This is more of a question for the GA guys.

In the pilot's operating handbook of any S-Tec autopilot (let's say the System 30 in my case), they outline a preflight check to perform with the engine running before takeoff. These checks include mode checks, disconnect checks, override checks, etc. which are common to all autopilots.

Flying Lears (or any jet for that matter) we obviously do similar checks on the autopilot at least on the first flight of the day. However, I am training someone for their instrument now on a C-172 with an S-tec autopilot. While performing the checks as outlined in the manual, I sometimes fear we're being rough on the system (even though we try to be gentle!), such as during override checks (i.e. holding pitch up/down until you get a trim up/down light). As an example, when we test the disconnect, even though we've neutralized the yoke (no stress) a "thunk" can sometimes be heard from the servo in the tail.

My question is, although the manual calls for it and it's good operating procedure..... from a practical/mechanical standpoint in terms of the system, is it still a good idea to do these checks? Although it's obvious for the bigger jets (can't get around SOPs), does the same hold for a light piston with a basic autopilot system?

Thanks

 

Not a mechanic, but I always teach my piston students to do these checks on the two axis units because you can get into trouble with bad pitch commmands. Roll-only units I do not bother with the preflight. Some of the older S-Tecs are quite unreliable, so if it has pitch control consider a preflight mandatory. And if stuff falls off the airplane from doing the check, then obviously you should taxi back and squawk it.

 

When I owned my Piper I never did those checks for just the reason you mentioned. Even if it failed in flight, I knew that the worst it could do would not cause the airplane to get away from my control of it.

 

Thanks for the replies. Cruz5350, when you say "When I owned my Piper I never did those checks for just the reason you mentioned." are you talking about the reasons I mentioned of potential damage to the system, or what Cubdriver was talking about?

 

I always tested the autopilot and trim disconnect/override but did not override the servos in any autopilot. The exception was after an annual...I would test everything I could physically think of to the limits.

 

I didn't want to damage the servos by over riding them before every flight.