Need help troubleshooting 60 Hz hum - SOLVED

[hummusdude]

This is the power amp board from a Kenwood 1035W powered subwoofer. It has an intermittent 60 Hz hum problem that happens suddenly from time to time. It can happen when there is signal and when it's quiet as well. The source of the issue is sensitive to vibration.... a bump to the side of the unit will get it to quiet down, at least for awhile but it always comes back.

I was able to localize it to one of the smaller power supply smoothing caps (#1)...at least that's my best guess so far. With the power on I can make the 60 Hz hum cut in and out by applying a small pressure to the top of #1.  But the board flexes a tiny amount and my experience has shown that it could be anything in the area that is the actual problem.

The power supply caps all have high DC resistance. The diodes in the high current rectifier appear to be okay... 0.6VDC in one direction and very high resistance in the opposite polarity. I inspected the solder joints under a microscope and didn't see anything suspicious. The way the hum starts and stops with a clear clicking noise makes me wonder if it isn't a failing solder joint. I reflowed the power caps anyway just in case. I haven't tried it yet however because I'm waiting on a fuse.

I should add that this thing is probably 20 years old, or more. I got it off craigslist for $25. The speaker needed new foam and aside from the hum issue it's working well.

Aside from the hum issue, I'm curious about the DC rectification. There is a full bridge, low current rectifier...I'm assuming that's for the 12 and 24 VDC relays. And there is a single high current rectifier chip as well. My best guess is that the design uses a positive DC voltage and a virtual ground circuit to drive the power transistors. Does that even make sense, and is it a common thing?

[bob91343]

This appears to be a poor connection.  Perhaps a cold solder joint or loose grounding screw or corrosion somewhere.

[abdulbadii]

First check if moving the AC wires has some effect on the hum.
60 Hz could mean already mentioned AC mains coupling to the amplifier circuit.

Try twisting any wires carrying AC in and out of the transformer, and run them far from the audio cables or boards.
Hum on speakers indicates either (caps in) power supply or ground problem.
Replace the balance of the caps in the power supply and if there are any on the amp board

[CaptDon]

Seems you have a poor connection in the power supply section causing a ground loop. It would also appear and be likely that the unit uses both positive and negative rails for the high power part of the amplifier as evidenced by the two large capacitors. Typically anywhere from + and - 40vdc to as much as + and - 80vdc or more. As you have indicated you have localized the problem to one area of the board. Check all grounding screws and add star washers if they are not already included. Abdul's suggestions of twisting transformer wires and shotgunning all caps is way off base. If the amplifier was ever dropped it is possible to have a microscopic crack in a trace on the PCB whose resistance can change and cause ground loop effects. I have on rare occasions seen electrolytic capacitors that had internal breakdowns and varying resistance where the internal leads were bonded to the terminals extending through the seals to the outside. Physically disturbing the capacitor caused changes in the ESR and hum levels. Cheers!!

[hummusdude]

That is my guess as well. There are no visible signs of corrosion and no obvious issues with solder joints either. I reflowed the power supply caps just in case.

[hummusdude]

It would also appear and be likely that the unit uses both positive and negative rails for the high power part of the amplifier as evidenced by the two large capacitors. Typically anywhere from + and - 40vdc to as much as + and - 80vdc or more.

I agree it seems odd to think it would use a single supply design but I'm not sure what to make of the fact that there is only one high-current rectifier chip on the board.

[hummusdude]

I have on rare occasions seen electrolytic capacitors that had internal breakdowns and varying resistance where the internal leads were bonded to the terminals extending through the seals to the outside.

I wondered about this as well. It seems  at least plausible that a cap could fail in such as way as the connection on one or both of the leads to the internal conductive material is less than 100%. I can imagine how external pressure on the cap could create enough movement to make/break the circuit. The symptom is clearly an on/off kind of thing. Pressing on the cap in question causes a distinct "click" in the speaker as well as sum hum. I'm thinking I will replace the two smaller power conditioning caps just in case.

[andy3055]

Ground the input and use a rigid plastic or wooden thing and prod various components till you find the culprit. Some times you cannot see bad joints just by looking with the naked eyes. If the fault can be triggered by tapping, it is a good sign of a mechanical problem rather than a component failure.
Also, check the input sockets.

[cvanc]

Check the soldering at your RCA input jacks - it's common to get cracked solder joints here.  Good luck!

[hummusdude]

Check the soldering at your RCA input jacks - it's common to get cracked solder joints here.  Good luck!

Good to know. I'm using the speaker level inputs and not the line level RCA jacks to drive the unit. Is it possible that the RCA jacks are still the source of the issue?

[hummusdude]

Ground the input and use a rigid plastic or wooden thing and prod various components till you find the culprit. Some times you cannot see bad joints just by looking with the naked eyes. If the fault can be triggered by tapping, it is a good sign of a mechanical problem rather than a component failure.
Also, check the input sockets.

I was able to narrow it down to the area right around cap #1 in the attached pic using that method. The way it cuts in and out, with a distinctive "click" noise in the speaker makes me think it is a mechanical issue as well. After looking at it some more it appears that the 2200 uF cap (#1) is more likely a part of the low-current rectifier circuit that powers the relays and not part of the signal path. But even if it were allowing some ac to pass through to the control side of the relay (and then granting that ac on the control side of the relay might bleed through to the signal side) it would still be 120 Hz. So not sure if that is relevant.

[cvanc]

Good to know. I'm using the speaker level inputs and not the line level RCA jacks to drive the unit. Is it possible that the RCA jacks are still the source of the issue?

Ah.  Then no, it wouldn't.  It would only affect signals coming in via that connection.

[hummusdude]

Good to know. I'm using the speaker level inputs and not the line level RCA jacks to drive the unit. Is it possible that the RCA jacks are still the source of the issue?

Ah.  Then no, it wouldn't.  It would only affect signals coming in via that connection.

That was my assumption so thanks for confirming.

[hummusdude]

I was able to find a schematic and as I thought the 2200 uF cap in question (C 812 on the schematic) provides low-current DC to the control board, it's not part of the signal chain. Also, since it is downstream from the bridge rectifier (D811-14) then it would be passing 120 Hz and not 60 Hz. So at this point I'm thinking the issue may be with the rectifier diodes since any ripple downstream would be at 120 Hz if they were all working correctly.

Sorry for my lack of understanding, but could it also be the case that a bad ground connection somewhere could be allowing 60 Hz into the ground and that is what is coupling back into the signal somewhere?

[hummusdude]

Once I was able to get a hold of a replacement fuse for the power supply (blew it when my probe slipped!) I put it back together and the hum issue was gone. All I did was reflow the solder joints the all of the power supply caps on a hunch and that did the trick.

Thanks to all for the input!!