First schematic for nixie clock

[Jarod Findley]

Hi all this is my second post and I'm still learning a lot  about nixies-neon tubes and how they work. This is my first project design i used this schematic to design a clock around in-28 neon tubes and some large size nixies like zin-70. I have not calculated values for the resistors and such nor have I tested anything yet. Id like to know if my design is correct before I move it into the real world. It will have a ring counter for the Hours, Minutes, Seconds, Am Pm, and a 60 ring counter to visually indicate the seconds. This will be a 12 hour clock and have Am Pm tubes. My plan is to design a nice case and have it displayed in my living room. I know its ambitious for a first time project and apricate any feed back, 

[Jarod Findley]

https://drive.google.com/file/d/1-OQk9pM2vFbIHDEv9-DlOhHWoDQwO6y0/view?usp=sharing Heres a link for the pdf i realized the picture is very low quality 

[gregebert]

I cant quite zoom-in enough, but it looks like you are using analog-ish ring counters, rather than digital ICs. Unless you have built this on a breadboard and tested it, or run SPICE simulations (hmmmm...might be tricky getting accurate modeling of neon bulbs and tubes).

For my first nixie clock, I extracted a netlist from the schematics, and ran Verilog simulations on the logic to confirm it worked correctly (it uses 4000-series CMOS), and I did SPICE simulations on the analog portions. I hate doing bench debug, so I always spend a lot of time simulating the design before fabbing the boards. You would be surprised how many stupid bugs I find from simulations, especially on things that should "just work".

[Joe Croft]

Hi Jarod.

This reminds me of my NixieNeon clock, https://www.nixieneon.com/main/nixieneon-clock-documentation/.  This clock was a lot of fun but the ring counters could  be hell! I had a lot of matching to do with the neon bulbs. Granted, I was using NE-2 lamps, not triggered neons for the counters. But the best I can tell, you are not using the trigger leads. Ig nore the processor, it is mostly for setting the time as well as giving a little eye candy every 5 minutes. The rings  will free run with out the processor.

I recommend you bread board one or two of the rings.Also,  I was not able to coulpe just using a capacitor and a diode. I ended up needing a transistor. Of course all of that could have been from my short comings in hardware design (I am principally a software guy who loves the smell of rosin) .

I hope your clock works out for you. I still have a love for ring counters like these.

-joe

Sent from Mailspring, the best free email app for work

On Jul 18 2025, at 5:14 pm, Jarod Findley <keys2...@gmail.com> wrote:

Hi all this is my second post and I'm still learning a lot  about nixies-neon tubes and how they work. This is my first project design i used this schematic to design a clock around in-28 neon tubes and some large size nixies like zin-70. I have not calculated values for the resistors and such nor have I tested anything yet. Id like to know if my design is correct before I move it into the real world. It will have a ring counter for the Hours, Minutes, Seconds, Am Pm, and a 60 ring counter to visually indicate the seconds. This will be a 12 hour clock and have Am Pm tubes. My plan is to design a nice case and have it displayed in my living room. I know its ambitious for a first time project and apricate any feed back, 

--

You received this message because you are subscribed to the Google Groups "neonixie-l" group.

To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.

To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/390fea2f-0fce-4557-80d1-4e6b8ed0813en%40googlegroups.com.

[Jarod Findley]

I hoping the design I used is going to be robust enough for my circuit.  If you haven't read the attached website to the pdf I linked he has some very good information on ring counters and why the ne2 is so difficult to use in them. Im hoping the in28 are similar enough to be adapted. 

My next purchase definitely is going to be a quality bread boat and scope for doing some diagnostic work. Thanks for info and help guys ill definitely be simulating and real world testing. Ill keep the group posted 

To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/90056DDC-CBF5-459C-87A9-C6B76DB1628F%40getmailspring.com.

[Joe Croft]

Hi Jarod,

I saw that about the ne-2 diodes. My solution was much more painful. I went through the neon lamps and matched them. I look forward to seeing the final clock!

-joe

Sent from Mailspring, the best free email app for work

To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/CAL_1Jahz7tHfe-ep4kuXwOdLNmdgn7%3DJJkwekbe9sSWVJ5Yupw%40mail.gmail.com.

[Grahame Marsh]

Hi Jarod

I skipped neon bulbs and and used trigger tubes. My first clock used XC18 tubes which originally had (long-gone) radioactive doping to get reliable operation in the dark. I then used Z700U tubes which have a primer electrode and do operate reliably in the dark.

Does anyone have experience with neon bulb ring counters operating in the dark?

Grahame

To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/1690C145-FF58-4C72-B0D4-CF28E4CEF42F%40getmailspring.com.

[Jarod Findley]

https://www.pa3fwm.nl/projects/neonclock/

In this write up you can see he addresses the problem of the neon bulbs not being able to light in the dark with blue leds. I wonder if you could use ir leds and keep it invisible

To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/2aadf306-d917-476b-893c-04199d53a2a3%40googlemail.com.

[Jarod Findley]

https://www.pa3fwm.nl/projects/neonclock2/

Sorry this is the write up I was referring to the other one is still very informative 

[Adrian Godwin]

I can't see IR leds working, though I haven't tried them UV leds are
more likely to work. Might be made invisible if filtered unless Nixies
have any phosphors in them.

Yes, I know they don't rely on phosphors but you might find some
cement or getter residue or something that fluoresces.

> To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/CAL_1Jah3t90OgjV7ib_sacwEWX8M9X57UQRFS72zdUwqX%2BSNGQ%40mail.gmail.com.

[Adrian Godwin]

Another invisible excitation might be a brief pulse of blue or UV
light just after a new digit is illuminated. As I understand it, it's
only needed to light the digit, not sustain it.

I think the blue backlight often used to overcome this problem is
quite a pleasant complement to the neon colour, but it's also a bit
garish. I can totally understand the desire to have pure neon orange.

[Jeff Walton]

I am using UV leds under MG-17 (segmented neon display) tubes to get them to work in the dark.  When ambient light is present, the UV leds are not needed and automatically shut off.   The tubes would not work right in the dark, even with higher HV than standard 170v DC supply.  They have been running for more than a year and still work in the dark as long as the UV is present.  

There is a small amount of visible light from the leds but most of the light is not visible. I can hold some uv reactive materials near the base of the tubes and can definitely see a bit of uv "spill" that causes some maretials to glow, so you can definitely tell that it is working.  You will need to experiment with different wavelength uv leds as some glass materials begin to become opaque at the really short wavelengths. 

Jeff

To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/CALiMYruMjjqwzt8GDWOe0s_chTjC87V57YyOGOWsvEhGUoLnjg%40mail.gmail.com.

[Dekatron42]

@Jeff Walton - did you try using a high value resistor connected to the ground and then to the decimal point, something like a few megaohms up to perhaps 10 or 20 megaohms? I do beleive that it was Burroughs who mentioned that in one of their documents to get proper ignition at low illumination situations without the decimal point actually glowing but giving off enough electrons to get the glow starting on the digit you wanted to light up.

Perhaps someone else has tried this for Nixies?

/Martin

[Jeff Walton]

Here are a couple photos of using a 385nM UV LED to help ensure that some tubes will light up in a darkened setting.  Without the UV assist, none of these tubes would fire reliably in a darkened room, especially if only a single segment was supposed to be lit.

3mm UV LEDs with 385nM output.  Not intended as a visible baselight, only to help for reliable firing in darkened room.

 

LED’s off when the ambient lighting is sufficient for reliable tube segment lighting.

 

Driven at ~2mA in dark setting.  Some visible light from UV LEDs. 

 

Here is the clock with the case in place and UV LEDs operating

 

The clock uses MG-17G tubes and is a version of Richard Scales Nixology Elfin 8-Digit clock.

[Dekatron42]

This thread: https://groups.google.com/g/neonixie-l/c/WutNWZBDMAo/m/AYv7pDYxmxsJ contains the Burroughs N101 Application note which discusses ionization, where they mention keep alive electrodes as well as Krypton 85 radioactive gas on page 5 & 6.

/Martin