mystery stories. Except, as you can tell from the picture, it is
not about diagnosis of a human. Rather, it is about diagnosis of
a machine. The photo shows the inside view of a Fellowes SB-87Cs
paper shredder. I bought this several years ago to shred several
boxes full of old charts. Given the intended purpose, I paid
extra for a heavy-duty model that could shred many sheets of paper at
once, cutting them into tiny diamond-shaped flecks. It worked
perfectly well, for many years. However, it eventually developed a problem. It would still shred,
but it would shut off after about five seconds of operation.
Due to the fact that the shredder mechanism itself was still fine, I
couldn’t bring myself to throw it out. But I also couldn’t
imagine paying for a repair.
After some investigation, I noticed that there is a sensor
inside, The sensor shuts off the machine when it detects that the
bin is full. The bin holds the shreds of paper. When the
level of shredded paper gets to the top, the paper pushes up a little
plastic flap. On one side of the flap, there is a light-emitting
diode (LED). On the other, there is a photo sensor. When
the flap goes up, it blocks the light, which causes the sensor to stop
conducting electricity. This causes the shredder to stop, and
causes an indicator light on the top to go on.
The indicator light would go on after about five seconds of operation,
even though the flap was still in the down position. Why?
After a little investigation, I found that the LED was burned
out. Aha! I checked my diagnosis by shining a flashlight at
the photoresistor. The machine worked when the sensor was
illuminated. Simple enough.
Even though the unit is not really built in a manner that would make it
easy to repair, I did manage to remove the LED, check the voltage (it
operates at 3 volts) and find a replacement. A quick trip to the
parts store was in order. I then managed to get the replacement
fitted back inside. I confirmed that the LED would light up, and
reassembled the whole thing.
Problem. The new LED did not make any difference.
Even though the sensor was getting light, the machine turned itself
off, and the indicator light indicated a full bin when in fact the bin
was not full.
I noticed something odd. The LED would remain on, even when the
machine was turned off. That explained, perhaps, why it burned
out. It had been on constantly for several years, with the
machine being plugged in. But it did not explain why the machine
would turn off after five seconds, with a functioning LED.
More investigation. I had to defeat safety mechanisms that
prevents the machine from being turned on with the door open and the
bin removed. Mindful of the possibility that the thing could
destroy my fingers, I turned it on and watched a piece of paper get
shredded. It would shred about a half sheet, then stop.
Bizarre. When the operating switch was turned on, the LED stopped
shining steadily. Instead, it would blink every five
seconds. I figured the blinking caused the machine to shut off,
although it was not entirely clear why it did not restart itself.
Perhaps it was designed to wait for some indication of human
intervention before it would restart.
After another trip to the parts store, I attached a little battery pack
to the LED, bypassing its internal power source. That way, with a
steady power source, the LED would not blink. (Also, I could
remove the battery after using the shredder, thus preventing the LED
from burning out.) That had to work.
Except it didn’t. I confirmed that a steady, bright, light came
from the LED, and hit the sensor. The LED no longer blinked,
given the power from the battery. Even so, The shredder continued
to malfunction in exactly the same way.
So, although I could not figure out exactly what the problem was, I did
at least localize the problem. The problem was in the circuitry
that powers the LED and interprets the sensor. In other words, it
was not a problem in the peripheral nervous system; rather, it was in
the central nervous system. There is a little circuit board that
connects to the sensor and the LED, which has some relays and other
stuff. The problem was somewhere on that board or its components.
For a few moments, I thought about how I was going to establish a more
precise diagnosis. But there were problems with that. For
one, I don’t know much about digital stuff. I don’t think I could
have figured it out. Two, I was not in the mood for a third trip to the
parts store, for yet another part. So I just took the two wires
that go from the circuit board to the photoresistor, cut them, and
soldered them together.
I was not sure that would work, but it does.