Roland Juno-106 - Part 4 (Heat Sink and Dead Key)

With the refurbished voice chips installed, and the original factory presets loaded in, I was really enjoying playing my Juno-106. I even decided to use it as the primary source in a synth-pop cover of Pink Floyd’s “Comfortably Numb.” But something kept nagging at me. What if it overheats? What if it just dies? What if all my work and expense is for naught?

It was time to open her back up and do a little more work before continuing with the music. Of  course, there’s a lot more work to be done—greasing the sliders, changing the tact switches, brightening the buttons—but none of that will mean anything if the power transistors fail.

There’s a big heat sink connected to the power supply board, with a number of transistors bolted to it. I’m no electronics expert, so my knowledge here is limited to what I’ve gleaned from the internet, but it seems like these transistors are pretty important. Because of long use, the solder joints connecting these transistors to the PCB can develop cracks, or so-called dry joints. This can result in all sorts of electrical problems, including serious damage to the machine.

To make sure this doesn’t happen, I removed the power supply board and checked the transistor solder joints. They looked OK, but then again my eyes aren’t what they used to be. I decided that it was better to be safe than sorry and reflowed the joints. It was a simple job and I soon had the board back in. No explosions, no smoke, happy days.

Next was the dead low C. I’ve had plenty of synths with dead keys and without fail, all have been fixed by cleaning contacts. Even my Korg Poly-61, which I was convinced was going to need key surgery, eventually came back to life with a little pencil lead. But not this Juno. I tried cleaning, I tried lead. I even tried sacrificing a chicken. OK, I didn’t sacrifice a chicken.

There’s not much happening in the PCB under the keys, just a diode for each note. Of course, the issue could also be in the cables connecting the keyboard to the main PCB, but since it was just one key and not a regular series of them, I figured it probably wasn’t that. I also had my doubts that it was the diode as apparently they’re not known for failure but if anything it would be a good opportunity to work on my diagnosing skills.

I tried checking the diode with my multimeter to see if it was working but it wouldn’t read in circuit. If I was going to yank it out I may as well put a new one back in, so I ordered a bunch (500, my god) from Wish. I got it out, checked it, and it seemed OK. I put a new one back in, held my breath, and gingerly depressed the key. Nothing. I thought maybe I heard a sad trombone playing in the distance but it was only the wind.

The next most likely culprit was a damaged trace. Actually this was the most likely culprit but I had been hoping it really wasn’t this. Unlike on some other synths, the keys on the Juno have to be removed one at a time to get at the PCB underneath. I had already taken them all out once to clean them. I wasn’t in any rush to do it again. But I really wanted to fix that key.

I started pulling them out, careful to set them down in the correct sequential order, and kept my old eyes out for any damage to the traces. Thankfully my fears were unfounded and I didn’t have to look very far until I saw it: a nasty little scar right on the trace leading away from the low C. Maybe someone jammed a knife in there doing an impression of Keith Emerson, who knows.

I grabbed an X-acto knife from my tool box and carefully scraped at the top green coating on the board until I could see the copper layer below. I used a soft artist’s brush to brush away the powder, and scratched off another spot on the other side of the scar. The goal was to expose two spots on the trace on either side of the break that I could bridge and complete the circuit. I snipped a leg off one of my many Wish diodes and soldered it to the exposed areas.

I turned it on, pressed on the rubber contact plunger with the (nonconductive) end of my screwdriver and it happily sounded a glorious low C. As with most of my repairs it won’t win any awards for looks but it works and that’s good enough for me. I should also mention that I’m pleased as Hawaiian Punch that I was able to successfully diagnose and repair the problem, something I’ve been working towards for almost two years now. It was a small problem, admittedly, but it’s a start.

So now my Juno plays without issue, and should hopefully not be electrocuting itself any time soon. There’s still a long way to go on this restoration but I’m very happy with my progress.

I should think about maybe getting some reading glasses though.