Components and supplies
CD4017
VFD Display
BC547
PC817
470nF Capacitor
1N4007 – High Voltage, High Current Rated Diode
Electrolitic Capacitors
CD4094
Arduino UNO
BC557
Dupont Jumpers
Perfboard
1/4W Resistors
Tools and machines
Wire Clipper
Solder Wire
Wire
Soldering iron (generic)
Project description
Code
Arduino code with working routine
arduino
Downloadable files
General Schematic
I avoided drawing ALL the BC557 transistors because they are just there connected on all the CD4094 outputs, all the same. There's a jumper selector (running on the red line) you must configure on the 4017 chips, it resets the two 4017 at the completion of the grids scanning, you must connect (or solder) the flying jumper on the output pin AFTER the last grid, so if your display has 10 grids the jumper goes on the 11th output. The Arduino connections are outlined in red, you can use the USB power for it but it's strongly advised to use a regular old-style transformer for the remaining supplies if you already have a grounded power supply like your PC. These supplies must be all DC except maybe the filament supply, I added some diodes because it's likely you need a separate transformer with a low voltage secondary. The filament power usually ranges around 3V 150mA, a 5V AC transformer will suffice. The "ballast" thing will possibly be a wirevound power potentiometer of like 100ohms, or a fixed resistor, or also some 1N4007 diodes to reduce the voltage. The BC557 transistors pull the anodes up to anodic voltage and the 100kohm resistors will let the voltage drive fall when the segment is off, while the CD4094 remain at 5V normally. The CD4017s will be powered with grid voltage and don't require additional transistors. There's a "ghetto" reset circuit for the 4017 that lasts like one tenth of a second, you have to wait for it to settle before running the code. The "original" schematic for cascading some CD4017 required other logic gates chips, I used instead an NPN transistor and the optocoupler itself to replicate an AND gate, it's fast enough at closing so it's perfect for the task, at releasing it's instead a bit slower but we don't care, especially because the clock inputs on the 4017 are schmitt triggered and the required speed is not "scary" there. There are various grounds for the chips and stuff as you can see, take a good look.
General Schematic
General Schematic
I avoided drawing ALL the BC557 transistors because they are just there connected on all the CD4094 outputs, all the same. There's a jumper selector (running on the red line) you must configure on the 4017 chips, it resets the two 4017 at the completion of the grids scanning, you must connect (or solder) the flying jumper on the output pin AFTER the last grid, so if your display has 10 grids the jumper goes on the 11th output. The Arduino connections are outlined in red, you can use the USB power for it but it's strongly advised to use a regular old-style transformer for the remaining supplies if you already have a grounded power supply like your PC. These supplies must be all DC except maybe the filament supply, I added some diodes because it's likely you need a separate transformer with a low voltage secondary. The filament power usually ranges around 3V 150mA, a 5V AC transformer will suffice. The "ballast" thing will possibly be a wirevound power potentiometer of like 100ohms, or a fixed resistor, or also some 1N4007 diodes to reduce the voltage. The BC557 transistors pull the anodes up to anodic voltage and the 100kohm resistors will let the voltage drive fall when the segment is off, while the CD4094 remain at 5V normally. The CD4017s will be powered with grid voltage and don't require additional transistors. There's a "ghetto" reset circuit for the 4017 that lasts like one tenth of a second, you have to wait for it to settle before running the code. The "original" schematic for cascading some CD4017 required other logic gates chips, I used instead an NPN transistor and the optocoupler itself to replicate an AND gate, it's fast enough at closing so it's perfect for the task, at releasing it's instead a bit slower but we don't care, especially because the clock inputs on the 4017 are schmitt triggered and the required speed is not "scary" there. There are various grounds for the chips and stuff as you can see, take a good look.
General Schematic
Comments
Only logged in users can leave comments
Anonymous user
2 years ago
A wonderful project, I have been searching for. I want to make use of my old alarm clocks using beautiful green VFDs. Thank you.
screwpilot
2 years ago
you welcome, if you are talking about vintage VFDs (tube or pack) i would really like to know if it works ok or not (it should, in my opinion) if you remember it, come back... maybe power the CD4017 with ~16 volts if they require more juice
Anonymous user
2 years ago
Liels paldies. Tieši sen meklēju šādu projektu ar šo displeju. Jūs esat unikāls. Labākais
Anonymous user
2 years ago
Liels paldies. Tieši sen meklēju šādu projektu ar šo displeju. Jūs esat unikāls. Labākais
Anonymous user
2 years ago
Very nice project. Not many engineers would take on such a more complex wired circuit. And the fact that you made a serial interface on top of getting a VFD to work is a big plus! Also for those who want to know how those displays work an interesting read.
screwpilot
2 years ago
thanks, actually i needed more time to turn it into a presentable project rather than developing it... what convinced me to publish this was the comparison with other more complex solutions, it went apparently lucky because i'm an optocouplers fan (even if it seems possible to use transistors instead), the voltage level shift and the 4017 thing came together, the "from top PNP" solution seems also a slimmer approach used inside late control chips\\schematics
Anonymous user
6 years ago
A wonderful project, I have been searching for. I want to make use of my old alarm clocks using beautiful green VFDs. Thank you.
screwpilot
2 years ago
you welcome, if you are talking about vintage VFDs (tube or pack) i would really like to know if it works ok or not (it should, in my opinion) if you remember it, come back... maybe power the CD4017 with ~16 volts if they require more juice
x_arduino_x
6 years ago
did you get the desk lights from vetco electronics?
screwpilot
2 years ago
not at all, these are chinese 10W leds, i bought em 2€ for five pieces shipped... honestly i'm not sure about quality and efficiency but they seem pretty ok, the "browned one" you saw may have 1000 hours of work (not impressive but it seems still 100% in shape), they anyways have the classic imbalance between chips when you turn em off or power em with microamperes
Anonymous user
6 years ago
Very nice project. Not many engineers would take on such a more complex wired circuit. And the fact that you made a serial interface on top of getting a VFD to work is a big plus! Also for those who want to know how those displays work an interesting read.
screwpilot
2 years ago
thanks, actually i needed more time to turn it into a presentable project rather than developing it... what convinced me to publish this was the comparison with other more complex solutions, it went apparently lucky because i'm an optocouplers fan (even if it seems possible to use transistors instead), the voltage level shift and the 4017 thing came together, the "from top PNP" solution seems also a slimmer approach used inside late control chips\\schematics
Vacuum Fluorescent Display Controller | Arduino Project Hub
laurissuper
2 years ago
Liels paldies. Tieši sen meklēju šādu projektu ar šo displeju. Jūs esat unikāls. Labākais