[sdiy] More vintage computing stuff, was: Static ADC and DAC recommendations?

[cheater cheater]

Hi Philip,

On Fri, Jan 18, 2013 at 5:56 PM, Phillip Harbison <alvitar at xavax.com> wrote:
> m brandenberg wrote:
>>
>> Phillip Harbison wrote:
>> > The professor who taught control theory at UAH did the
>> > control theory work for the Saturn V rockets using an
>> > array of analog computers and an IBM 1401.
>>
>> Would love to read more on that.
>
>
> His name is Dr. Carroll Johnson. He was a graduate of Purdue.
> Someone told me he was the #2 control theory man in the USA.
> When I asked the obvious question (Who is #1?) that answer
> was "the guy who taught him at Purdue". I would guess he is
> now in his late 70's or early 80's if he's still alive.
>
>
>> I'm always amazed at the  gimballing F1 engines and the tiny,
>> tiny little system making  that work and work correctly. That
>> was some first-rate engineering.
>
>
> It's amazing the Apollo/SaturnV went to the moon and back with
> less computing power than in a modern programmable calculator.
> My uncle worked on the computers for the space shuttle. Those
> were about 1000 times more advanced.
>
> I always enjoy looking at how engineers solved complex problems
> in an age prior to VLSI. That IBM 1401 (not absolutely sure I
> have the right model number, but definitely pre-360) had a
> novel solution to power conditioning. It was an electric motor,
> a generator, and a HUGE flywheel on one shaft. It was not a
> UPS by any stretch, but the flywheel had enough inertia to ride
> out a power glitch.

You do use flywheels for UPS nowadays! They amortize better than
battery-based ones because you don't have to keep replacing the
batteries.

> I looked at a much older IBM design using vacuum tubes. It used
> a rotating drum to store its REGISTERS. Can you imagine how slow
> that would be? I guess when the cost of a flip-flop is at least
> one (probably 2) vacuum tubes, you do everything you can to
> reduce the amount of state kept in the CPU.

I think the way you'd do this in vacuum tech would be to use a storage
grid (like in storage oscilloscopes). This is just one tube. But I
know very little about tube tech, so it may have been prohibitively
expensive.

> Someone mentioned the 10-turn pots common on the analog computers.
> I forgot to mention those. The pots were of the Vernier variety
> much like this example.
>
> http://www.ace4parts.com/images/products/503-0004_medium.jpg
>
> There were hundreds of these spread over about 30 computers. It
> is a shame there were not salvaged. Usually when equipment was
> sold for salvage the buyers were only interested in the copper
> and gold. *sigh*
>
> It annoys me when younger coworkers do not respect the work of
> engineers from the "distant" past. How do you know where you are
> going if you cannot appreciate where you've been?
>
> --
> Phil Harbison   <---- almost a fossil
>
>
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy