--
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/831d7192-f611-4b27-87c9-ac988c059ac4%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Another option (just to help troubleshooting but involves no soldering) is to change the driving sequence to avoid adjacent tubes to be driven one after the other. For example if you are driving the tubes in 1-2-3-4 sequence, you can try with 1-3-2-4, which should produce ghosting only between tubes 2 and 3.
It is not the transistor that keeps conducting or leaking. The voltage you see at the anode comes from leakage from floating cathodes from the other tubes. What you can do to kill the ghosting is connecting 10 diodes to all tube cathodes. Connect the anodes of the 10 diodes to the 10 cathodes of the tubes. Tie all cathodes from these diodes together, and connect that to a voltage divider made of 2 resistors of 10k, between your 180v supply and ground. That should kill any ghosting...
I can recommend reading the Burroughs document N101 on multiplexing Nixies which can be found here, if you haven't done so already: http://worldpowersystems.com/ARCHIVE/Burroughs/index.html ./Martin
It is not the transistor that keeps conducting or leaking. The voltage you see at the anode comes from leakage from floating cathodes from the other 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/951afbbd-ff69-4fa9-8ca9-2106ee3b5bbb%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
--
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/8aac8a90-6197-4d1e-8963-f8b673a12842%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Tomasz - which nixies are you using in this application? I seem to remember some discussion from years ago that some tubes (maybe ZM1000?) were particularly prone to ghosting and leakage effects even at microamp currents.Jon.
And use a 50/50 divider, to make the voltage between anode and cathode of the tube as low as possible. Do not set it at 120V, but go for 90V, using equal resistors. If your HV is higher than 180V, try reducing that to at least 180V or perhaps 170V.
From the datasheet, collector leakage current is about 100nA. Assuming all of the NPN leakage current is amplified in the PNP, you would get 4-5uA of actual leakage to the anode from the PNP. That doesn't seem like enough current to me to cause a visible glow in a nixie.Your scope probe (10meg) will "pull" 5uA of leakage current down to 50V. You can do a quick leakage test on the PNP by disconnecting the nixie anode, and any bleeder resistors, and measure the voltage at the PNP's collector when it's off (but while your power supply is on, of course). You should measure substantially less then 40v; anything near or more than that, confirms a leaky PNP or NPN. Be aware that if the NPN is leaky, it can cause a good PNP to appear leaky, because your driver is basically a 2-stage amplifier.Based on the scope pictures so far, I'm not convinced it's leakage. But we need to know if we can rule it out.The last set of scope photos show good turn-on, turn-off, blanking, and control signal behavior. So I think that can be ruled out.There's definitely a clue with the 'ugly' shaped signals; I just havn't figured it out yet.Can I assume your probes are properly compensated ? The leading-edge 'spike' at the anode when it turns on is hopefully the effect of the tube ionizing. But on the falling-edge I noticed some undershoot which goes significantly below 0 volts. That could be a clue.
I've created a properly optimised and searchable PDF with the correct metadata from the JPGs below which can be found in the group library under Nixie Tube Data -> Application Notes.
Nick
Nick