FISHERS (project pt.7)

 -WIRING/DEBUGGING:

Black and red wire for power connections

With all of the circuit boards aligned to their temporary faceplates, and fitting to the enclosure how they should, the next step is wiring everything up and turning the power on for the first time, which is always exciting. The wiring process was pretty simple since I took the extra time to label everything well, and even had a sort of plan for how everything will be connected so that the wires don't overlap too close to where they are connected.

I decided to use this thin wrapping wire for all of the board-to-board connections, with the exception of the keyboard connector. For that I used a 1.27mm pitch IDC connector cable. Unfortunately, the connector cable's connections broke pretty easily after a few times folding it over during debugging. I had to re-strip and solder it again, and secure it with hot-glue. I probably should have just used the wrapping wire instead. The wrapping wire was an impulse buy from years ago–probably from ebay–that is actually pretty decent wire. It is super thin, but pretty tough, and it is pre-tinned, so soldering it is easy. I seem to remember it being super cheap, and I have a ton of it, so using it for longer leads is great because it can be twisted together in groups for stability. Having a massive roll of 50-pin IDC ribbon cable on hand has been nice over the years, but it is bulky, needs to be ripped apart, and tinning the stranded tips can be annoying. I think I'll be using more of this wrapping wire in the future. Plus it's colorful!

Keyboard wired up

Capacitive module

Joystick depth control board

After making all of the board-to-board connections, I connected all of the peripherals like the capacitive touch module, the 2-axis joystick and its depth control/LED board. Once those were done, I wired up the output jack and the power-supply input jack, and then the CV and clock inputs and output jacks. I waited until this point to decide which parameters would get CV jacks. I designed the circuit so that almost all of the parameters could be modulated externally, or any modulation source could be output(24 in total), but would have to choose only so many to be practical. I ended up going with 12 that I felt were essential, including separate clock inputs for the sequencers. Some of the CV inputs needed to have some extra protection resistors and diodes added since I hadn't designed them in the original schematic. The 'Echo' CV parameters were not intended to have external modulation, but after plugging the Sequencers' and the EG's CV into them for a listen, it was clear that they had to be added!

Powering on for the first time can be stressful. In the past, I have been traumatized by powering on to find a plume of smoke, or exploding capacitors. The scariest thing though, is when nothing happens at all, and by the time I think to turn the power off, it is too late and everything is fried... This is why I have a sort of ritual before turning the power on for the first time. I go through and check all the polarity of any bypass capacitors or power diodes. Then I check for continuity between power rails all throughout the circuit, and at random points on different boards to be sure nothing is shorting on the power supply. Then I do a close-up visual ' tour' of every component and connection with a magnifying glass to find anything out of the ordinary, always cross-checking the schematic. Only when I am sure there are no visible errors, I turn the power on with my multimeter connected to the power supply. I also have my oscilloscope on, and ready to probe quickly, so that if I need to test something with it, and find that I need to quickly turn the power off, I won't be dilly-dallying(technical term) any longer than I have to.

Thankfully this time there were no catastrophes. The circuits powered on, and started to make sounds, but I could tell right away that there were a lot of bugs. Most of the bugs were solved by simply making changes in component values, or reversing the direction of a switch or pot here and there. There was a pretty substantial mistake in the noise-voice 'RAW' filter circuit that took the better part of a day trying to fix. It turns out there was nothing wrong with the design, the circuit was just connected in the wrong place from the signal to the filter. This was probably due to me trying to make my schematic look as clean and symmetrical as possible, and in the process botching the circuit. The other big bug was with the 'Key-gate' circuits that would not work–or if they did, would double trigger, or only trigger every other clock-step. I probably spent two days trying to re-invent the circuit to make it work as intended when I finally realized that the only problem was the value of the capacitor on the gate-to-pulse input circuit that I had mislabeled in the schematic. The value of the capacitor shouldn't have made such a difference, but the circuit is pretty touchy, and by simply changing the capacitor from 100nf to 220nf, everything was working just as it had on my breadboard, with no need to change anything else.

The next step after confirming everything works as it should, and fine-tuning all of the trim-pots–securing the wiring with hot-glue was important to do right away to avoid any stress on the wire connections. Hot-glue is great for holding wires. If used thoughtfully, it is actually easier to remove or alter than zip-ties, especially if you have a hot-air wand handy.

Then, more testing and playing extensively to find issues or make necessary changes to improve functionality. I like to have everything ready so that the keyboard is as close to finished as possible before adding the final faceplates, hardware, and knobs.

Part 8 coming soon...