> Michael Park wrote:
> > Or else, what happens? I've always wondered. F'r instance, if it's a 5V
> > regulator
> > and the input is 5V, what's at the output?
>
> Something less than 5V.

Definitely  !:-)

How much less is a function of luck and the cleverness of the designer.
Some regulators degrade their output quite gracefully for a considerable way
below their dropout point. ie the output voltage tracks down at some level
below but near the output voltage.

Others are extremely badly behaved and not only drop their output voltage
substantially or vary it in an ill defined manner as Vin falls, but may draw
much higher input currents, thereby hastening the discharge of the battery
in battery powered equipment.

Manufacturer's whose regulators behave well under these conditions usually
point this out in the spec sheet.

Note that the terms "well" and "badly" above are arguable. It could be
argued that a device should always be used "in spec" and that its
performance outside this range is irrelevant (something like the Copenhagen
quantum mechanical view of devices :-) ).  HOWEVER, if the manufacturer does
spec a regulator outside the constant voltage range of operation then it is
legitimate to make use of this for your purposes. For example, you may have
a processor that you wish to operate on 5 volts but which is specified to
operate under your applicable conditions down to 4v2. In such a case you may
use a 5 volt regulator but want the system to fail gracefully when the
dropout point is reached. This may eg allow you to operate the cpu even when
certain external i/o is inoperable due to the low Vdd. As long as you can
formally design for this condition it is a legitimate one and a "well
behaved" regulator is a bonus.



        Russell McMahon

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.