Bose 321 Gs Series II DVD hum

[R_J]

I also see the main filter capacitor C22 is 15,000uF?
I know those hardly ever fail but I imagine that could be a source of hum if it did.

That is possible, but if Q1 does not get turned on fully, C22 could be floating slightly above the bridge rectifier negative. however, it is quite common for these large value caps to fail in sub woofers causing hum.

[Agent24]

That is possible, but if Q1 does not get turned on fully, C22 could be floating slightly above the bridge rectifier negative. however, it is quite common for these large value caps to fail in sub woofers causing hum.

Do you think that's a vibration issue causing failures?

[R_J]

Do you think that's a vibration issue causing failures?

Not really, it is likely cheap caps and also these subs are basically ON 24/7. the audio gets muted, but the power supply is usually always running but at a lower current demand when muted.
You could check the voltage across Q1 Drain and source, it should be near zero volts when operating. If the hum is not too bad, and you don't have a replacement just add a second cap maybe around 4700 µf across C22 and see the hum goes away or is less noticeable.
Also Check the DC voltage across - and + of BR1 and compare it to the voltage across C22, it should be the same, likely around 16.5vdc

[Uranium-235]

I also see the main filter capacitor C22 is 15,000uF?
I know those hardly ever fail but I imagine that could be a source of hum if it did.

I replaced the main filter capacitor on the refurbished board just to be sure (with Nichicon KT Audio). Used non conductive RTV to stop vibration issues

[Uranium-235]

I used AI (Gemini Pro Extended) with schematics I managed to find (original ones for this version, version I). Gave it the specific symptoms and this is what it gave me

Me: So, considering the symptoms, and the fact the mute does take the buzz away, but if inserting a dvd (no sound until the DVD initializes), no matter what input, and the hum stays at a constant level despite volume, what is the biggest culprits

Gemini: Those three specific symptoms are the diagnostic golden ticket. By mapping how the signal traffic flows through this system, we can instantly box the fault into one tiny physical section of the board.

Here is the exact physical layout of the signal path and why it isolates your culprit:

1. The Volume Valve (Digital)

Volume attenuation in this system happens digitally inside the main processor before the audio is ever turned into a physical analog wave. Because the hum stays at the exact same physical volume whether the system is cranked to max or set to minimum, the hum is entering the traffic flow after this digital volume valve.

2. The Hardware Gate (Mute) Looking at the schematic, the MUTE command is a hard, physical gate. When you press mute, the microcontroller sends a logic signal directly to pin 22 of the TDA7385 amplifier chip and out to the subwoofer, commanding them to physically shut down their outputs. If the hum stops when this gate is closed, the hum is originating before the final amplifier chips.

3. The Holding Pen (DVD Initialization)

When you insert a disc, the system's microcontroller clamps that physical Mute gate shut. It holds the audio path closed while the digital chips sync their clock timing to prevent loud pops from hitting the speakers. Once the DVD is fully initialized and spinning smoothly, the microcontroller opens the Mute gate. The fact that the hum floods in the exact millisecond the initialization finishes proves the hum is constantly pooled up against that closed gate, waiting to be released. The Culprits


By overlapping those three behaviors, the leak has to exist in the exact physical middle-ground: after the digital volume, but before the final hardware mute gate. On this board, that leaves only two primary suspects:

1. The Analog Power Rail (The SMD Capacitors) The system relies on the CS4228 CODEC to convert the digital audio into analog sound , and NJM3403AM quad op-amps to buffer those analog waves before sending them to the final amplifiers. Both of these chips require a perfectly flat, clean analog DC voltage to operate.

If the surface-mount electrolytic capacitors filtering that specific analog power rail have baked dry (exactly like the 330µF 16V SMD can you found sitting next to the linear regulator), they will dump 60Hz A/C ripple directly into the power pins of the CODEC and op-amps. Those chips will seamlessly mix that A/C ripple into every single audio channel they process. When the Mute gate opens, that hum flows straight to the speakers.

2. The CODEC's Local VA Capacitor

The CS4228 CODEC has a highly sensitive, dedicated analog power pin (labeled VA on the schematic). Manufacturers usually place a small, dedicated electrolytic capacitor (often 10µF to 47µF) millimeters away from this pin to act as a final shock absorber. If that specific local capacitor fails, the CODEC itself becomes the noise generator, pushing 60Hz out to the op-amps.

Your plan to grab replacements for that 330µF surface-mount capacitor and its surrounding cluster is dead-on. That aging analog power stage is physically the only place a steady, unchangeable 60Hz wave can park itself behind the mute gate.

[Uranium-235]

here is the sheet

https://drive.google.com/file/d/1aSM...ew?usp=sharing

[Agent24]

Or, use your brain instead of AI and scope all the power rails for excess ripple

[Uranium-235]

Or, use your brain instead of AI and scope all the power rails for excess ripple

*this level of work is entirely new to me
*I don't have nor can afford a scope (not do I come across this level of work often enough to justify buying one)
*I don't have the subwoofer to connect it to turn it on. The guy lives like 45 mins away.

[Agent24]

*this level of work is entirely new to me
*I don't have nor can afford a scope (not do I come across this level of work often enough to justify buying one)
*I don't have the subwoofer to connect it to turn it on. The guy lives like 45 mins away.

Fair enough. But as we've already said, mains hum in audio equipment is very often dead power supply filter capacitors.
Or a grounding issue.
If the system was working fine for 20 years then suddenly developed issues, failed filter capacitors are more likely.
And those SMD electrolytics are always suspect, in my opinion.
Did you check the 10uF capacitor C25 that R_J suggested?

If you don't have a scope that will make things a little harder/slower.
But even an extremely basic one would work since you're only looking for slow 120Hz signals.

Or you can do a basic check for ripple on the rails by measuring them with your multimeter in AC mode.
Ideally you should read zero AC volts if the rail is smooth DC.
If you read several volts AC, there is likely a problem.

Then check the capacitors supplying that rail for ESR and capacitance.
If you don't have meters for that, you can try the "tack a known good capacitor across a suspect one" trick.

It won't be easy to test much without the main power supply/subwoofer, though....