. Only 1 segment at a time is turned on and the digits are multiplexed. You can check out the code on GitHub and see how it's done. @@Enigma758
@@jayhamlin9991 Thanks for the responses and sorry for all the questions (sometimes I think too much ☺) . I think the fact that you used 6 pins made it easer to multiplex since you could simply break out each intersection. That way you wouldn't have to hard wire the additional connections required for 4 pin charlieplexing. And considering that was probably decided before you added the backing PCB, it seems like a reasonable decision. But now that I think about it again, the digits are driven by the shift registers and not the microcontroller I/O pins so you really can't do charlieplexing since you can't set an individual shift register pin to high Z. So I believe you are actually doing "traditional" multiplexing at the segment level.
At any rate, you did an amazing job on the clock and it inspires me to do something similar!