The high speed cycling makes a lot of sense when the "slot machine" effect 
isn't desired.

I guess the harder question is how much a rapid cycling is enough. 

>From the standpoint of creating a look and feel for the clock, I would 
implement 20ms of cycling (10 x 2ms) done every second when second digit 
transitions. This way the clock has a small flicker on every transition 
(second, minute, hour transitions will all have 20ms of cycling). Keeping 
flickering consistent and tied to the transitioning seconds digit makes the 
flickering easier to explain to those that notice. The most reductive 
answer to those that notice is "Nixie tubes are less stable than modern 
displays this is why there is a little flicking". This leaves out cathode 
poisoning as a concept and the fact that the flicking/highspeed cycling 
used to reverse cathode poisoning. Yet the answer still gets across the 
truth that because Nixie tubes are flawed there will be some flickering.


Generally if a tube has heavy cathode poisoning, over driving the tube with 
2 to 3 times the rated current is often needed to try and recover / reverse 
cathode poisoning in a nixie tube. Yet assuming a nixie tube doesn't have 
any cathode poisoning the going theory is; if every cathode is used 
regularly, cathode poisoning never should never occur in the first place. 
Simply creating neon plasma around every cathode on a regular basis should 
prevent cathode poisoning all together. What qualifies as "regularly" in 
duration and frequency is a tricky question but doesn't change the theory.

On Wednesday, March 7, 2018 at 8:48:24 AM UTC-8, Spirit's lab wrote:
> While designing my own clock, I decided to investigate the cathode 
> poisoning prevention methods utilized by most clocks, and I discovered that 
> there's room to improve and experiment.
> Here's a short video:
> If you can't or don't want to watch: Switching between cathodes with a 
> delay of 1-2 milliseconds is going to provide the same cleaning effect as 
> the "slot machine", except without the extreme flicker which may be 
> annoying to some.
> As for my design - it's two HV5522s in the PLCC package connected to an 
> ESP8266(for driving the HV5522s the 3v3 signals are shifted to 5 - that 
> works up to a supply of 12.9V) and Yan's NCH6100HV boost board.
> It would be interesting to see what everyone here thinks.

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 post to this group, send an email to
To view this discussion on the web, visit
For more options, visit

Reply via email to