Sunday, 12 May 2013

How to Save 200 Quid (or Repairing Standby Fault on Mordaunt Short MSW20 Subwoofer)

Having moved three times in two years I have only just last week bothered to set up my hifi for the first time since 2010. Unfortunately, this period in storage and bouncing around in removal vans doesn't seem to have been good for my decade-old system and amongst other issues I found that my MSW20 sub woofer had stopped working. I didn't fancy spending hundreds of pounds replacing kit I was perfectly happy with so I decided to have a go at fixing it myself, despite component-level fault finding not being a forte of mine.

I wasn't sure whether the problem lay with my surround amp (which was also suffering from other faults), the line cable, or my subwoofer. I thought I had proved the sub because when I connected the line in cable with the sub's volume on full while music was playing from the surround amp, it gave a good "thump" from the small surge generated on connection. This was misleading. After confirming the amp line out and the cable with my multimeter (volts fluctuating during music play) I came across this post at AV Forums and realised that my sub appeared to be suffering from the exact same problem - it was stuck in standby.

Dismantling the sub was a little fiddly, but as always, it's easy when you know how; I will explain that in this post. Finding the fault was a LOT harder and, frankly, the fault was discovered more by luck than anything, especially as there were no obvious signs of damage to any of the components.

Idiot Disclaimer: It should go without saying (as should most safety warnings) that you shouldn't attempt the below unless you are confident with electronics, electrical safety, and safe use of soldering irons. You do this work entirely at your own risk and if you break anything, hurt yourself or someone else, set anything on fire, or generally do anything stupid, you're the one responsible, not me or anyone else. If you're not sure then seek professional advice, etc etc.

To remove the woofer amplifier, first switch it off, remove the power cable and wait a while until the capacitors inside have discharged. Then unscrew the outer ten screws from the back panel:


Gently begin to remove the plastic box containing the electronics. Before removing it entirely you will need to put your hand in the speaker box and disconnect the red and black speaker wires by unplugging two spade connectors at the back of the speaker. Wobble them while pulling them gently but firmly and they should come free.

Then remove the next four screws in:


Now, gently remove the plastic housing from the metal chassis with a flat-head screwdriver. I'm not sure if they're glued together or if the rubber has just gone sticky with age but they will come apart with a bit of persistence.


Unclip and unplug the speaker wire from the PCB so that you can completely separate the two parts of the housing without breaking the glue sealing the hole for the speaker cable. After you've done this be careful where you rest the metal chassis as the edges are sharp and scratchy. (Antique wooden tables are best avoided - I'm still in the doghouse for that one!!)



This is where things got interesting. After a lot of staring blankly at the circuitry, some further fruitless research online and some poking around with the voltmeter while the circuit was powered resulting in a massive spark when my probe slipped while testing the power supply, I nearly gave up. Fortunately, my stepfather was visiting so we decided to have a bit of fun with some father-son teamwork. We identified an area of PCB discoloured by excessive heat as a likely candidate for the fault. We didn't think the fault lay with a capacitor as they usually show visual signs of damage and we thought a resistor fault was unlikely as they rarely go faulty at all and, when they do, are likely to show some sort of burning damage. My dad therefore took an educated guess that we should begin by investigating the transistors, as they both don't tend to show visual signs of being faulty and are capable of generating sufficiently large amounts of heat to create the PCB discolouration we could see. We needed (or thought we needed) to find out the exact type of transistors used, but couldn't see the markings - so I used my iPhone to get the camera where I couldn't get my eye. Note how the plastic of the PCB in the surrounding area is darkened:



We then wanted to see behind the nearest transistor (circled red in the above image) to see the markings on the second one, so my dad bent the near transistor a bit to move it out of the way. While he was doing this he realised that it was wobbly - and, sure enough, the solder joints had cracked.

The solder joint in question is here:


It's a little hard to demonstrate with photos so here's a video of the solder joint wobbling. Make sure you switch to HD so you can see it properly:



I'm not sure whether it is only this particular transistor that suffers from cracked solder joints or whether you might have a problem with the neighbouring transistor or one elsewhere, so you'll need to ensure that the fault does indeed lie here.

Also, take care when re-soldering the joints - my first attempt resulted in a flawed join that cracked again as soon as I touched the transistor, and I've had professional training and experience at this sort of thing. It may be that the surface of the solder point on the PCB is flawed or damaged with heat, or it may be that the transistor acts as a heatsink making it harder to get the solder sufficiently heated through for a good join. Either way, your first join of the three is likely to be a bad one as it's pretty much impossible to stop the transistor moving at all while the solder cools - so once you've soldered all three joins go back and re-do the first one. When soldering, make sure you apply enough heat to thoroughly melt all the solder in the join without over-heating it. As always, don't forget to add a touch of fresh solder just before you remove the iron to prevent a dry (cloudy) joint forming.

(If you need help with good soldering practice I'm sure there are tutorials online somewhere).

I hope this post is helpful. Please comment if you've anything to add or just if this has helped you fix your sub. I was pleased to have saved the £200 cost of a replacement - I'll be even more chuffed to know if, by taking the time to write this, I've saved you some cash as well as helping to avoid perfectly good hifi gear going into the skip!

13 comments:

  1. Hi there, I just wanted to post to thank you for saving me from buying a new sub! I was suffering an intermittent standby on mine until I moved properties after which the msw20 did not survive. I had taken mine apart before but until seeing your guide had not notice the discolouration around those transistors.

    When I tried to desolder mine ( it was lose from the board like yours) unfortunately i found the track had detached from the board so had to bridge 2 of the pins back to the board using wires but it works a treat!!

    Thank you very much for taking the time to write this great article and make the Internet that little bit more useful than it already was!

    Lee.

    ReplyDelete
    Replies
    1. Awesome! Thanks for letting me know, you've made my day - it's good to hear that the time I spent on this wasn't wasted as I'd been wondering whether this would ever be of any use to someone :).

      Delete
    2. Thanks for this great write up - looks like I've got the exact same problem. Need to wait for my broken collar bone to heal before I can wield a soldering iron, but now I know what to do! Don't suppose you've come across an issue with the 'level' knob?

      Delete
    3. No probs :)

      I'm afraid I don't remember having any issues with the level knob, but it is fairly common for level knobs on ageing amps to cause crackling either when being turned or left in certain positions. I'm not sure what the cause of this is though - I would guess at either a bad solder joint that only manifests itself when the knob is touched or a faulty knob. If it's the latter, finding a like-for-like replacement might be impossible; you might be able to find an equivalent but that's likely to be quite difficult I imagine.

      Delete
  2. Hmm, the knob was faulty before - it seemed to be quite temperamental and if you wiggled it a bit and got it in the right position it was OK. not sure if the sub not working is a progression of this or the transistor solder like you had. Hopefully the variable resistor is a standard part I can get easily.

    ReplyDelete
  3. This comment has been removed by the author.

    ReplyDelete
  4. I would also like to thank you for taking the time to put this up where other can find it and benefit. I pulled mine apart, all solder looked fine until I touched the transistor with almost no pressure and I heard it come loose. As with Batman my track had come away from the board. As a make do I scratched down to the copper next to the pads on the track and put a little bit more solder and all seems good so far. Later I may go all the way and add wires as Batman did. Thanks again, Tristan

    ReplyDelete
    Replies
    1. Good to hear! It's funny, this was up for over a year and a half before suddenly in the last few weeks people have found it useful. Really glad it's helped.

      I've been wondering about doing a write-up on eHow - I had thought I'd keep this blog going with other little tips but since I haven't this post is looking a little lonely.

      Delete
  5. I have just acquired an old MSW20 from my son & it didn't work so I followed these instructions, did both transistors just in case and it is now working. Excellent, saves me a trip to the tip.
    Laurie

    ReplyDelete
  6. Great post this for accessing the workings. My sub suffered intermittent sound from the potentiometers. I sprayed some switch cleaner in the pots and turned them at same time, this flushed any tiny dust particles out. Let it dry for 5 mins( it does evaporate as my switch cleaner is alcohol based) and all working as new.

    ReplyDelete
  7. I've just repaired my brother's Mordaunt-Short MS-907W which was "motor-boating" - a medium volume throbbing independent of the volume control. Probing while powered up led me to the 220 Ω, 1/2 W resistors above J225 beside the transistors in the original post. One read 220 Ω and the other was about 250 kΩ. I replaced both with 3 x 680 Ω, 1/3 W resistors in parallel (giving 227 Ω nominal and a power handling of 1 W). The unit seems to be fine now.

    ReplyDelete
  8. I have removed the 10 screws but the back panel does not seem to budge. Wanted to know how to remove the box. Appreciate some help here. Thanks.

    ReplyDelete
    Replies
    1. This comment has been removed by the author.

      Delete