Air bubbles
 

I have a tower 75 and gms 46. The 75 has been trouble free for 10 flights. Then last time out. It started giving me problems. It runs fine on the ground with the nose in the air and flat. But when nose down the fuel line gets air bubbles in it. Then when I bringing the nose back to horizontal it appears that the fuel flows back into the tank. This has caused several dead sticks. I have replaced all fuel lines with no luck. As for the 46 it's new and has simular problem from day one.

RE: Air bubbles
 

Holding the plane nose-down for an extended length of time is not a valid test. All the fuel flows to the front of the tank, where the clunk cannot, and should not follow. Of course, the line goes dry and the engine quits unless the tank is completely full. That's perfectly normal.

Frankly, I don't think your "nose down test" is telling you a darned thing about why the engines are quitting in the air. The fact that two completely different engines do it indicates to me that it's user-related, as in how you tune your engines, and not a mechanical problem. You are probably running the engines too lean. They're just conking out in straight and level flight, right?

RE: Air bubbles
 

Hi!
The nose down test doesn't prove anything as the clunk could not and should not follow (as Matt said).
99% of all dead stick flights is due to running the engine too lean in some way..too large prop carb too lean , wrong fuel, wrong glowplug, hole in the fuelhtubing or tank ...etc.

Regards!
Jan K
Sweden

RE: Air bubbles
 

Thanks Matt for the info. They are quiting when returning to level flight from my up lines. the engines run good at full throttle but when throttling back on the down lines the engine dies. Would this mean I need to richen the low side?

RE: Air bubbles
 

Yes try going a little richer on the low end. As these engines break-in, they have a tendency to run slightly leaner. This could be the reason your engine quits. Suggest you use the pinch test at both the high and low end to determine correct mixture. Much has been written about this method and you should be able to read plenty - just search for "pinch test", or "pinch".