Bearhawk Patrol Trim

[N3UW]

There has been a lot of discussion on the trim system of the 4 place and a little bit discussion of the trim system on the patrol. People talk about how sensitive the trim is and far aft CG loading the pitch becomes a bit sensitive. There has also been discussion about how the trim becomes very sensitive at higher speeds. Most of this information has been around the 4 place.

I have a few questions for you Patrol flyers.
1) How much is the trim lever typically moved from takeoff to cruise and then cruise to landing?
2) How much trim change is needed for power changes?
3) At cruise speed how sensitive is the trim lever?


I am in have been considering building an electric trim for the patrol but I am not willing to give up the manual trim. I agree with Mark Goldberg that the aircraft would be a real handful and respect his opinion that it might be uncontrollable with a run away trim situation if you had not manual override. I prototyped up an electronic trim system that also allowed me to use the manual trim lever. I am using a Dynon autopilot servo as the donor power control unit. I am attaching this to the trim cables in the ceiling behind the baggage compartment. The Dynon servo was the choice because it uses a stepper motor and has a fairly low power off idle resistance so that I can still use the manual trim lever when the power to the servo is off. As an added advantage the idle resistance is not zero so it adds a little drag to the system so the trim will not slip when the servo is off. Also with stepper motors and good stepper motor controllers like the one I am using (TC78H670FTG) I can control both the torque and the speed from very very slow to fairly fast. I can also still keep the torque so that I can always override it manually if I have to. I ripped out the electronics and replacing it with my own circuit board. The circuit board has two trim switch inputs. One for the trim on the stick and one from an auto trim on an auto pilot. There is also a RS-232 input for GPS ground speed so I can scale the speed of the trim to the airspeed (shower at cruise and faster at slow air speeds). It also allows the safety mechanism of maximum of 3 seconds of switch activation to help catch a runaway trim situation. It worked well on the bench with proto boards and I should get my custom bare circuit board that replaces the Dynon board Friday. I will post pictures when I get the board populated.

[schu]

Interesting idea. What microcontroller is on that board you posted? Looks like it has 50 pins at least.

Anyway, my thoughts, which is worth $0.02 American, and dropping fast:

I decided to go with electric trim for a number of reasons. I get buttons on my stick that are convenient, it's much easier to have the autopilot system trim the airplane thus not working the pitch servo hard or always nagging you or handing you an out of trim airplane when you cancel, and because I don't like the servoing trim setup as I'd rather have a more stable airplane with a more cessna feel to the pitch and trim as my airplane will be flown in weather and at gross frequently.

Given my desires, I just mounted the trim servos in the elevators themselves like many other experimentals . I think it makes the trim system simple. No need for cables and bellcranks and torque tubes and bolts and joints or cables twisted or tightened to a G on the bass clef. The trim servo directly pushes a rod that's only 7-8" long attached directly to the trim tab. There is absolutely no slop or play at all in my system.

As for controlling it, I have a microcontroller that gives me some features like a go to neutral button and I have mounted an overhead trimwheel that is nothing but a rotary encoder. The idea is that the co-pilot doesn't get trim controls on the stick, they can use the wheel, it has a cool retro style, and I can detect run away trim in software and still allow trim on the wheel.

Because I don't totally trust my controller, I also have an overhead switch marked Trim: Normal/backup and Backup Trim: Nose up/down. If my controller fails, I switch to backup trim, and then I have a switch that doesn't pass go or collect $200, it applies power directly to the trim servos. The only single point of failure is the Trim Normal/Backup toggle switch, but it uses screw termination is rated for 30X the current, is a quality toggle switch, and will nearly never be cycled. The backup trim switch uses a different fuse as well.

To those that want an electric trim servo driving the manual trim, an even easier answer is Garmin. The garmin servo can be configured as a trim servo on the can bus, and with a capstan inline with the trim cable, it can drive the trim back and forth and there isn't really any engineering needed, you just make a mount for it and wire it up.

schu

[N3UW]

Interesting idea. What microcontroller is on that board you posted? Looks like it has 50 pins at least.

Anyway, my thoughts, which is worth $0.02 American, and dropping fast:

I decided to go with electric trim for a number of reasons. I get buttons on my stick that are convenient, it's much easier to have the autopilot system trim the airplane thus not working the pitch servo hard or always nagging you or handing you an out of trim airplane when you cancel, and because I don't like the servoing trim setup as I'd rather have a more stable airplane with a more cessna feel to the pitch and trim as my airplane will be flown in weather and at gross frequently.

Given my desires, I just mounted the trim servos in the elevators themselves like many other experimentals . I think it makes the trim system simple. No need for cables and bellcranks and torque tubes and bolts and joints or cables twisted or tightened to a G on the bass clef. The trim servo directly pushes a rod that's only 7-8" long attached directly to the trim tab. There is absolutely no slop or play at all in my system.

As for controlling it, I have a microcontroller that gives me some features like a go to neutral button and I have mounted an overhead trimwheel that is nothing but a rotary encoder. The idea is that the co-pilot doesn't get trim controls on the stick, they can use the wheel, it has a cool retro style, and I can detect run away trim in software and still allow trim on the wheel.

Because I don't totally trust my controller, I also have an overhead switch marked Trim: Normal/backup and Backup Trim: Nose up/down. If my controller fails, I switch to backup trim, and then I have a switch that doesn't pass go or collect $200, it applies power directly to the trim servos. The only single point of failure is the Trim Normal/Backup toggle switch, but it uses screw termination is rated for 30X the current, is a quality toggle switch, and will nearly never be cycled. The backup trim switch uses a different fuse as well.

To those that want an electric trim servo driving the manual trim, an even easier answer is Garmin. The garmin servo can be configured as a trim servo on the can bus, and with a capstan inline with the trim cable, it can drive the trim back and forth and there isn't really any engineering needed, you just make a mount for it and wire it up.

schu

Hey schu. I am using an AtMega 2560 because I have a few of them. It would not have been my first choice normally but many of the MCU I would choose are unavailable for a while because of the chip shortage. This is the first time parts supply has driven a design for me,

interesting idea on the Garmin Gsa-28 being used as a trim motor. I was not aware of that. I knew that a gsa-28 could drive an external trim motor and the G3 could be configured to adjust the speed, with PWM, of the external trim motor. DO you have to have a g3 to make it work?

[schu]

No, I’m not using a gsa-28 for trim since I wanted to git rid of the servo action, I only mention it so that people that want an electrically driven manual trim know that they can get that pretty easily if they choose garmin.

As for a micro controller, I’m using one from rugged circuits with extended temp range and building a custom “hat” for it that contains the power supply, digital pot, and motor driver. The power supply will diode OR from my aux and main bus (another thing garmin stuff does) and has the power protection circuit from the speeduino project:

https://cdn.hackaday.io/files/4413188805792/Speeduino%20v0.3.5_schematic.png

This should produce something very reliable that has the protection needed to last in an airplane. I’m including a digital pot so that my controller can relay servo position to my EFIS as the linear resistors in the servos are terminated on analog pins so that I can drive them to neutral with a button push. My code will also compare servo positions and if they aren’t very close to each other then I’ll pull a pin low on my EFIS that will bring up a trim fault TCAS message.

As for the auto pilot driving the servos, it can only drive one and I don’t want conflicts so the output from the auto pilot pitch servo will go to the trim controller which will drive the servos.

Any interest in peer review or sharing notes? I’m not an electrical engineer but I play one on TV.

schu

[Light&Sweet]

Interesting discussion. I understand Carbon Cubs use electric trim—type and reasons are unknown to me. Weight savings and simplicity, I suspect. But, from a pilot’s perspective, trim on the stick is great when in critical phases of flight a gross arm movement to locate and adjust a manual trim is awkward.