There are more tubes in my clocks because I'm using 6-tube and 10-tube rings to divide 60 Hz to 1 pulse per minute. My cloches are not as tall as yours so I used fifteen rows of five tubes instead of seventeen rows of four, giving plenty of space for more tubes. I figured the life of the tube was related to its on time, so having more tubes in a ring should increase the time before replacing tubes. I'm using four brass posts for support. The ladder in the front is +300v, the ladder in the back is ground. The rungs in the back are offset vertically so the bottom of the tube rests on the front rung and the cathode is connected to a slightly higher rung in the back. Two front-to-back diodes interconnect the front ladder and the rear ladder, a pair of diodes on each side. On each side is a filter capacitor between the two ladders. High up between the two ladders are discharge resistors, one per side, giving the four posts more stability. The AC mains are connected through a fuse and a current-limiting resistor to the center connection of the front-to-back diodes on each side. I left off the bottom two rungs of the rear ladder and built a third ladder slightly narrower with only two rungs. The smaller ladder in the rear is connected to the current-limited AC line. The first divide-by-ten ring sits in the bottom, using the AC line as its counting input.
I did several tests to see if having 300v near the front of the tube would affect firing. 300v near the back of the tube did cause the tube to fire earlier, but had no effect at the front. I did find that the front ladder isn't needed to support the MTX-90 tubes. The rear ladder is sufficient for the MTX-90 tubes, their wires are stiff enough for support. The Nixies at the top do need the support at the front. Mike On Sunday, May 2, 2021 at 3:45:12 PM UTC-4 Pieter-Tjerk de Boer wrote: > Nice that you managed to duplicate my MTX-90-based clock and get it to > work! > > If I saw it correctly, your clock has 59 tubes while mine has 56. So > apparently you changed the circuit a bit? > > My clock became unreliable in darkness a week or so after completion. > Adding some blue LEDs fixed this, but not for long. One tube needed to be > replaced. It turns out this tube was an odd one out: it already at the > start had the lowest striking voltage among all tubes in that particular > ring counter, and this had drifted down by some 10 volts, while the other > tubes had drifted less and mostly up. Since replacing that tube, the clock > has worked reliably, except for one (so far) hickup. > > The blue LEDs are still in the clock, it seemed prudent to leave them in > even if they were perhaps not needed anymore after the tube replacement. > > As for powering the LEDs, the easiest is to put them in series with the > entire clock circuit (with some protective components, otherwise they may > not live long - I learnt this the hard way in my earlier NE-2 based clock). > I've updated the schematic on my website to show how I connected the LEDs, > see https://www.pa3fwm.nl/projects/neonclock2/ > > Regards, > Pieter-Tjerk > > > On Sunday, May 2, 2021 at 1:59:44 PM UTC+2 Mike Mitchell wrote: > >> I've built two clocks out of the MTX-90 tubes, following Pieter-Tjerk de >> Boer's schematic. The longest I've gotten one to run is about a week, at >> which point I have to change out tubes. The clock runs fine in the >> daylight but some random tube will stop firing in the dark. I'm thinking >> about sprinkling some blue or green "neon" tubes throughout the clock just >> to provide some extra photons. Something like these: >> https://www.amazon.com/Othmro-Pieces-6x16mm-Bright-Indicator/dp/B07WFNSKSM/ >> I do have some near-UV LEDs (400nm) but the clock is line powered and I >> don't really have space for a buck converter. I hate to drop nearly 300 >> volts across a resistor just to light a string of LEDs. >> >> Does anyone have any other ideas? >> >> Thanks! >> Mike >> >> >> >> On Thursday, December 31, 2020 at 8:21:16 AM UTC-5 Pieter-Tjerk de Boer >> wrote: >> >>> Yes, at first I tried to use them as proper trigger tubes, in multiple >>> ways, but I couldn't find or come up with a circuit that worked reliably >>> over a range of supply voltages (which is also an indication for how robust >>> the circuit is for variations of the tube properties). So I went back to >>> the neon tube counter circuit, and found that to work more robustly, thanks >>> to the very large difference between striking and maintaining voltage of >>> these tubes. Somehow, this seems "wrong", >>> as the trigger electrodes are there precisely to make this kind of >>> circuits easier, but... >>> >>> Regards, >>> Pieter-Tjerk >>> >>> On Thursday, December 31, 2020 at 12:00:07 PM UTC+1 Dekatron42 wrote: >>> >>>> Really nice! >>>> >>>> Thanks for sharing the circuit diagram! >>>> >>>> Did you try to use them as proper trigger tubes and not just as neon >>>> tubes? >>>> >>>> /Martin >>>> >>>> On Thursday, 24 December 2020 at 19:13:10 UTC+1 Pieter-Tjerk de Boer >>>> wrote: >>>> >>>>> >>>>> I recently completed a nixie clock without any transistors, >>>>> microcontrollers or other ICs, instead using trigger tubes to implement >>>>> the >>>>> digital counters: >>>>> https://www.youtube.com/watch?v=E3PBJUcKeoo >>>>> >>>>> Actually, I'm not even using the 'trigger' function of these tubes, >>>>> they are just used as neon lamps with a large difference between strike >>>>> and >>>>> maintaining voltage. It's this difference which allows one to build >>>>> counters with them. I'll publish a more detailed description on my >>>>> website >>>>> later on. >>>>> >>>>> Some may know that I built a similar clock over ten years ago: >>>>> http://pa3fwm.nl/projects/neonclock/ >>>>> That clock used regular NE-2 style neon lamps for the logic; >>>>> unfortunately it became unreliable, as the lamps' properties changed with >>>>> time. I hope the new clock will turn out to be more reliable... >>>>> >>>>> Regards, >>>>> Pieter-Tjerk >>>>> >>>>> -- 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 [email protected]. To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/fa339978-3373-44da-9fb6-f53f778d9843n%40googlegroups.com.
