With 10 turns wrapped around the core, that construction
             would exibit a minimum of about 58uH, but the highest could be
             20 times that value depending on the grade of core material.
             Also, since the dimensions are very critical to the calculation
             of the inductance, even a tiny error could lead to a big error
             in determining the inductance. Thats why i will suggest the
             following test:

             Wrap 10 turns around the core and call it an inductor :-)
             Connect a 100 ohm carbon composition resistor (5%)
             (or metal film, 1%) in series with this inductor,
             and power the two with a frequency
             generator set to put out 1 volt ac at 100kHz.
             Measure the output of the generator with a
             high impedance digital voltmeter
             first and set it to 1.000 volt on the meter.
             Now using the same meter on the same scale,
             measure the voltage across the inductor.

             Here is a list of the theoretical ac voltage measured across
             the inductor for various inductors close to 100uH:

             L(uH) ...... vac
             ----- ...... -----
             120uH => .602
             110uH => .569
             100uH => .532
             90uH => .492
             80uH => .449
             70uH => .402
             60uH => .353
             50uH => .300
             40uH => .244
             30uH => .185

             If your voltage reads low, add a few turns and measure again.
             If your voltage reads too high, remove a few turns and measure
             again. Anything between about .4v and .6v will work ok.
             I used a 200uH inductor in the circuit (Brinkmann) and
             it worked ok.

             I hope you have a frequency generator, if not what test
             equipment do
             you have? There are some other ways to measure the inductor
             also.

             If you dont want to be bothered with measuring the inductance
             at all,
             perhaps you can start with 10 turns and see if that works in the
             actual circuit. If not, add a few turns.

             When you test the circuit for the first time, if you want to
             connect
             a 5 ohm resistor in series with the inductor
             that will protect the transistor in case something goes wrong.
             When you are sure it works, you can always remove this 'test'
             resistor and fire it up full steam :-)

             In order to save on the more expensive white leds, if you would
             like
             to, you can do all your testing by replacing the white led with two
             standard red leds in series. That way if something blows the leds,
             you are only out two cheap red ones instead of a $3 white one.
             When you are sure it works ok, pop in the white one :-)
             The two cheap standard leds in series drop about the same
             voltage
             as one white one and are ideal for testing out new designs.
             See that they have a forward drop of about 1.7 volts each.

             Good luck with your LED circuits,
             --Al

             --------------------

             LED's vs Bulb's, the battle is on.