Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: circuits



Tina Fanetti wrote:
Does anyone know a good basic electronics reference book?

I doubt there's anything better than Horowitz and Hill.

I was hoping to make a radio at some point (simple one)

I can do the theory but watchin me wire one of
these things is distressing.

The right way to wire "these things" depends on just
sort of thing we're talking about.


*) Is it temporary? Temporary means you would rather
rebuild it the next time you need it, rather than buying
the parts, documenting what you've build, storing it, and
finding it again when you need it. It's also temporary
if you'll be wanting to change it.

If yes --> you might want to use a matrix board, which
makes wiring a snap.


*) Are you going to make more than one or two?

If yes --> you might want to make your own printed
circuit boards. There are techniques for doing this
that are a lot simpler than you might think.

There's some set-up cost involved to acquire the equipment
and the expertise, but somebody in the physics department
and/or the EE department ought to spring for it.


*) Does it involve high frequencies?

If yes --> you might want to make your own printed
circuit boards. It gives you better control of
the geometry.


*) Does it involve lots of chips?

If yes --> you might want to use wire-wrap. Get
good sockets. The socket probably costs more than
the chip that's in it, but it's worth it.

(Hint: Have you ever tried to debug a circuit with
200 chips, 3 of which are in intermittently-defective
sockets? It doesn't matter how logical the design is;
the implementation is defective and there's no good
way to fix it.)


*) Does it involve a small number of non-expensive
non-fragile components?

If yes --> then you can probably solder things in
place.

Hint: A piece of vectorbord with the three-hole
solder-pad pattern will probably lead to a quicker
and better result. In particular it allows you
to copy a design from matrix board to vectorbord
with no great brain-strain.


===================

In general:

++) Get somebody to show you how to solder properly.
It's not easy to figure that out from a book alone.
If you want to supplement the lesson with a document,
what you want is
NASA-STD-8739.3 SOLDERED ELECTRICAL CONNECTIONS

which tells you how to make a spaceflight-qualified solder
joint, and how to inspect such things. At various times
in my life I've paid people to solder stuff for me, and
I always insisted that they do it right. It ain't that
hard once you know how. And if you don't do it right,
the result is a nightmare. Undebuggable.

++) Get a good soldering iron. You can't make a good
joint with a crummy tool. (Alas the converse is not
valid.)

++) Similarly, get somebody to show you how to wire-
wrap properly. And get a good tool.

++) Put it in a decent chassis. Good chasses aren't
cheap, but there's nothing more uncool than a good
electrical design that's mechanically unsound.

++) Horowitz and Hill has a chapter on construction
techniques.

=======================

I was hoping to make a radio at some point (simple one)

How simple? Five components? (Crystal, L, C, headphone,
antenna)?

At the other extreme: You can get a radio on a chip these
days, for a few bucks. Wiring it up isn't very educational,
and the result isn't very exciting unless you're embedding
it into something else that's exciting. Designing your
own radio is hard; soldering together a typical kit is
mindless. Typical radios, even "hobbyist" kits, are what
I call overdesigned: you can't change a single thing
without redesigning the whole thing, and changing 17
other things at the same time. This is like jumping in
at the deep end; it's not a good way to learn to swim.

Note that Horowitz and Hill doesn't discuss RF stuff
until the next-to-last chapter, waaay in the back of
the book.

There is no royal road to electronics. It's not a trivial
subject. A BS in electrical engineering is not particularly
easier to get than a BS in physics. Don't expect to read
Horowitz and Hill as quickly as you read Treasure Island.