The Incredible Shrinking Capacitor
I had always heard that when a piece of electronic equipment fails, the electrolytic capacitors are often a prime suspect. I got a much better sense of that today as I tried my hand at fixing an old digital projector I had gotten from a friend.
The unit wouldn’t even power on and while I suspected the power supply, I wasn’t able to figure out why it wasn’t working. That is, until I went online and found a post on Fixya.com where other people had run into a similar issue. One poster pointed out that there are four aluminum electrolytic capacitors on the power supply board which are rated at 85 degreesC and, because the board gets so hot in normal operation, these are often the first to fail.
I found the capacitors and they looked ok; their tops weren’t bulged nor did I find a short circuit across their terminals. I needed a way to test their capacitance (since I don’t have a multimeter with this function) and so I set up your typical 555 timer circuit where you use an unknown capacitor with two known resistors and you calculate the capacitance from the resultant frequency. When I did this, sure enough, I found the capacitors in question were significantly below their rated capacitance values and were in need of replacement.
One capacitor in particular, a little 100uF one, was particularly strange in the fact that it’s capacitance changed dramatically while I was watching it on the oscilloscope. I watched it start from around 40uF (still too small) and dwindle down to mere picofarads in the span of a minute or two. I thought it was so neat that I captured a video of it.
(a definition of the word “neat” you weren’t previously aware of, I bet)
I later found out that the whole shrinking thing was a function of it having being heated by the soldering iron and cooling down. Still, I thought it was cool.
Anyways, I replaced the capacitors and plugged the projector in and…. nope, still doesn’t work. However, the fans *do* spin up which is way more life than it had previously and also I’m seeing some stable voltages on the power supply for once. Looks like there’s some other issue at work as well.