Bill Daniels wrote:

Experts, correct me where I'm wrong.

I'm not an expert, but I'll answer anyway, this is RAS, after all.

1. An operating transponder, not being interrogated, is simply a receiver
with a low current drain.
2. When interrogated by a ground radar or TCAS, a transponder transmits a
short "squitter" at 175 or 250 watts which is a BIG current drain.
3. Flying in an area with many interrogations per minute is likely to be a
congested area where the transponder is needed and a wise pilot would keep
it on despite the current draw.
4. Flying away from a congested area toward a remote area with few
interrogations, the transponder automatically cuts back on its current draw
by operating less and less as a transmitter and more as a receiver.

So, where is the need to turn it off? Doesn't the transponder effectively
manage its own current draw to match the level of congestion? Move away
from congested areas and the current draw is minimal.

You are assuming that areas of low traffic are also areas with few
interrogations, which is not necessarily true. In my case, the primary
concern is the area within 30 miles or so of Reno, where there is high
likelihood of encountering airliners in the 10 to 18K foot range.
Outside of the immediate area of Reno, my transponder still gets
interrogated, due to enroute and military radar.

I haven't read of a case where the transponder is sucking batteries flat
and if that happened, how much additional battery capacity is needed to
keep it running for the whole flight?

Our Duo came standard with dual 7ah batteries. A combination of an
LNAV, GPS-NAV, radio, and transponder sucks a single 7ah battery dry in
a little more than 4 hours. We were replacing the batteries roughly
once per year, since some of the partners (not me, of course) would
forget to either switch batteries or turn off the transponder before
draining the first battery completely. We switched to using dual 12ah
batteries, and the problem appears to have gone away.

Marc