Weak output from Roland RS101 string synth

[servicenote]

I tested the output of transistor Q23 at the base, it's strong and modulated.

Out of curiosity, I also tested the base of transistor Q22; there's no signal there.

[Dr. Phibes]

So the opto-coupler was not the issue? Bummer. Sorry for dragging you down that blind alley.

But did you say the base on Q22 is not showing much? Did you replace Q22 along with Q23?

[Yekuku]

As you reported , you get a low pulsing noise out of Q22's collector. It could be a suspect.
Do you have output from q23's Collector ? It is connected directly to the base of of q22, so they should have the same output. if not , check for broken tracks/joints in their path

[servicenote]

So the opto-coupler was not the issue? Bummer. Sorry for dragging you down that blind alley.

But did you say the base on Q22 is not showing much? Did you replace Q22 along with Q23?

That's okay, Yekuku, your help (and Dr. Phibes') is very valuable to me. I never would have gotten this far without you. The solution is getting closer and closer.
I need to re-test the signal at the bases of Q22 and Q23. I think the reading I took was from the collectors (the middle legs) of these two transistors instead of at the bases. I thought the base of a C945 transistor was the middle leg, which I just now realized is not -- it's the rightmost leg (with the flat side of the transistor body facing upward or outward). So to answer your question about output from the collectors (the middle leg), Q22's collector had no signal, and Q23's collector gave a strong, modulated signal.
I'll test the bases of these transistors again tomorrow and let you know the results.

[servicenote]

Oops, sorry, Dr. Phibes, I thought I was quoting Yekuku in my response a few minutes ago.
I'll also check for broken tracks/joints in the paths of Q22 and Q23.

[Dr. Phibes]

Yeah I've replaced C945 transistors in Roland stuff before, it's something to remember.

If the collector of Q22 is not showing any appreciable signal I think it would be a good idea to replace it. Take it out and do a basic diode test on it and then test its gain (Hfe on most multimeters). Bear in mind that a faulty transistor can still *appear* to work with a basic test but it will be all buggered up in the circuit. C945s are cheap anyway.

[servicenote]

I did audio testing with Q22 and Q23 today, and both are showing a strong, modulated signal at the base and emitter (the left and right legs), but nothing at the collector (the middle leg). I had replaced these two transistors about a month ago with NTE199's, so they're both brand new. I looked online for some C945's but couldn't find any. After some research, I found that NTE199 was recommended as a replacement, so I used those. The general output signal is the same with the NTE199's as it was with the C945's. Do you think I should do some more searching for C945's?

[Dr. Phibes]

Nah, if the problem remained after replacing them then it must lie elsewhere. To be honest, having no direct experience of the RS-101 I'm somewhat out of ideas. That said, I did notice there are two 'soft volume' switches in the schematics that I presume alter the loudness of the preset group they're allotted to. Do they still alter the volume (what you're getting anyway) as they should?

[servicenote]

Yes, the soft volume switches are working normally.

[servicenote]

Also, I examined the Modulation board for broken tracks and joints but didn't find anything.

[servicenote]

One thing I noticed with the trimpot R101 (for adjustment of output) -- adjusting it doesn't seem to change the output any, no matter what position I turn it to. From what the manual says (p. 10), I assume this pot adjusts the amount of modulated signal to unmodulated signal, but I can't hear any effect on the sound when I adjust it. (I replaced it with a new one a few weeks ago). Should adjusting this trimpot make a noticeable difference to the output?

[EthanAndrews]

Hi, Did you ever figure this one out? I just got an RS 101 with the same problem, or at least one that presents the same way. I'll probably try some of the steps mentioned in the thread, but I'd be interested to know if you eventually nailed it.

[servicenote]

Hi,
Unfortunately, I never did find the exact location the problem, so I stopped working on it soon after the October 29, 2014 post. I had to return the multimeter I borrowed. Maybe one of these days I'll get hold of another multimeter, and have another go at it.
But I wish you good luck in your endeavor!

[Mooger5]

One important thing about audio probes: they must be AC coupled, that is, they must have a capacitor in series with the signal probe to block whatever DC is present in the audio. Just a 100n polyester cap will do. You´ll get more accurate readings with it, notably at the junction between the collector of Q23, R110 and the base of Q22. Without it you risk damaging the headphones.

If there´s a strong signal at the emitter of Q22 and the main output is still weak, I´d look at the capacitor C61, the resistor R108 and the vactrol.

To calibrate R101 there are instructions on page 10 of the service notes.

A few minor things: C945 and A733 are short for 2SC945 and 2SA733. That should be easier to google.
Those are general purpose japanese transistors. Equivalent to the 2N3904 and 2N3906 respectively, only the pinouts are different. Any old japanese CD player or transistor radio will have plenty for salvaging.

Did the OP replace the two-prong cord for a three-prong "to get rid of hum"? As long as the Earth isn´t wired to the metal chassis it´s OK. Otherwise there might be hum as it will make for a ground loop.

Cheers.

[servicenote]

Thanks, Mooger5. I'll take it out and have a look at C61, R108, and the vactrol.
I was thinking that the fault might be with the vactrol, and that I should try a different one, as there are several similar modern types available. The original one isn't made anymore, and there's no exact equivalent.
I replaced the two-prong plug with a three-prong one, mainly for safety. I don't think there was a hum before, but there's definitely no hum now.