<snip>
I come from a computer science background and I'm now looking to get
into electronics as a hobby.
I'm mostly self-trained, as the only courses on electronics I've taken
were one year of electronics drafting (at Tektronix, at a point when I
understood rather little about anything electronic) and one single
short class on BJT circuit basics (also at Tektronix, mostly a short
course of testing one's ability to get roughly close to the right
voltage values on some small, sample circuits.) These, at roughly age
27, around 1982. Before this, I had had to learn digital logic basics
and a vague idea of some analog rules of thumb in order to fix some 4k
dynamic RAM cards I paid dearly for from MITS and didn't work, when
assembled. I had spent just about everything I had in getting those
two boards, and then some, and was terribly pained to have them not
work. Since I'd assembled them (sold as kits), I didn't feel I could
send them back, either. After I figured out how to make them work,
I'd learned a lot and _then_ received a blanket letter to all owners,
from MITS, saying that they had some design problems and noting the
fixes (which I'd already handled.) That hard lesson on dynamic ram
and the few courses I mentioned are all that I had at the time.
Not many years later, in 1985 I think, I took on the personal
challenge of converting my IBM Electronic model 85 typewriter into a
printer. For that, I used an 8031 processor, a static ram chip, an
EPROM, and set about figuring out the relay combos used and set up a
table. I wire-wrapped my design and then wired the 8031 system to the
keyboard in a reversible fashion (didn't want to destroy the
typewriter) and set up the software for handling RS-232 input,
buffering, timed driving of the various relay inputs to the
typewriter, and so on. Worked first time! My first real, practical
design. A printer I used for quite some time, too. And it still
worked as a nice typewriter, as well. When I sold it, I simply
removed the wiring (I'd used a ribbon cable pinched by the typewriter
case, without cutting the case in any way, for the RS-232 connector so
the operation was completely reversible) and I still have the box and
processor board in a box in a closet.
However, I'm still able to only muster up a modest understanding of
most things. I have a few things I know well, but the broad
experiences I still lack as this isn't my business and, perhaps, I'm
not as smart or creative as some of those who make a profession of it.
That's okay.
Anyway, my experiences above are NOT a recommendation for how to
pursue electronics as a hobby. It's just part of how my own
experiences played out. If you can find it, an education is worth
having and a good education is priceless. Look at community college
courses as a possible option and take them, if you can manage it. I
still imagine taking courses on electronics for pleasure, someday.
I have a few ideas for circuits I'd like
to construct eventually, most of which revolve around various sensors
and audio processing. One specific circuit I'd like to build a sample
& bit rate reducer for an electric guitar or bass signal with a
CV-controlled LFO to control the intensity of the effect.
It's really important to have some specific applications in mind, just
as you mention, because trying to achieve them will drive you. I
learned more math than I was getting in high school, as a teenager,
because I wanted to design my own telescope. I had built a Newtonian
from someone else's design and I wanted to learn the details of design
and come up with one more to my specific tastes, at the time. This
meant a cassegrain, as my second; and then for my third 'scope it was
a Gregory/Maksutov, without complex curves that take more time to make
and test.
But aside from specific end-results, are you also interested in how
things work, too? Because if you don't have an abiding interest in
dismantling doorknobs, TVs, your car, and just about everything else
you see around you -- just to see what makes them tick -- then you may
find yourself tiring of the endeavor over the long-haul. At least, if
no one is paying you for your interest (which means 'hobby'.) You
really need reasons other than just a few end applications themselves
to stay involved.
I enjoy dismantling things. And the nice thing is that when I put
them back together again, I have all these nice, spare parts left
over, too!
Anyway, I've been reading books by Forrest Mims and also /Electronics
for Dummies/ at the book store as well as various internet sources for
additional information.
Used to be good articles found in several electronic periodicals, once
in a while. I'm not sure what modern magazines might now offer,
though. I used to follow Ciarcia's column in BYTE. But a lot of what
goes for electronics projects in magazines these days doesn't spend
the time to teach as much as once they did, I think. Maybe I'm wrong
about that.
One source for developing basic knowledge came from either Heathkit
(back in the day) or similar training kits that usually started with
DC basics, then went on to AC basics, then to ICs and general design.
Also, the training books developed for military (the ones I saw and
spent some time with were for WW II) weren't too bad, either. Lots
and lots of material to keep slogging through, if nothing else. I
suspect that there are good training systems still available -- though
at what cost, I don't know.
Some sources claim /The Art of Electronics/ is
a necessity while others say it has entirely too much information and
the wrong approach for a hobbyist.
It's both and neither. It will have way too much information in many
places and then also just the right amount elsewhere. I'd recommend
it --especially if-- you also get the student manual for it. And
maybe still, if not. But some things in AofE are a big help and will
be there later on when you are ready for it, too.
I don't want to just build pre-made
kits or copy other people schematics -- I'd really like to make new and
original things happen. I have plenty ideas, but I seem to be stuck in
the CS / software point of view when it comes to construction. I can
build an entire program in my head for the aforementioned sample rate
reducer, but I just can't see envision it in components. I assume I'd
need some sort of analog signal to digital signal converter and an IC
chip to do the real work of the problem... but I really have no idea.
Because of your software background, you may also eventually want to
consider the idea of doing projects that are "embedded" and use a
microcontroller, instead of just analog electronics. But don't
neglect analog, either. But take it slow and don't envision the
really big ideas until you've managed to completely master some fairly
basic things.
Just as in software design, electronics is often more easily handled
by breaking down the functions into tight groups of parts with a small
number of inputs and outputs. Within the function block, there may be
many intimate relationships, but between them the concepts should be
fairly simple and easily understood. You will need to master these
various function blocks a few at a time and build your repertoire. As
you do that, you will find that you can tease apart more complex
circuits by first recognizing sections and eliminating them while you
focus on the rest. To begin, most of it won't make any sense as you
won't have enough blocks to make any headway at teasing apart all the
rest. Later, you will find it much easier to tease them apart and
will then know which areas you _don't_ understand or know anything
about and you can then focus your questions to others onto those
areas. Much faster for everyone, then.
In addition to other possibilities, you might consider the idea of
getting a prototyping board with a microcontroller on it already and
with access to pins on which you can "add stuff." Sometimes, that is
a good way to get thinking about BJTs and MOSFETs, for example,
because you can use them in one of their simplest modes -- a switch.
And with the software, you can try out different ideas. A 'scope
starts to be important, about this point in time, though.
Some things you can easily learn yourself. Ohm's law is like that.
The conservation-law based ideas of Kirchoff are easy to understand
the theory of (sum of currents in and out of a node must be zero [or
else electrons would be jammed together or sucked out, over time] and
the sum of voltages around a loop must be zero [the voltage must be
the same voltage when you get back to the same place]), but it took me
a little practical work to develop a practical feel for actual cases
and to start thinking more in that mode. A detailed understanding of
BJTs will take more time -- starting with the forward biased active
mode and some gross simplifications that get you often close and then
proceeding to more refinements as you are able to assimilate them. For
example, it wasn't until later that I began to realize the cause and
effect of B-C and B-E capacitances in the BJT and some methods of
overwhelming them quickly when speed is needed. And getting deeper
understandings of various parts will probably never stop -- at least,
hopefully it won't.
I already have some preliminary equipment (solder iron, desoldering
braid, thin pliers & cutters, wire stripper, 20 AWG solid wire, digital
multimeter, etc) but no actual components or breadboard. At this point
I am sort of sold on the dead-bug style for educational / experimental
prototyping.
There are some nice proto boards for cheap, recently. They are global
specialties solderless boards and the ones I found were similar to the
PB-102 and cost me, just last month I think, about $14 each. I forget
where, but I could look it up. They are nice to use.
I also find wire-wrapping useful. So you might consider the idea of
that, as well.
An oscilloscope will be useful, if you can get one. And these days,
they are a LOT cheaper than they used to be when I was desperate for
them. They are available readily as used, too.
So does anyone have any suggestions for me at all in anything I've
mentioned? I find myself wanting an expert electronics guy around to
guide me, but I don't know anyone, so I turn to internet folk. Any
material to read, sources to look up, comments regarding /The Art of
Electronics/?
It's always better if you can find someone to help. Lacking that, you
will just have to try. There is the internet now, so that is a plus
that didn't exist when I was struggling hard in the mid-1970's. Take
advantage of it, as you are. Other than that, just get started doing
some simple things and try and master them and various ways of looking
at them before going on to much more complex things.
Jon
