David,
What kind of lights are you looking at? Florescent lamps will
display a sinusoidal variation of the mercury emissions on top of a plateau
from the time delay in the phosphorescence, but I doubt that incandescent
lamps will as heating of their filament entails a time lag that should
average out the signal.
In the case of fluorescent lamps I would expect to see a 120 hertz
variation, as the mercury vapor in the tube will reach maximum emission with
the current peak no matter which way the current is flowing, that is you
will get maximum output for both positive and negative peak currents that
happen twice per cycle. Typically the power company deviates less then 0.1
of a hertz from 60 hertz and are required by law to ave the same number of
cycles in each day. This keeps the clocks using synchronous motors keeping
the correct time.
The shape of the curve is probably close to a sine wave for two
reasons the power emitted goes like the voltage squared which if you plot it
looks like a sine wave of twice the frequency, and the phosphorescence of
the phosphor will provide a time delay that will smooth out any rough
corners and keep the illumination from going to zero.
I am surprised and impressed that a CBL will sample this rapidly. At
power line frequency one cycle is 16.67 ms or 8.33 ms per half cycle. If
your only reading to the nearest 0.1 ms then I would expect a frequency
between 119 and 121 Hz, and not as far off as 124 Hz. This corresponds to
about a 8.06 ms period.
I hope this helps.
Gary
Gary Karshner
St. Mary's University
San Antonio, Texas
KARSHNER@STMARYTX.EDU