I recently purchased a the PV Electronics Spectrum 18 IN-18 Nixie clock kit. The problem is that after about two weeks I am beginning to see cathode poisoning, mostly on the 8's. I have had it set to shuffle the digits for 5 seconds and display the date every minute since I assembled the kit. I purchased the tubes from a reputable seller on ebay, and am reasonably sure they are new old stock. They are all date code matched (05-85), show no darkening of the glass and didn't show any cathode poisoning when they were first used. Could having too low of current be causing this cathode poisoning? When I measure the current with my multimeter I get 3.1 mA, but when I researched these tubes online I seem to get the impression they should run at around 6 mA. The anode resistor provided with the kit is a 10k ohm resistor. If the current does need to be increased, how many ohms anode resistor would you recommended installing?
My personal experiences (and the IN-18 datasheet specs) suggest that anything less than 4mA is too low and poisoning will occur quickly. I went thru this when initially testing out my chiming IN-18 clock design. I began with 15K resistors on my IN-18 anodes and by the time I was finished testing I was down to 6.8K resistors.
IN-18 forward voltage drop (Vf) per data sheet = 150VDC
Assuming your HV power supply is 180VDC.
180 - 150 = 30 volts dropped across resistor
I = E / R
I = 30 / 10000
I = 0.003A or 3mA which is about what your meter said.
but if I = 30 / 6800 (6.8K resistor)
I = 0.0044A or 4.4mA
So give 6.8K or even 5K a try. I promise even with 5K the current is nowhere near enough to destroy your tubes, and it may well solve your problem. It did for me... I have zero cathode poisoning issues on my chiming IN-18 clock, I only "spin" for 30 seconds once an hour, the tubes do not run hot, and they look as good today as they did when I first lit them up 4-ish years ago.
Look into it later when the dust is clearing off the crater.
Last edit: 5 years 11 months ago by Ty_Eeberfest. Reason: Fix typo: 150Vf not 105