SZ-8 operating life span

[Mike Mitchell]

I'm using SZ-8 nixie tubes in my dekatron-based clock as shown here:
https://www.youtube.com/watch?v=pK276xZWyo4

I find I have to replace the nixie tubes more often than I expected.  Most of the failures are in the '7' digit.  The wire from the insulator stack to the '7' cathode starts to glow faintly, and brightens over time until it is full-on bright and the '7' is not glowing at all.  It leaves a dark smudge on the glass where the wire sputters away.
The other failure is with the '1' and '5' digits.  They start to glow simultaneously, first dimly shadowing each other but again brightening over time.

I direct-drive the tubes, with a 170v supply and a 20K anode resistor.  Individual cathodes of the A101 dekatrons are tied through resistors to the base of MPSA42 NPN transistors, the transistors switch the cathodes of the SZ-8 nixies.

Should I be using a higher-value anode resistor?  Would a different nixie (QS30, ZM1022) work more reliably?

Thanks!

[Nicholas Stock]

That's a beautiful clock Mike. Any documentation anywhere? I've run ZM1022's for a couple of years with no issues (multiplexed however). I have no experience with the chinese tubes....

--
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 post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/1b491d6d-0a0e-43b3-b981-0e1ee3c6ac3f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[gregebert]

Yes, a better-quality nixie tube will work. Burroughs are my favorite, and I dont use any other brand.

I found that with IN-1 tubes, when 2 or more cathodes glowed at the same time, a small metallic filament was shorting them together. You can measure it with an ohmmeter, and you can also zap it with a few hundred mA of current. Yes, it *will* glow when heated, until it burns out. I suggest a good bench power-supply with an adjustable current-limit, rather than voltage, to blow the filaments. Then your tube is usable again, at least for awhile. Be warned, though, that after zapping, your tube will probably fail again for the same reason, but perhaps in a different location, and with different digits. In my opinion, it's usually a sign of a poor-quality tube (though I did have 1 Burroughs 6091 tube fail like this).

On a related note, NASA has done research into tin whiskers. They form in an electric field and grow at exponential rates as the whisker gets closer to shorting. I've actually *seen* one in the CCFL driver board of on of my laptops; it had the classic symptom where the screen goes dark when moved, then works again (for awhile) after rebooting. It was a smaller diameter than typical bondwires in ICs. I suspect the shorts in nixie tubes are a similar mechanism and dependent upon the type of cathode material.

[Mike Mitchell]

My last two posts in the "Source for Dekatrons" topic has some details on the clock.  It uses Russian A101 dekatrons with MOSFETs as interstage couplers. All counting is done by the A101's, there are no integrated circuits.

Soon after I posted my clock, this all-vacuum tube clock showed up: http://wv7u.com/cwc/hourglass.html