On Sun, 2007-04-01 at 20:17 +0000, Peter P. wrote:
> Herbert Graf <mailinglist3 <at> farcite.net> writes:
> 
> > True, although personally I'd still never connect the PIC output
> > directly to the relay, it's just asking for trouble.
> 
> Ok. But then why stay with 10-20 mA outputs (that and the CMOS 'B' style IO pins
> were the main attraction of using PICs vs. other micros) and not degrade to 1.4
> mA which seems to be the rule, and require external boosters for all uses. 

Because not everybody wants "higher" pin currents for connecting to
inductive loads.

20mA is VERY useful for many other uses, LEDs, driving MOSFETs,
displays, ETC.

> After
> all, they are after die shrinks. I think that moving from 20 mA (haha) IO
> drivers to 1.4 mA drivers on a 13 IO chip could halve the die area (just
> assuming). Then I'd expect 25 cent pic16f54b's.

Working for a company that makes chips, die size is by far not the only
(or often even biggest) cost for something like a PIC. Packaging is a
HUGE cost. Binning, testing and distribution are also very large costs.

> > BJTs are stupidly inexpensive, if you're already using a relay, the
> > additional cost of the BJT is barely visible.
> 
> A pic chosen correctly usually has enough spare pins to use 2 or 3 in parallel
> to drive such a small load reliably.

But in my experience driving an inductive load with an MCU is pretty
much ALWAYS a bad idea, it has nothing to do with current carrying
ability in my mind. Yes, you can do it, and often get away with it, but
it can be risky, and hasn't in my limited experience been worth the
effort.

TTYL

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist