Thorp Air Command - T18.net

To anyone that has installed a Garmin G3X autopilot (with GSA 28 servos,) I am seeking input...

A few months ago, I installed a G3X system, with the Garmin autopilot (GSA 28 servos) coupled to a Garmin GPS175 navigator. I now have about 35 hours on the airplane post-install, and use the autopilot on every flight.

The autopilot works well, mostly. The Heading / Roll / NAV portion is perfect. The pitch works fairly well, but not perfect. In altitude hold mode, the plane will, after a few seconds, begin to hunt about 25 - 50' ( + -) from the altitude set in the window. Then it might settle down at the correct altitude for a few seconds, and then start wandering again. The autopilot is not coupled to any trim servo.

I've spent quite a bit of time going through the Autopilot Tuning steps outlined by Garmin. I've bracketed the gain settings trying to narrow down what works best, and have it optimized, I think, but the hunting continues.

If you've installed one of these systems, and the altitude hold (and VS mode, IAS mode) work well for you, could you share the various gain settings for your elevator servo?

Thanks a lot,

Peter

Thanks for the reply, Bruce.

The roll servo works very well- zero problems, smooth and precise navigation. Although I've yet to do any approaches with it, I think it will work fine.

But the pitch is still hunting a bit. If I set it to vertical speed = 0, or altitude hold mode, is will gently porpoise to either side of the set altitude. It's tolerable if I'm not looking at the vsi or the altitude tapes. The pitch change is subtle. But I can hand fly it better than that, so I'm sure the autopilot is not quite right.

I've spent hours adjusting the gain settings and thought they were pretty well optimized- yet the hunting continues. After seeing a YouTube video of a Mooney with the exact same behavior (that guy replaced the servo which fixed his problem, I swapped in another servo. That didn't change things. I mounted a camera in the back to see where the crank arm was in relation to the pushrod in level cruise flight. That all looks good.

I've now reached out to G3X Expert at Garmin, and they suggest it is the vertical gain setting. I'd be very interested to learn what your pitch gain settings are. In the meantime, I'll fly more and work on it.

Thanks again,

Peter

It sounds like static lag to me. I recently helped a friend with a Bendix/ Trutrack system and the static line was either too long or it had other instruments hooked up to it. A separate line fixed it.

Regards,

Jim

Thanks Jim.

The static port / line has been in my mind after reading other's reports. But I'm not sure what I can do, other than ensure there are no leaks. I have a Piper style pitot / static mast on the port wing. That is all plumbed into the G3X ADAHRS (GSU 25.) So, straight from the static source to the GSU 25. There appear to be no issues with ADS-B reporting, etc....

When you say, "A separate line fixed it." Can you elaborate on that a bit?

Thanks again,

Peter

Peter,

Basically you want to reduce the volume of air in the static system to a minimum. Since the air is compressible it acts like a spring and the static air pressure will fluctuate a bit and the A/P sees this. This results in the oscillation you are seeing. In the case I worked on we reduced the size of the tubing and tee'd in a separate line at the static source. Maybe you could try another static source with a shorter line length.

regards,

Jim

Thank you, Jim. Your explanation makes perfect sense.

I think I will dig into the static system, as I'm exhausting the other possibilities. The gain settings definitely have an effect, but at best, I can only minimize the oscillations. I can't seem to eliminate them.

I'll report back after checking / sorting the static system, etc.

Thanks again,

Peter

Big thanks to Jim for suggesting the static line as a potential problem!

I disconnected the static line and flew for about 45 minutes (mostly in a holding pattern.) With the static line disconnected, the altitude hold was perfect!

But of course, that affects the indicated airspeed and altitude, so I'll have to figure out the static solution. Not sure if the Piper mast (which is a combination pitot and static port) might be causing the problem. Or, as Jim suggested, reducing the diameter / length (volume) of the static line might work to resolve the hunting.

I guess I can cut holes in the fuselage sides for static ports, if necessary. I'd prefer not to do that....

But in any case, I'm happy to be making progress, at last!

Thanks,

Pete