SE/30 Sound Distortion

[Hurry]

Context: SE/30, fully recapped logic, analog, and psu. Using (alternately) a BlueSCSI and RaSCSI device for boot. No internal hard drive or floppy drive. Other known but likely unrelated issues at the end of this post in case I'm missing a connection.

Problem: Sound is garbled/scratchy/distorted. This symptom existed before the recap. Though the speaker is probably junk after all these years, I don't believe this is that simple, as sound distortion through the headphone out is identical. Software based volume control works, for what it's worth.

What I've tried: Based on research I focused my attention on the ASC and the two Sony sound chips (using the se30_sound_chips.pdf doc as a guide). This board was covered in cap-gunk(TM), but just that - I don't see damage to the pcb in general - no acid burns or exposed copper. I don't have a microscope, but using what magnifying lenses I have the traces seem to be fine.

I don't have an ultrasonic cleaner and I have no experience with dishwashing boards (though given instructions I'd we willing to try) - the board isn't factory like-new clean but I did the best I could. That flux residue...

Anyway - maybe I should have completely removed the three chips but I didn't - I played cautious and just reflowed them. Sound quality *did* improve, but not much.

Until or unless I can obtain replacement chips, I'm thinking the only other thing I can DIY right now is completely remove the chips, clean, and resolder. My mental risk/reward radar is up... am I just peeing in the wind and doing things for the sake of doing them?

Does anyone have any experience with a similar issue or ideas for things I should try (or just validation that completely removing/rewelding the chips is worth the effort)? I did see one post on the internet where someone said all it took was a thorough enough cleaning to get the gunk out from under the chips to resolve the issue.

Other issues with this SE/30 that are probably unrelated but you never know section:
1) The horizontal width pot is damaged but it failed safe - width is "full screen". I think I just got lucky here, the "dial" in the center is just floating in there. I expect the solution will be to replace the entire board at some point if it gets to me.
2) Strange issue on shutdown, "Sorry, a system error occurred. "Finder" bus error" followed by lock-up running (it seems) 7.5.3 or newer - I think this is a software related issue and my lack of knowledge about System 7, as it doesn't happen on 7.0 or 7.1 but it does happen with multiple different images using RaSCSI or BlueSCSI, I don't think it's a problem with "disk".

 

[Bolle]

Broken address lines to the ASC are a common issue. I have seen a single broken data line as well but that is way less common. it's not totally impossible that this might as well cause your shutdown woes with 7.5 and above reading back values from the ASC for whatever reason which is going to fail if data/address lines are broken.

You might also want to look at the crystal that's right next to the ASC. I had two of those die resulting in no sound at all when the board was cold.
Once it warmed up there was distorted sound and after warming up the crystal some more using hot air the sound started to work just like normal. Replacing the crystal helped in that case.

Suggestion:
#1 check all connections to the ASC as per the schematics
#2 take a look at the crystal

 

[caver01]

I had distorted sound and bus errors on my SE/30. It is cleared up after the following:

• Removed, cleaned, resoldered ASC
• Found broken trace for D31 between SCSI chip and CPU. Bodged and that fixed sound distortion
• Recapped analog board and PSU which cleared up bus error on shutdown.

There are some thin/fragile traces that go between the pads of the ASC, and several used by the ASC are prone to breaking inside/underneath. I think it was worth the effort to clean residue and verify these are all OK. I used Chipquick to remove the ASC which worked OK. Had to resolder it twice though to get it back on straight.

 

[Hurry]

Thank you both. This afternoon I am to get in and under there - I never thought about the crystal, either. Reading that this is resolvable is enough to get me in there - was starting to have that "lost cause" / "am I just going to break something else" feeling and needed prodding

 

[caver01]

I totally get the “lost cause” feeling. I actually put my SE30 on the shelf for 9 months in frustration before I found the broken data line, which was actually easier in the long run than how I was feeling about it at the time. Progress was slow to uncover multiple overlapping issues, but I had encouragement from others which was also helpful! Stay with it. You will figure it out.

 

[Hurry]

Yeah, on and off for the past three years I've been doing electronics stuff with direction, and every time I start a new project I see my work getting better and better. Sometimes I put something away and it just seems to... stay away (I'm looking at you old Atari 400 with the CTIA chip), other times I pull something out and bam, in no time based on giving it a break or leveling up my skills it is back to working. When I have trouble is on a project like this - I feel so close - I want to just bring it across the finish line (and I want to learn and play with the actual computer). That said, I've made some of my biggest and most costly errors when I start to feel that pressure. That's the way I was starting to feel with this - keep going or become shelf (or worse, basement shelf) candy. I think just knowing that at least two other people have encountered this issue and resolved it in the past is more than enough to refill my enthusiasm tank.

 

[caver01]

In my case, I had already removed the ASC, cleaned it up, put it back (a little crooked) and then I started having SCSI issues. I had not recapped the AB or PSU yet, so as I was moving from issue to issue it felt like problems were evolving so while I might fix one thing, something else would break. I was starting to think I was maybe lingering too long with the soldering iron here or there, or while cleaning up corrosion on a via I might have caused it to stop working.

I know this—initially, my audio was distorted and would crash the system if I tried to adjust volume, or change the alert to elicit a beep. At that time, however, SCSI was working OK. Later, after troubleshooting the ASC, my control panel interactions cleared up, but then SCSI stopped working. It was fix one thing, break something else. I now know that there were multiple problems that were changing as I worked, or more likely, I was finally clearing out problems that led to new bottlenecks.

I also had a poorly seated ROM SIMM and a bad UE8 chip which I replaced. . . but I digress.

I was pretty happy to find that broken D31 and have it fix both the SCSI and the audio distortion. But, with that success, I moved on to recap the AB to cure some CRT shimmers and power issues which led to a black screen, white vertical line! That was a 3 week setback, but I knew I made things worse by trying to improve them. . . so I pressed on knowing I had an otherwise working system.

I will be curious about your results. Do followup with another progress report!

 

[Hurry]

I will, particularly because threads with no resolution are the bane of the internet! However my chintzy multimeter died and stopped giving me proper continuity alerts so I gave in and ordered a Fluke. Might not be till the weekend when I make any progress. I know I wouldn't have done anything differently, but man I wish I could go back and lecture 2019 me about buying a cheap soldering irons, multimeters, etc.

 

[Hurry]

It took a while to complete, but the issue is resolved. I meticulously traced everything identified in the Branchus sound troubleshooting guide and found not a single bad trace. I mentioned that I don't have an ultrasonic cleaner and I have no experience attempting the dishwasher method, so I was relying on good old fashioned elbow grease. I went through and did a second scrubbing, much IPA and many q-tips died in the process. When I booted after, I heard a slight improvement in sound - if it was at 50% quality, it had lept to 75%.

I started inspecting under the chips and sure enough, there was something "blocking" my view beneath one of the two sony audio ic's. This afternoon I pulled both chips from the board and found a lot of gunk under one. I scrubbed everything down, cleaned away all of the old solder, and reinstalled the chips.

Boom - problem resolved. Ditto for the bus error... so I can only guess the two were related.

Thanks everyone for your advice!

(In the meantime I obtained a "for parts" SE and stole the floppy and analog board from it - at this moment my SE/30 is 100%)