I finally got done wiring up my 32x32 matrix, and got it all up and going with some basic test software, but something was not right

or to get to the point my brilliant self wired up the cathodes wrong, so instead of being able to address 1024 points, I can only control an 8x8 matrix and mirror it 15 more times across the screen


luckily its wire wrapped so it will be easy(ish) to remove and redo, but dang that was a lot of time in the crapper

moral of the story, no matter how much you think and plan, there is really nothing from jamming your head in your arse when you go and make it

Ouch! Man I feel your pain…I’ve made way too many mistakes like that :(.

Hope you get that worked out relatively quickly ;D.

took about 4 nights, couple hours / ~64 twist’s a night (and I wasnt going hardcore on it, so couple hours with smoke, tea, restroom breaks)

When i build one of the VCO’s for my analog synth i spend half a weekend troubleshooting some strange intermittent problems it had.

In the end the only problem was a defect crocedile clip test cable :slight_smile:

My worst test problem was a TV text display circuit in the 70s. After a lot of tracking down I knew that there could be no pulse coming from a monostable but when I looked at my scope there was the pulse.

This had me puzzled for quite some time until I discovers that the action of putting my scope lead on the monostable pin caused it to work and that the display had actually sprung to life, but I wasn’t seeing it as I was looking at the scope.

Monostable chip replaced, problem solved, but it had me going for some time.

In the end the only problem was a defect crocedile clip test cable

These used to be known as “wheel-tappers’ faults”, after the tale of the wheel-tapper in the days of steam locos, who sent fourteen locos to the repair sheds to have their wheels replaced, before they found out that it was his hammer that was cracked.

We’d call them “Heisenbugs” - things where the observer changes the behavior…

well I am in a bit of a tug o war with myself, I need more wire

radio shack sells it, but I have to be honest, its garbage. It constantly breaks and its thinner than decent stuff

but the digikey stuff I used for the cathodes is much much better, and alot faster due to it not breaking + its a shade thicker (your 30 vs my 30 crap)

now the problem is, I want to work with the good stuff, but its nearly 10 bucks + shipping + 4 days wait

the radio shack stuff is only 4 bucks, I can drive 40 min and pick it up now, but it was a massive headache to work with

so what should I do, spend more, wait more and have a 100X easier time, or cheap out and start cussing today

I dunno

Go with the quality.

One of my favorite stories is one of two bulls an old one and a young one, stood at the top of a hill looking down at a field of cows.
The young bull says “lets run down the hill and have some of those cows”, to which the old bull says “no, lets walk and have them all”.


yea I went to digikey and got the “snotty” wire, but its night and day difference

the silver coating is slicker so it doesnt get jammed up in the bit, its thicker, it has slightly less capacitance in its insulation (really tho all this stuff sticks like packing peanuts in a darn Tesla coil, who thought that up???) blah blah blah, its just worth it

And here i was, impressed with my cheap RS wire :(.

Ah well ;D.

rs wire is fine, until you have to put it in a twist bit and wrap it around a square post

guess who just messed up some of the anode/cathode stuff in their own 8x8 matrix?

points to self this one.

Wasn’t too hard of a fix though, now it’s working and i’m getting into code for it.

Well its rewired correctly, tonight is the first time I have tried it but it was done on Sunday

I do have 3 columns and 1 row out, but A) that much better than last time and B) that is pretty darn good considering I removed 100% of the wirewrap and redid all of it backwards (so I would not have to reverse logic in script) without a model, from 10am Sunday morning until 11pm Sunday night

anyway I am going to go fix these 4 wires and write some basic test routines, Ill post pictures in a few days


I once had a problem like that with my 2n3055-based h-bridge that I have breadboarded; I kept having a problem where I could get the motor to turn one way, but not the other. I figured the maybe one of my transistors were fubar’d, so I swapped a couple around - and I still had the problem!

WTH? I started thinking my breadboard was working right (or I had a short somewhere); I tried everything for a few hours - what was even more maddening was the sometimes (but only briefly!) the circuit would work - then stop; sometimes swapping the transistors would help - it didn’t make any sense.

It wasn’t until I decided to try out a different set of aligator test leads (which I used to connect the collectors together on the TO-3 cases) did I find out, and confirm with a multimeter, that one of them had an intermitent break near the clip - once that was sorted, I resoldered the clip, and haven’t had a problem since.


well after checking my connections, the parts of the screen that are out seem to be connected just fine

I notice i get a ton of noise on my data lines (~3 inch unshielded wires), I am wondering if I should try shielded wire, or is there a way to, I dunno decouple the data lines