The cleanest roll servo install I’ve seen has a servo under the seat attached to the seat pedestal using a push rod to an arm clamped onto the torque tube. A longer arm would provide more power to be applied but the servo travel would dictate that. The popular pitch servo install is capstan type to an elevator cable in the aft section. Frankly I think that’s better than my install. More authority is available there than my push rod with a relatively short arm. Mine works for me but in turbulence I’d need to hand fly the pitch, and that sorta defeats some of the potential.
Anyone have autopilot in their cub?
- Thread starter don d
- Start date
frequent_flyer
Registered User
Arizona, USA
That sounds like the standard CubCrafters installation for the Garmin GSA 28 smart servos. Mine worked fine until the pitch bridle cable clamps slipped and the bridle cable wrapped in a knot around the pitch servo capstan. Installed a new bridle cable and better quality clamps and has been ok since. Ref SK SI004.
https://cubcrafters.com/c/wp-content/uploads/2020/09/SK-SI004-Rev-NC.pdf
With as delivered gains this AP works fine in still air. Give it a bit of turbulence and it grossly over-controls like a demented student pilot. Sure it holds altitude and heading, or NAV and Glide Path, but I can't tolerate that stick movement and turn it off. One day I'll try to optimize it for turbulence.
Farmboy
MEMBER
Middlebury, VT
There are a number of revised gains for the install in the CC, F/EX, /2 and /3. Fine tuning makes a difference.
The servos under the pedestal seem to work well although can get crowded if you have Amphibs installed as well.
Just make sure the pitch servo is aligned.
Transmitted from my FlightPhone on fingers… [emoji849]
The servos under the pedestal seem to work well although can get crowded if you have Amphibs installed as well.
Just make sure the pitch servo is aligned.
Transmitted from my FlightPhone on fingers… [emoji849]
Attachments
frequent_flyer
Registered User
Arizona, USA
That's horrible!
BritishCubBloke
SPONSOR
Bellingen, NSW, Australia
Is there a certified AP for the PA18-150 yet?
Fortysix12
Registered User
RanchAero Grand Vista, Brooksville
Check the older S-TEC's STC's
farm_boy252
Registered User
Norman wells, NT
That seems like simplest was to do it. Does it feel any heavier on the controls when flying by hand/autopilot disengaged?
frequent_flyer
Registered User
Arizona, USA
The Garmin GSA 28 servos have essentially zero resistance to being back driven when not engaged. I suppose any linkage will have some resistance but I doubt it is detectable.
One thing you should check is the worst case force required to overpower the engaged servos. I found that the force was unacceptably high with controls at full deflection but easily overpowered with stick close to center. I thought that was unsafe and reduced the maximum servo torque. Worse case scenario is an AP hardover that has driven controls to the stops and cannot be disconnected. Must be able to overpower the servos.
wireweinie
FOUNDER
Palmer, AK
This is auto pilot 101. You MUST be able to overcome an engaged servo. As frequent_flyer brought up, if an uncontrolled action happens (any movement of a control surface that you did not tell the auto pilot to do) the pilot has to be able to push the controls against the servo. Even if you are able to disengage the servos later, that first moment may require you to move the controls to keep from kissing the ground.
Web
Web
Utah 12 Driver
SPONSOR
Jax Beach, Florida
Experimental Cubs may want to take a look at Levil Autopilot, for a total new approach. Carbon-fiber trim tabs are added to the elevator and aileron, and then can be driven thru Foreflight program, $3000 for the entire system. I would love to get some feedback from the SC group.
https://shop.levil.com
https://shop.levil.com
It’s hard for me (Garmin autopilot owner) to imagine not being able to overpower the servos, and my plane has heavy controls so the servos are set accordingly. The gain is easy to adjust and test on the ground, and in the event that it’s uncomfortable, reduce it. The AP disconnect switch should be in an easy to reach place regardless.
Fortysix12
Registered User
RanchAero Grand Vista, Brooksville
Couple of notes about my experience with installing Experimental super cub servos. I had visited Airtronics, Billings MT, and observed the professional installation of a roll servo. One point that was made to me is that as a general rule is that the servo mounts are heavy duty ready for the maximum torque of the servo and hold it rock solid. This one had supports for all directions and was fabricated from .090 if not larger. The Dynon Servo is a stepper motor type and expect to feel the magnetic indexing when controlling the stick. At first it was concerning. Now I don't even think about it. The servos will send a signal to the D10A annunciating when they have been overpowered by human or flight conditions. Because I was the second Beta project for the Dynon system we started out with the smallest servos and progressively installed larger and larger servos. Even in the lightest of turbulence the small servos would light. FedEx Lou was the first Super Cub experimental Dynon install as I recall and he provided valuable installation information way back when.
The servos don’t need to be any more powerful than necessary to control the airplane. If you can operate the controls, overpowering the servos should require only marginally higher effort, and only if the AP has gone wonky, which is unlikely.
How the servos connect to the system should be considered. If I’m being honest I don’t think my installation is the best for pitch. My servo drives a rod to the bottom of the stick. To achieve full range we made a longer drive arm for the servo. Leverage puts the servo at a disadvantage. I don’t think the pilot overpowering it is any concern. I’m not sure the elevator wouldn’t overpower it in rough air, but haven’t found out in actual flight. I’d have to be out of trim to require much force so not a big concern. For a wing leveler on a normal day it works fine. Anyone else using a rod for the pitch servo? Where is it attached?
How the servos connect to the system should be considered. If I’m being honest I don’t think my installation is the best for pitch. My servo drives a rod to the bottom of the stick. To achieve full range we made a longer drive arm for the servo. Leverage puts the servo at a disadvantage. I don’t think the pilot overpowering it is any concern. I’m not sure the elevator wouldn’t overpower it in rough air, but haven’t found out in actual flight. I’d have to be out of trim to require much force so not a big concern. For a wing leveler on a normal day it works fine. Anyone else using a rod for the pitch servo? Where is it attached?
Attachments
Last edited:
Bearhawk Builder
FRIEND
In the woods
kestrel
PATRON
Southern NH
I can't comment on any experience with this system. I love the concept, but I'm not a fan of the implementation. It is a VFR only system run by a tablet instead of a wired in, "real system". The tabs look like they want to flutter, though anyone can choose to install better tabs. If it hooked up to a real EFIS instead of Forflight, I'd be interested.
Steve Pierce
BENEFACTOR
Graham, TX
What about Notice Number NOTC2741 on the Garmin GFC 500 telling certified users to pull the breaker until Garmin comes up a with a fix.
| Garmin GFC 500 Autopilot Notice Notice Number: NOTC2741 Garmin has identified an issue with GFC 500 autopilots with optional auto trim installed in certified, experimental and Light-Sport Aircraft, including certified installations installed with optional GSA 28 pitch trim in accordance with Garmin STC SA01866W. This issue may result in an uncommanded automatic trim runaway when the autopilot is first engaged. This condition could result in a sudden and significant deviation from the intended flight path. Garmin Service Alert 22109 Revision A requires operators of certified installations to pull and collar the autopilot circuit breaker and placard the autopilot as “inoperative” prior to further flight. Operators may contact a Garmin dealer regarding reactivation of the GFC 500 autopilot with the optional pitch trim configured off in accordance with recommended service bulletin 22110 Revision A. Garmin is working to fix this issue for certified installations in a future software update expected in Q4 2022. Garmin Service Bulletin 22112 Revision A requires operators of Experimental and Light-Sport Aircraft to update system software or disable the autopilot or trim control prior to further flight. Service Alert 22109 Rev A is available at: https://support.garmin.com/en-US/aviation/faq/mZwVwMogJ09ZLIPJgWTBJA The recommended Service Bulletin 22110 Rev A is available at: https://support.garmin.com/en-US/aviation/faq/s7SeFk9oh96I8qVfjtNbV8 For non-certified aircraft, Service Bulletin 22112 Rev A is available at: https://support.garmin.com/en-US/aviation/faq/QFnVz92cFZ5AvovG4LF6l5 If you have any questions or comments, contact the Wichita ACO Branch at: Phil Petty 1801 Airport Rd. Wichita, KS 67209 (316) 946-4139 |
Attachments
frequent_flyer
Registered User
Arizona, USA
Only applies if optional auto trim is installed. I don't know how common that option is but it's not used by CubCrafters.
With the basic pitch and roll servos the Garmin system gives a "trim up" or "trim down" alert when the pitch servo is holding torque against an out of trim condition. The proper response is often a change of thrust rather than a change of trim.
frequent_flyer
Registered User
Arizona, USA
Thanks for the alert notice though. It told me that experimental G3X Touch firmware 9.12 is available. it not only fixes the auto trim problem but includes several other useful system updates.
sub3
Registered User
KANE
No dice on the full travel. it’s too bad would have been a great location, concealed and easy to access inspect/adjust. Im going to try making a longer arm but im not optimistic, it’s a long swing to cover the elevator travel. maybe someone else smarter than me has a hack to make the arm work vs the pulley?
Gordon Misch
MEMBER
Toledo, Wa (KTDO)
Might run out of available torque with the long arm, even if adequate travel can be achieved ????
It would limit travel available overall though right? Hard stops on the servo
