Here's a nice little time-waster for you all...

You have an opamp circuit breadboarded on the workbench. It's showing=20
symptoms of oscillation, and the oscilloscope FFT shows a spectral peak=20
at around 30kHz and ~25dB above the noise floor. There's a second=20
(smaller) peak at around 9kHz, ~14dB above the noise floor.

After several hours of head-scratching, circuit simulation,=20
pen-and-paper recalculation and so forth, you decide to give up and take=20
a breather. You turn the desk lamp off...

And the circuit springs into life, the MCU boots, proclaims that=20
"powerup diagnostics [have] passed", and proceeds to display perfectly=20
valid A/D measurements on the serial port. The oscilloscope has also=20
decided to behave itself -- the peaks are gone, replaced with the=20
near-DC waveform that's supposed to come out of a well-behaved current=20
sense amp.

The lamp in question was later found to be emitting a 30kHz peak, plus=20
harmonics beyond several MHz, and was more than capable of overloading=20
an RF field sensor at a distance of a foot.. Its effects were still=20
measurable on the other side of the room.....

Some days, I hate this hobby. Other times, I just hate the offshore=20
manufacturers who fill the market with garbage which doesn't even meet=20
EMC regulations.

Errgh.

For anyone interested -- this is a Maplin A29FF "magnifier/desk lamp".
   http://www.maplin.co.uk/22w-fluorescent-daylight-magnifier-lamp-47980

=A340 for *this*...

--=20
Phil.
piclist@philpem.me.uk
http://www.philpem.me.uk/

--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.