[TriEmbed] TriEmbed Digest, Vol 31, Issue 5

[Brian]

On 12/2/2015 11:33 AM, Chip McClelland via TriEmbed wrote:
> Brian,  Mkaes perfect sense in steady state but could there be transient
> differences between the board's Vcc rail and the internal voltage inside
> the Arduino?  Is this something I need to worry about?

Shane also replied to this, but I thought I would expand on it a little.

Take-home message: Vout physically cannot exceed Vin on a 
microprocessor.  Not "must not" but "cannot," i.e. it is impossible due 
to conservation of energy (a law of physics).

Application:  If Vcc sags, Vout from the digital pin will sag right 
along with it.

Further Discussion:
No silicon junction has a zero "on" voltage.  Even if the micro has 
highly efficient internal drive circuitry that connects its digital 
output pin as directly to Vcc as possible for a logic "one", it's still 
flowing through at least one silicon junction and incurring a voltage 
drop across that junction.  It may be a drop on the order of a few tens 
or hundreds of millivolts, but the drop is there.  The output pin's 
potential will *always* be lower than Vcc.  No current at all will flow 
through a driven-high pin connected directly to Vcc, because the drive 
transistor won't be forward-biased.

The only way for an output pin to exceed Vcc would be through some 
catastrophic internal event involving catastrophic external conditions, 
and the pin-to-Vcc connection would be the least of your worries!

HTH,
-Brian