In article ,
says...
In article ,
Ned Simmons wrote:
In article ,
says...

[ ... ]

The ones which I have taken apart (after failure) have a printed
circuit board hosting a high frequency inverter to ionize them, instead
of depending on filaments to start the ionization. So -- the variation
in ionization will probably be too fast to show up relative to the decay
time of the phosphor.


The compact fluorescents with electronic ballasts typically run in the
tens of kHz range. Surprisingly, the phoshpor is not persistent enough
to smooth out the peaks. I was able to measure the frequency of several
lamps with a phototube and scope.

Hmm ... are you sure that you weren't seeing ultraviolet coming
through the phosphors.

It seems unlikely. If that were the case I'd have expected to see a high
frequency ripple biased by the persistence of the phosphors. This was
several years ago, but my recollection is that the trace made a full
excursion.

This was working on an illuminator for
a high speed machine vision application. With very short exposures it
was possible for the vision system to get confused if the exposure did
not average enough on-off cycles of the lamp.

Was the system sensitive to UV? Perhaps a haze filter would
have helped.

The image sensor itself may have been, but any UV would have to have
passed thru not only the lamp phosphor and envelope, but about 2" of
acrylic, a 1/8" glass window, a 1/8" clear urethane belt, and finally
thru the the glass elements of the camera lens. Even acrylic optimized
for UV transmission passes very little at the wavelength Jim cited.

The illuminator backlit the parts under inspection (wooden clothespin
halves travelling at 15-20 feet/sec) and it was easy to see random
variations in the apparent brightness of the illuminator when the
exposure was short enough to include only a few peaks of the light.

Ned Simmons