Again the FAQ!

James Newton wrote:

> The safest thing to do is to tie all unused pins to either ground or
> the positive supply voltage using resistors, and leave them tri-stated
> (if they are programmable i/o).
> This way, if the pins get set to an unintended state (due to software
> bugs or electrical noise or whatever), the chip will not be damaged
> because the current will be limited by the resistor.

  *But*...  If you leave them unconnected, you certainly guarantee no
damage from unintended states.

> You could get rid of the resistors by just leaving the pins
> unconnected and set as outputs, (high or low)

  Best idea.  Allocate them pads for possible later use.  This satisfies
the design sensibilities of most of us who abhor waste!  And saves
resistors!

> but this leaves the possibility of the pins floating while the chip is
> initializing, or in a fault situation in which the pins are tristated
> unintentionally.

  Oh no!  You mean to say if the pin floats for a millisecond, the chip
will go up in smoke?  Shock!  Horror!

> It also introduces the (low but real) possibility of the pin being
> shorted to the other supply rail and causing functional failure or
> damage to the chip or supply.

  Why, but why, would this happen only to unused pins?  *Surely* it is
far more likely to happen to *used* pins?

> The important thing is to avoid having a tri-stated (input) pin
> floating,

  This is ludicrous!  So it might add a few dozen microamps to the
current draw.  Not half as much as if your suggested pull-up resistor
is accidentally pulled down by the same circumstance.

  Or do you propose this will "damage" the chip?  If so, why is it such
a popular concept, frequently advised, to use I/O pins to measure
analogue resistances or voltages?

> and to avoid having an output pin driven against an external
> connection so that it exceeds its rated source/sink capability (or the
> combined source/sink capability of the chip).

  Surely far more likely to happen to pins wired to other components?

> Pull up or pull down resistors eliminate both possibilities in all
> conditions.

  Rubbish!  If you can seriously conjecture a shorted pin, it's just as
likely to happen to a SMD resistor or SIL pack.

>  The worst than can happen is accidentally driving an output against a
> pull up/down and thus wasting power.

  33 k ohm resistors?  140 ľA?  I suspect that's rather more than the
switching current of an input held at ¸ Vcc.

  Anyone got data on that though?
-- 
  Cheers,
        Paul B.