What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Runaway Trim

Thanks Dan for the data.
I was trying to indirectly indicate to those building that shortening the stick does have some consequences. The RV stick forces are very light in trimmed flight so a shorter stick seems like a good idea. Like everything, when cutting down the stick length there needs to be a conscious decision on what you maybe trading off. One possible thing traded away is landing with runaway trim.
 
Good data, best advice

Reading through this on a cold morning while drinking coffee and learning my daily lessons from folks that are way more knowledgable than I am about airplanes.... The one take away for me on most safety items is what Dan H just stated. Go fly your plane, try it on your plane, and know what it does so you know what to expect. Thank you to the OP for bringing this concerning issue back up for discussion so that we all can learn our planes better, even the ones who have built their own time machines and have been flying them for years and years. You can always learn something new.

Also, congratulations on the way you handled the emergency.

Happy Holidays everyone. Fly Safe.
 
Like everything, when cutting down the stick length there needs to be a conscious decision on what you maybe trading off. One possible thing traded away is landing with runaway trim.

Change in required force is proportional to change in length. I've attached a quick cheat sheet for those contemplating shortness ;)

Go fly your plane, try it on your plane, and know what it does so you know what to expect.

Yes, running the exercise is eye-opening. Although the forces are manageable in controlled test, the surprise factor would be huge given an autopilot disconnect. Shoot a landing or two also, as the unusual stick force is distracting. My thanks to Gary for prodding me to go do it...and Mr. Weyant, well done sir.
-
 

Attachments

  • Stick Force vs Length.jpg
    Stick Force vs Length.jpg
    22.3 KB · Views: 193
Thanks Dan. Interesting experience huh? Not sure how any individual six/seven, ten or whatever with whatever various trim systems as installed would compare to your airplane, but at least we have one good data point.

From memory and as I imagine it now looking back, the 26 pounds pull is in line with what I experienced. For those who haven’t done it, go grab a Nearly full five gallon gas can and pull it to your waist for comparison. Dealing with that kind of stick pressure is not trivial. I was not too comfortable subjecting my control system to that pressure.

With the way I have my trim servo biased, and the addition of the TCW controller (at low speed and short duration activation), I am comfortable with the measures I have taken to mitigate this risk.

My point was that I hope others have given this some thought and are comfortable with whatever measures they have taken. The consideration is warranted

Happy Holidays and fly safe.
 
Last edited:
I've seen posts in this thread and others about having circuit breakers on the trim servos so that you can pull them in the event such as this. I'm actually at the point in my panel layout were I'm locating some of the breakers and in fact on this iteration I have breakers for the pitch and roll trim to the GAD27 power inputs. My question is how does this solve the problem if the servo has run to an extreme limit?

It depends what is making the trim run. There have been several posts in this thread about a bad hat switch causing the motor to run. A kill switch on the panel would let you disconnect all functionality on the grips, letting you run the trim back to the right position using the panel-mounted switch. Or, if you had a kill switch for each grip, one grip could undo the damage caused by the disabled grip.

If it is a wiring problem, motor problem, or the panel switch/relay fails, then yes, pulling the breaker wouldn't help much once it was at full stop, except it would make things stop shorting, if that is the cause of this all. Wires shorting, or a motor running non-stop and overheating, might end up causing other really bad things, so the ability to stop that is invaluable.

Either way, I've drawn two conclusions from this thread --
1, the original poster did a fantastic job getting down in IMC with control problems.
2, I will be mounting a kill switch for my grips (Kevin at Tosten said a single switch on the common would do it all) when I get to that stage.
 
I’m not flying an airplane with electric trim just yet but my question for the wise is....if you’ve had runaway trim and it’s a full deflection and you pull the CB or somehow kill power to the servo, how do you get trim back to neutral if there is no manual way to do it in an RV?

Please educate me.

Thanks
Keith
 
You could have a reversing switch with its own power source.

Also, if using the Safety Trim, you will not get to full deflection unless you repeatedly try to run the trim. It limits the trim speed in cruise, and the length of time the now slower trim is moving.
 
Last edited:
You could have a reversing switch with its own power source.

Also, if using the Safety Trim, you will not get to full deflection unless you repeatedly try to run the trim. It limits the trim speed in cruise, and the length of time the now slower trim is moving.

Not necessarily. The VPX can be set to only allow a run time of three seconds then it will cut off. Due to the nature of my failure, a short in a wire, that was intermittently shorting it managed to run all the way to the stop.

The VPX did eventually, correctly sense an issue and disable the trim. However, and this is what I find most concerning, the trim was mechanically stuck at full deflection.

No amount of switches, or other “interesting” schemes that folks posted on this thread would have prevented this situation.

-Dan
 
I’ve been following this thread for a few days, and the thing I haven’t yet seen mentioned is “EAA Flight test Manual, Test Card 18 - Runaway Electric Pitch Trim”.

Yup - if you have thoroughly tested your airplane, then this is the kind of thing that you should have tested. Granted, the FTM is only a few years old now, and many folks did their phase one in the years before that. But....many probably never considered that this is a test that should be performed.

Remember - Phase 1 is not just “flying off 40 hours” - it is testing the airplane - including failure and off-nominal conditions. And there’s nothing that says you can’t do this test on your next flight - regardless of what phase you’re in!

BTW - I have flown all of our RV’s at full trim extremes, and they all survived. All flyable - just not terribly comfortably at all edges of the envelope. Doing it unexpectedly on an IFR flight plan - good job OP!

Paul
 
I’m not flying an airplane with electric trim just yet but my question for the wise is....if you’ve had runaway trim and it’s a full deflection and you pull the CB or somehow kill power to the servo, how do you get trim back to neutral if there is no manual way to do it in an RV?

Please educate me.

Thanks
Keith

See post #39 in this thread.
 
Keith - read all the post by Paul Dye aka "Ironflight" he seems to "think" differently than most on this forum and uses objective logic to explain things with unique clarity.

When these servos are commanded to move, they will move to the position and hold that position. If an external force pushes against the servo while the servo is holding a position, the servo will resist from moving out of that position. The maximum amount of force the servo can exert is called the torque rating of the servo. Servos will not hold their position forever though; the position pulse must be repeated to instruct the servo to stay in position. So if the servo doesn't receive the signal then the pressure on the surface is designed to overcome the torque of the "dead" servo. In other words - the servo is designed to return to a position of "least resistance" when power is removed. If anyone feels the need to dispute this information - walk outside, look up and argue with clouds in the sky.

https://www.jameco.com/Jameco/workshop/Howitworks/how-servo-motors-work.html

I’m sorry, but I need to jump in here - the Ray Allen trim servos that we use (we’re not talking autopilot or model airplane servos) are basically DC motors and gear boxes. Calling them a servo is being kind of generous.... You reverse their direction by reversing the polarity of the input power to the motor. The gearbox is such that you pretty much can’t back drive them if there is no power on the motor - I’ve worked with them quite a bit, and I certainly can’t push the output end to back drive them, and I have one airplane with a pretty large trim tab, and I have flown that with the tab locked in the full “up” position by the motor, and there was no way the airflow was going to push it back to neutral.

Building or incorporating a reversing circuit will solve some stuck trim tab problems - but not all, so it really pays to test the airplane with the tabs stuck full deflection (as has been discussed earlier in this thread) - because you CAN get there. And RV’s are all flyable in that condition.

Paul
 
Keith - read all the post by Paul Dye aka "Ironflight" he seems to "think" differently than most on this forum and uses objective logic to explain things with unique clarity.

When these servos are commanded to move, they will move to the position and hold that position. If an external force pushes against the servo while the servo is holding a position, the servo will resist from moving out of that position. The maximum amount of force the servo can exert is called the torque rating of the servo. Servos will not hold their position forever though; the position pulse must be repeated to instruct the servo to stay in position. So if the servo doesn't receive the signal then the pressure on the surface is designed to overcome the torque of the "dead" servo. In other words - the servo is designed to return to a position of "least resistance" when power is removed. If anyone feels the need to dispute this information - walk outside, look up and argue with clouds in the sky.

https://www.jameco.com/Jameco/workshop/Howitworks/how-servo-motors-work.html

You are wrong. That is not at all how they operate.
 
Dear Mr Cloud::D
Ray Allen trim motors are small jackscrews, not servos.
AP servos are real servos.
 
I'm as confused about this as when I asked the question in my earlier post. I think I know the answer but without actually trying it as a test I can't prove it to myself based upon this thread. There doesn't appear to be consensus on what the cause and effect is on the trim servos or the actual autopilot servos.

While I have circuit breakers on my panel for trim servos. I think this will do nothing to fix the problem of a trim servo which has run to its limit. A reversing switch has been proposed as a possible solution but I don't plan on adding that complication to my system since it may actually do nothing as well.

The autopilot servos appear to operate differently and removing power to them may allow them to be overpowered or they may even return to a neutral position with the loss of power.

Certainly someone actually knows how these things work on this thread and isn't just guessing. For something like this I think the folks who are guessing should limit their comments.
 
Ray,
As Paul said, above, all RVs are controllable by a pilot of average strength even with pitch trim at a full limit. The forces are lessened if you slow down, but still, it may be tiring and you should land asap. All well-designed autopilot servos also have safety provisions. Some use a shear screw, designed to break if the pilot applies sufficient force, which frees the servo from the control loop. Others (Trio, I know, for one) use a spring loaded clutch to allow the pilot to overpower the servo.
 
While I have circuit breakers on my panel for trim servos. I think this will do nothing to fix the problem of a trim servo which has run to its limit.

The autopilot servos appear to operate differently and removing power to them may allow them to be overpowered or they may even return to a neutral position with the loss of power.

Entirely correct, both statements.
 
Dan, Bob;

Thanks for the comments. Sometimes there is so much commentary its difficult to sort out the real facts. At least for someone who isn't knowledgeable in this area.
 
Not necessarily. The VPX can be set to only allow a run time of three seconds then it will cut off. Due to the nature of my failure, a short in a wire, that was intermittently shorting it managed to run all the way to the stop.

The VPX did eventually, correctly sense an issue and disable the trim. However, and this is what I find most concerning, the trim was mechanically stuck at full deflection.

No amount of switches, or other “interesting” schemes that folks posted on this thread would have prevented this situation.

-Dan

Dan, I'm glad you found the issue and you're safe on the ground. There IS a way to "unstick" the trim at full deflection. We put a lot of thought and testing into the VP-X trim system because of this very failure mode. I'll say with full conviction that it is the safest trim system for exp aviation on the market.

The trim system hardware goes through a self-test every time you boot it up and will flag a fault if something is wrong. Additionally, the hardware has two controllers in series so that if one fails the other will shut down the system. There's more but I digress.

It's important to understand that when on autopilot (and you have auto-trim system) that the autopilot takes over control of the trim system from the VP-X as it is "downstream" of the VP-X.

If I understand correctly, an external wire (prob between the VP-X and the stick) shorted that then commanded the trim to run. If this occurred when the AP was on, nothing would happen because the VP-X was cut out of the control loop by the auto-trim system.

If it happens during manual flight, there are a couple of things built into the system that would have helped:
1. If the trim is running one direction in an uncommanded manner, move the trim switch in the opposite direction and the system will immediately stop the trim motor movement. After a bit (I forgot, I think 2 seconds) if the system sees both up and down trim simultaneously it will then ignore the trim switch inputs for the rest of the flight. The shorted wire will have no further effect. As you mentioned it will also ignore the switch inputs if one is stuck.

2. Go to the EFIS on the VP-X page and select pitch trim. The soft keys will say up and down and you press them to move the trim back to a safe position. The trim only runs while the key is pressed. You can do this at any time, not just when there is an emergency.

More details are in the operating manual.

Again, glad you made it back safely and I hope this helps.
 
Dan, I'm glad you found the issue and you're safe on the ground. There IS a way to "unstick" the trim at full deflection. We put a lot of thought and testing into the VP-X trim system because of this very failure mode. I'll say with full conviction that it is the safest trim system for exp aviation on the market.

The trim system hardware goes through a self-test every time you boot it up and will flag a fault if something is wrong. Additionally, the hardware has two controllers in series so that if one fails the other will shut down the system. There's more but I digress.

It's important to understand that when on autopilot (and you have auto-trim system) that the autopilot takes over control of the trim system from the VP-X as it is "downstream" of the VP-X.

If I understand correctly, an external wire (prob between the VP-X and the stick) shorted that then commanded the trim to run. If this occurred when the AP was on, nothing would happen because the VP-X was cut out of the control loop by the auto-trim system.

If it happens during manual flight, there are a couple of things built into the system that would have helped:
1. If the trim is running one direction in an uncommanded manner, move the trim switch in the opposite direction and the system will immediately stop the trim motor movement. After a bit (I forgot, I think 2 seconds) if the system sees both up and down trim simultaneously it will then ignore the trim switch inputs for the rest of the flight. The shorted wire will have no further effect. As you mentioned it will also ignore the switch inputs if one is stuck.

2. Go to the EFIS on the VP-X page and select pitch trim. The soft keys will say up and down and you press them to move the trim back to a safe position. The trim only runs while the key is pressed. You can do this at any time, not just when there is an emergency.

More details are in the operating manual.

Again, glad you made it back safely and I hope this helps.

Marc,

First, thank you.

You are of course correct on the VP/X operation. And that behaviour is one of the reasons why I used a VP/X for my build. It is also part of what made this situation so odd.

I sent the logs to Garmin, and they took a pretty close look at them.
The short in the trim was intermittent. The first trim command disconnected the AP. Garmin verified that the VP/X did as designed and stopped the trim after three seconds. However since it was an intermittent short it kept resetting the three second timer.

Second, like you said opposite direction trim input will cause the VP/X to lockout the trim. That also happened, but by the time that happened the trim was already at full deflection.

I then went to the G3X soft keys. Unfortunately, and this is the part that still really bothers me, the trim had become stuck and would not move. I was able to confirm once on the ground that it was a mechanical issue. The trim was actually bound up. I had to partially disassemble the trim to get it working again. I have tried numerous times to recreate this and have not been able. The only thing I've been able to do is remove some metal from the bottom of the elevator in the area that it was bound up.

One other point to make, although I might have already said this. I called both VP/X and Garmin once on the ground to help with troubleshooting. Both provided great support both for a short term fix to get to my destination, and will analysis to figure out the root cause.

-Dan
 
Beep....beep...

Thread drift...a trim "beep" tone when the trim motor runs has been pretty much standard on (almost) all the large aircraft I’ve flown. Where would an experimental aircraft equivalent fall on the technical/practical/Human factors/“worth the effort” scale?
Initial thoughts (in answering my own question:) I doubt the "beeps" would help much in a runaway situation where the trim was commanded to run at full speed while on autopilot...but if you're at cruise and you hear more than a very few and very occasional beeps, you’d likely know something is wrong before it got ugly. If the VP-X (or similar device)is set up to run trim speeds at substantially slower rates when above 100 knots, a lot more beeps would have to be ignored before things got ugly. ?

(Thanks to the OP for starting the thread)
 
Thoughts on Runaway Trim

Much of my flight time before starting my -8A build was in the Navy A-4 Skyhawk, where runaway elevator trim was an immediate action emergency. Gave a lot of thought to that in my panel design.

Forward three switches on my switch panel: Elev Trim Power; Aileron Trim Power; Autopilot Power. Have never had reason in 1300+ hours to activate the first two in flight, but have used the A/P power cutoff. If building now, would eliminate the first two. Didn't think the power cutoff switches could be activated in time to stop full runaway, but could prevent further random trim inputs caused by uncontrolled electrons.

Didn't have the Flight Test docs in Phase 1 in 2012 but one of the early tests I conducted was full nose up trim and full nose down trim over slow to fast cruise and landing speeds. For nose down scenarios I let the nose drop to various attitudes before initiating recovery. For slow nose up scenarios, immediate unload to zero g, stop any roll, get the nose down then sort things out.

The -8A was controllable - admittedly sometimes uncomfortably so - at all test points.

The key - just as was shown in the 737 Max debacle - was to control the nose-low airspeed by reducing power to idle as soon as possible. With a constant speed prop, airspeed reduction is pretty quick. I never got near Vne.

This was for an -8A with fairly forward CG. Your results may vary.
 
Don't forget that if you are like the OP, and you have a Garmin G3X or G5-based autopilot with GSA 28 autopilot servos that drive the aircraft's DC trim motors, then simply pressing the AP DISC button on the control stick will interrupt power to both the autopilot servos and the trim motors.
 
Sam,

The VP-X and other modules that run trim don't normally directly connect to the audio system. In the case of the VP-X, it sends a signal to the EFIS whenever the trim is moving, and then it is up to the EFIS (which is normally connected to the audio system) to generate a tone if that vendor so chooses.
 
Runaway Trim vs Disconnected Manual Trim

Manual trim malfunctions can be really scary too (and yes, they do occur)!

During my first two years of ownership my RV7A had manual trim. Twice during that period the trim cable became disconnected from the trim knob where the two join. Both times this occurred immediately after takeoff when I was adjusting for climb out speed.

The aircraft immediately acted as if it has full nose up trim and became a handful to fly and created a real interesting landing problem. Of course the faster one flies the greater the out of trim pressures. There’s nothing like landing an airplane where you release forward pressure on the stick to touchdown. Not a pretty landing!!!

Those two experiences convinced me that “runaway trim” couldn’t be any worse than a manual trim cable disconnect. At least with runaway trim I should probably have some indications that trim is moving in a direction that I don’t want. With a manual trim cable disconnect it happens suddenly!

Those two incidents convinced me to convert to electric trim and I haven’t regretted the decision. I did use the Pat Hatch UTRB module with Tosten military style grips. I’m also getting ready to install the TruTrac Autotrim Module that a friend (who has manual trim) didn’t use when he installed his autopilot.

Let’s face it. There isn’t a system designed by men that isn’t subject to unexpected failures. Just be prepared to handle them when (not if) they occur. As the saying goes: sh** happens!
 
Don't forget that if you are like the OP, and you have a Garmin G3X or G5-based autopilot with GSA 28 autopilot servos that drive the aircraft's DC trim motors, then simply pressing the AP DISC button on the control stick will interrupt power to both the autopilot servos and the trim motors.

That is not correct.

In my case the AP had nothing to do with the trim issue. They are two different motors, and while the G3X will drive the trim.

Disconnecting the AP would have done nothing. In fact the GARMIN setup immediately disconnects the AP if it detects a trim input from one of the switches. That was in fact my first indication that I had a problem.

-Dan
 
Thanks Dan for starting this instructive post on runaway trim, something that until now I didn't think much about. I haven't purchased/wired my grips yet but my panel is fully installed and wired. Here's what I took away:

- Get to know how your plane handles with fully deflected trim at various airspeeds so your prepared when this happens.

- If this happens, slow down immediately and focus on flying the airplane

- Consider this an emergency and land asap. Dan's interaction with ATC was a great example of cooperation and keeping your cool under pressure, very impressive.

- I won't rely on back-up systems and complicated electric solutions. Assume some fault, probably a short, caused the trim to fully deflect and stays put, certainly if power is eventually cut to the trim motor

- I imagine any short is most likely at the molex connecting the stick to the main harness. This connection moves continually so I will double check that all terminations are securely made and that there is sufficient strain relief. Probably also add this to my maintenance inspection plan.

- I still plan on the Tosten grip with the 4-way hat switch but only for the pilot side.
 
Is anyone aware of a good grip that includes split switches for pitch trim? Also, can either the VPX accommodate needing parallel switch input to run the trim? As this switch or the wires serving it are the most likely culprit on a runaway, I'd love to plan for this on our 14.
Edit: I think the pinout of the VPX suggests it will support split switches. There's poles for the trim position and trim power circuits...
 
Last edited:
Thread drift...a trim "beep" tone when the trim motor runs has been pretty much standard on (almost) all the large aircraft I’ve flown. Where would an experimental aircraft equivalent fall on the technical/practical/Human factors/“worth the effort” scale?
Initial thoughts (in answering my own question:) I doubt the "beeps" would help much in a runaway situation where the trim was commanded to run at full speed while on autopilot..
(Thanks to the OP for starting the thread)

This is an old thread but as I am planning to rewire my trim connections on my RV8, I discovered a variety of things that I overlooked when I started this wiring journey 3 years ago.

Here is the Dynon partial solution to what you are looking for. Its SV-AP-TRIMAMP can provide alerts to the Dynon HDX display and audio alert when it detects the trim motor is abnormal. Not sure what abnormal but it is probably better than finding yourself in an abrupt nose down at 1000 AGL. I am sure the level of trim annunciation isn't on par with the big birds you fly.

Another advantage to the Dynon trim motor wiring is that it provides proportional manual trim at different speed levels so that you don't get the full trim motor travel at high IAS. The Dynon Autopilot also has an autotrim function so not to overload its large pitch servo.

I annotated the Dynon block diagram for clarity. At least this is my understanding of the Dynon system as I am rereading the manual
 

Attachments

  • Capture.JPG
    Capture.JPG
    116.9 KB · Views: 80
Last edited:
Don't forget that if you are like the OP, and you have a Garmin G3X or G5-based autopilot with GSA 28 autopilot servos that drive the aircraft's DC trim motors, then simply pressing the AP DISC button on the control stick will interrupt power to both the autopilot servos and the trim motors.

My Dynon system does the same when the AP is disconnected using the CWS or the autopilot disconnect button. However, Dynon recommends installing a circuit breaker to cut off power to the autopilot servos. And if the trim motor is installed with the Dynon Autopilot panel to provide the autotrim function, it recommends a fuse or circuit breaker to disconnect power to the trim motor. Probably out of prudence. Incidentally, this was how I wired my servos after this thread was started.
 
Back
Top