I finally finished my ringlight project. I can't take photos thru my
scope with my camera, but I have put up a brief webpage with a couple
of pix of the light.

http://users.goldengate.net/~dforeman/ringlight/

I've not used a fiberoptic ringlight, but I think this light is
markedly superior to a fluorescent ringlight as Aristo. It exceeds
my wildest expectations. I'm delighted with it.

It uses four 1-watt Luxeon LED's with 30-degree collimators, held in
a ring around the microscope objective. The diameter of the ring is
about 4", but it isn't really a ring, just four lights that are each
about 2" displaced from the objective axis.

Each of the lights is aimable so their beams will converge at the
workplane about 6-1/2" below. They are also each individually
dimmable at the control box, using thumbwheel pots. There is no
color shift when dimming. Maximum brightness is almost too bright to
be comfortable even at max zoom. The light in the field of view is
bright, white and very uniform.

Tools in the work region (tweezers, soldering iron, solder etc) cast
only faint shadows, just enough to give 3D cues. Nothing anywhere
runs even warm, much less hot.

Ths is not a particularly complex project, but it's considerably more
than a one weekend project. For me, it was absolutely worth it.
I am going to enjoy using this tool a lot.