Yamaha CS-5

I decided I wanted to learn more about how synths work. I can solder and replace tact switches and internal batteries, but the hows and the whys, well, they’re still elusive. I wanted to know more. I asked around on Facebook for a good starter synth and people suggested the Korg MS-10 and Yamaha CS-5, as their circuits are relatively simple. A CS-5 in bad shape came up on Yahoo Auctions for not much money, so I pulled the trigger.

The CS-5 is a single-oscillator monosynth from 1978. It was released a year after its big brother, the CS-10, and was likely intended to be one’s first synthesizer. As an entry-level synthesizer you could do a lot worse. Although it only has one envelope pulling double duty for both VCA and filter, it has mixable waveforms (sawtooth and square), as well as pulse width modulation. Of course, there’s only one LFO (for now) but with the CS-5 you get sample and hold, and it’s a great one to boot. But specifications don’t mean much if the sound isn’t there. Have no fear, the CS-5 is very much worthy of its CS name. It’s smooth and musical, and the filter—creamy with just enough sizzle—is to die for.

I actually prefer the CS-5 to the CS-10. Although the 10 is slightly more complicated (check out that filter envelope), they’re nearly identical in many other ways. And yet the CS-5 is slightly different. Or perhaps it’s just my two units and their own particular idiosyncrasies. Do I really need both a 5 and a 10? Probably not. But don’t forget that you can sync them up with CV and gate and suddenly you have a two-oscillator monosynth.

Cosmetically, my new CS-5 had seen better days. There were accumulations of dust gunk around the pots, plenty of rusty screws, and more scratches than my floor. Sound-wise, the keys worked, but most every pot was noisy like my neighbors. I rolled up my sleeves and got to work.

First, I pulled out the rusty screws and tried to soak off the rust in vinegar. It took the black paint off as well. Oops. I went down to my local hardware store and they didn’t have any black screws, so I bought some silver ones. The three replacement screws on the back didn’t matter so much, as I couldn’t see them, but the four on the front were pretty conspicuous. Good enough for now.

Next I pulled off the knobs from the front panel to get at the gunk under them. I use a guitar pick and plastic wedge tool to pry them up gently. Small movements work well. Next I had to unscrew the bolts around the pots that are holding the two boards to the chassis, and soon I was looking at a very grungy front plate. Water and a sponge took care of most of it, although some gunk was really caked in and required a little isopropyl alcohol to dislodge it.

With all that coagulated dust around the pots, some had surely gotten inside, so I squirted some DeOxit into all the pots, and into the few switches just to be sure. After working them all back and forth a bit, the noise was soon gone. There was more cleaning to be done though.

The keybed on the CS-5 hinges up like the hood of a car, so I was able to get under it and tackle the carpet of dust that lay waiting. I must have spent a good 10 minutes vacuuming that out using the long thin attachment on the machine. There was some corrosion under the thickest dust carpet spots, mostly up near the front of the chassis, but not enough to worry about so I let it be.

Although the keys seemed to be working OK, I figured I may as well clean the contacts while I had the thing on the operating chair. The contact board was easy enough to get off, and geez was it dirty. I’m surprised any of the keys were still sounding. Just a little isopropyl alcohol on a Q-tip—well, actually quite a few Q-tips—was enough to get it looking good. Lastly, I set the whole thing, keybed still propped up, out on the veranda to soak up some sun and hopefully lose a little of its old synth smell.

To tell you the truth, I was a little disappointed that that was all I had to do to get the thing working again. Just a clean. Hats off to Yamaha for making such a well-built machine but I was hoping for more of an educational experience. Maybe it was this desire, or maybe I was just looking for issues, but I started to realize that the envelope release was super long. Everything else seemed OK—the attack was fine, it decayed and sustained normally—but that release. Like 30 seconds when fully open! Maybe this would need some work after all.

I downloaded the service manual and was pleased to find that it was very well written, and detailed. I’ve looked at service manuals before and usually the only part I can decipher is the parts list. But this seemed to be written with the beginner in mind. It has a legend explaining the symbols on the circuit diagram, and even has a break-out section with explanation of how the envelope circuit works. Yes, circuit. It’s not a chip!

I spent a few morning sessions at a cafe before work looking at the manual and trying to figure out how the circuit worked. I also discussed it with a helpful member of the CS group on Facebook, and he suggested replacing the transistors in the circuit. I’m still afraid of poking around with my multimeter with a synth on, and I could still not get my head around how the circuit works in terms of flow of electricity. It’s apparent there’s still more I need to learn, but replacing the transistors would be a start. After a trip to Akihabara I was back with the necessary parts and soon had them swapped out.

Now was the moment of truth. Had I made a difference? The short answer is no. It was exactly the same. I decided to check the envelope timing against the one in my CS-10, and while compared to the regular envelope setting, the CS-5’s is ridiculously long. However, it’s exactly the same as the CS-10’s Time X5 envelope setting, which allows you to use an envelope that’s five times as long as the regular one. Well, what do you know. I still don’t know if this is correct but it’s good enough for now.

By now, the Kenton filter socket mod I ordered had arrived. I wanted to try my hand at some easy modding. I figure it’s a good way to learn more about how electronics work and also add functionality to my synths. (I draw the line at mods that interfere with the cosmetics of the synth though. I realized how important this was for me when I decided to abandon adding pulse width modulation to my MS-20 Mini. I don’t think I could stand having jacks up on the top of the thing.)

The actual mod itself was easy. The jack came pre-soldered with a lead connected to a resistor, as well as a ground wire. The lead went to a point on the board in the filter cutoff circuit, and the ground to a ground post. No, the hard part was drilling the hole without destroying anything. I live in a small apartment in Tokyo and my “work bench” is a table between a dresser and the ladder up to the loft. There isn’t a lot of room to work.

I decided the best way to do this, given the shape of the chassis and the size of my rabbit hutch, was to drill from the inside out. I got two boards from a hardware store and covered them in plastic packing material to create a raised area for the chassis so I could hold the metal down with one hand while I drilled with the other, and not drill straight through my table. I used a center punch and hammer to get started, then the smallest bit I had, and then used a tapered drill to reach the correct width. It worked well enough but I ended up with metal splinters all inside my CS-5. Luckily I had covered the PCBs with plastic wrap to protect them. I used a moist paper towel, then a vacuum, and finally a magnet to pick up all the wayward metal splinters.

The mod adds CV control over the filter cutoff, so I ran a cable from the LFO out of my MS-20 Mini to the newly installed filter CV jack. And it worked! I now had two LFOs controlling the filter cutoff of my CS-5. Two LFOs, just what the kids today love.

How how about those screws? I didn’t mind the silver ones but black would be better, for sure. I bought some cheap spray paint from the dollar store and rigged up a cardboard box in my kitchen to spray into. It took me a couple tries to get it right—the paint kept drying and sticking to the cardboard—but I realized that the only parts that mattered were what was visible: the head of the screw and one side of the washer. Everything else would be under the chassis and thus out of sight.

Although it doesn’t seem like I did very much, especially where the envelope is concerned, in reality I learned a lot. I plan on adding the Tubbutec ModyPoly to my Poly-61 at some point, so I wanted to make sure I could drill through metal without making too much of a mess. Next time I’ll drill from the outside to minimize metal splinters.