The general idea is a homemade desk clock that is pretty and retro as f**k. I will also use this as an excuse to bring my fresh assembler skills to the real world, away from a lame training board with seven segment LEDs. LEDs! What is this, the 21st century?!
Main features are:
- hackability (programming header accessible without opening the box etc)
- opensourcery and easy to reproduce (no specific ICs unless necessary)
- sexy looking (retro enclosure)
- precise (radio synced)
- just one 12V input voltage
- noob techniques used (no 3D printers, no CNC), PCB designed for low price
Russian single VFD in a tube
Something arduino compatible
because I'm a lazy bastard
HV5812 or similar
VFD latching shift register driver
Well, it's not really procrastination that kept this project on the backburner for the last year and a half, but more like total demotivation to do anything creative. Well, ladies and gentlemen, I'm back to and I'm ready to make stuff!
There'll be a small change in direction in that I won't try to build everything from scratch. Having PCBs made and then finding out that it just won't work and having to troubleshoot tens or hundreds of components sounds like it would ruin the whole project for me. I think you can still learn a lot even if you use the easier routes.
I'm currently designing the schematic with KiCAD. My workflow is as follows:
1) identify challenge
2) look at Mouser for integrated solutions or use random internet reference designs
3) study datasheets, create components in KiCAD, do the wiring
4) repeat for next challenge
I'll try and upload the KiCAD files after small milestones.
EDIT: I didn't know shit about vacuum tubes when I wrote this. The
cathode is basicall both pins 2 and 3, but between 2 and 3 there's the
filament heater. Basically, you gotta ground one of the pins and put
1.5V on the other. EZ! Also I'll be using a PWM controlled boost
converter to be able to control the segment voltage.
I translated this (link to datasheet in links section) from russian into english using some online azbuka to alphabet converter and google translate. Here's the important stuff I gathered:
1 anode e
3 cathode, wire layer inner surface of the container (difference from 2 being...?)
5 anode d
6 anode c
7 anode b
8 anode a
9 anode g
10 anode f
filament voltage 1.5V
segment anode to net voltage 25 V
filament current 90mA-110mA, typ. 100mA
net current N/A-17mA, typ. 12mA
aaand the rest are candelas/metres squared and I can't be arsed to translate that as well
So the thingy that will have the outputs of a BCD to DEC converter and the 28V power supply fed in and put out the high voltage when switched by the BCD to DEC has been optimised from the one used in http://web.jfet.org/inGrid/ to my specs - higher current, lower voltages. See the Yenka simulation schematic snaps: