dekatron brightness

[Martin Kaufmann]

hi everyone. stumbled upon this group while researching dekatron tube stuff and decided to join. will be probabily more on the receiving end, since i just started with them tubes. i am in the middle of building my own nixie/dekatron clock – currently working on the dekatron part. yes, for my first tube clock, i really have drawn a requirements joker; not widely available hv supply, no off the shelf control modules and code examples one more cryptic than the other. haha. my goal is to have a dekatron animation with increasing glow points growing to full circle for a one minute cycle. well, after a lot of tinkering, i more or less got the desired result on a dedicated pcb.

used circuit:

what i am not yet happy with, is the brightness and uneven glow on the cathodes of my og-4 dekatron. the uneveness is probably coming from using «windscreen-wiper» principle in the arduino code to drive the tube back and forth between k0 (12 o’clock position) and seconds-position, which means not every cathode gets the «refresh» in the same amount of time (currently driving the tube at 200us). beside detecting ndx-postion (k0), i also tried to integrate a reset circuit, so i could get rid of going «back&forth» and «jump» instead, but was not very successful – tried a few circuits i found online, but whatever i tried, it just did not reset to k0. My current lack of understanding in electronics (being more a code-person), is hindering me to analyze these further. first i thought, i just have to drive the tube faster, but this does not really help much either.

maybe one of you who truly understands dekatrons, can explain to me or give me a hint, what’s crucial to maintain high brightness in the tube? the single lit k0 cathode is very bright, but i experience a brightness drop for every aditional seconds-point i cycle inbetween:

single lit:

half cycle:

any ideas, help, hints and suggestions are appreciated.

thank you, martin

[Mac Doktor]

On Oct 5, 2024, at 4:50 PM, Martin Kaufmann <mka...@gmail.com> wrote:

maybe one of you who truly understands dekatrons, can explain to me or give me a hint, 

Just to be clear, have you seen/devoured this site?

https://threeneurons.wordpress.com/

I take it that Mike is one of leading lights in the Dekatron realm.

Terry Bowman, KA4HJH
"The Mac Doctor"

https://www.astarcloseup.com

"If only you could see what I've seen with your eyes."—Roy Batty, Blade Runner

[Martin Kaufmann]

https://threeneurons.wordpress.com/ - yes, this was/is my main info site regarding these tubes. being a noob, I did not fully understand everything though. Also cannot recall to have read something about controlling tube brightness, but I might have missed it and will check again.

[gregebert]

Brightness is a function of current; more current=more brightness, but also causes more wear out of the tube. It's important to stay within datasheet recommendations, otherwise your tube will fail prematurely. Brightness is increased by increasing the high-voltage supply or reducing the series resistor value, or a combination of the two. I have a single A-101 dekatron in one of my clocks, and it's been running 27/7 for 11 years now with no visible degradation. I think my supply voltage is around +500VDC, and the series resistor is around 470K [sorry, that's from memory from 2013...]

Brightness is also a function of how much (or little) the pattern changes. If you are trying to get a full-circle on a dekatron, it will need to be spinning about 30 times/second to trick your eyes. There are 30 dots per revolution on a dekatron, so that's around 900Hz. Since the individual dots are only glowing 1/30 of the time for a full-circle pattern, they are going to be noticeably dimmer than a stationary dot; you can see this in your photos.

If brightness is paramount, then you would be better off using 30 LEDs (or 30 neon bulbs....) in a ring and drive them individually so they can be on 100% of the time for max brightness. It's not as cool as a dekatron, but it is brighter.

[Martin Kaufmann]

Thank you very much for your explanations.

The OG-4 type draws 400-450 uA according to specs. I'll try to see, if there is potential with optimizing the series resistor.

As for the drive speed. Currently I use 200us - if my math is right, I am more in the range of 5000Hz with this (probably divided by two, due to the nature of my isr stepper routine). Maybe I have to reduce speed as much as possible without getting into "flickering" in order to max. out brightness. Gonna try this too.

Alternative tubes or leds are not an option - it just matches perfectly with the in-4 nixies I will combine the dekatron with. I really love the optics here :-)

I've seen this sort of dekatron-animation with clear bright glows on all of the cathodes - this tells me it should be feasible...

[Martin Kaufmann]

Have tried now with the series resistor (used a pot). but it does not affect brightness. Also current stays between 280-300uA, regardless which value the series resistor has. Did expect current to either rise or fall when changing the resistor value (ohm's law), but it just stays the same. funny.

I have a feeling my "problem" is related to the hv supply, which is probably better suited to charging a capacitor than supply a load.

Have yet to try with reduced driving speed...

[gregebert]

Measuring the current is tricky if you are doing it while the dekatron is spinning (need a scope); a meter will only give an accurate reading if it's stationary. Even then, the apparent brightness as judged by your eye is not linear relative to current because there is a plateau effect where large current increases have small visible impact on the brightness. Only at the low-end is the brightness vs current effect most visible. So, if the current is in the right ballpark, your only other option is to experiment with the timing.

If you are using 200usec steps, and the pattern uses overlapping (best option in my opinion), it will take 60 steps for 1 full revolution. That works out to 2.5kHz, which is near the upper limit for most dekatrons. I think the A101 is rated at 1Khz, but I've never run mine anywhere near that. I would try running it at lower speeds to the point where the flickering just starts to appear.

[Martin Kaufmann]

I've now tried with slower frequency, but to no avail - brightness decreases with every additional lit cathode. Gonna try a different power supply next.

[Martin Kaufmann]

Have now tried with a proper power supply, which does not sag voltage under load - does not make any difference. Seems, I've to live with the current reduced brightness...