What's new
Van's Air Force

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

Check Your Autopilot Servo Arm Screws!

Reply to Dynon Support

Thanks for checking my six.

My thoughts run like this:

1. The failure I saw actually happened to ME. That makes it a more immediate threat (to me).

2. I don't run the autopilot very often - usually only for long flights. The failure you mention could only happen with the autopilot turned on. My failure could have happened with the autopilot turned on or off.

3. The failure mode you mention can only happen with a failure of the shear screw, which will cause a failure of the autopilot to control the aircraft. I would be acutely aware that a control jam could follow if the autopilot should stop controlling the aircraft, while the previous failure mode provided no warning whatsoever. Odds are, I just won't be using the autopilot until a better solution is determined.

TruTrak is investigating several fixes for this event, and when they come out with their final determination, I will probably be incorporating that modification. Until that happens, I will limit or eliminate my autopilot usage. The safety wire will insure no problems with the autopilot off.

Thanks again,

Pat
 
Glider,

Glad you are OK. Sent you a PM.
I think the Top Gun movie said it the best.

"I will fly with you anytime":D

Thanks for sharing the experience and a LOT of people will learn from
this.

See ya soon,
BB
 
NTSB Reporting

Guys, for what its worth, I did report this to the NTSB (required by regulations and all that).

I got the distinct impression that they have better uses for their limited resources than tracking down problems with experimental airplanes. The investigator I talked to (who is an RV'er, by the way) told me that for an isolated instance like this the NTSB will not do much of anything. If we had a bunch of airplanes all having the same problem, that might be a different story.

One of the best things about experimental aviation is, in my opinion, the lack of close governmental supervision. Of couse that does place an extra burden of responsibility on us as pilots and builders, but if you do not wish that extra responsibilty, then that is what certificated airplanes are for, and hence the reason for the tight regulations on them.

Pat
 
servo installation

Bob Mills, go back out to the airport and take a close look at your Dynon servo installation (I like taking out bulkhead and floor screws!). From the picture in post #34, it appears that the 2 lower bolts attaching the servo to its mounting bracket have a negative safety.:eek:
 
Matt,

Found and corrected...thanks. Pulled pic, and will repost with bolts re-installed as recommended herein, safety wire done, etc. Thanks again.

Cheers,
Bob
 
dynonsupport said:
We thought about doing exactly what you did with saftey wire, but it has a bad failure mode too:

If the shear screw breaks, the servo will spin the arm off, or it will re-lock the shaft, rendering the shear screw ineffective.

When the shear screw breaks, the purpose is that the shaft and the arm are no longer related to one another. The arm should be able to spin 360 degrees all day around the shaft without putting any torque on it.

If you break the shear screw, the servo has no idea of this. With the TT AP, the servo motor will just run one direction all day, since it's trying to command the aircraft with no effect, and it has no idea of its position. Now the shaft is spinning, the arm is not. No big deal if the screw in the shaft is not hooked to the arm, But in your case, the screw is now hooked to the arm, so as the shaft spins, the screw will tighten or loosen. If it loosens you're in the same situation where the arm falls off, and if it tightens, it will re-engage the servo to the controls. Even without the servo moving, it's possible for bad things to happen because the pilot is moving the controls and thus the arm, and this may re-tighten a screw down onto a jammed servo.
Click to expand...

I'm thoroughly confused. Please clarify. If a TT servo locks up and the shear screw subsequently breaks as desgined, then I thought the shaft will spin independent of the SCREW and ARM. And if that is correct, then safteying the screw to the arm shouldn't matter. I'm missing something. I simply cannot believe that the spinning shaft after a shear screw break is also spinning the screw. That seems like a major design flaw. I'm missing something here, please clarify.
 
tobinbasford said:
What? Why is TT saying they've only had one failure?
Click to expand...

Don't get the servo control arm screw falling out failure confused with the shear pin/screw breaking. The shear pin is there to allow the pilot to break the controls free in the event the servo locks up.

It seems that over time these can fatigue and break for no reason. I would assume that TT will come up with a standard time to replace these before they fatigue.
 
Servo

Pat:

Looking at your earlier photos (on post # 25, I should have quoted that post here), at least in my pea sized brain, I wonder if a temporary solution might be a clamp around the elevator push-pull tube directly above the servo push-pull rod end bearing where it is connected to the servo arm with a spring attatched to that clamp extending down to connect to the servo push-pull rod end bearing. Then if the screw backed out, it would pull the servo arm up tight against the elevator push-pull tube to keep it from jamming down in the bracket as it appears yours did.

Am I making sense here????:confused:

TT: If this works as a temporary solution---------Royalties accepted!!
 
Last edited:
Pat:

I'm glad to know that you're ok and that you were able to fix the problem. I met you last Aug/Sep '07 at the GPM terminal; I was in a CAP flight suit. You invited me to check out your RV-8...I believe you were making finishing touches to the canopy at that time and had planned to make your first flight a couple of months later.

Anyway, you're a first-class test pilot in my book.

Take care,
Randy
 
With the off the shelf TT and Dynon design the SHAFT and SCREW always rotate together. Or at least they are supposed to, since Pat's whole problem is that they didn't and the screw came loose. The screw goes into the shaft. If the shear pin breaks, the ARM can rotate around the shaft, but cannot fall of the end of the shaft due to the screw.

Safety wiring the screw to the arm means that now the arm and the screw are one. Now as the shaft rotates, the screw will unscrew or tighten. This is bad. It's the exact opposite of what is supposed to happen when the shear screw breaks.

On the other note, there are two screws, the arm screw and the shear screw. What we are discussing here is the arm screw. But the shear screw is what keeps the arm in relation to the shaft. Failure or loss of a shear screw should be no big deal.

Breaking a shear screw does happen, but it normal situations, all it causes is lost use of that servo (like it did to Paul/Louise today). But if you safety wire the screw to the arm, then it becomes a very serious situation.
 
Last edited:
dynonsupport said:
With the off the shelf TT and Dynon design the SHAFT and SCREW always rotate together. Or at least they are supposed to, since Pat's whole problem is that they didn't and the screw came loose. The screw goes into the shaft. If the shear pin breaks, the ARM can rotate around the shaft, but cannot fall of the end of the shaft due to the screw.

Safety wiring the screw to the arm means that now the arm and the screw are one. Now as the shaft rotates, the screw will unscrew or tighten. This is bad. It's the exact opposite of what is supposed to happen when the shear screw breaks.

On the other note, there are two screws, the arm screw and the shear screw. What we are discussing here is the arm screw. But the shear screw is what keeps the arm in relation to the shaft. Failure or loss of a shear screw should be no big deal.

Breaking a shear screw does happen, but it normal situations, all it causes is lost use of that servo (like it did to Paul/Louise today). But if you safety wire the screw to the arm, then it becomes a very serious situation.
Click to expand...

Should the arm screw just be locked in place with thread lok ? There doesnt seem much more one can do. Maybe a bracket attached to the servo which acts as a gate to the arm and screw is the only solution. Or TT can send out shafts with a castelated setup.
 
Only TT can answer what they believe is the proper fix in this case. When we designed our servo we didn't feel there was any other way to make the setup completely safe except for the castle nut solution.
 
help

Where do I find the drawings above dates 8-29-08? I had a pitch servo installed about a week ago. I just took a look at the servo and noticed there is NO safety used on the install and on the rod linkage there is not a nut on both sides of the shaft. The drawings on Trutrak's website do not show these items. I need to know where the drawings for RV-6 Pitch servo install as show below are located.

Thanks,

Jason


rvmills said:
Dan,

I understand what you're saying (place the bolt so that the pressure from the servo arm and rod are applied to the shank, not the threads, correct?), and I see that Pat has his TT servo installed as you are suggesting.

I do appreciate you watching my six, and after reading this, I went back to the install guides to see if I had followed them correctly. The diagrams actually show it as I have them installed, with the bolt head on the side of the connection rod bearings, so if I may, let me bounce a Q off you and the group, as I do want to do this right (best practices), and really respect your opinion:

The way the diagram has it, the bolt head is on the rod bearing side, which is captured by the large washer. Then there is a small washer on either side of the servo arm, followed by the nylok fastener. In looking at it, that puts the bolt shank under the push rod. It shows the same assembly at the bellcrank.

I guess my question is, where do you think the most pressure is applied to the bolt, at the connecting rod bearing, or at the servo arm and bellcrank? Since the servo arm is the driver, I sure can see your point. What do you think after looking at the diagram, posted below?

pitchservodiagramrt4.jpg


The diagram for the Dynon roll servo shows the same orientation for the servo arm connection (bolt head on the rod bearing side, bolt shank under the rod bearing), but the opposite orientation at the stick attach point (bolt shank under the aileron push tube connection, nylok at the rod bearing side). Here's a pic of that diagram:

rollservodiagramdq7.png


Any thoughts after seeing the diagrams? Like I said, I respect your opinion, and would like to have this put together in the strongest and most reliable way! Thanks again very much!

Cheers,
Bob
Click to expand...
 
Last edited:
Dynon vs. TruTrak

Jason, the drawing you asked about for the servo installation that includes a castle nut and cotter pin on the output shaft of the servo is a Dynon servo, not a TruTrak.
 
Ops check prior to take-off??

Do most of you guys with TT's or trio's engage the autopilot prior to takeoff and move the controls to ensure you can overcome the autopilot? This procedure was drilled in to me when learning to fly Cessnas many years ago. I've looked over some of the POH's listed on this website but don't see any reference to checking the autopilot.

Chuck Olsen
Tehachapi
RV-7A
 
cbo111 said:
Do most of you guys with TT's or trio's engage the autopilot prior to takeoff and move the controls to ensure you can overcome the autopilot? This procedure was drilled in to me when learning to fly Cessnas many years ago. I've looked over some of the POH's listed on this website but don't see any reference to checking the autopilot.

Chuck Olsen
Tehachapi
RV-7A
Click to expand...


Chuck,

I can speak concerning the Trio servos. The pre-flight check is certainly good practice but not particularly critical due to the design of the Trio servo. If you remove power from the Trio control head (flip the power switch) a solenoid in the servo drops out and completely disconnects the gear train from the aircraft controls. At that point the servo has "disappeared" from the control system. Even if the servo isn't released (sudden evasive action needed) the servo has a clutch that is very easy to overpower, and no damage is suffered by the servo. In testing the Trio devices I have intentionally hand-flown the plane through aggressive maneuvers with the servo engaged to make sure I can easily fly the plane in that mode if necessary.

The Trio servo drive system is similar in design (but much beefier in execution) to the old Navaid servo which no doubt has the safest failure mode of all the current servos. Another safety feature of the Trio system is how the autopilot will automatically disengage once the aircraft starts moving if you inadvertently leave it engaged following a pre-flight check.
 
Last edited:
which washers

In the previous Dynon diagram and in the photos, you have large penny washers on the rod end bearing. My Trutrak drawings don't show this "large" washer. Why such a large washer?

Bevan
RV7A wiring
 
The "fender washer" is often used on ball rod ends to maintain the integrity of the connection if the the ball drops out of the rod end.
 
nuts

If someone could tell me (or give me a spruce link) what size nuts are used on both sides of the shaft it would help me out a bunch. The guy that installed my linkage did not put the nuts on. These would be the nuts on the fore and aft side of the tube that is female threaded.

Thanks,

Jason
 
Sam Buchanan said:
The Trio servo drive system is similar in design (but much beefier in execution) to the old Navaid servo which no doubt has the safest failure mode of all the current servos. Another safety feature of the Trio system is how the autopilot will automatically disengage once the aircraft starts moving if you inadvertently leave it engaged following a pre-flight check.
Click to expand...

The TruTrak and Dynon servos use stepper motors that can always be overpowered, no matter what the control head tells them to do. No clutch needed, no power removal required. They're both safe systems, but I fail to see why the Trio system is superior in the safety area. The Dynon and TT servos only have 2 moving parts, and mechanical simplicity means a lot when trying to analyze failure modes. On top of that, the Dynon servos have a hardware disconnect of power to the motor whenever the disconnect stick is pressed. So even if everything in the system is yelling "MOVE" to the servo, including the processor inside the servo, if you press the disconnect button, the disconnect signal goes right to the last hardware step in the servo and fully disconnects it.

The Dynon system disengages when it gets airspeed as well.
 
Last edited:
Question for Pat Tuckey

First of all kudos for keeping your head. I have been asking myself if I could have done the same.

My question is this. Had you ever removed and reinstalled the #8 pan head screw that attaches the control arm to the servo? The screw is ?secured? with blue Loctite when delivered. Once broken loose, it should be cleaned and new Loctite applied before reassembly to reestablish the thread lock.
 
Just an idea

I do not have an autopilot installed but have been following this thread for its educational value.

In following the "safety wire vs. not saftefy wire the nut" discussion it occured to me that perhaps a small bracket could be riveted to the control horn such that, if the nut came loose and the arm fell it could not jam the controls. That way you have all the inherent safety of shear pin override plus prevention of catostophic failure mode associated with jam of servo arm in primarly control path.

I don't know if the geometry would allow it and it would entail removing elevator control horn from aircraft and careful check for clearance of said bracket but it might be worth considering.
 
Note from TruTrak Support re: Servo causing jam-don't safety wire

I was in the process of ordering screws with drilled heads so I could safety wire my TruTrak servos. However, after reading comments from Dynon support, I decided to contact TruTrak support by email through the Forum. I recieved this reply within an hour!!! Now that is support!

I thought before we all crawl back in the tail, maybe we better wait!

Here is TruTrak's reply:

dynonsupport is correct. I suppose I will need to make a post confirming his/their warning about using the safety wire. It could invariably cause a dangerous situation.

When we have a solution that we feel comfortable with, I will make a post on here and we'll put some info on our website.

However, please don't be worried about your servos. As I stated in my post, we have over 10,000 servos in the field operating right now. This is the only one we know of that has happened. Most likely, the arm was removed at some point in its service and threadlocker was not reapplied.

Thanks Cleve!
__________________
Lucas
Trutrak Tech Support
 
Threadlocker

Before my incident, I did not remove, modify or alter either servo arm screw.

I never did find the screw from the elevator servo. After the incident, I removed the screw from the aileron servo. It had no threadlocker on it whatsoever. If, "the arm was removed at some point in its service and threadlocker was not reapplied," it was done before the servo was shipped to me.

Pat
 
More info...

I removed the screw on my servo 3 days ago, there was no thread locker at all on it. I added some and screwed it back in. Would be nice to get to the bottom of this. My servo is pretty new, I would say it was shipped to the guys doing my install about 3 months ago.

Jason
 
Well done Pat!
Just another view talking specifically about installations on the RV7. Looking at the picture in post #38 it is clear that if one uses a long enough rod this will and can never happen. At least this is the way we have installed it. The rod is about 4 times the length of the servo arm.
 
Great job flying the airplane, Pat. You instinctively did what is written in some flight manuals, if a control is jammed - apply pressure until it breaks loose.

Can't comment on the screw issue except it sounds like someone failed to apply the locktite. I will be checking the servo arms although I am using an old Navaid devices autopilot.

Thanks for a great write up.
 
Examine screw carefully

Look carefully at the screw before you decide there is no Loctite present. It is dark blue in color and may be hard to see. I removed mine to check just how well secured it was and initially thought there was no thread locker present, however, on closer inspection I found it quite plainly there. (Do not attempt to remove the brass sheer screw, it is also inserted with Loctite and will break. It does not contribute to securing the servo arm, which slips over the head of the screw.)
 
Given the (now understood) undesirability of safety wiring the screw, I'm wondering if simply fashioning a small "stop" clip and either screwing or riveting it to the servo support bracket in such a way that in the unlikely event the servo arm screw worked its way loose, it would butt up against the "stop" clip long before migrating completely loose from the arm? Would this be a reasonable safety measure? Any unanticipated downside?

2mmfnko.jpg
 
Rick6a said:
2mmfnko.jpg
Click to expand...

I don’t have one so easy on the flaming :eek: but isn’t the large washer suppose to be on the bolt head side of the installation to prevent the pushrod from falling off in the event of a bearing failure? I suppose you have it installed as shown to prevent interference at the working angles of the arm and rod end?
 
Last edited:
Rick6a said:
Given the (now understood) undesirability of safety wiring the screw, I'm wondering if simply fashioning a small "stop" clip and either screwing or riveting it to the servo support bracket in such a way that in the unlikely event the servo arm screw worked its way loose, it would butt up against the "stop" clip long before migrating completely loose from the arm? Would this be a reasonable safety measure? Any unanticipated downside?
Click to expand...

At first glance, I don't see any potential problem with that idea Rick - maybe just a piece of angle with the vertical face attached to the bellcrank mount.

(BTW, I think Cytoxin is right - check your "big washer" locations - maybe htis was just a mockup?)

Paul
 
How 'bout this....

......method, Rick. Why not add a light spring, like the aileron trim, from the forward jam nut of the servo pushrod back up to a small hole in the upper part of the bellcrank. If the screw ever falls off, the arm would simply be lifted out of the way, unable to jam anything.

A thought,
 
Wouldn't that screw...

pierre smith said:
......method, Rick. Why not add a light spring, like the aileron trim, from the forward jam nut of the servo pushrod back up to a small hole in the upper part of the bellcrank. If the screw ever falls off, the arm would simply be lifted out of the way, unable to jam anything.

A thought,
Click to expand...

..up the forces that are balanced in the autopilot control system - an extra force in an unknown direction could upset the control lop design.
 
Rick6a said:
Given the (now understood) undesirability of safety wiring the screw, I'm wondering if simply fashioning a small "stop" clip and either screwing or riveting it to the servo support bracket in such a way that in the unlikely event the servo arm screw worked its way loose, it would butt up against the "stop" clip long before migrating completely loose from the arm? Would this be a reasonable safety measure? Any unanticipated downside?

2mmfnko.jpg
Click to expand...

Go back to post 14, in this thread.
 
STOP - DON'T TOUCH THAT SCREW

I just got of the phone with Lucas. I was calling about an altitude hold issue and we talked about the arm issue.

If you try to tighten the arm screw, you can break the Locktite seal just the same as unloosening the screw. He also recommended the medium strenght (blue) type.

Have a nice day and fly safe.
 
cytoxin said:
......isn’t the large washer suppose to be on the bolt head side of the installation to prevent the pushrod from falling off in the event of a bearing failure?........
Click to expand...
Since the screw issue arose, I had no choice but to post an early detail photo to illustrate the potential fix I was thinking about. I could not find a decent photo of the correct washer orientation, and I was not to keen to drive out to the airport, open up the baggage compartment then crawl back there to take a picture. Nothing gets past the good folks at VAF! Well...almost nothing. I'm surprised no one has mentioned the lack of torque stripe on the jam nuts at both ends of the pushrod. :)
 
Last edited:
I think Cytoxin is right also, but it is an understandable error in that TT does not supply good drawings for this installation. I looked back at my install diagrams, dated March 2006, and no washers are shown anywhere nor is the bolt orientation depicted. And Loctite -- nope, no mention.

For those of us who are not A&Ps nor engineers, we are left to our own devices. When I saw the drawings from Dynon in an earlier post I thought to myself, now that's what I needed when I was trying to figure out how to allocate the washers and orient the bolts supplied with my ADI PILOT II. TT has a wonderful product but I think the clarity in their directions and drawings could be enhanced for those of us new to the building game and not so versed in proper mechanical techniques. Yes we should know these things, but practically speaking these are very easy items to overlook in my opinion.

Chris
 
And that is a really good...

chrispratt said:
..... Yes we should know these things, but practically speaking these are very easy items to overlook in my opinion.

Chris
Click to expand...


....plug for having your local EAA Technical Counselor make detailed inspections as you build... and especially in the final stages of construction...

Join your local EAA Chapter to get access to their TCs...
 
az_gila said:
....plug for having your local EAA Technical Counselor make detailed inspections as you build... and especially in the final stages of construction...
Click to expand...

I never needed the technical counselor. And I never called Van's either! :D

Between the Matronics list (in the old days) and this site, as well as numerous builder websites..............I found everything I needed to know; and got it right! :D

But then I read a heck of a lot..:)

L.Adamson ---- RV6A
 
Rick6a said:
Since the screw issue arose, I had no choice but to post an early detail photo to illustrate the potential fix I was thinking about. I could not find a decent photo of the correct washer orientation, and I was not to keen to drive out to the airport, open up the baggage compartment then crawl back there to take a picture. Nothing gets past the good folks at VAF! Well...almost nothing. I'm surprised no one has mentioned the lack of torque stripe on the jam nuts at both ends of the pushrod. :)
Click to expand...

I figured as much, but you know when something doesn?t look right you feel obligated to point it out at the risk of being ridiculed. I hope others would view my pics with the same scrutiny...but then that?s why I love this place . It the only place a not flying yet airplane can get a conditional inspection....;)
 
A sidebar to the servo inspection discussion for RV-8 owners (If you're not an RV-8er you can pass on this post).

I inspected the servo screws in my RV-8 this evening and all is okay. While it took an hour or so to unscrew the front floors, inspect and re-install, the elevator inspection was a breeze thanks to the battery inspection door I installed. If you haven't put one of these in you might want to consider it. I only need to open the hatch, reach inside and feel the screw on the servo arm. There is a separate thread on this site describing the door install that shows several variants. Below are few snapshots of mine:

First is the rear baggage area with the access hatch door closed:




Next with the access door open:




And lastly you can see the bellcrank and servo area. Easy to see and reach from here without taking the entire baggage area out.




Chris
 
Chris,

This is the first time I've seen the mod, an excellent idea and one that I am sure to copy! :)

dfby80.jpg
 
Just checked my servo arms. The roll servo screw was barely finger tight and had no evidence of locktite. The pitch servo was tight and the screw clearly had been loctited. I've removed both servos and the attaching linkage and pulled the fuse for the control head. Until a means of positively locking these screws in place (not loctite) is developed I don't feel this is a airworthy installation. I like the looks of the Dynon more and more.

CHECK THOSE SCREWS!!

Doug
RV6
 
I have two of the TT servos so have followed this thread with interest.
A point that does not seem to have been mentioned is that the Arm retaining screw and the C/S screws that hold the motor mount plate on are both "NOT ABLE TO BE CHECKED FOR SECURITY" i.e. there is no way of confirming that they are secure!
Visually you cannot tell whether they are Loctited or not, nor done up or not.
If you test them with a screwdriver you will break the Loctite seal, so that is out. A proper dual control inspection as required for certified aircraft cannot be done.
The mount plate can be made visibly secure by inserting a couple of short AN3 screws in the thread in the body and then lock wiring them but the only really secure, inspectable answer to the arm retaining screw would appear to be the solution used by Dynon. Threaded shaft and nut.
Hopefully these will turn up soon!
Brian
 
I'm just finding out about this issue a bit late. Good reason to stay in touch with the VAF.

One thing I haven't seen anybody mention is a caution about reinstalling the servo shaft screw. For people that have been removing the screw to check for Loctite, or just cranking on the screw to make sure it's not loose, DO NOT OVERTIGHTEN THE SCREW!! The whole purpose of the shear screw is to break if the motor jams for some reason. If you tighten down the motor shaft screw excessively, the arm will not be able to move independently of the motor, thus defeating the whole purpose of the shear screw.
 
I looked at my arm last night and it has a nylon washer on the shaft screw, this should help prevent over torquing to the point of locking the arm. The screw was originally set with locktite. Some one suggested torque paint on this screw after it is reset with locktite. This will help the next time you inspect it. The RV7 and 9 and probably the 6 look less likely to get in a "locked" up elevator mode if the arm falls off. I looks like it would fall down and just hang there (assuming you are positive Gs when it falls off). The photos of the RV8 appear to have more hardware there to hang up on.
 
AP on trim?

A basic question that goes somewhat along this thread: Why not AP on the electric trim? Avoid the potential for this issue all together. Am I wrong, or is that the way my Father's C172, C210, C340 were all set up?

Thanks,

Forrest
 
RV6JOY said:
A proper dual control inspection as required for certified aircraft cannot be done.
Click to expand...
Brian, FWIW, duplicate inspections of flying & engine controls don't exist in the US. The concept is not even recognised. Only the English speaking world seems to insist on them. Pete
 
You must log in or register to reply here.
Back
Top