1. Things you need to know about the new ‘Conversations’ PM system:

    a) DO NOT REPLY TO THE NOTIFICATION EMAIL! I get them, not the intended recipient. I get a lot of them and I do not want them! It is just a notification, log into the site and reply from there.

    b) To delete old conversations use the ‘Leave conversation’ option. This is just delete by another name.
    Dismiss Notice

Nixie cathode poisoning help sought!

Discussion in 'd.i.y.' started by lsinclair, Dec 3, 2015.

  1. lsinclair

    lsinclair pfm Member

    I’m asking for a favour – I have a lovely (to my eyes) Nixie tube clock that displays information (time, date, location) received by GPS. I have been aware for a while that some digits have been suffering from cathode poisoning, visible as darkened areas on the figures. This is not noticeable on the digits used to display time -


    But here I have swapped the third and fourth tubes around so the fourth tube is showing an ‘8’ rarely used in the third position. The darkening is obvious.


    I have only now found out that this can be reversed

    http://www.tube-tester.com/sites/nixie/different/cathode poisoning/cathode-poisoning.htm

    and I wondered if there is someone with a variac or similar and more confidence in their abilities than I have in mine would be willing to try to fix the affected digits. By my calculations there should be 15 bad digits spread across three tubes. I’d like to bring them over personally rather than trust the post, so it would probably be best to ask someone in the wider Midlands – say between Stoke and Oxford and Northampton and the Welsh border. I have some nice Rhones and clarets to offer by way of thanks!

    Unfortunately new tubes are not a sensible option any more.


    Thanks for looking!
  2. lsinclair

    lsinclair pfm Member

    I'll bounce this thread once in case there is a kind soul out there who missed it, then I'll let it slide ignominiously down the pages, unwanted and unloved...
  3. orangeart

    orangeart KJF Audio Ltd.

    Hey Leigh, sorry I can't help you but have another bump on me.
  4. audioflyer

    audioflyer No going back once you heard a ceramic cartridge

    Are the Nixie tube solder on to the pcb or are they in sockets?

    I may be able to help.

  5. lsinclair

    lsinclair pfm Member

    Thanks Stefan, hope all is well with you.
  6. lsinclair

    lsinclair pfm Member

    Hi Sharif, they are in sockets and easily removed.
  7. audioflyer

    audioflyer No going back once you heard a ceramic cartridge

    That's good I should be able to to regenerate them using my AVO CT160 valve tester to set the voltage and current.

    Not sure which county you in, I live in Northamptonshire but I work in Warwick.

  8. lsinclair

    lsinclair pfm Member

    I never thought of a valve tester - I guess if you used flying leads from the CT160 it could work quite well.

    Here's a link to a page with the data for the nixies - it's the Z568M I'm using. (Yes, they're big!)


    If you are happy to give it a try after reading that, let me know. I'm only 10 miles from Warwick, so getting them to you would be easy.

    Thanks a lot!
  9. Retrovert

    Retrovert pfm Member

    This comes down to increased heater voltage to burn the coating off the cathode. As others have noted, you can do this with a tube tester set to a higher than normal heater voltage.

    Some have built special purpose gadgets for this. Here's what I elsewhere wrote about this:

    Oxide cathodes (barium or strontium) often have miniscule amounts of impurities such as silicates in the nickle, or residual oxygen which takes time to migrate to the surface. These kill the tube over time. So can applying B+ prior to plate voltage, or using bad bias circuitry which stresses the tube. Temperature cycling also is a problem, as is outgassing and migration of impurities to the surface from any of a number of components.

    The causes for cathode damage are legion, and many can be cured with high voltage to burn off the crud. The disused (i.e. turned off) cathodes of Nixie tubes are known to suffer from these sorts of problems.

    The technique of higher filament voltage to burn off contaminants dates back a very long time. Here are some patents describing how to do it for radio tubes, as opposed to CRTs.

    NB: the OCR for the text was a bit off and I had to manually correct errors, some of which may have been missed. I also didn't quote that much text, since lots of it needed to be fixed.

    The first patent I found on the subject was filed in 1925:
    From 1947:
    One patent dating to 1953, explains how it works and includes an explanation to build a "fool-proof" adapter which removes any shorts as well:
    If anyone builds this, I hope you will post some experimental data.
  10. audioflyer

    audioflyer No going back once you heard a ceramic cartridge

    The Nixie tube has no heaters it work like a Neon indicator.
  11. Retrovert

    Retrovert pfm Member

    Yes, right, of course. I have cathode poisoning on the brain, having been researching B+ issues in output tubes for an amp upgrade. And there it's all about heaters and delayed B+ application.

    But the point is still that a cold cathode gets deposits on it and you need to get the cathode nice and hot to burn them off. So overdriving the cathode does it.
  12. audioflyer

    audioflyer No going back once you heard a ceramic cartridge

    Hi Leigh

    Hope you have had a good Christmas.

    Just letting you know I have recovered your Nixie Tube by using the AVO CT 160 valve tester as the power source and by setting it up for testing Cold Cathode Tubes.

    The AVO CT 160 HT supply is AC (setting this to 300v) I've used a single diode for half-wave rectification this gave me 148v DC. I then set the current using a bank of resistors to give a current range of between 3.8mA - 9.2mA.

    Just like to say your Nixie Tube would make a good reference generator at 80.2v.

    It has take about seven hours to recover the No. eight digit.

    Here are a few pictures.





  13. lsinclair

    lsinclair pfm Member

    Fantastic! I reckon it took 12 years to get to that state. In future I promise to operate a strict rotation policy.
  14. misterdog

    misterdog Not the canine kind

    Loving the Christmas story with a happy ending.

    Cool Yule.
  15. martin clark

    martin clark pinko bodger

    That's all very interesting - nice pics and well done audioflyer!
  16. nickds1

    nickds1 Member

    Resurrecting a long-dead thread, but just a few things to add here - this is a Jeff Thomas NixiSat - Jeff unfortunately passed away a short time ago, but support is still available - the main issues with this clock are related to the GPS unit expiring - before he died, Jeff issued an upgrade kit:

    So, the old OnCore unit, CR2032 holder & antennae BNC were removed, the original BNC hole was reamed out to 13mm and an in-line 6-pin mini-DIN female connector mounted into the cavity (its a tight fit), the space for the MAX232 was populated and the PS2 connector wired in (Vcc, GND, RXD, TXD) - a new Haicom HI-204III 20-channel high sensitivity GPS puck plugs into that - the anode resistors were changed to a slightly lower value, the current limit on the HV SMPS was upped by 30%, the PIC was replaced with one with new firmware (more features and support for the new GPS unit), plus a few other minor electronic & cosmetic changes etc.

    I documented the process a bit on the NEONIXIE-L forum at https://groups.google.com/d/msg/neonixie-l/evzX0noEGXg/JrVaZG3YFogJ

    You will, of course, be aware that this is a very valuable clock - the Z568s are silly money now (I paid USD 25 each for mine back in about 1998) - USD 400 each is not uncommon, USD 250 or so is the target price - the clock should be insured for at least USD 3,000 (one sold a couple of years ago for USD 4,000).


    Nick (neonixie-l moderator)
  17. lsinclair

    lsinclair pfm Member

    Thanks for the post, Nick. I bought the clock kit from Jeff around 2000 and the cathode poisoning is the only problem I have had with it. Sharif cured almost all of it - there are just a couple of digits with the problem now - and as he has shown me the process I now only need to find the time to finish the job.

    I think I saw the post you linked to above when I was researching the poisoning problem. If ever the GPS fails I will be in contact!

    I think I paid a similar amount for the Z568s - for the last few years I have been kicking myself for not buying a couple of spares. The only other thing that worries me is the possibility of the processor failing, and I wonder if I should get a copy made. Although the Nixisats change hands for crazy prices I can categorically state that mine will not in my lifetime!
  18. Michael J

    Michael J pfm Member

    I feel a little bit warmer because of this topic. :)
  19. nickds1

    nickds1 Member

    The CPU firmware is copy-protected so you'll need someone with a HEX file really - I believe that Jeff's co-developer on this project is still active on neonixie-l, so that would be the place to ask - it's worth noting that (obviously) if the old OnCore unit dies, the replacement Haicom unit requires completely different firmware - the old firmware will not work. I've never heard of the CPU on one of these dying.

    The OnCore unit is really a 1st gen device - the BNC-connected puck is just an antennae - with the new puck, the complete GPS unit is in the puck and it talks RS232 serial comms back to the NixiSat, hence the need for new firmware and populating the MAX232 etc.The new puck acquires satellites really quickly and is very sensitive - lock time is now just a few seconds from start.

    If anyone needs replacement nixies, let me know - I still have about 100 types in stock - maybe 10,000 in total (including sets of all the rare ones!) :)
  20. YNWOAN

    YNWOAN 100% Analogue

    Well done Sharif - top job :).

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice