• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Repair JBL SCS 125 Subwoofer Capacitor leak

RonSteel

Member
Joined
Feb 6, 2023
Messages
6
Likes
0
Hello,

I have a JBL SCS 125 which is 5 satelite speakers and 1 subwoofer connected to a Harman Kardon AVR 300RDS.
Recently, the subwoofer kept turning on and off rapidly for 1 to 2 minutes before staying on.
It's only after that clicking on and off for 1 to 2 minutes that I could start using the subwoofer.

I'm new to repair and willing to learn so this is going to be my learning project.
On a youtube video you can find the problem I had:

In the comments section, someone suggested that the problem could be the 2 big capacitors 6800uF 50V 85°C that might have leaked.
When I opened up my subwoofer, there was some corrosion on the components that were near these 2 capacitors.
One of the capacitors leaked. So I removed both of them and installed 2 new ones 6800uF 50V 105°C .
And cleaned with isopropanol alcohol all I could clean and removed as much corrosion from near by resistors.
As well as removed as much brown glue as possible.

I tried the turn on the subwoofer, it turned on right away which felt like a victory but the speaker from the subwoofer wasn't reacting.
I thought almost there then.

After opening the subwoofer a second time, I tested with my digital multimeter surrounding components new the capacitors.
Resistors were giving almost the right values, the transistors C1815 and A1015 seemed to work well as well.
But I decided to change these 2 transistors with new ones which were for some reason smaller.
I swapped C1815 GR 0C with C1815 GR 331 and A1015 GR 0A with A1015 GR 331.
Now the subwoofer turns on and off repeatedly every seconds and make the subwoofer's speaker bump at the same pace.
The nob to increase and decrease bass has no effect on the speaker.

Now I'm stuck, not sure what to do next.
I could replace all caps and transistors but it would be better to be able to identify all the components posing problem first and replace those first.
There are only 2 boards in the subwoofer. One is the small amp I guess, the other one is the psu.
On the back of the PSU board, there are 2 yellow marks, one near the big capacitor C119, in the area around the big resistor R129.
And the other one near the blue relay around the resistor R136.

I cleaned all the links in the area as well as the resistors in the area.
For the resistors, I wasn't sure if I should have replaced them with new ones since they tested ok on the multimeters outside of the board.
And I'm not sure with what I should replace them, in terms of size, I wasn't sure what to buy.
R302: yellow, violet, orange, Gold – looks oxi - 47kOhms 5% - 1/4w? or 1/2w? or? (5mm)
R129: green, blue, brown, gold – looks oxi - 560Ohms 5% - 1w or 2W or 3w? (11mm)
R127: orange, orange, red, gold – looks oxi - 3,3kOhms 5% - 1/4w? or 1/2w? or? (5mm)
R128: red, violet, red, gold – looks oxi - 2,7kOms 5% - 1/4w? or 1/2w? or? (5mm)

Let me know if you need any additional info.

Do you have any suggestion as to what I could be testing in order to troubleshoot the problem and find a fix?
Thank you in advance
01-PSU-Amp-Boards-JBL125Sub.jpg
02-PSU-JBL125Sub.jpg
03-Back-PSU-JBL125Sub.jpg
03-Cap6800uF-Leak-JBL125Sub.jpg
05-Cap6800uF-Leak-JBL125Sub.jpg
06-afterCleaning-JBL125Sub.jpg
07-Transistor-C1815-JBL125Sub.jpg
 

AudiOhm

Senior Member
Joined
Oct 27, 2020
Messages
409
Likes
410
Location
London, Ontario, Canada

JayGilb

Major Contributor
Joined
Jul 22, 2021
Messages
1,383
Likes
2,344
Location
West-Central Wisconsin
From your video, the blue relay (RY01) is powering up momentarily and then shutting down and going into power cycle.
According to the repair manual link generously provided by @AudiOhm, that relay coil is being driven by (page 18) Q121.
the base of Q121 is driven by Q122. The schematic has voltage test points that you can measure to see if they are out of spec.
Trace backwards from those transistors looking for incorrect voltages levels, since it's continuously cycling, some safety feature is causing it to shutoff
and then restart. Since it does come back to life after a few minutes of clicking, it might be an active component that requires it to be heated before proper function.
You might want to see if a slight heating of any those transistors will hasten the resurrection. Sometimes just placing a soldering iron near a transistor case will heat it up enough
to help. Or after it comes to life, cooling down an active component with component chiller spray might cause it to fail once again. Compressed air in a can when the can is inverted usually works well as a component chiller.

Voltage should be able to be measured between power cycles to get a general idea if anything is completely out of whack.
Visually inspect for cracked solder joints and resolder any that may look bad.
 
Last edited:

DVDdoug

Major Contributor
Joined
May 27, 2021
Messages
3,016
Likes
3,966
Resistors and non-polarized capacitors rarely fail.

Sometimes electrolytics will "degrade". They can leak electrically (like having a resistor in parallel) and I think it's normal for the capacitance to drop a bit, and maybe eventually drop the point where it's ineffective.

Sometimes a blown transistor can send excess current through a resistor and fry it, but a resistor would be visually burned. You also can't reliably measure the resistance in-circuit. You have to remove at-least one-end so that nothing else is in parallel.

"Large" transistors & MOSFETs are usually the most-stressed components and the most-likely to fail. And when one fails, it's not unusual to take a resistor or another transistor with it. (I don't see any burned resistors.) You can test a transistor with your multimeter after removing it from the circuit. It's not a "complete test" but in my experience it's been reliable and usually it's a short.

I'm not sure about checking MOSFETs. You can probably find a short. But you have to be careful handling MOSFETs once you remove it from the circuit because they are easily damaged by static discharge.

Or, you can get a crack (or corrosion) in a PCB trace. If the copper is broken or corroded and not making a connection you can solder a wire in parallel.
 
OP
R

RonSteel

Member
Joined
Feb 6, 2023
Messages
6
Likes
0
From your video, the blue relay (RY01) is powering up momentarily and then shutting down and going into power cycle.
According to the repair manual link generously provided by @AudiOhm, that relay coil is being driven by (page 18) Q121.
the base of Q121 is driven by Q122. The schematic has voltage test points that you can measure to see if they are out of spec.
Trace backwards from those transistors looking for incorrect voltages levels, since it's continuously cycling, some safety feature is causing it to shutoff
and then restart. Since it does come back to life after a few minutes of clicking, it might be an active component that requires it to be heated before proper function.
You might want to see if a slight heating of any those transistors will hasten the resurrection. Sometimes just placing a soldering iron near a transistor case will heat it up enough
to help. Or after it comes to life, cooling down an active component with component chiller spray might cause it to fail once again. Compressed air in a can when the can is inverted usually works well as a component chiller.

Voltage should be able to be measured between power cycles to get a general idea if anything is completely out of whack.
Visually inspect for cracked solder joints and resolder any that may look bad.
Thank you! I will explore that route and report back.
 
OP
R

RonSteel

Member
Joined
Feb 6, 2023
Messages
6
Likes
0
Resistors and non-polarized capacitors rarely fail.

Sometimes electrolytics will "degrade". They can leak electrically (like having a resistor in parallel) and I think it's normal for the capacitance to drop a bit, and maybe eventually drop the point where it's ineffective.

Sometimes a blown transistor can send excess current through a resistor and fry it, but a resistor would be visually burned. You also can't reliably measure the resistance in-circuit. You have to remove at-least one-end so that nothing else is in parallel.

"Large" transistors & MOSFETs are usually the most-stressed components and the most-likely to fail. And when one fails, it's not unusual to take a resistor or another transistor with it. (I don't see any burned resistors.) You can test a transistor with your multimeter after removing it from the circuit. It's not a "complete test" but in my experience it's been reliable and usually it's a short.

I'm not sure about checking MOSFETs. You can probably find a short. But you have to be careful handling MOSFETs once you remove it from the circuit because they are easily damaged by static discharge.

Or, you can get a crack (or corrosion) in a PCB trace. If the copper is broken or corroded and not making a connection you can solder a wire in parallel.
Thank you! Super useful. I'll pay special attention to transistors then. Do you know if there is a difference for these transistors? C1815 GR 0C vs C1815 GR 331 and A1015 GR 0A vs A1015 GR 331. The 331 ones are smaller but maybe it's just new technology that makes them smaller and they actually do the exact same thing perfectly.
 

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
3,377
Likes
3,325
Location
.de
IMO the suspected "leakage" is all killer glue - the big caps have their vents on top. J117 and J116 still look a bit sus. Also, where even is the relay power supply? (The symptoms suggest a bad filter cap in that.) I would eye some of the caps of a few hundred µF scattered around the relay, I suspect one of those is our culprit.
 

JayGilb

Major Contributor
Joined
Jul 22, 2021
Messages
1,383
Likes
2,344
Location
West-Central Wisconsin
IMO the suspected "leakage" is all killer glue - the big caps have their vents on top. J117 and J116 still look a bit sus. Also, where even is the relay power supply? (The symptoms suggest a bad filter cap in that.) I would eye some of the caps of a few hundred µF scattered around the relay, I suspect one of those is our culprit.
There is a bridge rectifier next to the relay.
 
OP
R

RonSteel

Member
Joined
Feb 6, 2023
Messages
6
Likes
0
IMO the suspected "leakage" is all killer glue - the big caps have their vents on top. J117 and J116 still look a bit sus. Also, where even is the relay power supply? (The symptoms suggest a bad filter cap in that.) I would eye some of the caps of a few hundred µF scattered around the relay, I suspect one of those is our culprit.
Thank you!
I disoldered J117 and J116, removed the little corrosion on them as well as the "killer glue" which is turning black.
I read somewhere that with time that kind of glue becomes conductive and starts shorts.
Could it be both, leakage + killer glue? because there was some green on the legs of some resistors.

There is also D103 the bridge diode, its pins look very dark. So I will test that as well. Not sure how.

What is a relay power suppy? I thought the relay would be the blue RY101 thingy near the bridge diode near one of those big capacitors.
Sorry I'm new to this. Learning a lot, brain melting but I'm hanging in there.
The only part from the PSU not shown on the pics is a big heavy square block. The power transformer.

I will spend some time near the blue thing and look at the caps of a few 100uF near it.
And report back maybe upload a few more pics.

Right now I'm trying to learn how to use test points and test bit by bit the pcb.
 

JayGilb

Major Contributor
Joined
Jul 22, 2021
Messages
1,383
Likes
2,344
Location
West-Central Wisconsin
There is also D103 the bridge diode, its pins look very dark. So I will test that as well. Not sure how.
The purpose of the bridge rectifier is to convert ac to dc.
Bridge rectifiers are not prone to failure, but to test it put your DVM into ac volt mode and measure across the two inner pins (-HC and +HC)| to verify it is receiving ac voltage.
To check it is converting ac to dc switch your DVM to dc voltage mode and measure the + and - pins. You should see a dc voltage slightly less than the previously measured ac.
 
OP
R

RonSteel

Member
Joined
Feb 6, 2023
Messages
6
Likes
0
The purpose of the bridge rectifier is to convert ac to dc.
Bridge rectifiers are not prone to failure, but to test it put your DVM into ac volt mode and measure across the two inner pins (-HC and +HC)| to verify it is receiving ac voltage.
To check it is converting ac to dc switch your DVM to dc voltage mode and measure the + and - pins. You should see a dc voltage slightly less than the previously measured ac.
thank you for the info and how to test the bridge rectifier.

DVM on ac volt mode, pins -HC and +HC = 0v (guessing that's normal since the thing is not plugged)
DVM on dc volt mode, pins - and + = 2,6v
So I'm guessing I should plug the thing first then do the test.

I found another test with DVM on diode mode, pcb unplugged:
red lead on minus, black lead on -HC = 0,560
red lead on minus, black lead on +HC = 0,540
red lead on minus, black lead on positive = 0,970

black lead on positive, red lead on -HC = 0,540
black lead on positive, red lead on +HC = 0,560
black lead on positive, red lead on minus = 0,970

since there is no shortage on it then I'll assume the components is still good.
Still fun so far :)
 

JayGilb

Major Contributor
Joined
Jul 22, 2021
Messages
1,383
Likes
2,344
Location
West-Central Wisconsin
thank you for the info and how to test the bridge rectifier.

DVM on ac volt mode, pins -HC and +HC = 0v (guessing that's normal since the thing is not plugged)
DVM on dc volt mode, pins - and + = 2,6v
So I'm guessing I should plug the thing first then do the test.

I found another test with DVM on diode mode, pcb unplugged:
red lead on minus, black lead on -HC = 0,560
red lead on minus, black lead on +HC = 0,540
red lead on minus, black lead on positive = 0,970

black lead on positive, red lead on -HC = 0,540
black lead on positive, red lead on +HC = 0,560
black lead on positive, red lead on minus = 0,970

since there is no shortage on it then I'll assume the components is still good.
Still fun so far :)
The unit needs to be powered up to measure the voltages. Any voltages being measured when it's not powered up are just stored voltage in the caps.
 
Top Bottom