[Collins] new subscriber with a 75A4

[Dr. Gerald N. Johnson, electrical engineer]

Is that true of ALL A4? Remember they were graded by serial number by
the dealers and the later versions sold at higher prices even when new
because of upgrades that weren't retroactive.

>From my experiences working at Collins on projects monitored by Art, I
am certain that not all circuits were better performing, just they
pleased Art more. He tended to start projects, then check on them,
indeed at the times I worked there the design procedure was to write a
"working paper" and submit it to the "system." In a few days one's boss
would receive a call from Art approving or disapproving of the design in
the working paper. But at that time Art hadn't been in a class room or
designed a circuit in decades, yet he decided equipment details by such
oversight and edict.

One day in the transmitter lab in Texas, I was measuring coil Q's in
tapped coils suitable for medium to large transmitters. The coils and
Q's were far too large to use a Boonton Q-meter, so I had a transmission
setup with very light link coupling to a coil and very low loss vacuum
variable capacitor. With a vernier added to the dial of an HP606 signal
generator and a magnifier over the -3dB point on the RF VTVM, all
mounted on a copper bench top to minimize interference from
transmitters, I had to use a line voltage regulator to keep levels
adequately stable. HP tried to sell me a 5110 synthesizer but my level
measurements showed it far too crude. Anyway (this being a project
commissioned by Art) he came by to see how it was doing. As I was taking
a small difference between large numbers (coil Q over 1000 at 3 MHz) I
was using an abacus for the difference then my slide rule for the
division. Only business managers had calculators and those
electromechanical monsters would have upset my sensitive setup. That was
long before pocket calculators existed. Art didn't understand the
measurement because he asked why I didn't do all the calculations on my
log-log duplex vector slide rule. I just handed it to him and asked him
to show me how to take very small differences between large numbers. His
reaction was to call for a computer type to computerize the data. Which
slowed data taking because I STILL had to make the computations to find
the next frequency (because the unused part of the coil causes a suckout
in Q) based on what I'd found. Then I had to enter the data into the
data entry form for the programmer to have it plotted. And the first
plots included circles because I bounced back and forth across the
suckout. But Art was pleased with the pictures in the report. Art was
pushy, he had the money and company ownership to run, but he was not an
electronics genius at that point. So radios were shipped with design
defects even though approved by Art himself. Not every A4 has been
perfect or all would have had the same dealer value. And there would be
only the 75S1, never the S3, or 3B (with extended coverage S2, 3A and
3C).

Art's finances were equally troubled, else he'd not have sold the
company to Rockwell for one years gross income. He was in charge, but a
mortal that made mistakes and had not learned everything there was to
know about products his company sold.

73, Jerry, K0CQ

-- 
Entire content copyright Dr. Gerald N. Johnson, electrical engineer.
Reproduction by permission only.