Well said.

My TruTrac is pseudo "instant on" in that, within a second or two, it
shows the turn even when switched on in the turn. It is "rate based"
whatever that means. When a gyro instrument is turned on in a turn it
will erect itself to the vertical g-loading of the aircraft unless it
has some fancy way of telling up from the load experienced. This I know
from partial panel practice whereby the front seat pilot would perform
aerobatic maneuvers while the back seat pilot, under the instrument
hood, would hold the attitude indicator caged. The AI would show level
flight regardless of attitude until the cage knob was released and, when
the aircraft was recovered from its unusual attitude, the AI would show
some strange attitude which was, of course, incorrect.

It would be very interesting to try one of these new digital gyros and
see how it erects in an unusual attitude.

On 4/20/2015 1:51 PM, wrote:
A few more thoughts after reading the other related thread:

* Have you every watched how the magnetic compass behaves in a turn? Even with a bank angle as shallow as 30 degrees, it's hopeless, especially at low airspeed. Random freeze-ups and reversals in rotation are par for the course. Nonetheless there are specific ways to use it in cloud-- by establishing a southerly heading before entering cloud, or by using another instrument to level the wings before attempting to "capture" the southerly heading via a very shallow turn. Again, these comments are for mid-to-high latitudes in the northern hemisphere.

* Is there really such a thing as an instant-on artificial horizon? Is it instant-on even in a spiral dive? Which models are these? The turn rate indicator I linked to above has an initiation time of less than 5 seconds and is not affected at all by the aircraft's orientation and flight path.

* Re the dirty vs clean question-- I'd vote strongly for "dirty". Anything that creates drag tends to dampen the pitch phugoid. Sudden changes in bank angle-- due to turbulence or wrong pilot control inputs-- pump energy into the pitch phugoid. This is a huge problem when flying "partial panel" in an aircraft with relaxed pitch stability. Drag is your friend even if the G-load at which the airframe breaks is reduced by several G's. If you are pulling more than 3 G's or so in cloud, you are unlikely to be getting the aircraft back under control anyway-- especially if it is clean.

S

--
Dan Marotta