Trying to hand-assemble a free form SMT transistor clock. Likely fail.
I posted the KiCad project files on GitHub if anyone is interested.
Here's a photo of the "spine" with air wires.
This took me about 30 minutes to solder.
it seems fairly sturdy when I bend around on it, so I think it'll hold up.
Next is to make the two outer three-component groups, attach them to the spine, and then..... TEST!
After moving things around and building test circuits for just about forever, I think I've finally got a layout I can live with.
I suppose I should put the binary files on GitHub or something.
In the mean time, here's the schematic and layout for one flipflop.
The white lines are "air" wired that will be magnet wire. They are things like ground +9V0 and some other signals that I couldn't get routed perfectly. I also will wire up an input signal and two output signals.
I like having the central "spine" of 10 parts and the two "leave" segments of three parts each separated by wires.
When I connected them all together "hard" the unit was too fragile.
I'm going to rig up a wire structure that I can glue the three parts to to hold them together securely.
That's the next step.
I spent about two hours trying to figure out how to rip up traces, watching Youtube videos, looking at websites that didn't match my vesion of KiCad.
I even BUILT THE DAMN THING FROM SOURCE and I just could NOT figure out where the damn ripup tool was!!!
Then, about 15 minutes into a Youtube video, there, right in the middle of everything, was......
(you hit the delete key to ripup. There's no graphical deal or pull-down menu item!)
So, I feel like a complete dunce, but I got some things done.
I built a schematic of one flipflop.
I added footprints to the schematic.
A created a PCB.
I moved parts around to get the best layout I could.
I'll look at it a bit more again and then I think I can start soldering flip flops together and it'll go pretty smoothly.
I'll add the kicad files wherever I can find to add them. ;-)
I spent a few days soldering parts together trying to rearrange the schematic in my head to make building easy.
I made plenty of nonworking flip-flops, so I went back to basics.
1) Install KiCad
2) Enter the schematic of one flip flop
3) do a PCB layout pretending I don't have a PCB so i can get all the parts in the right order and connected together properly.
4) Start soldering!
I'm on step #3 and I'll post the KiCad files soon.
I got a chance to power-up the first divide-by-two and it actually works!
I've got both output wires soldered on the wrong place, but when I scope the right place, I can see it flipping and flopping! Top trace is a 9V 60Hz signal from the signal generator. Lower trace is one end of the divide-by-two running at 30Hz just as expected. The amplitude isn't right since I've got the wire on the wrong place, but it's working.
Here's the monster wire-nest of testing:
Here's a three-trace image showing the input at 60HZ, the normal output at 30HZ and the inverted output at 30HZ:
Sorry about the photo quality. I'll do better in the future.
So, IT'S WORKING!!! Now I'm really in trouble.
Oh, GOD, this was crazy. It took me about two hours to get this together.
now I have to test it and see if it actually works.
With all the darn SNOW here in Cambridge, I haven't been able to get to my lab at Artisan's Asylum, and I'm just buzzing waiting to get back to finish and test the first divide-by-two unit.
Did anyone notice I soldered that transistor in wrong?
Here's what it should look like.
I can see that testing each flipflop after assembling and before adding to the final circuit is going to be critical.
Well, I got the first FOUR parts together and it went fairly well.
I decided to put C12 and R43 on top of one another since they are in parallel.
I stacked them in the reverse tweezers and added a little dollop of flux.
Then a touch of solder.
Next I arranged R41 and Q11 in two more "third-hand" with reverse tweezers and more flux.
Add heat and solder.
This is the result. The first complete net of the project.