[sci.electronics] Building from Scratch -- a Reminiscence

rsd@sei.cmu.edu (Richard S D'Ippolito) (02/27/88)

(This is somewhat long and has little technical content.)


In article <1988Feb23.175954.5465@utzoo.uucp> henry@utzoo.uucp (Henry Spencer) writes:
>> And everything is in the IC's today--so you aren't going to be building
>> circuits like the hobbyist of 20 years ago, only wiring up IC's to
>> switches, etc. with a few other discrete components.
>
>"Why, sonny, hobbyists twenty years ago didn't really *build circuits*.
>Oh sure, they strung together components here and there, but they bought
>all them components at a store.  Back when hobbyists was real hobbyists,
>we had to *make* most of our components...  You can't call it "building"
>circuits when you just buy the parts and wire them together!"
>
>Yes, ICs do more for you than transistors did.  And transistors were just
>a bit easier to work with than a hunk of galena and a catwhisker, too.
>Having ICs to use just means that you don't have to spend two days building
>and debugging a NAND gate before doing something *useful* with it.
>-- 
>Those who do not understand Unix are |  Henry Spencer @ U of Toronto Zoology
>condemned to reinvent it, poorly.    | {allegra,ihnp4,decvax,utai}!utzoo!henry


One of the reasons we built components was money. The other was fun. What I
couldn't buy, I salvaged from old radios (vacuum bottles with 2-digit
numbers, etc.)

Call me an old fart if you want (Is 44 old), but I had a lot of fun as a kid
making things from scratch like motors (even a synchronous one from
permanent magnets), hot-wire ammeters, galvanometers, spark coils, geiger
counters (still have the detector tube!), and human-hair hygrometers.  I
suppose that's where I learned to like engineering and design.

The project I remember with most fondness was making lead-acid storage
batteries to use as a DC source for experiments. I charged the thing with a
selenium rectifier. It was neat -- my Dad brought me scraps of lead sheeting
plumbers used under shower stalls and bought the lead oxide and sulfuric
acid from Fisher Scientific. I drilled holes in the plates, filled them with
the oxide paste, made separators from thin strawberry basket strippings, put
the bundles in jelly jars, mixed the electrolyte, charged it and was in
business. Good thing my parents didn't know how dangerous that stuff was! --
or maybe my Dad just trusted me to be careful. I won't get into the
chemistry set, bombs, and rockets...

To those with kids or having similar interests -- I say go for it. The
thrill is seeing the finished design work. Don't apologize for buying what
you can. Right now I'm working on an train signal control system for my
model railroad. I haven't decided whether to use small logic modules at each
block or run the sensors into a SBC. The I/O fan-in and fan-out can be quite
huge. Consider: 30 blocks on five major routes with detectors and
three-light signals at each end and some 25 turnouts. I have to design the
block occupancy detectors, too.  The damn thing would be easy if I could buy
a custom chip for the local block logic. A PLA with two dozen gates and four
JKs would do the job for each block. Hell, I need a small SBC that runs Ada!
Who wants to make me one?

Then there is the cab control (much later). I have a neat idea for that which
doesn't use messy, interfering rf or complicated carrier signals on the track.

Time, time .... I just need the time! As a kid I had the time and little
money -- now I have the money and little time. Isn't it always this way?

				-----0-----

Oh, and let's stop knocking Radio Shack -- they're filling a gap. I know all
about JEDEC specs, Mil standards and AQL's as a professional designer. The
stuff they sell is fine for breadboarding and concept testing, isn't it?


Yours,

Rich D.
--

--------------------------------------------------------------------------
The facts expressed here belong to everybody, the opinions to me. The dis-
tinction I attempt to make obvious...			   RSD@sei.cmu.edu
--------------------------------------------------------------------------

phil@amdcad.AMD.COM (Phil Ngai) (02/29/88)

In article <4367@aw.sei.cmu.edu> rsd@sei.cmu.edu (Rich D'Ippolito) writes:
<block occupancy detectors, too.  The damn thing would be easy if I could buy
<a custom chip for the local block logic. A PLA with two dozen gates and four
<JKs would do the job for each block. Hell, I need a small SBC that runs Ada!

Sounds like a large PAL (23S8) would do. If not, surely an LCA would.
-- 
I speak for myself, not the company.

Phil Ngai, {ucbvax,decwrl,allegra}!amdcad!phil or phil@amd.com