Re: How do you design a circuit?

Well, yeah, that's it, more or less. But it rather trivializes the whole
process. One thing you can't trivialize with that concept is how a limited
number of people in the world ( << 1%?) make a reasonable amount of money
doing, according to you, apparently random, combinational meddling and not
real work.

Heh, and besides, random variation completely obviates the mathematical
influence some solutions have. Do you try every resistor in your box, from
1 ohm to 1Mohm, before you decide it is well placed, and which value works
best? Didn't think so. ;-)

Tim

--
Deep Fryer: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

"Kevin Aylward" <see_website@xxxxxxxxxxxxx> wrote in message
news:zCagg.263321$tc.14644@xxxxxxxxxxxxxxxxxxxxxxxxxxxx
ku7485 wrote:
This might be a dumb question to some. I've been reading a lot about
electronics, but never find a good piece of text that properly tells
about how to design a PRACTICAL circuit. I've seen some basic examples
such as connecting a LED, resistor, and a battery and using Ohm's Law
to determine to right resistance, but that's about it. I never have
heard why an engineer explain how they built their own circuits and
why they chose these parts. We'll anyway, I'm just wondering how
people are able to design.

All circuit design, as everything else, can be resolved to the Darwinian
Algorithm. That is, a process that utilises, Replication, Random
Variation, and Selection.

The basic starting point for any new circuit is to *copy* an existing
circuit. This is than varied. These variations are than selectedfrom to
see if it meets desired goals. If it does not, a new variation is made
from the copy and more selection done.

Most variations of a particular circuit are not actually random, but cross
over (genes) from other circuits, given the notional view that it is a new
circuit. However, it isn't new as this *new* circuits is actually
*derived* from existing circuits. The only way to obtain a truly new,
original circuit, i.e one not derived from existing circuits is to
generate a variation randomly. That is something that is random, is by
definition, not derivable from an existing state.

I have been designing for er.. some while now, and it took me some while
to actually understand what it was that I did. Rather disappointing I made
add on being enlightened:-)

Kevin Aylward B.Sc.
431infoEXTRACT@xxxxxxxxxxxxx
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

"There are none more ignorant and useless,than they that seek answers
on their knees, with their eyes closed"




.

Quantcast