Cambridge SoundWorks CSW 250 subwoofer not working

[martinlink]

Hey there,
New member here with an old speaker set I found in my new apartment. It is a 2.1 system from maybe year 2006.

What I have done: checked the fuse (fine), looked over every piece under a magnifying glass to spot any visible damage (none), tested the subwoofer element on another amp (works great), tested the sub's amp pins with a voltmeter for any signal (got none). So it must be the amp.

The 2 satellites connect to the subwoofer and as I've found, they are run from a separate amp, because I get perfect sound from the little speakers.

A few threads I ran across in google suggested that a resistor in position R24 (47 ohm, 1/2W) is the main culprit for this kind of problem, so I replaced it without testing it first, just to be sure - I later found that the resistor was fine...

Then I got a multimeter and tested most of the resistors in the area surrounded by the heatsink. I did get one odd reading where the value kept changing, but I think it's because I didn't remove the resistor from the circuit - too lazy and I don't have a solder removal pump at hand...

So my question is: What should I look at as the most probable suspect in this case? I have no experience in "tracing" a broken component, so a step-by-step walkthrough would make my day

Images included: 1) top down view with the rear facing up
2) bottom view with the rear facing up
3) close up of the area surrounded by heatsink
4) close up of the area behind the heatsink

Thanks a ton and if you need any clarifications, please ask. This is a good project to keep me "entertained" for a while, so any suggestions etc. are welcome!

[999999999]

It may need the remote control plugged into the DIN socket to work, were you testing with that plugged in? If not, disregard the following as I think the audio path may be an open circuit without that remote volume control plugged in. If you don't have the volume control there may be some way to hack it if you have a matching DIN connector and a POT, or to jumper across some pins on the PCB with resistors for a fixed/non-adjustable volume level.

Resistors don't generally go bad without some sort of visual evidence like charring or cracks, or delaminated traces or broken solder joints on their leads.

Generally speaking when there is no visual evidence of any fault it's probably the output power amp IC or transistors, they are the two mounted on the metal heatsink. Pull the chip #'s off those, trace the circuit to see which is output for the satellites and which for the subwoofer. Other times when there's transistors for output, there could instead be a leaky or failed transistor before them then a protection circuit kicked in but that doesn't appear to be the cause with this amp, it seems to just one two power amp ICs, one for L & R satellites and one (probably bridged) for sub output.

Having the chip #s, get the respective datasheet that shows pinouts for the sub chip and (with it turned on, but be careful if it's high voltage as some chips energize the heatsink they're mounted on) while feeding it an audio signal, measure for power at the chip's power pin(s) on a DC ~ 100V meter range or so (or 200V, etc) and signal on AC 20V range. If you have both power in a remotely plausible range and signal, the chip is bad.

If you don't have both power and signal, trace the one that's missing backwards in the circuit, each time you come to a component check that the power or signal didn't stop there. Once you get back to the opamp (if tracing the signal) you'll need the datasheet for it too.

[imp]

I have an identical set which had several bad capacitors inside (CapXon), especially the large one on the first of your photos was bad. It had dried out without any visible sign but the bass was quite wague.

Also, I changed the Rifeking in the PSU, which was quite bad as well.

[martinlink]

1) I have the working remore controller for it and I use it to turn the speaker set on, so no problem there.

2) All resistors and caps are visually fine.

3) The 2 chips attached to the heatsink are Creative ct1975-sam which I looked for in google, but found absolutely no good info about the pinouts
I sent a few e-mails to suppliers asking for pinouts (and free samples just in case ) so we'll see how much they know about their products...

4) What do you mean by "If you have both power in a remotely plausible range and signal, the chip is bad"? What are some plausible ranges for them? And why do normal values mean that the chip is bad?
Also, I think I can trace the output signal by looking which pins are connected to the + and - of the pins where the subwoofer element connects to the board? Or do you mean something completely different which my lacking tech knowledge cannot comprehend yet? :P

Thanks for the thorough input so far!

EDIT @ Imp: My set doesn't really suffer from "lacking" bass, it's more like missing completely. Also, I tested the big cap with the multimeter and it is holding at least some juice after a bit of charging - otherwise I think the satellites wouldn't make such great and loud sound as they are also connected to the big cap...

[999999999]

imp, what is a Rifeking?

martinlink, what are the specs for the amp? Like wattage to satellites and sub? That determines plausible range of voltage. For example in amps around that size I'd expect maybe +-20VDC rails. Audio input pin (with a signal playing, best to use a continuous tone input) on the chip might be around 1VAC, but consider anything in tenths of a volt to a few (single digit) volts acceptable audio signal.... it depends on where your volume is set, your source signal strength, and if there is any gain from the opamp before it or if it's just a buffer.

Normal values for power and signal would mean the chip is bad because if it has power input, and has audio signal input, but nothing is coming out of the chip, that is the fault point in the circuit. However you need to measure that there's nothing coming out at the chip output pin not just at the speaker connector.

You wrote the two chips are both CT1975. Good! You can compare one to the other since one is working to power the satellites, though the other may be configured to bridge the output for the sub.

Either way, you can probe the working chip pins one at a time. I don't know your experience or knowledge level in general electronics to know how much to describe. Basically put the multimeter negative probe on a ground somewhere. If all else fails you can use the star ground point they made on the back of the PCB. In the middle of the PCB several traces come together at one point, that point will be ground as are all traces connected to it.

With the other meter probe - keeping in mind that the board may have power voltage levels high enough you should not touch it with your hands until you know whether voltage is low enough to do so safely - amp on and audio input continuous signal, probe each of the chip pins to see if you have a higher DC value and around 1V AC value as mentioned above.

Even without a schematic or datasheet we may be able to surmise the pinout of the chip and if it's bad, a replacement chip. These chips are basically like an opamp with some larger transistors also in the same package. The significance is it's going to have the same pins.

Since the same chip can drive the two satellites we know it's a two channel chip. At a minimum it's going to have the following pins with the following measurable properties. It gives me headaches trying to trace circuits from pictures so I will refrain from that.

- Positive power input

- "might" have negative power input, trace the PSU subcircuit to see if it has both positive and negative rails, or probe all pins to see if you get a negative DC value relative to ground that corresponds to a positive DC value relative to ground on another pin.

- if it doesn't have a negative power input, one pin will be ground(ed)

- channel 1, ~ 1VAC signal input

- channel 2, ~ 1VAC signal input

- channel 1, amplified signal output ~ multiple of the input 1VAC depending on volume set

- channel 2, amplified signal output ~ multiple of the input 1VAC depending on volume set

- channel 1, feedback pin ~ will have a trace leading to a voltage divider comprised of at least one resistor to the output pin and one resistor to ground. These are going to be small resistors not power resistors.

- channel 2, feedback pin

I've listed 8 possible pins (not 9 because "usually" it would have either a ground pin OR a negative power rail pin, not both. Looking at the PCB, it does appear to have a ground pin not a negative rail pin). There appears to be 11 pins total on these chips so it might have a mute pin or some pins are unused or some other function I can't think of at the moment.

Using measurements and deduction, map out which pins do what and whether the power and signal are present, then with the pinout it's just a matter of finding another power amp chip with a compatible pinout which may take a little while or might not. The supply voltage will also tell us if a replacement chip has a compatible voltage range.

Keep in mind, I wrote a lot about the power amp chip but it may be something else. After the power amp chip I would do similar probing on what look like two opamp chips, one looks like a DIP 16 (pin) an inch away from the power amp chip and the other a DIP 8 twice as far away from the other power amp chip.

[999999999]

Oh, degraded capacitors will usually make it sound bad/distorted or shut off, not just have no output like exhibited on this unit.

[imp]

Originally Posted by 999999999

imp, what is a Rifeking?

Sorry, I meant the capacitor made by Rifeking in the PSU

[martinlink]

Got tired of the useless and long testing so I popped the PCB under a microscope and thoroughly checked every connection of every leg on the IC's. And fair enough, managed to find 2 that had cracks on them. Kind of skeptical at first but after re- soldering them, I popped the set together and lo and behold... All is fine and working again!
The potentiometer for the bass was a bit wobbly and the contact was loose in it, so I opened it up and found that the little metal strips were bent, so i rebent them and it worked too!

Amazed at my technical skills now, as I have a great speaker system now!

Thanks to everyone who helped!

[budm]

Iwould resolder all the pin and any thing that get hot such as large resistors, Power AMP IC pins, very common failure in poor wave soldering works. You can see small ring develop around the pin.

[f-bone]

After so many years Id like to thank you guys and the forums for helping me solve intermittent subwoofer sound for my soundworks digital 2.1 speakers.

It seems it was some soldering joints plus some corrosion.

Glad I still have them running from SPDIF electrical (SB Live! USB on Hackintosh with Yosemite).

BR,
George

[Bungz]

To bump a old thread.

I have a set of these in the pipe line, they have been in storage for some time.

I plan to recap and re-flow the solder work on the "hot" areas.

Will post up some pics in due course.