Nixie clock circuitry finally complete!
It may have looked like I fell off the earth, got distracted by shiny things again and dropped my electronics projects – again. Well… NOPE. I’ve been furiously, wildly and tediously soldering this beast together.
Things I learned along the way
Well, the first thing I learned is that building anything beyond a simple project is a right pain in the arse on these solderable prototyping boards. I do like their convenience, and previously I had assumed I would be limited to using this and other perfboard-type prototyping boards, but from here on out, I’ll be making proper PCBs. You’ll see why below.
A right rat’s maze
One big drawback to making your projects on perfboard or solderable breadboard or similar is the sheer number of solder joints you have to make. Not only do you have to seat all of your components and solder them in, but you need jumpers or longer wires to anywhere it has to connect to. This is fine for a simple design, but in this case, it got obscene on board two.
As you can see, I’ve had to use so many jumper wires that it almost completely obscures the active components beneath. The 4017 counters are nicely and properly socketed – but good luck getting them out of there. It was a small miracle that I didn’t fry any of them during the build, narrowly avoiding a rather annoying process of getting the part out from under that wire blanket.
What is all that mess anyway? Well, apart from connecting the 4017s, its support stuff, and connections between, each digit I intend to use with the nixie tubes basically needs its own low-side transistor (MPSA42), which the pulses from the 4017s will activate, lighting the digit. So add it up. I’m using 10x minutes digits, 6x 10s minutes digits, 10x hours digits, and 3x 10s hours digits. That’s 29 transistors, plus one for the neon colon, making 30. Each of those needs a connection to ground on the emitter (fortunately just a jumper to the rail), a connection from the collector to the pcb connector for that digit, and a 33kΩ resistor connection to the appropriate 4107 output. So, that’s 90 connections, or 180 solder joints. Jeez!
So, that took me from the last post I made until today basically. That and mounting the nixies, wiring them up all nicely and completing the project save for having a nice enclosure for it. I’m thinking a nice wooden box.
Here’s the result
Well, here it is!
Please excuse the quality of these. For those that know me – I am the absolute worst photographer on the planet. I should take a class or something.
One thing any readers should be aware of is do not use the older schematic I posed earlier in this chain of posts. I’ve made some modifications, which I detailed, but I have yet to make a revised schematic for the whole thing. I will do this very soon. In short, here are the changes I made:
- Moved the decorative blue LEDs from the 5V circuit to the 12V circuit
- Added a 9V battery backup with a protection diode
- Swapped the darlington transistor I was using for the LEDs for a MOSFET and tweaked the RC timing circuit there
- I removed the switch for turning off the nixie’s as redundant
- Adjusted the neon colon current limiting resistors from 10kΩ to 100kΩ VERY IMPORTANT! That was a typo on my previous schematic
So not a lot of changes, but some very important ones. I hope you agree that the end effect is pretty nice indeed.
On the whole, while the build was tedious as holy hell, It went mostly smoothly. I managed to not burn or electrocute myself, nor melt any components to release the magic smoke. All that remains now is to find a nice case for it.
As mentioned above, I’m unlikely to continue using perfboard/proto-board for future projects (unless they are stupid simple and tiny) and will be having PCBs made properly for them. I’ll be posting about how in future posts involving two more projects I’ve been developing for some time.