SX-535 Re-Capped, but Buzzing Returned

Natalya

New Member
SX-535-After-Cap.jpg


I have a SX-535, it wasn't sounding right. I read up on here about it, and I replaced all the caps and transistors on the power amp board, and once I did that (and De-oxited the Pots) everything worked fine! It sounded great!

Unfortunately, about 2 or 3 months later it suddenly developed a buzzing noise on both channels. Initially the volume knob had nothing to do with it, but now I get a hiss too at higher volume levels. A or B doesn't matter, and whether Balance is right or left the sound persists even in the speaker that shouldn't have any noise.

Because it was everywhere at all times I suspected the power supply board, so I re-capped that, replaced the diodes, and replaced its transistors. Unfortunately, that had no effect at all, sound didn't get better or worse. I know the wiring over there looks pretty crispy, but I soldered the wires to the posts and again the sound quality didn't change so I think those connections are adequate. Poking them didn't make any noises.

I'm not sure what I need to look at next. I did the power amp adjustment and got the 15mV on each side after 10 mins w/ nothing connected to the AUX input and speakers off, as specified in the manual. I also went and poked around all the transistors and caps, nothing buzzed from being touched, so I think my solder joints are okay.

I can replace more stuff if I need to, I just don't know what I should look at next. Maybe I didn't do a good enough job with the De-Oxit or something?
 
Out of curiosity, why is there an extra mica insulator stuck to the heat sink between the first two output, but no insulator under the third output?

Nice looking job on the recap, too.

Can you guess at the frequency of the buzz/hum? If it is in the 60-120 Hz range it could be a power supply or shielding issue.

A constant hiss may be a transistor upstream generating noise, such as the FM or PHONO sections if the noise is occurring in those modes only.

As for the hiss at higher volumes, does the LOUDNESS control have any effect on the level of noise? Even with the BALANCE set fully to one channel or the other, the noise floor of the AWM-068 board will be amplified to the level set by the volume control.

The 535 used some suspect devices in the audio path which can be noisy, like the 2SC1318 (thermal stress, thermal flex) and 2SC1344 (static burst). There are also 2SA720s in the audio amps. These could all be replaced with modern, low-noise devices, such as 2SC1318 => KSC2383Y or KSC2690AY, 2SC1344 => KSC1845F, 2SA763 => KSA992F, and 2SA720 => KSA1013Y. NOTE: If the 2SA763s are replaced, they should be gain-matched in pairs as closely as possible using KSA992Fsfor the differential amp installations in either channel. Getting extra stock for this would be useful to enable selecting the best pairs.

There are 6 total 2SC1344s, 4 total 2SC1318s, 2 of the 2SA720s, and 4 of the 2SA763s. For robustness, the 2SA489s could be replaced with MJE15031Gs with its B-C-E pinout, and the 2SC789s could get MJE15030Gs with the same lead arrangement. For the TO-220s a few extra parts might be handy, such as a new insulator, mounting screw, and insulation bushing since both replacements have a "hot" collector tab connection. Basically, you would have about $15 in parts (if you get a dozen and a half 992Fs for matching) and probably have as much in shipping, so it would be reasonable to evaluate further and build a larger order to save shipping charges if it is warranted.
 
Out of curiosity, why is there an extra mica insulator stuck to the heat sink between the first two output, but no insulator under the third output?
Great question! I never noticed that! This one had some transistors replaced before I ever had a look at it, I think that might have been one of them, but I replaced all 4 of those again just as precaution. (Along with all the other transistors.)

Nice looking job on the recap, too.
Thanks!

Can you guess at the frequency of the buzz/hum? If it is in the 60-120 Hz range it could be a power supply or shielding issue.

A constant hiss may be a transistor upstream generating noise, such as the FM or PHONO sections if the noise is occurring in those modes only.
Negative, noise occurs in all modes. Hissing goes away as volume goes down, but the buzzing is always there.

As for the hiss at higher volumes, does the LOUDNESS control have any effect on the level of noise? Even with the BALANCE set fully to one channel or the other, the noise floor of the AWM-068 board will be amplified to the level set by the volume control.
Loudness ONLY affects hiss, hiss shows up only when volume is mid//high. Hiss is a new phenomenon too.

The 535 used some suspect devices in the audio path which can be noisy, like the 2SC1318 (thermal stress, thermal flex) and 2SC1344 (static burst). There are also 2SA720s in the audio amps. These could all be replaced with modern, low-noise devices, such as 2SC1318 => KSC2383Y or KSC2690AY, 2SC1344 => KSC1845F, 2SA763 => KSA992F, and 2SA720 => KSA1013Y. NOTE: If the 2SA763s are replaced, they should be gain-matched in pairs as closely as possible using KSA992Fsfor the differential amp installations in either channel. Getting extra stock for this would be useful to enable selecting the best pairs.

There are 6 total 2SC1344s, 4 total 2SC1318s, 2 of the 2SA720s, and 4 of the 2SA763s. For robustness, the 2SA489s could be replaced with MJE15031Gs with its B-C-E pinout, and the 2SC789s could get MJE15030Gs with the same lead arrangement. For the TO-220s a few extra parts might be handy, such as a new insulator, mounting screw, and insulation bushing since both replacements have a "hot" collector tab connection. Basically, you would have about $15 in parts (if you get a dozen and a half 992Fs for matching) and probably have as much in shipping, so it would be reasonable to evaluate further and build a larger order to save shipping charges if it is warranted.

I replaced everything in the POWER AMP, TUNER, and POWER SUPPLY sections in the attached spreadsheet with the stuff on the right from Mouser, except I think 2 or 3 small caps on the tuner board.

I did NOT touch the equalizer board or the control board, and some of the transistors on your list are in there. Maybe those shall be my next target!

ALSO -- Forgot to mention that it did blow one of the 3 Amp fuses on the back around the same time that the buzz started.

Thanks so much for the help!
 

Attachments

  • sx-535.zip
    1.1 KB · Views: 17
All those soldered wirewrap pins on the power supply look pretty bad. Lots of exposed wire to short out. Solder connections look pretty iffy.I'd go back and check all your work for cold solder connections and solder bridges.
 
I don't see any heatsink compound on the outputs. You know the messy white stuff.:dunno:
 
In your photo. RH side of the heatsink. The small horseshoe-shaped STV3H diode fastened to the heatsink doesn't look like one lead is soldered to the circuit board. Might just be the photo quality. :idea:
 
Last edited:
Okay thanks for noticing that mica insulator, I had no idea what that was! I had bought extras of the transistors when I placed the original Mouser order, so right now I've replaced the transistor that was there and I put the insulator there with it.

Power supply wires:
Yeah they look terrible, I don't want to keep them that way, when I unwrapped them to do the power supply board they were brittle and kind-of falling apart as I undid them and the ends broke off a couple of them. I think I should replace them with new wire. I checked the connections though and re-melted the solder just in case, but that had no effect on the buzzing.

Heatsink Compound:
There is (not visible in photo) some kind of residue or substance at each of the heat sink transistors, but maybe it's not enough? I have extra silver-based thermal paste (you know for a desktop CPU) on hand, would that be acceptable, or is there a certain type of compound I should get?

STV3H diode:
I inspected it, the diode is soldered to the board, the insulator just got pushed up and the exposed wire is kinda gray from oxidizing.


Okay so where I'm at now:
- Have replaced transistor that didn't have an insulator/
- Put the insulator back in position
- Re-melted all solder on Power Amp board
- Have not had chance to touch Equalizer or Control boards yet
- Hiss sound is GONE but buzzing still occurs same as before

Thanks for taking the time to help me with this, I am a complete noob and have no idea what I'm doing beyond the stuff I've read on this site.
 
I have extra silver-based thermal paste
Some of that arctic silver CPU paste is electrically conductive. And is not suitable for this app. If it is arctic silver ceramique it is suitable and not electrically conductive. Look at the compounds data sheet before using.
 
Look at the big main caps where the wires are connected. Look for a bulge or leaking fluid. If so, they probably need replacing.
 
it all needs a damn good cleaning and remove that glue for a start . then install the components properly and replace main filter caps . then check out those wires that look to be not connecting very well especially the black one . near the fuses ..

p,s welcome to AK
 
Last edited:
Buzzing:
It's very consistent, one speed, uniform across all speakers, volume unaffected by volume knob.
- Lamps do not flicker
- Volume of the buzz kinda goes up for a second when the stereo is turned on, but then settles into its normal state and stays there until it's turned off.

Big Caps:
- No fluids leaking
- Will check for bulge
- Any other tests I can do on them? I've got a multimeter

Component Leads:
I didn't know they were supposed to be pushed in to be made short, is it important to do this? I can do so if necessary.

Power Supply Wires:
Yup, very crispy, I probably need to replace the wires because they're brittle. I'm not going to leave them all janky as they are right now.

Glue:
I'm not sure what you're referring to. : (
 
Buzzing un-affected by volume control suggest that the problem is isolated to post-volume audio circuit or power supply.

Is there any tan goo or residue on the bottom of the main filter capacitors (The big black ones) or any bulges on the bottom about the size of a BB? You can check the big caps with a multimeter if your meter has a capacitance setting, but it's not really a very accurate test because the caps aren't tested under real operating conditions.

Glue is often used to hold capacitors down on the circuit board before wave-soldering in a bath of molten lead occurs. It isn't conductive and is highly unlikely to be the source of your problem. That said, removing it allows components to seat better (down close to the circuit board) which makes for a neater rebuild and easier isolation of problems later by reducing visual anomalies.

Can you post a picture of the bottom of the large capacitors? We can tell you if they're leaking electrolyte.
 
Back
Top Bottom