RSG BPS Meterr

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "S & J Young by way of Terry Fritz <twftesla-at-qwest-dot-net>" <youngs-at-konnections-dot-net>

This is mainly for the few of you who use ASRGs (e.g. with DC supplies).

I use an angle grinder run from a variac for my RSG in a DC powered TC.  I
like calculating stuff, and one parameter that is important is rotor RPM or
gap BPS.  My previous method of using a strobe to set RPM then measuring the
AC voltage to the angle grinder was not very repeatable.  After the grinder
runs a while, grease & bearings warm up, and one gets more RPM for the same
voltage.

So I decided to make a direct reading BPS meter.  To avoid the horrendous
electrical noise problem of a RSG switching high power, I decided to try
fiber optics.  It turned out to be really simple.  Here's how:

Couple a small opaque disk (e.g. 2-3 inch diameter) to the rotor shaft.
Drill the same number of holes or slots inside the disk perimeter as you
have rotating electrodes.  I have 16 electrodes, so I drilled 16 3/16 inch
holes spaced so the holes and "no-holes" are spaced the same.

Jameco sells inexpensive 10 foot lengths of cladded fiber optic cable.  They
also sell a mating NPN phototransistor that accepts one end of the fiber
cable.  Arrange a bright red LED to shine through the holes in the disk to
the other end of the fiber cable.  My red LED only required abut 3 ma to
give good results, so I just power it from 2 1.5 volt batteries in series
with a current limiting resistor.

I connected the phototransistor in series with a 33K resistor and applied
about 4 volts to this combination.  One can then hook a frequency counter (I
use a Wavetek 27XT) across either the phototransistor or the resistor.
Presto, it reads out directly in BPS - very nice!

I improved this a bit by adding an emitter follower to the phototransistor
so the output impedance to the 27XT was lower and less succeptable to EMI.

The 27XT gives 3 digits resolution (measures Hz/BPS).  Some other less
expensive DVMs have a 20 kHz max frequency measuring capability, but they
only give an output in kHz.  163 BPS will read .016 - you loose the 3rd
digit.

--Steve Young