Vintage Synth Explorer Forums

I see you´ve had already replaced the capacitor. And also replaced the vactrol with a modern version? In that case we should look at the power supply to the LED inside the vactrol. Just a minute...

Here. Adjusting the trimmer R101 is maybe all that´s needed to increase the audio level. Inside the vactrol, as the LED lights up, the photocell decreases its resistance, letting more audio through. A bad Q21 would not let the current feed the LED. And of course a bad Q20 would not let Q21 feed the LED. It never ends...

I remember adjusting the trimmer in October of 2014, and it didn't change the output level.

Then I´d replace Q21, then Q20, then Q19...

I'll try it and see. Thanks.

Forgive me for resurrecting an old thread, but I believe it is for a good cause - please read on....

This is my first post here, and possibly not my last. A couple of months ago a friend asked me if I would look at a keyboard for him. Having a background in electronics (ten years in an electrical engineering lab), I thought this would be a fun thing to try to repair.

Lo and behold, when doing some online research on this keyboard, a Roland Strings RS-101, I happened upon this thread. As the keyboard was given to me to work on, it also came with a lot of marked up documentation, where the keyboard had been recapped and other work performed, dated in 2014 (note previous post dates).

I read this thread with great interest, because, and I would bet large dollars on this, that the keyboard I have is the exact one that was being worked on and asked about in this thread. All of the same problems described herein are exactly as I found them, and some of the marked up doco schematics exactly matches pictures posted in this thread. What are the odds, eh?
Nonetheless, I have the keyboard fully up and functional, and returned it today.

I looked at the work that was previously done, and whoever did it was fairly skilled at soldering. The re-capping had no ill effects on any of the PCB traces (none were lifted), and the joints all looked good. Kudos.

Here is a synopsis of what I had to do to make it functional again:
The keyboard had no output whatsoever when I started.

First things first, I disconnected the power supply from the internal PC boards and turned it on. The output rails are supposed be -15V, +10V and +20V, but were more like -3, +4, +0. The first problem was the diodes D15 and D16, that together, form a bridge rectifier. One of the diodes in D15 went bad. I removed them and replaced both of these "dual diode" pairs with a single bridge rectifier. After that, the output was -15V, +10V, +30V. Was not able to figure out why the +20V rail was 30V, so slightly re-engineered the power supply by disconnecting/removing Q31, a 2SD234 pass transistor, and installing an LM317 voltage regulator with the appropriate adjustment resistors for a 20V output. Power supply done.

Next was the no output issue.
I saw that the original analog optocoupler ("Vactrol") had been replaced. I noticed that the output side directly controls the entire output, so I just put a jumper/short across the resistor side of the device, and the thing started working! Not completely, but I now had sound coming out of the output jacks.
I removed the optocoupler and tested it out-of-circuit - it was good, so that means that the LED side was not being turned on for some reason. I also discovered that the Vactrol had been installed incorrectly. It was installed with the writing on the device facing up (an intuitive way to do it), but doing so installs the device with the LED reverse biased and it will never work that way, so I turned it over. Since the output is just a resistor, that side can go in either way. I now had the LED installed correctly, but still no output. That put the problem somewhere in the output controller section. The trimpot R101 had been replaced, and checking it revealed that it was bad. Replaced the trimpot. Still not working. I was not getting any signal to the base of Q21 (the transistor that drives the Vactrol LED), so looked there. Suspected that the 10ufd/25V tantalum capacitor (C53) had shorted out, and it was indeed shorted. A common failure mode for tantalum caps is to short out. I replaced this with a 10ufd/35V cap instead. I felt that running a 25V cap at just under 20V (20V is on the emitter of Q20). This won't affect the sound at all.
I now had output without having to short the resistor side of the Vactrol. Getting there......

There were two non-working keys. I first verified that the keyboard switches were good, then verified the wiring back to the tone generator board. All checked out OK. Because I found a shorted out tantalum cap in the output controller section, and there were 61 identical caps on the tone generator board, I started there. I tested each one to see if any were shorted. Sure enough - two caps were shorted, and they happened to coincide with the non-working keys. These were replaced with identical spec tantalum caps - 10ufd/25V.

I did notice that the power cord had been replaced, but I was not confident about the installation, so shored that up a bit with some good strain reliefs. Internal wiring was additionally secured using Velcro strapping. The wood at one corner of the lid had separated and the aluminum pulled away, so fixed that too.

All in all, I would have to say that this is one of the most satisfying repairs that I have done in a long time.

Hey thanks for solving this years old mystery and letting us know !