Edson, Alden, 1942-1968

              Dear Pat, Clyde, & Annettei
As usual, I'm starting a letter with very little idea
of what I'm.going to write. There seems to be a general
scarcity of news here. Work at the lab. is very brisk now
and there seems to be always something requiring my attention
at home. I've been trying to finish up an unfinished
telephone table and a night table, paint the kitchen and a
million other things of the sort. This week, to add to my
miseries the lizzie snapped a starter spring bolt and I
had to put in a couple of evenings on that.
In common with everyone else I know, I howl eternally
about the small size of my paycheck, but otherwise things
are going pretty well at work. I am busier than the traditional
one-armed paperhanger with the hives, but that's a reasonably
healthy situation. E. F. Cone, the editor of Metals and Alloys,
informs me that my weld hardenability paper is being prepared
for the printer, so I guess it will come out in two or three
months now. I feel that I am rather sticking my neck out on
that, for it will probably evoke a riot of discussion. My
story seems to be pretty well substantiated by evidence,
though. My gear testing program, which goes through cycles
of aotivity and inactivity is about set to come out of -the
dormant stage once again. The main headache just now is to
establish a reliable criterion of gear failure. Initially, we
expected to be able to determine gear failure by deorease in
mechanical effioienoy, but (and here's one for the books) the
mechanical efficiency of the gears increased steadily until
so much metal broke out of the teeth that the tests had to be
stopped to.replaoe ruined shaft bearings. With shielded
bearings we just gave up the ghost; the efficiency didn't
decrease even when the gears were so badly chewed up that
you couldn't stand the noise near them. They were stopped
after days of service under compressive stresses on the order
of twice their nominal ultimate strength. Currently, the
argument relative to"when is a gear not a gear,but a piece
of junk" is raging hotly. The next series of tests will be
run with a wattmeter sensitive to about 0.01[percent] change in
gear effioienoy, thermocouples and a precision potentiometer
to measure-the oil temperature rise as the gear teeth pass
through mesh, and a trick noise meter with a oontaot microphone
fastened to the gear box. I'm banking on the noise meter
(which I built), Pilling is banking on the oil temperature
rise, and the wattmeter is going along for the ride, mainly
because it was purchased for the machine and we haven't any
better use for it. My mein pride and joy, though it is a