[Milsurplus] Design for a cheap 400Hz power supply

[Gregory W. Moore]

John, I was just a little "tongue in cheek" about this. I am both an aeronautical
engineer and electrical engineer, as well as a commercial pilot. I know exactly where
you are coming from. However, I tend to get a bit cynical at times, because I have seen
some poorly written specs from the end user, who happened, just like your payload, to
be NASA.
    The specs I refer to were in the mid-'70s, and the components were both the body
flap and speed brake position transducer for the Shuttle. I worked on the design, and
the problem I kept running into was NASA's evidently blind faith in things mechanical,
versus electronic. Both of the transducers were 4 linear transducers on a common shaft,
and were driven through 2 plastic gears. I was more than a little perturbed about the
plastic, and was lobbying for metal. I couldn't see any real redundancy in 4 pots
driven off one shaft, other than the voting by the computer systems in the spacecraft.
Incidentally, I won the argument eventually, with the gears being hobbed from
stainless, but, the same 4 on 1 arrangement remained, as far as I know, to this day.
    The second requirement was in the metal finish of the cases. Bear in mind, that due
to minimization of outgassing, NASA decreed that nothing more than alcohol was to be
used to clean the units. However, they also decreed that the units were to be finished
with zinc chromate primer, which, as you probably know, is thinned with toluene. That
is the way they left the facility, sprayed with zinc chromate, and the date codes and
inspection symbols stamped with epoxy screen printing ink (Wornow Cat-L- Ink, as I
recall).

Believe me, though, I agree wholeheartedly with you about the "held together with Home
Depot Glue". Having had the experience of flying what I worked on and shooting ILS
approaches down to minimums with said equipment, one tends to be extremely careful
about procedures and inspections. The RTV story is a rather famous one, and as far as I
know, is true, as it made the rounds of engineering back in the mid '70s, and was also
semi-confirmed by the manufacturer.
73 de Greg WA3IVX

"J. Forster" wrote:

> Gregory W. Moore wrote:
>
> > Third is "mil-spec" and "avionics grade". This costs about 1000% more than the
> > other two per unit. Comes with voluminous specification sheets and inspection
> > records, and most definitely has an expiration date, usually of only about 2
> > months, after which it must be discarded (ALL the RTV caulking and sealing in my
> > home has been done with this grade ROTFL).
> >
> > The interesting thing is, of course, that all three grades come out of the same
> > vat, at the same time, and only the tube is labeled differently.
>
> I have to disagree with the business of mil-spec stuff. It really does not matter a
> lot if your home's window leaks a bit. It does matter if the seals on a plane leak.
> I built spacecraft payloads for a number of years and you absolutely must use the
> most carefully  selected and QC'd stuff. I saw $55,000,000 in taxpayer money and
> three years of my work go poof because of some wrong paint.
>
> Would you fly in a plane held together with Home Depot glue? Not me.
>
> -John
>
> _______________________________________________
> Milsurplus mailing list
> Milsurplus@mailman.qth.net
> http://mailman.qth.net/mailman/listinfo/milsurplus