I recently had a student that I sent to the FSDO for a private pilot checkride with an actual fed (he wanted to save the $300 fee). During the emergeny engine out check, the examiner killed the engine using the mixture (they were at about 3000' right above an airport). Upon reaching 1500' AGL the examiner pushed the mixture back in and used the throttle to simulate the engine out the rest of the way. I discussed this with the fed and he told me he has seen way too many unsuccessful actual engine out forced landings, and it could be possibly attributed to the fact that noone ever practices gliding with the engine actually not producing any power. Some airplanes idle pretty high, and that can extend the perceived glide distance. And yes, it is legal to simulate an engine out with the mixture to a certain altitude (1500' AGL I believe).

My philosophy on this subject has changed. I used to never think about killing the engine in flight as a part of training. However, if done in the correct situation (5000' AGL right above an airport with a long runway, mixture pushed back in at a safe altitude like 2500', not busy airport) I think it is a beneficial demonstration. My Grumman does glide a little (not much, but noticeable) differently with the prop windmilling on a dead engine rather than the throttle just being pulled. In my opinion this is still safer than killing an engine on a multi in flight for training. I had a friend that instructed at ATP who lost his life conducting multi training:

http://www.ntsb.gov/ntsb/brief.asp?e...16X00876&key=1