Am 11.05.2016 um 10:05 schrieb scott:

> As clipka mentioned, a transistor used in digital circuits is either in
> a state where the current flow is zero ("off") OR the voltage drop is
> zero ("on"), so heat output is usually zero. It's switching between
> those two states, when both voltage and current are non-zero, that
> significant heat is dissipated within the transistor.

Um... no?

The fact that modern digital circuitry only draws power when switching
between states has nothing to do with the voltage drop across the
transistor; as a matter of fact, the type of transistor used in modern
digital circuitry acts pretty much like a low-ohmic resistor when
switched on, and will happily draw as much current as you happen to
allow it to, accompanied by a corresponding increase in the voltage drop
and a merry self-heating, up to the point of destruction.

The fact that modern digital circuitry draws virtually no power in
stable state is that (1) they use FET (Field Effect Transistor, aka
unipolar transistor) technology, in which the transistor's control
terminal (the "gate") is isolated from the other terminals (the "source"
and "drain"), controlling the source-drain voltage exclusively through
the presence of charge (held there by a sufficiently high
gate-source/drain voltage); as a result, the DC component of the gate
current is zero, i.e. the only time when current flows to or from the
base is when the base voltage is changed. And (2) the FETs are always
wired in such a way that in a stable state each switched-on FET connects
the power rail with nothing but other FETs' gates, or switched-off FETs
that would connect those same gates to the other power rail in the
complementary state.

(So yeah, the voltage drop across a switched-on transistor in a modern
digital circuit typically /is/ zero, but only in the same sense as the
voltage drop across a simple resistor with one terminal wired to a power
source and the other terminal floating in mid-air is also zero.)


Pre-FET-era digital circuitry was instead based on bipolar transistors,
in which a current must flow through the control terminal (called the
"base" in that case) to switch the transistor on. Those transistors did
in fact have a constant voltage drop in the "on" state, but even that
was distinctively non-zero.


(As an interesting side note, the earliest transistor prototypes were
actually FETs; however, they were not feasible to manufacture back then,
so bipolar transistors took the lead for a while.)

Post a reply to this message