ATM Re: Dwight's tester

["Bob Walker" <tac@i1.net>]

Innovation rears its ugly head!  Think of it--a slitless tester with a
stationary source!  As it happens, Mark's surmise about my tester was
correct: it does have a moving source.  My LED is less than 1/2" long, so it
doesn't interfere with getting my eye into the right position.  I would
suggest that if you're getting enough heat from an LED to be objectionable,
you may be running too much current through it.  Check the specs, and keep
the current well below the Absolute Maximum ratings.  There will probably be
maximum and minimum current ratings.  I have found that a value about
halfway between will give good brightness and low heat, which translates
into long life.  As an example, consider an LED rated 3.7 Volts, 30 mA max.
A good guess at a reasonable current rating for long life, low heat would be
20 mA.  Powering it with a 9 V battery,  we need to insert a resistor in
series that will drop the excess voltage:  9 - 3.7 = 5.3 Volts.  Calculate
the value of the resistor with Ohm's Law:  I (current)= E (voltage) / R
(resistance).  Plugging in our numbers:  .02 A = 5.3 V / R.  Rearranging,
R= 5.3 V / .02 A  or 265 ohms.  To determine how much power the resistor
must dissipate, multiply current by voltage: .02 * 5.3 = .106 watts.  So, a
quarter-watt resistor will do the trick.  Actually, a 1/8 watt resistor
would work, but I believe in safety factors.
Back to the layout of the tester....I made mine with a very simple spring
clip to hold the knife edge in place in front of the moving source.  This
allows me, in a matter of approximately 3.27 seconds, to switch from
Foucault to Ronchi testing.  This is important, because the latest lesson I
have learned is: If two different tests give significantly different results
for your mirror, You're Doing Something Wrong!

 *    Best regards,
*          Bob
         *          *    *

           *         *