All extremely interesting - perhaps I should not worry about it so much and be grateful that they work at al!
For my project I am using a nice A/C for the filament drive (a custom designed board) then DX for the grid. The segments are being controlled by an HV5812 VFD driver. The filament has a 10R resistor in series to help deal with the situation where filaments are started from cold. There is also a suggestion that the grid and segments should be driven via series resistors too. I have not seen this in any other designs but am more than happy to have a go! So far then hand wired bench testing is working nicely (apart from the patterns I noticed) so I'll be pushing ahead with an 8 digit versions (using two displays as separators/colons) though my design does allow for a 6 digit (or any multiple of 2) version that will also allow the use of an IV-26 as a colon separator (with the ability to control all 7 dots if desired!). Essentially I have a board that accepts two ILC1-1/7's and has a single HV5812 driver. These boards will daisy chain together as desired. I also came up with a small 'link' board which has an IV-26 (there are 3 types - I propose the use of the one with all 7 dots controllable) That board has to borrow spare 'bit's from the adjacent ILC1-1/7 boards. (The HV5812 has 20 bits output - 14 needed for two displays - 6 spare so the IV-26 board has to borrow some from the left and some from the right - if you see what I mean). For software I'll be using the same code base that I use on my panaplex clocks so it will have all the usual bells and whistles (PIR control, web gui for settings, date/temp/pressure/humidity displays as well as a definable scrolling message - albeit subject to the limitations of 7 segments! Anyhow - it's all on the bench and I'm about to push the button on the main tube boards so I can test them more properly. I'll keep you posted. - Richard On Sunday, 23 November 2025 at 16:15:30 UTC gregebert wrote: > I saw similar on the smaller ILC-1/8, so I applied rectified AC on the > grid and the problem went away. In my case, only the vertical filaments > caused dark bands, not the mesh. > > I concluded there are some electrostatic patterns that are, well, static > when the grid is pure DC. With the grid varying from the rectified AC, the > field changes and the phosphor averages it out. The other departure from > normal vacuum-tube behavior is that the grid current can be surprisingly > high, over 100mA compared to microamps for a vacuum tube, because the grid > is biased positively. > > If you look at aging VFD's, you can clearly darker bands closer to the > filament (phosphor degradation) which I concluded were caused by electron > bombardment. > > On Sunday, November 23, 2025 at 6:35:38 AM UTC-8 Tom Katt wrote: > >> Almost looks like the fluorescent material was unevenly applied or >> delaminating somehow.. Like thick / thin spots or similar thickness or >> distance issue. >> >> Does the anomaly align with the cathode heater wires? >> > -- 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, visit https://groups.google.com/d/msgid/neonixie-l/080b43bf-e707-4693-9e08-88c4d638a6b8n%40googlegroups.com.
