Nathan House wrote:
> From what I've read, it doesn't look like it's mandatory to put a
> resistor in series with digital/analog inputs, since all of the I/O
> pins have a high input impedance.

Right, sortof.  The inputs are high impedence.  But that means they are mor=
e
able to tolerate a resistor in series than if they were low impedence.

There is one additional wrinkle with analog inputs though.  All input pins
have some leakage current.  This leakage current times any series resistanc=
e
causes a voltage difference accross that resistance, which is a difference
from the actual analog voltage to what the PIC sees.  For digital inputs,
this only matters if the error got far enough to get out of the guaranteed =
1
or 0 region.  Let's say the pin leakage current is guaranteed not to exceed
1uA.  That means a 100K resistor can cause up to 100mV error.  That's
generally not going to be a problem for a digital input since you only want
to know whether the voltage is representing a high or low.  However, when
you are trying to measure the real voltage, 100mV error is a lot.  If the
PIC is running on 5V, then that's one part in 50, or not even 6 bits
effective A/D accuracy.

This is why there is a minimum source impedence spec for driving PIC analog
inputs.  For most older PIC 16 with 10 bit A/D, this is 10K Ohms.  Some
newer and PICs with higher speed and higher resolution A/D have
significantly lower input impedence specs.

> But is it a good idea to do it anyway? When an I/O pin is
> configured as an output it's not high
> impedance (right?), so a current limiting resistor could theoretically
> save the PIC if, for instance, I happened to connect an output pin in
> a high state directly to ground.
>
> I'm just trying to make my projects as idiot-proof as possible :-)

It is easy to go overboard with this.  In environments with high noise and
spikes, like in a car, you have to think about this carefully.  Most of the
time you'll cause more trouble than you solve trying to make things idiot
proof.  Idiots can be very clever.

I would not add any series resistors on the generic board.  Add them as
needed in particular applications.  In reality, PICs and most other digital
logic can tolerate a output tied to the opposite rail without getting
damaged, at least for a few seconds.  Of course you should never rely on
this in a design, but the occasional accidental short of a output pin while
messing with a one off personl project isn't going to be a problem the vast
majority of the time.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.