RE: [linux-audio-dev] OT: Electronic advice for PC.

New Message Reply About this list Date view Thread view Subject view Author view Other groups

Subject: RE: [linux-audio-dev] OT: Electronic advice for PC.
From: Bob Colwell (bob.colwell_AT_attbi.com)
Date: Thu May 02 2002 - 02:44:23 EEST


There are a few other things to worry about, and I hope that when you add
them
all up you conclude that this isn't a good area for a novice to design in.

U/L labs approval...if you plug something into the wall, your insurance
carrier
should insist on it having U/L approval. There are a lot of really
interesting
specs that U/L has conjured up over many years of evaluating the safety of
various designs, and they will apply ALL of them to you. Give the process
from
6 weeks (a rush job for which you pay big $$) to 6 months. U/L cares about
what
kind of batteries you're using, whether they're rechargeable or not, and if
they are, the actual schematics and code running any microcontrollers inside
the recharge unit. For Li+ batts you need a pretty capable recharge unit to
stay out of danger (overcharging being one of the dangers). Many people turn
to
wall-warts because those can be independently U/L evaluated and approved,
and
then it may look as though your entire product has been U/L approved even
though
it actually wasn't. (Don't believe me? Check out the back of almost any
recent
rechargeable device and notice there's no U/L on it, only on the wall-wart.
Yes, U/L has noticed this end-run and are not happy about it.)

Also, current and voltage are not the only things you must get right.
Computers
contain some digital devices that draw drastically different amounts of
current
depending on what they're doing, microprocessors being among the worse
offenders.
You must make sure that not only does your supply provide the peak current
needed, it must also respond fast enough to current changes and the
resulting
dI/dT induced voltage swings must be restrained by appropriate decoupling.
If
you get this part wrong, you probably won't notice it until you've shipped
enough machines that somebody out there ran some software that drove the CPU
through a big enough current swing to droop your supply and cause a
malfunction.
It takes major engineering talent to identify and resolve this if it
happens.

The best advice you got so far is to dissect a laptop and do what they do.
Even
better, hire yourself a consultant who knows this area, and she can keep you
out of trouble.

-BobC


New Message Reply About this list Date view Thread view Subject view Author view Other groups

This archive was generated by hypermail 2b28 : Thu May 02 2002 - 02:43:01 EEST