I have an Oriel 77501 fiber optic illuminator. It’s tungsten based. I have a 0.4 mm single quartz fiber inserted in it (this is much skinnier than the 1/8" or 1/4" glass fiber bundles it’s usually used with).
I just spent an hour fooling around with various means to estimate how much light is actually sent down the fiber - basically, I pointed it at a sheet of paper and photographed the circle of light. When I hold the fiber more than an inch or so from the paper, the circle (though bigger) is so faint I can’t quite see it. I then analyzed pixel values to see how much brighter the circle was than the surrounding paper (crossing my fingers about issues like gamma correction in my iPhone camera) and measured its diameter. Then I estimated the brightness of the room to establish a reference brightness. I went back and forth between lumens and watts, using published conversions for “ideal white light” and similar things, for light itself (not for lamps which also have inefficiencies and generate heat). I used lighting guides, and also counted my fluorescent tubes in the ceiling, and also used photography exposure guides with lists of EV values in various settings. These different approaches all agreed within a factor of two or three or so (that’s actually surprisingly good I think). All of this is stuff I feel pretty comfortable doing.
What I came up with was 6e-6 watts or 1.5e-3 lumens, for the light traveling along this fiber. WOW! Could it really be that small? The source has (I think) a 150 W tungsten lamp inside and its label says 120 VAC 2 A. Could its overall efficiency converting electricity to fiber-born light really be just one in twenty million?
Just looking for a gut check here, from one of our experts in optics! Thanks!
How is the light source being coupled to the fiber? The end of the fiber is just pointing at the light source, it’s not going to have much of a solid angle at .125 sq. mm. area.
The average efficiency of a tungsten lamp in delivered light relative to the energy it consumes is only 1.9% to 2.6% to begin with and then only a very minuscule portion of that is going to be captured by the receiving end of the little fiber.
Beowulff, the fiber has a pretty standard collimating fixture on the end that’s about 12 mm or so in diameter. The fiber is 0.22 N.A., so the angle isn’t all that wide. There are lenses of some sort inside of the source, a box somewhat bigger than the largest shoe boxes. There’s also a filter slot, a gate, an aperture, and a little hardware for alignment like spring loaded adjusting screws. I may well be mismatching things, but I did play for a while with the insertion depth and the alignment until I got a pretty even and relatively bright solid cone of light coming out the other end.
Astro, yes, I anticipate pretty poor efficiency. But we’re dropping seven orders of magnitude. Figure two, maybe, for the lamp itself. The lenses ought to be able to refocus the lamp emission to a surface brightness that’s a significant fraction of the filament brightness - and I am sure the fiber end is not five orders of magnitude smaller in area than the filament. I think I’m having sticker shock - it’s HOW MUCH???
I wonder if there is some simpler approach I am missing to figuring this out.
I do have some thermopiles I might be able to use to get a total power, or at least put a small upper limit on it. They’re pretty small but I could estimate whether they should handle what I think I have, and I could sneak up on it with the voltage control and the gate and aperture. Maybe I’ll try that tomorrow.