What is independant?
For those of us who fly IFR Stein is on the money here. The VFR folks can get along just fine with a single source fully integrated system, but as long as I am flying IFR, I am going to have some kind of standalone autopilot capability.
I fly lots of "real" airplanes that don't have that luxury, but it is cost effectively available in the homebuilt world, and the Rocket would be a busy airplane to hand fly with the backup instruments.
Tailwinds,
Doug Rozendaal
Doug... I the point I was trying to make is that two separate EFIS units are not necessarily an "integrated" unit. I would look at them as two separate units, with either one having the capability to drive the AP servos. This is not a "single source" system (where source = gyro reference, not vendor).
The "which one is right" problem comes up in all of the two (not three...
...) unit discussions... you get to vote with your eyes and butt.
If you are concerned at this level, then you need three gyro reference units.
As to similar technology, once electronic units are burned in and operating, the chances of an identical failure happening at exactly the same time to both units are extremely rare. When I worked on satellite programs, the reliability folks used voting and multiple identical units, not different technologies...
This does leave open the chance of software errors, but flying IFR, you are relying on a lot of software in multiple boxes anyway...
If you are looking for errors and failure rates, I bet the much more mechanical autopilot has a much higher probability of failure than the all solid state electronic EFIS gyro units.... which is why "Autopilot Disconnect" switches were invented....
gil A