"Michael" wrote:
If it does this, something's wrong. It's not a difficult control task.

I don't doubt it - but every installation I've seen does it. I know
full well it's not a difficult control task, but I also know it's not
the same control task for all airplanes. The more roll stability the
plane has, the easier the task.

Nevertheless, the S-Tec autopilot is not much different from an industrial
PLC. Such devices --even cheap ones-- can handle all sorts of PV upsets with
rapid overshoot damping. The S-Tec should do as well.

Admittedly, my airplane is quite stable in roll, but really, so is a Mooney.
And if things are a bit bumpy, the Mooney might even be better, due to higher
wing loading.

A friend of mine who flew 727's for a while says that the autopilots on
those were the same way - they never used nav track enroute because
twiddling the heading bug (or even the roll trim) worked better.

Well, it depends on what one is tracking. Attempting to track a wavering CDI
off a distant VOR will certainly produce oscillations. Tracking a CDI driven
by a GPS is quite another matter.
--
Dan
C-172RG at BFM