Van's Air Force

The definitive Van's Aircraft support community! Buying, building or flying an RV? Join our exclusive family of mentors and enthusiasts!

G3X Touch - AP Disconnect Mystery

bjdecker

Well Known Member
Ambassador
Greetings folks,

I thought I would share my findings of a recent issue that kept me from traveling to AirVenture this year.

After overnighting in Alabama last week, and at the start of the *second* flight of the day, the Autopilot (G3X Touch - GMC 307 Mode Controller) would not engage. Pressing the "AP" button would cause the "Bee-Boo, Bee-Boo" tone and a red "X" through the AP status bar. Cycling power to the AP servos had no effect. Same result pressing the AP option on the autopilot controller screen.

I contacted Garmin to arrange for replacement servos since mine were circa 2015 and subject to the AD/SB. (Service Bulletin 23038)

I removed and inspected the original servos today, and found the following corrosion inside the connector cavity on the Roll Servo GSA 28.

Screenshot 2026-07-22 at 4.08.23 PM.png


The corrosion appears to bridge the Trim In 1 & 2 pins (Pins 11 & 12). In my system, the Roll Servo is NOT connected to the TRIM Motor In / Out; no Auto Trim for Roll.

I suspect that the GSA 28 was sensing activity on the Trim Motor inputs and disconnected; this is expected behavior if the trim switches/coolie hat is moved. Or alternatively, the CWS input was being pulled down (corrosion around Pin 15).

Here's the interesting thing -- there's no visible means for water (or any other electrolytic fluid) to get to this connector -- it is located above the outboard wing access panel, adjacent to the aileron bell crank, and it is facing UP (it's looking at the underside of the top wing skin). The other odd thing, the D-sub connector on the harness appears to be a cheap-electronics-surplus type and not a M24308/4-x-F connector (F is cadmium plated).

Something to add to the CI-checklist I guess.

Cheers...
 
This has popped up once or twice before.
I think the consensus is If you fly in rain, water migrates up from the tie down hole and through the spar web tooling hole.

I do a moderate amount of IFR so on both my roll servo and my wingtip magnetometer I have sealed Dsub shells.
IMG_6308.jpegIMG_6307.jpeg
 
This has popped up once or twice before.
I think the consensus is If you fly in rain, water migrates up from the tie down hole and through the spar web tooling hole.

I do a moderate amount of IFR so on both my roll servo and my wingtip magnetometer I have sealed Dsub shells.
View attachment 123958View attachment 123959
Interesting -- The hole is plugged with a tie down ring (Cleaveland) and nylon area washer...But I'll put eyes on it.
 
How long has that servo been in service?
Installed 2016-17. First flight July 10th, 2018. 379TT

Aircraft hangared 99.9% of the time, and I can count on one hand the number of times I've flown in/through/under precip. (including the incident flight)
 
Installed 2016-17. First flight July 10th, 2018. 379TT

Aircraft hangared 99.9% of the time, and I can count on one hand the number of times I've flown in/through/under precip. (including the incident flight)
Ever taken off on a hot humid day and climbed up to where it’s cooler? Say below the dew point where you took off?
 
Tell us about how it was painted and prepped.

Upward facing connectors are just asking for trouble. Garmin failed on this installation design. Once water gets in it just sits there for a long time. FOD susceptible too.
 
Last edited:
Tell us about how it was painted and prepped.
It was bare aluminum until painted by Evoke in 2021. Not sure of their chemical wash (methylene chloride? dunno...). But they do a really good job in removing panels, covers, then recovering/sealing up holes...
 
Interesting. I also wondered if it was condensation.

Thanks for the heads up, some periodic inspections of these electrical connectors will go on my list.
 
Upward facing connectors are just asking for trouble. Garmin failed on this installation design.

Did Garmin design the servo installation? Where does Garmin say that the servos should be mounted with connectors upwards?

I know for a fact that at least one aircraft design does not install the GSA 28 pitch or roll servos with connectors facing upwards.
 
I know it sounds counterintuitive,but clean it and pack it full of dielectric grease. The contacts will
still contact and it will keep the moisture out of the connector. There will be naysayers but I have never had this fail me.
 
Did Garmin design the servo installation? Where does Garmin say that the servos should be mounted with connectors upwards?

I know for a fact that at least one aircraft design does not install the GSA 28 pitch or roll servos with connectors facing upwards.
Yes. Servo's are installed IACW Garmin's Installation Guide(s), attached for your reference and information:

Screenshot 2026-07-22 at 7.56.52 PM.png
 
Greetings folks,

I thought I would share my findings of a recent issue that kept me from traveling to AirVenture this year.

After overnighting in Alabama last week, and at the start of the *second* flight of the day, the Autopilot (G3X Touch - GMC 307 Mode Controller) would not engage. Pressing the "AP" button would cause the "Bee-Boo, Bee-Boo" tone and a red "X" through the AP status bar. Cycling power to the AP servos had no effect. Same result pressing the AP option on the autopilot controller screen.

I contacted Garmin to arrange for replacement servos since mine were circa 2015 and subject to the AD/SB. (Service Bulletin 23038)

I removed and inspected the original servos today, and found the following corrosion inside the connector cavity on the Roll Servo GSA 28.

View attachment 123949


The corrosion appears to bridge the Trim In 1 & 2 pins (Pins 11 & 12). In my system, the Roll Servo is NOT connected to the TRIM Motor In / Out; no Auto Trim for Roll.

I suspect that the GSA 28 was sensing activity on the Trim Motor inputs and disconnected; this is expected behavior if the trim switches/coolie hat is moved. Or alternatively, the CWS input was being pulled down (corrosion around Pin 15).

Here's the interesting thing -- there's no visible means for water (or any other electrolytic fluid) to get to this connector -- it is located above the outboard wing access panel, adjacent to the aileron bell crank, and it is facing UP (it's looking at the underside of the top wing skin). The other odd thing, the D-sub connector on the harness appears to be a cheap-electronics-surplus type and not a M24308/4-x-F connector (F is cadmium plated).

Something to add to the CI-checklist I guess.

Cheers...
Why would that prevent you from going to OSH???
 
Risk v Reward. I’ve been there / done that, and the thought of a 10+ round trip with an aircraft system on the blink didn’t make sense to me.
Agree completely. Also been there, done that. People who don’t routinely travel long distances in RVs underestimate the importance of a functioning autopilot, in my experience. If you plan to cover a lot of ground, it’s an important part of the equation.
 
I get it’s more fatiguing. But I’ve flown coast to coast in an RV twice with no AP. It just seems like it’s not a show stopper. But I get everyone has their personal limits. All good.
 
I get it’s more fatiguing. But I’ve flown coast to coast in an RV twice with no AP. It just seems like it’s not a show stopper. But I get everyone has their personal limits. All good.
Yup - I have a low (no) tolerance for suboptimal/off-nominal operation. e.g. Landing light inop ? Grounded! CHT’s more than 20°F apart? Grounded! Fingerprints visible on the leading edge of the wing? Grounded!

It’s tough to be me…
 
I get it’s more fatiguing. But I’ve flown coast to coast in an RV twice with no AP. It just seems like it’s not a show stopper. But I get everyone has their personal limits. All good.
I am impressed!

I’ve found that copying a huge amended clearance while popping in and out of building cumulus over JFK, sans autopilot, is more than I want to be doing. Doable, but more than I want to be doing. 🤣
 
Forget not having an autopilot or personal limits, I don’t see how anyone could question not continuing a long cross country flight with a new anomaly of a system that is connected to the primary flight controls. What if it was a bearing that was corroded and seizing instead of a connector. It’s not just the autopilot disconnecting. It’s why it did.
 
Last edited:
I am impressed!

I’ve found that copying a huge amended clearance while popping in and out of building cumulus over JFK, sans autopilot, is more than I want to be doing. Doable, but more than I want to be doing. 🤣
Oh, if I were flying on an IFR clearance in the Wx - NFW w/o Autopilot. But I went VFR all the way on a nice day.
 
Back
Top