Garmin G3X autopilot with GSA28 servos

During the test flight after installation, I have a couple of observations and questions.

The vertical modes and altitude hold seemed to work fine. When you've selected the vertical mode/altitude on, is the autopilot also turning on a lateral mode as well?

The roll mode was a different story. The autopilot followed the heading bug, but a couple of times in light turbulence, the stick whipped back and forth after rolling out on the selected heading, left to right about 1/2 a dozen times before settling down. I assume this is a function of the gain setting, but which way should I go? Reduce the gain or increase the gain? The initial setting was .50

Thanks for your response,

Mike

Mike, you didn't say what type aircraft, but mine is an RV-6; I used a Roll Servo gain of 0.50, same as you, set the Max Torque value to 25%. So far I have not seen anything that would want me to change these settings. Actually, the autopilot has been performing flawlessly.

As for your other question, the initial (default) lateral mode is Roll Hold (annunciated as ROL), until one of the other lateral modes is selected. What I do is, before takeoff, I'll select HDG mode and set runway heading on the bug. For a pitch mode, I select IAS and bug in a cruise/climb speed of 120 kts (personal preference). After takeoff, I'll engage the autopilot and change the lateral mode to NAV if necessary for enroute navigation. I'll also set the initial climb altitude in the ALT window and set the minimums for the approach under the BARO softkey if necessary.

Pat,

This setup is in an RV-7A.

How does yours react to light turbulence in heading or nav mode?

Thanks for response.

Mike

Mike, check your Max Torque value on the roll servo. You might consider setting it at 25%--I have not had any of the stick overshoots that you describe even in moderate turbulence.

On page 9-41 of the G3X install manual (Rev. K), they show roll servo max torque as an in-lbs value. Mine is in percent. I got my settings from Steve at Garmin who said they were from an RV-8 that had had success with these values. Let us know how this works out for you.

Tuning the roll axis.

Hello Mike,

Congratulations on getting your autopilot installed and flying. We are sure you will enjoy flying with it.

Most RV-6/7/8/9 aircraft have been able to use the default gains and have reported that they can't see any need to change anything because these gains work so well, so you might do a quick double-check of your roll servo installation just to make sure nothing is wrong. It is probably fine, but worth checking.

Section 9.15.1 of the G3X Installation Manual explains the process for tuning the autopilot roll servo in flight at a safe altitude and speed and you can get some tips there.

Please try turning the roll servo gain down to 0.4 or 0.3 and see if that doesn't make it stable. If it feels lazy, then increase it slightly. It seems the 0.5 gain is just aggressive enough that it is on the verge of being overly active in certain conditions.

We agree with Pat that you should find that 25% roll servo torque and 50% pitch servo torque should work well.

In my plane I find that I can halve the standard roll servo gain that I use and it still performs well, but is a little "lazy".

It starts behaving in a manner similar to what you describe when I increase the gain to something around 1.6 times the normal setting.

We think you will find that tuning the gain to your liking is quite easy.

Please call or send an email if we can help further.

Thanks,
Steve

Pat and Steve,

Thanks for your responses. I'll give them a shot.

My short exposure to the autopilot control through the GDU has me focused on getting the GMC305 installed as soon as possible. The GMC305 interface is much more natural for me as it mirrors many of the flight director panels that I'm familiar with.

Thanks,

Mike