I'm not familiar with this particular device, but I'm curious if the RCA jacks and headphone out are run by separate pre-amps. If that's the case, I'm wondering if a cap in the pre-amp phase for the RCA outputs went bad.

Someone else on the board may be more knowledgeable, but whatever went wrong must have existed on the board in more than one form for one output to still function, which probably rules out any of the chips.

Just my take. Best of luck figuring it out.

I might have to look at that. What a pain in the but if that happened. Thanks for the idea. I'll wait a few days and see what others have to say about it, just in case they have run into this exact problem.

Good call. If someone else has had the same problem, it'd save you the trouble of poking around looking for it yourself.

On the plus side, it's awesome that you still have a functional output, even if it's not the ideal one.

Referencing the SC-55 Service Notes, you'll notice that the output stage for the headphone and RCA jacks is common all the way through the VR encoder board. Since the headphone output is functional, this rules out all of the ICs except for the single M5218FP op-amp (IC11) driving the RCA outputs. Rather than that or capacitors being to blame, I wonder if you just have cracked/broken ground or signal joints at the RCA output assembly. If you loosen the screw holding it to the chassis, and then slightly wiggle the output cables during playback, does the behavior change?

In this case, a continuity tester is your best cheap friend. Those Service Notes also establish that the Roland SC-55 has 3 Mbytes of sample ROM and a 512Kbyte external firmware ROM which may be coupled with the firmware internal to the CPU.

I'll whip out the 87V and give it a check this evening. Thanks for the link to the service manual Cloudschatze, I'm sure it will come in handy.

Great Hierophant wrote:

Those Service Notes also establish that the Roland SC-55 has 3 Mbytes of sample ROM and a 512Kbyte external firmware ROM which may be coupled with the firmware internal to the CPU.

Yes, although this isn't related to the topic at all. 😀 By comparing with the SC-50 Service Notes, you'll also come to the conclusion that 1MB of that 3MB sample ROM is used for the MT-32 compatibility instruments, leaving just 2MB for the capital tones, GS variant instruments, and drumsets.

If I had to bet then the problem is broken joints on the RCA plugs... for a quick just reflow and see if that helps, had the same problem with a SC-55 I bought as spares/non-working

I resoldered the whole pad area to the RCA jacks and tested for continuity afterwards. That seemed to solve the problem. I have no idea how a joint could have gone bad just sitting in one place for several years, but it clearly did. The temperature in the room is fairly constant, so thermal movement should have been minimal. My only guess is it was a poor joint to begin with and time, use, and temps just caught up with it. That or the level of dust that had collected on the RCA cables was just to much added weight, 🤣. The last time I have moved the unit was going on two years ago to replace the coin cell inside. Use it and my MU2000, MT-32 old, and CM-32L all the time though.

On a side note, I noticed while I was used the headphone jack on the front of the SC-55 that there was a fairly large amount of line noise. Very "hissy" sounding compared to the RCA jacks. Something to keep in mind if somebody finds a unit that the RCA out is dead (and a simple fix cant be found) and they have to resort to the headphone jack.

Thanks for all the help guys. I never would have thought to just resolder the RCA joints. I had just assumed that it was a much larger problem, and skipped right over the simple solutions.

P.S. Cloudschatze, is the official Roland SCB-55 only 2MB? I would assume so if the SC-55's own soundbank is only 2MB.

Glad you fixed the unit Kodai! 😀

Kodai wrote:

P.S. Cloudschatze, is the official Roland SCB-55 only 2MB? I would assume so if the SC-55's own soundbank is only 2MB.

3MB Waveform ROM

• CM-300/500 - 3 x 1MB
• GPPC-N(A) - 3 x 1MB
• SC-55 - 3 x 1MB
• SC-55mkII - 1 x 2MB, 1 x 1MB
• SCB-55 - 1 x 2MB, 1 x 1MB
• SCC-1(A) - 3 x 1MB

2MB Waveform ROM

• SC-50 - 1 x 2MB

1MB Waveform ROM

• RAP-10 - 1 x 1MB
• SC-7 - 1 x 1MB
• SCB-7 - 1 x 1MB

Given the above, it seems reasonable to assume that:

• 1MB of waveform memory is utilized by the 128 capital tones and most of the drumkits
• 1MB of waveform memory is utilized by the sub-capital and variation tones, and additional drumkits (SFX)
• 1MB of waveform memory is utilized by the MT-32 compatible tones

So it would seem that the SCB-55 has an MT-32 option (I assume that's what the extra 1MB is for). I thought it was only SC-55. Good to know. 😄