The Major has expressed a desire to have some speakers for her home office so she doesn't have to use headphones all of the time. So an ancient set of Labtec LCS-2612 speakers were located and extracted from a bedroom of one of the tykes. They had been using them for playing music from their phones until they demanded Bluetooth speakers instead.
When I say ancient I mean it, these are from back when beige was king and arrived with an equally beige desktop computer way back when (20+ years ago). If there was a manual with them when they arrived it's now long gone.
Labtec LCS-2612 Speakers. Never the knees of the bees. |
It should have simply been a matter of setting them up and connecting them to her laptop, but all was not well with them. They sounded like they were playing from the bottom of a well. The problem seemed to be at the volume controls, as when they were pressed in and turned the sound came back to normal levels. Dodgy potentiometers are a common issue.
Potentiometer diagram |
Over time either crud builds up on the track and wiper or one or other of them can get oxidised. Both issues lead to patchy contact across the pot, which creates the characteristic pops and crackles that you get from a bad pot. Fortunately options are available for cleaning and restoring them so I didn't have to go straight to pulling them out and replacing them. First though I needed to get at them.
Labtec LCS-2612 external front and rear views |
Removing the rear and side covers was just a matter of taking out the four cross-head screws at the back of the subwoofer to reveal more discrete components than I have seen in a while.
Labtec LCS-2612 subwoofer internal view |
The pots were located at the front of the box and the front cover was secured by another set of four screws in a much more difficult position to get to than the first set. However I managed to remove them without too much bother.
Volume, treble and bass pots |
Looking at the pots, the external shells were tarnished, so it seemed a safe bet that the internals were likewise affected. The first choice to make is to replace or clean. Replacement means identifying the part then sourcing a supplier. For some more specialist pots this can be a tall order, but these are not high-end parts so it was possible to locate a picture on the web.
Panasonic dual-gang 50K pot |
I couldn't quite tie down a part number but at about £8 each that would come to £24 plus p&p to replace all the pots. As a similar set of secondhand speakers would come in at about the same price it didn’t seem worth either the expense or the trouble to replace them.
The next choice was to clean in situ or desolder and clean; in situ is obviously best/easiest. I needed to accomplish three actions, clean up the wiper and track of the pot, remove oxidation if present and lubricate the cleaned surfaces.
And this is where I fell down the rabbit hole. There are many contact and control cleaners out there and twice as many opinions on what is the best product or strategy to use. The market leader is about £30 a bottle, which again is a bit toppy given the replacement cost of the speakers.
So I had to venture even further into the rabbit hole. In the end it looked like trying to clean and then lubricate the pots would be the first strategy to try. Most contact/control cleaners (the ones you can buy now anyway) use naphtha or white spirit as a base, both to solvate their active ingredients and for its cleaning action. So I took a fine plastic pipette and dropped just a couple of drops into the pot housing via the notches at the tops of the pots and worked the controls for a bit.
Notch on the top of pots provides solvent access without having to desolder component |
White spirit isn't conductive so it shouldn't matter if there was a bit left behind. Even so I dried the pots out with a hair drier as sparks and white spirit aren't a good mix. Once that was complete I plugged the power in and tested the system and everything was back to nominal. So I wouldn't need to remove any oxidation this time. The next job was to lubricate the controls to stop the dry surfaces abrading the track of the pot and knackering it. Again there was a wide range of lubricants on offer…
WD-40 was mentioned quite frequently and appears to be quite a polarising topic, with some saying that it has worked fine for them for 20 years and others saying that it will wreck controls. For me it has a couple of major advantages, in that A) it is cheap and B) I have it to hand. I don't think the speakers will get that much more use and if they fall over they don't owe us anything.
So I put a couple of drops on the joint between the shaft and case of the pots and let it wick into the body of the pot. Then I rotated each of the pots from fully open to fully closed ten times and put the hair drier back on it.
Now to get the stuff back together. I ended up using a magnetic catcher to replace the screws, As you can see they were well buried and it required a Phillips screwdriver with a 10" shank to get them in and out.
Buried access to front cover fixings on a Labtec LCS-2612 subwoofer |
I also noticed that one of the struts protecting the subwoofer was broken so I solvent welded it back together with MEK as the case had ABS stamped on it. No problemo.
Broken strut on Labtec LCS-2612 subwoofer cover |
So I now had the pots working smoothly and the case reassembled. But the sound was not really that good and the high ranges seemed to be missing? Back to troubleshooting mode again.
This time it looked like a problem with the feed in. Either there was another bad connection or the system wasn't processing the stereo input properly, as it looked like it was expecting a pair of L/R mono inputs. After a vain search of the cables box for a stereo to mono splitter lead it was off to eBay.
Female stereo 3.5 mm jack to a pair of male mono 3.5 mm jack plugs |
Once that arrived I duly plugged it into the rear of the box. But whilst doing so I found a bad connection in one of the jack plug housings on the lead I had been connecting the laptop and speakers with. So it was at this point that I found out that the stereo-to-mono splitter wasn't required. Doh! But that is what comes of working off no manual. At least the next time they will be to hand and they weren't expensive. So the inputs labelled 1 and 2 are for two different stereo inputs. Noted.
I then swapped out the dodgy patch cable and the volume came up to expected levels. Better, but the high ranges still weren't right. Next I plugged in a pair of headphones to the jack at the front of the box and everything was good, indicating that there wasn’t a problem with the stereo output of the amplifier.
The only culprit left was the speaker and after a few runs with a stereo test program and some swapping around of the speakers it was obvious that the left speaker unit wasn't working properly.
On opening it there was nothing obviously wrong (there isn't much to go wrong), so it's either the cable or the speaker. The impedance was 4.6 Ω so it wasn't an obvious fail. I would now have to replace the cable and/or the speaker, but at this point the law of diminishing returns loomed large. The Major decided that the speakers were good enough as is, so I called it at that.
JOB DONE (sorta)
Time taken: 1 hr (plus literally 2 days for contact cleaner research!!)
Cost: Cleaning materials, pennies - stereo to mono splitter cable, £2.95