Nixie All In One- 2 tubes not lighting up

More
2 years 2 weeks ago #8646 by Grundlenuts
I just finished the All in one Nixie clock and everything works fine except for the ones digit in the minutes and the tens digit in the seconds. All other tubes are working correctly. I am able to connect to wifi and get the correct time. I can't see any cold joints or solder bridges. I tried switching the tubes with ones that I know work and that didn't work either. I am getting about 179-183 volts at the test point. I've tried switching around the octocouplers, and they still won't light up. The wifi module was removed for testing. What else should I be checking?
Attachments:

Please Log in or Create an account to join the conversation.

More
2 years 2 weeks ago #8647 by Ty_Eeberfest
Okay, you already did the first things I would have suggested: swapping optos and swapping tubes. I have a couple things to try. Remember, I know nothing about your experience or skills so these things may already be obvious to you.

Got a scope? If so, look at the waveform (referenced to DC ground) on Pin 1 (the anode) of several optocouplers and compare results. Should be a string of rectangular (not square) pulses.Does the waveform on optos OK4 and OK5 look different from the rest? Totally missing? Normal? If OK4 and OK5 have weird or no waveform on Pin 1 then you want to be looking for problems between the micro and the optos. Like maybe on pins 3 & 4 of the micro (see schematic). If the waveform looks okay then the problem is on the HV side (toward tubes) side.

If you don't have a scope you could still do a half assed version of that test using a DC voltmeter. Pin 1 voltage should be the same on all optos. Hard to say exactly what voltage to expect (varies depending on dimming etc.) but on my clock, immediately after a factory reset, I get 0.26vdc.

Any chance there's something nasty going on underneath the optos socket (or maybe even the micro's socket)? I've heard of cases where solder creeps on the socket pins so badly while soldering that a bridge forms - on top of board but underneath socket - invisible. What made me think of this is that it's 2 adjacent positions failing. You could try various pin to pin continuity checks, using schematic as a guide, to figure out if anything's shorted. Or if you're good at de-soldering you can just remove the socket and look... not fun.

Look into it later when the dust is clearing off the crater.

Please Log in or Create an account to join the conversation.

More
2 years 2 weeks ago #8648 by Ty_Eeberfest
Also. If you end up doing continuity checks to search out shorts under sockets be sure to take the micro and all 6 optos out of their sockets first. Otherwise you'l just end up going in circles chasing false readings caused by sneak paths.

Look into it later when the dust is clearing off the crater.

Please Log in or Create an account to join the conversation.

More
2 years 2 weeks ago #8649 by Grundlenuts
ok, so I pulled out the optocouplers and the micro and checked for shorts, pin by pin looking for continuity, but didn't find anything. I tested pin 1 of each opto per your recommendations, as follows: OK3 225mV, OK2 226mV, OK1 225mV, OK6 .68V, OK5 474mV, OK4 306mV. At this point I am stumped. The opto voltages are way off and I have no idea why. Any other ideas?

Please Log in or Create an account to join the conversation.

More
2 years 2 weeks ago #8650 by Ty_Eeberfest
With those readings it seems like you should have 3 dead digits and yet you only have 2. That's rather puzzling.

Last thing I can think of to check is the ground on pin 2 of each opto. Do it this specific way: First, with clock running, tie your negative meter probe to ground and probe pin 2 with positive probe. All pin 2s should be the same - a solid zero or really close to zero. Second, pull out the optos, tie positive probe to +5, power up, probe each pin 2 with negative probe. All should read very close to +5v.

If any of those readings are off, you're somehow not getting ground to the opto. Maybe look for a cut trace on the board. Just so you know, I only have a Modular here to look at right now and it's not 100% the same as the All In One. What I see here is that the pin 2s of all the optos are "daisy chained" with a real thin trace.

Look into it later when the dust is clearing off the crater.

Please Log in or Create an account to join the conversation.

More
2 years 2 weeks ago #8651 by Ian
OK, we can attack it from the other way as well. Pull all the optos and while the clock is running short pins 3 and 4 together of OK6, and then OK5. Be careful, there is HV on those pins. I would suggest making a small jumper and installing it before turning on.

The tube in each case should light up. Don't run it for long like this, we are over-driving the tube, but it will survive the test with no problems.

If a tube does not light up, we know to look on the tube/R27/R28. If the tube lights up, we need to look at the controller and R19/R20.

If it turns out to be the controller side of things, please try to re-seat the controller and check the soldering around pins 3 and 4 of the controller socket. Also inspect the socket carefully around these pins. The sockets sometimes get a bit bashed around in transit.

Please Log in or Create an account to join the conversation.

Moderators: AccutronTy_EeberfestIan
Time to create page: 0.197 seconds

Search

Tube Suppliers

Go to top
JSN Boot template designed by JoomlaShine.com