What's new
Van's Air Force

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

Dynon AP pitch issue

The datalog was built to record the sorts of data that would be useful to customers in their own analysis. When you get really into it, there are lots of deeply technical things that COULD be logged that we choose not to (the resistance of some random circuit, for example). There also a size constraint too - if you add too many things to the log, you end up not having a whole lot of recording time at 1Hz. Anyway, in troubleshooting your case, having that data isn't valueless, but also probably wouldn't be a panacea. We'll be looking into whether we should add that data to the log.

Also, while we're happy to see this thread out there, please make sure to send technical data or things you're trying directly to us via the email thread with the case number in it. This lets us track our progress in getting your issue resolved directly.
 
Dynon A/P Pitch Issues

I checked my Pitch Servo numbers today on the ground.
Centered -00001
Forward +00070
BAck -00107.

I can try data logging Friday, weather premitting.
I have traveled about 50 miles @ 3500' with steady altitude
no hunting.
 
dynonsupport said:
Also, while we're happy to see this thread out there, please make sure to send technical data or things you're trying directly to us via the email thread with the case number in it. This lets us track our progress in getting your issue resolved directly.
Click to expand...

I have been doing this and of course will continue doing so. I'm just gathering data at this point. Obviously there is something funky going on...just trying to gather as many data points as possible. :cool:
 
Jamie
My servo's are Pitch up -108
neutral +004
down +081

Roll left +101
neutral +011
right -098

My initial reaction to my autopilot performance is disappointing.
With that being said, I have had only 3- 15 minute flights with it and have not been able to go thru the inflight adjustment program.
With the default settings 100% torque, sensitivity 10.
Altitude hold +/- 60 ft., VSI rates to recapture 100-400 fpm.
Heading hold +/- 10 deg.
Nav hold +/- 10 deg. constant heading changes.
Reset sensitivity to 5 on both axis's not much change, and then
I had to land at the paint shop.
So, now I'll have a couple of weeks to explore the Dynon manual and see if there's going to be something to adjust. I sure hope there is!!!!

Dynon has been a great company to deal with. I have faith that their autopilot will eventually perform as well as the Tru Tracks I've become accustomed to.:D
 
Dynon AP RV6

Just did a 30 min test flight.
My altitude is rock steady 2350' +/- 5'
We turned on the Data Logger but it will be Friday before I can send.
The roll does seem to hunt some.


We watched the servo data while in AP mode.

Pitch data was DISC:N POS: -47
REV:N FRC: -00023 to -00028 varies

Roll Data was DISC:N POS: -00001 to -00003 varies
Rev:N FRC: +13 to +2 varies
 
After spending some time at home reading the Dynon D180/AP installation manual, regarding flight tuning. It appears that my brief flight and attempt to tune the servo's was exactly the opposite of what I needed to do. Upon initial engagement of the autopilot, there was some abrupt movements of the stick. I took that to be a "twitchy" movement and decreased the sensitivity.
I should have increased the sensitivity. :eek:
The weather here in Cincinnati has been crappy, my 15 min. flight to the paint shop was at 500 AGL, scud running with good vis., but towers between HAO and 3I7(Phillipsburg). So my attention was not on tuning the AP.
The plane is now disassembled at the paint shop, so I am hopeful that I'll get it "dialed in" in a couple of weeks.
 
A possible culprit

I may have found my problem tonight.

I engaged the pitch servo and then tried to move the stick fore and aft WITHOUT making the servo slip. The elevators were visibly moving several degrees. Going to the SETUP->AP->STATUS page I could see that the motor was staying on the same step.

Next I put the servo in TEST mode....the reason I did this is because the servo will not hunt in this mode -- it just stays on one step. Then I went back to the elevators and used my Craftsman digital level (similar to a Smart Tool). I was able to measure ~4 degrees of travel without the servo slipping. This is not good. I was able to 'wiggle' the elevators with very little force about ~3 degrees. Again, not good.

So I pulled the baggage wall and crawled (yet again) into the baggage area. Pushing and pulling on the pushrod I noticed that the servo arm was rotating slightly on the shaft.

I pulled the access cover on the right wing to inspect the behavior of the roll servo. It looks like it's doing the same thing.

I already spoke with my contact at Dynon and he's sending me two new servos. I will try them out to see how it works.

For comparison's sake I went over to my buddy's hangar where he's getting ready for his first flight (hopefully it'll fly this weekend). He has a fancy Trutrak with vertical speed select, etc. I tried the same exact test. His elevators are LOCKED into place when the servo is engaged.

In my opinion, the problem has been found. I'm not quite sure where to go from here. I will install the new Dynon servos, but I'm very curious why both of my servos have arms rotating on the shafts. They have consecutive serial numbers, so maybe it's just a fluke and the new ones will not do it, but only time and a couple more hours swapping out the servos will tell. I'm not ready to declare a design flaw....yet.

So can someone please do this test for me?
Go to SETUP->AP->SERVO TEST.

Get the AP test to move the stick to a particular location (doesn't matter where) and then put a Smart Tool or similar tool on the elevator to see how much you can move it without having the servo slip.

I'm curious to see if other folks have the same slop in the servos.

Oh yeah, I did the same test on my ailerons...checking the amount of movement I can get out of them with the servo on one step. I was able to move my ailerons 5.5 degrees without the servo slipping......and I know there's no slop in that control system -- the servo is at the bellcrank...very easy to check for slop there since there's only two pushrods involved (aileron->bellcrank and bellcrank->servo).
 
Jaime,

Did my first test flight today. Started out at the default torque and sensitivity settings (100%/10) on both servos.

For Roll, heading seemed to wander quite a bit, so I upped the sensitivity to 15, then backed it down to 13. Still wandered some, though a bit less. It was a bit twitchy as well (little bumps in the stick), so I need to play with it more. Then my Dynon compass started swinging, so I stopped the roll test, and need to re-cal the compass (Dynon instructions say that may be necessary, and though heading looked solid on the ground, it must need the re-cal...it may be the true cause of the heading chase).

And Jon, in reading, I think you were doing the right adjustments...wandering calls for increasing sensitivity, and overly twitchy (agressive) calls for decreased sensitivity.

For Pitch, the initial test went pretty well. Held altitude pretty smoothly for several minutes. When I engaged, I was about 70 feet higher than I intended and holding (8,570), so after a while I reset the ALT selector to 8,500, and trimmed slightly down to capture as shown on the DN annunciator. As it captured, it started to give me UP, then DN indicators, as it started some mild up and down pitch occillations. I have manual trim, and was making really small corrections, but it couldn't seem to dampen itself, so I disconnected and re-set it up. Held ALT pretty smoothly when I set it up carefully (trimmed, etc.)

It was a short test, and I'll redo, and expand the test next time. Will also turn on datalogging, and will record the position and force numbers in flight, like Al did.

My position numbers on the ground are:

Pitch: +73 to -117
Roll: +77 to -133

Forgot to check neutral...will also get that next time.

I just saw your last post Jaime...very interesting. Just to clarify, is what you're saying that you can see the servo arm(s) rotate on it's shaft back and forth a bit before the servo slips? And that translates into the wiggle or play you found in the control surfaces, right?

Will try the test you asked about, but wondering something. The test puts it near full deflection in two axes, so I'll check for play there, but seems like a good follow-on (especially if there is no play at full deflection) would be to engage ALT and HDG hold with stick neutral, and check for play there, as that is where (or near where) the surfaces will be in flight. That way we can look for dead bands or play near neutral, both in the servo arm and the control surfaces.

Will report back with what I find, and good luck with the new servos!

Hey, if it makes ya feel any better, I had to open the baggage bulkhead (again), and fiddle with the wiring connectors...seems my pitch servo went undetected, and I had to test all the connections in the "spaghetti" of my quick disconnect. All was good, and when I put the QD back together, lo and behold, there it was (I got lucky on that one...did not want to re-wire!!)

Good luck...we'll get there!

Cheers,
Bob
 
Last edited:
Bob,

Couple of quick setup tips:

- Try 15 to 18 on pitch sensitivity.

- HDG mode requires a good compass cal, but you can set up the AP in TRK mode if your compass isn't in cal.

- The trim up/trim down annunciations arent necessary for the autopilot to hold altitude so long as the servo isn't slipping. If you follow them however, the airplane will be in trim when you turn the AP off.
 
Hi Bob...excellent info...thanks!

Yes, the AP TEST mode puts it in the extreme positions, but the reason to do it that way is because if you just engage the autopilot on the ground it will continuously 'hunt' and will be changing steps occasionally so it isn't really a good test. If you use the test mode it doesn't hunt. What I did was just engage the test then using the stick put the control surfaces trail.

Yes, the servo is sliding slightly on the shaft. Of course a tiny movement at the shaft equates to a much larger movement at the end of the arm where the pushrod connects. The pushrod then connects to the elevator bellcrank which then magnifies the slop.

Also, as mentioned in a previous post I taped the elevators into trail and tried to move the servo pushrod by hand and it wouldn't budge...so that leads me to believe that the 4 degrees of travel I'm seeing at the elevators is the result of the arm slipping on the servo itself. I can also visually observe the slipping, so it seems to me like a reasonable conclusion to come to.

It's as if the sheer pin does not provide close enough tolerance and is 'wallowing' in the hole through the shaft.
 
Jamie
I just got back from the airport. While my plane is now at the paint shop and I'm unable to do any tests on it, my hangar mates RV8 is available.
This is the twin to my RV8 that was completed in April 08. This plane has a Tru Trak ADI II in it instead of the Dynon.
I engaged the AP and measured the "play" in both the elevators and ailerons.
1 degree at the elevator, and .5 degree at the aileron.
Good Luck
Jon
 
Today's test

Jaime (and DS):

Did the following today:

Ran the servo test and checked for slop in the control surfaces and at the servo arm. Found the following:

With very light fingertip pressure at the outer corner of the aileron, it moved about 1/16th to 1/8th inch in each direction (eyeballs only, no protractor, so I don't have degrees, but it was small). With the same very light pressure on the corner of the elevator, it did not move. However, if I applied a little more pressure at the center of the aileron or elevator (just enough pressure to keep the servo from slipping...a little more and it ratcheted), each surface moved about 1/4th to 3/8th of an inch in each direction (more than I had hoped for, but the flight test went OK, as I describe below).

Then I looked at the roll servo arm through the viewing window I made for the floor, and with the servo in test, I could move the stick enough to make the servo arm move about 1/4th inch in each direction, and the ailerons moved about the same as the second test above (all without servo slipping). Pitch movement at the stick seemed about the same...just a bit of slop in the stick...but I didn't pull the baggage bulkhead to look at the P servo (couldn't bring myself to pull those screws again! ;)).

I did try it in HDG/ALT Hold on the ground, and the results were about the same (as in both of the first tests). I did note the hunting you described, and it made the test less conclusive, just as you described.

I then did a compass cal, and then flew a test hop (and Dynon was right, the cal did wonders for HDG hold). Things went pretty well on the test. Felt a little more twitchy in roll than in pitch, and I ended up decreasing the sesitivity to 7 for roll and 8 for pitch, and it performed pretty well. A little overshoot and hunting in roll, and an occasional pitch occilation, but overall pretty good. Turns to heading were pretty good, with a little twitchiness in the turn and a few corrections to nail the new heading. Climbs went pretty well too. One thing I noticed is that if I left the pitch trim alone if it was hunting the altitude a bit after level off, and didn't chase the UP/DN annunciators with manual trim, it did better than if I reacted to each trim annunciation and trimmed (unless the annunciation was on for a while). I also noticed that having the ball in the center and being smooth with the rudders made a big difference in HDG hold (go figure! :))

I won't call it locked in, as it still seems a bit jiggly (good technical term), but it's probably just me getting used to the difference in a light duty AP on a nimble little plane, versus the heavy iron with the high dollar AP.

Some background info for comaprison:

- SV-32 servos
- RV-6, with clipped wings and a big gas guzzler up front.
- My seller and F1 Rocket driver said it's lighter on the controls than his F1
- Test at 8,500 to 9,500 MSL (3-4K AGL), with light winds, very occasional light chop.
- Flown at about 1,700# (1900# GW Airplane)
- Speed was about 165 KIAS / 180 KTAS (a bit less in climbs)

Today's position and force readings were:

Ground:

Roll...Left -90, Neut +11, Right +110
Pitch...Fwd +116, Neut +30, Aft -71

In Flight, Straight and Level, HDG and ALT Hold engaged:

Roll...Pos +15 +/- 5 (+10 to +20); Force +18 +/- 5 (+13 to +25)
Pitch...Pos + 25 +/- 5 (+20 to +30); Force -10 +/- 5 (-5 to -15)

Those in-flight numbers are a bit of a SWAG, as they move around a lot, as you know. Looks like the inflight positions were close to neutral in S&L flight, so that's probably a good indication (and the off-neutral numbers were probably a combination of whatever deflection the controls have at that speed, as well as perhaps an indicator of a very slightly out of trim elevator, and maybe an indicator of when I did not have the ball centered or was too touchy on the rudder).

I'll keep working it, especially if I get a chance to fly on a little X-C, and see how I can dial it in. Not sure how my control surface movement results compare with yours, but I'll keep an eye on it as I work to tweak it all.

Best of luck with your new servos, very interested to hear if the servo arm and surface movement remains, or decreases/goes away!

And DS, thanks for the tips. Will try various settings and see how they each work. And my compliments on your manuals, they do a great job throughout. If the last DS poster was Ian, thanks to you and Mike for all the help along the way, and Ian, congrats on the new baby!!

Cheers,
Bob





dynonsupport said:
Bob,

Couple of quick setup tips:

- Try 15 to 18 on pitch sensitivity.

- HDG mode requires a good compass cal, but you can set up the AP in TRK mode if your compass isn't in cal.

- The trim up/trim down annunciations arent necessary for the autopilot to hold altitude so long as the servo isn't slipping. If you follow them however, the airplane will be in trim when you turn the AP off.
Click to expand...

Jamie said:
Hi Bob...excellent info...thanks!

Yes, the AP TEST mode puts it in the extreme positions, but the reason to do it that way is because if you just engage the autopilot on the ground it will continuously 'hunt' and will be changing steps occasionally so it isn't really a good test. If you use the test mode it doesn't hunt. What I did was just engage the test then using the stick put the control surfaces trail.

Yes, the servo is sliding slightly on the shaft. Of course a tiny movement at the shaft equates to a much larger movement at the end of the arm where the pushrod connects. The pushrod then connects to the elevator bellcrank which then magnifies the slop.

Also, as mentioned in a previous post I taped the elevators into trail and tried to move the servo pushrod by hand and it wouldn't budge...so that leads me to believe that the 4 degrees of travel I'm seeing at the elevators is the result of the arm slipping on the servo itself. I can also visually observe the slipping, so it seems to me like a reasonable conclusion to come to.

It's as if the sheer pin does not provide close enough tolerance and is 'wallowing' in the hole through the shaft.
Click to expand...
 
Last edited:
Wow, you guys are fantastic. Thanks so much for going the extra mile to help a fellow RVer.

Bob: The movements you describe are certainly far less than what I'm seeing. It's looking more and more like all of the evidence is pointing toward slop in the system (the servo in this case). Every data point I have so far points to that...including being able to fly out of trim and have the AP hold altitude.

I'm very hopeful (but not tooooo hopeful) that the new servos will solve my problem.

Thanks again, guys.
 
Jamie,

I checked my Dynon AP installation on my RV-9A today and the arm of the servo definitely has a sloppy fit on the servo shaft allowing the arm to move around a bit when the shaft of the servo is locked or holding on a step. This slop allows about a +/- 1 inch movement at the trailing edge of my elevator, similar situation for my ailerons.

My plane has only flown twice and I do not know if the AP is working properly or not but there is definitely slop in the arm connection to the servo shaft. I don’t know if this is correct or not, it appears there is some kind of nylon bushing near the arm so there may be some intentional dampening or cushioning designed into the connection of the arm to the shaft but what I see appears to be a bit much. My servo serial numbers are 1095 and 1131, they were shipped 10/9/2008. Perhaps there was an undetected assembly problem at Dynon at that time.

I am anxious to hear if your new servos also appear to have loose fitting arms.
 
I'll double check my servo arms tomorrow, but as I remember there is NO slop in them. I do know the elevator is tight, i.e. no movement/slop.

Jim Shannon
RV-8 N52VV
Charlottesville, VA
 
rvmills said:
I won't call it locked in, as it still seems a bit jiggly (good technical term), but it's probably just me getting used to the difference in a light duty AP on a nimble little plane, versus the heavy iron with the high dollar AP.
Click to expand...

Keep dialing it in, you will be amazed at how well a "light duty AP" will fly our little planes! Once the system is debugged it should fly the RV very smoothly with no detectable oscillations or jiggles. Other AP vendors have it figured out and no doubt Dynon will crack the code, too. :)

Getting a new autopilot sorted out can be a circuitous path--you fix one issue and break another. But the digital AP technology works splendidly in our planes and we are fortunate to have several solid systems from which to choose.
 
Last edited:
Sam Buchanan said:
Keep dialing it in, you will be amazed at how well a "light duty AP" will fly our little planes! Once the system is debugged it should fly the RV very smoothly with no detectable oscillations or jiggles. Other AP vendors have it figured out and no doubt Dynon will crack the code, too. :)

Getting a new autopilot sorted out can be a circuitous path--you fix one issue and break another. But the digital AP technology works splendidly in our planes and we are fortunate to have several solid systems from which to choose.
Click to expand...

Thanks Sam, appreciate the words of encouragement! The comment wasn't meant as a slam on the AP or the RV at all (just noting the differences so far)...of course, I would much rather be playing in the RV, and I'm actually having a lot of fun doing the install and testing, after doing a panel upgrade. It's been enjoyable to see it all come together and start working, and when I get it tweaked in, it'll be really satisfying to hit the button(s), watch it do it's thing, and sit back and say, "I did that"! (with the added benefit of X-C flights with my boys and being able to relax and point out neat stuff along the way and watch their eyes light up). Pretty cool (not as much fun as wringing the RV out with the autopilot off...which is most of the time in the air...but another kind of fun, just the same!)

I think you're right about the circuitous path...the roll and pitch modes probably interact a bit, and dialing it in will be a fun project. Very pleased with the process and performance thus far! And learning and troubleshooting a new product as a community of flyers is pretty fun too! Good stuff!

Cheers,
Bob
 
I installed the new servos yesterday. I have to say that swapping them out wasn't too bad. Trutrak really did a good job when they designed those mounting brackets. ;)

The worse parts were removing the baggage wall and safety wiring the AN3 bolts on the new servos.

Here are my initial findings. The elevator travel is now reduced to about 1 degree or so. That's definately an improvement. I will occasionally perform this check to see if the throw increases over time. If it does...well there's something wearing in the servo.

I went out and flew the new servos yesterday. All of the area metars except my home field were reporting scattered 1400ft, overcast 5500, but I went out to have a peak and everything was completely overcast at 1400ft and it was quite turbulent down below.

In other words, I wasn't able to get in a proper test flight. The altitude hold seemed to be working OK, but it's impossible to measure any improvement in performance unless I can fly in smooth air. Forcast this week is looking grim -- seems that Thursday will probably be the earliest day I can test properly.
 
Jamie,

It sounds like the new servos also have a bit of slop in the connection of the arm to the servo shaft. Could you please estimate the free travel of the elevator trailing edge now when the new servo is locked on a step, you mentioned 1 degree or so, about what would that be in +/- inches?

Perhaps DynonSupport could provide a brief description of the connection between the servo arm and shaft, apparently some slop is acceptable and possibly designed into it since multiple installations have reported it. Perhaps this is to provide a dampening effect? I would sure feel better if I knew the slop in my installation was normal rather than spend a lot of time trying to adjust a system that is out of tolerance.
 
Dynon Servo Slop

There should be no movement between the output shaft and the output arm. (Some play in the output shaft bushing is normal.)

If you believe you have play between the output shaft and the output arm, please contact us via email at:

[email protected]

Or call us at
(425) 402-0433.


Re:

647jc said:
Jamie,

It sounds like the new servos also have a bit of slop in the connection of the arm to the servo shaft. Could you please estimate the free travel of the elevator trailing edge now when the new servo is locked on a step, you mentioned 1 degree or so, about what would that be in +/- inches?

Perhaps DynonSupport could provide a brief description of the connection between the servo arm and shaft, apparently some slop is acceptable and possibly designed into it since multiple installations have reported it. Perhaps this is to provide a dampening effect? I would sure feel better if I knew the slop in my installation was normal rather than spend a lot of time trying to adjust a system that is out of tolerance.
Click to expand...
 
Here's the latest update in my saga.

I installed the new servos and returned the old servos to Dynon.

Dynon replied to me that they did not find any play in the servos. I find that quite strange, since I know that my eyes weren't deceiving me. You can't see the movement with the servo off...only with it on.

Anyway, I installed the new servos but the servo I installed in the pitch position seems to have not had proper factory calibration. It is reporting servo friction on the ground with no load on it.

The result is that as soon as the AP is engage on the ground it reports that I need down trim. In the air I have to add a significant amount of down trim to get the down trim indications to show level flight.

At first I thought this was done as a one-off build of the firmware just for me since I previously reported that when flying the airplane with nose-down trim the AP worked a lot better. My contact at Dynon is telling me that my servos have the latest official build.

I measured elevator throw again. It's still about 1 degree of throw with the servo locked into place (previously it was 4+ degres), and every bit of movement I can see in the system (which is definitely smaller) is at the servo.

I trimmed the airplane and engaged the autopilot. It still oscillates around target altitude same as before. Out of trim it works perfectly, including during climbs and descents.

Maybe I'm just too anal and am expecting too much from this autopilot?

You other guys that have it working properly, do you have the VSI enabled on the EFIS and does it indicate level flight with the ALT hold engaged? Mine is back and forth, back and forth, and you can definitely feel it in your rear. Altitude is varying +/- 20 feet at about .25hz I would imagine.
 
Jaime,

Some interesting results, to be sure! Did you see the post by another gent in the "Dynon AP Photo" thread, in which he found the shear screw backing out? I looked on the Dynon forum and did not see any discussion on it, so I asked him a couple questions, and it's possible this is all related. I definitely do not want to jump to any conclusions on that, so I'll ask you similar questions here:

In the servo arm movement you are seeing, is the black arm moving around the silver mounting bolt, or is the entire servo arm, mounting bolt and castellated nut "system" moving together? If it's the former, maybe the shear screw is an issue...but like I said, I don't want to jump to conclusions, and I'm hesitent to start taking the servo arms apart.

How much linear travel are you seeing in the servo arm when locked (in test)? If I recall, I could move the stick a little left and right, and a little fore and aft, and saw about 1/4 inch travel of the servo arm in each direction, while in test, before the servo slipped. Not sure how that compares to yours, or to other APs, or if that is pretty typical (first AP install for me).

On the ground, is it just the trim annuciator you are seeing, or is there another indication of the friction you describe? I've not seen that on the ground at all (for feedback).

For feedback on current performance, I flew on AP a bit last week, and most of the time it was pretty smooth. Got about 15 minutes S&L with ALT and NAV engaged, and didn't have to touch the manual trim. I did see some slight VSI changes, and little altitude changes (perhaps similar to your +/- 20 feet) but ususally did not feel it, and I couldn't begin to put a frequency on it...it really wasn't bothersome. Once dialed in, I don't really know what to expect in altitude hold, but even at work I see some minor deviations (perhaps low freq fughoids) that are pretty much imperceptable unless I'm staring at the VSI and ALT readouts (FWIW). So far it seems pretty capable. ALT hold seems pretty good, HDG hold and NAV seem pretty good too, though I think it still chases those a tiny bit more, and hopefully some more Pitch and Roll "sensitivity training" with the servos will dial it in.

However, there have been a few occasions when I did get some pitch occilations that did show +/- 500 fpm or more on the VSI and were clearly felt, but I thought that they were the result of me not having the airplane stable when engaging. They would not dampen, so I would disconnect and stabilize and re-engage. The fact that it's good when you are out of trim and not so good when you are in trim seems odd. On the occasions when it is flying really well, what happens when you disengage (ie, how out of trim are you?)

I'm still just feeling it out and getting it tweaked in small bites, and feel I need to work it and test it quite a bit more. I won't get back up till next week, but I'll try to measure things a little more closely for you.

Would be interesting to hear from those that have been flying it for a while, hear what they see and feel with respect to these issues, and how tight the Pitch and Roll servos hold assigned parameters! (edit: Just saw Al's post...looks like the bar has been set! :))

Cheers,
Bob
 
Last edited:
Dynon autopilot settings

Wow, it took me about an hour to get through all these posts and I really feel for Jamie. I think if we all have patience and work together with info, then we will eventually end up with an autopilot that is satisfactory if it is possible. Dynon wants to get this right, but so far it hasn't happened. This is a new system and they are learning too. I also have the AP74, 2 servos, and FlightDek180 and so far autopilot performance is unsatisfactory. My altitude varies by up to 100 feet at vertical speed correction rates approaching 1000 fpm. Heading hold simply doesn't work within reason. Nav trak is a lot better, in that the deviations are more gradual. Heading wanders back and forth continuously in heading, but is a little better in trk or nav. One thing I tried today that helped dampen the altitude deviations was to set a max veritcal speed at 300 fpm. I had 1500 fpm set initially because I wanted to use it for climb, but setting 300 - 500 works better to control altitude deviations. Jamie said he starts out OK in altitude hold but after a minute or so it begins to deviate and the excursions just get worse (at least I think that's what you said). Well, I've found exactly the same thing. Take a look at your vertical speed setting and see if changing it makes it better. As far as heading hold goes, I don't know what to think because when I engage heading it usually starts to wander imediately, even though the heading bug doesn't move. Right now I have both servos torque setting at 100 as recommended by Dynon. My roll servo sensitivity is set at 9. 10 or higher and it gets real 'twitchy', rocking the wings back and forth rapidly. My pitch servo is set at 12. I had it at 20 when I started today as recommended by Dynon, but with 12 the deviations aren't quite as abrupt - still deviates though. I think the TruTrak autopilot - which was rock solid right out of the box without any adjustments in my RV-6 - is a totally different kind of autopilot, so comparisons probably aren't valid. This AP74 is really cool and having an autopilot integrated in an EFIS is too. Plus that female voice that barks out 'autopilot - off' is neat, although I don't really need it right now because when the autopilot is disengaged, everything gets smooth again... (ha ha - a little levity there Dynon). I just hope we can get it figured out. It must be a software detail that needs to be fine-tuned. The beta testers apparently didn't have these problems, so something is up. Dynon wants their new autopilot to work and I believe they are concerned. Mike Huff called me as I was typing this to see if I was doing any better with my upgrade. I had a real ordeal trying to get the EFIS and servos upgraded to 5.0 as many of you did. I told Mike about my trouble with autopilot performance and he wrote down the info I gave him so they could try to help. I think the biggest problem with the upgrade is using a PC that is pretending to be a computer. For us MAC users, using a PC is like trying to start a fire by rubbing two sticks together. It's amazing anything ever works with them. If they could make their software work with an Apple, they wouldn't have these upgrade problems.

Scott
RV-8 Fastback (and I mean fast)
 
Sure hope they get it fixed as I'm installing a system in my -8. It's funny the beta testers were happy - would one of you please chime in with your experience? It would also be nice if Dynon would chime in with their take on the issues and potential solutions.

I agree....once you go Mac you never go back....been a happy Mac user for 4 years now....just wish my company issued Macs :(

Jim Shannon
RV-8 N52VV
Charlottesville, VA
 
Is it possible the Dynon altimeter is not sensitive enough which makes the airplane oscillate when operating with the autopilot? Here's the reason for that question: I have a D100. It is not installed in the aircraft yet, but if I turn it on, let it stabilize, and then carry it from the basement to second floor of my house, the altimeter does not move as I climb the stairs (~20' change). It will eventually show an altitude change, but that takes a while. Maybe what I'm seeing is only because it's not installed in a moving aircraft, maybe not.

Jim Shannon
RV-8 N52VV
Charlottesville, VA
 
flickroll said:
Is it possible the Dynon altimeter is not sensitive enough which makes the airplane oscillate when operating with the autopilot? Here's the reason for that question: I have a D100. It is not installed in the aircraft yet, but if I turn it on, let it stabilize, and then carry it from the basement to second floor of my house, the altimeter does not move as I climb the stairs (~20' change). It will eventually show an altitude change, but that takes a while. Maybe what I'm seeing is only because it's not installed in a moving aircraft, maybe not.

Jim Shannon
RV-8 N52VV
Charlottesville, VA
Click to expand...

The Dynon EFIS must have pitot/static info to function properly. You will like it in your plane. :)
 
Sam...are you saying...

Sam Buchanan said:
The Dynon EFIS must have pitot/static info to function properly. You will like it in your plane. :)
Click to expand...

...that the Dynon will not work at zero airspeed?

When you have it as a hand held, the open port in the back should work exactly like a static port up and down the stairs....:)

I'd try it with my D-180 but I haven't bought the battery option yet...:)
 
I don't have much to input other than to say if Mike Huff is on it, it has a darned good chance of getting fixed. Dynon managed to snag him when Chelton closed down their Idaho facility, and what a catch he is. For those of you who don't know, Mike is one of the good guys - really good guys and Dynon was darned smart to get him. He's one smart feller.....

Regarding airspeed and the Dynon...I don't know about their AP, but the EFIS itself won't without it - been there done that.

My 2 cents as usual.

Cheers,
Stein
 
Be patient and keep feeding Dynon the info, they are smart people, they will figure it out. Some report no problems and some report problems. If the no problem reporters are telling the truth, the issue will eventually turn up!

Bet that the next major release of firmware will have more parameters for tweaking! I do process control loops for a living and there is a reason why the I and D belongs in PID. Seems Dynon originally thought that P was good enough or that a fixed value of I and D set at the factory would work across the board.

Elsewere I saw where the HDG hold funtion was not working well for someone and after they did a very good compass calibration, the problems with that went away. Might be something to try.

If Dynon is using a 12 bit A-D converter, they should have a resolution of 7.6171875ft on the altimeter.
 
Scott,

Sorry the issues continue...keep at it! I'm the guy Brian referred to on the compass cal issue. I recalibrated mine after the first flight (in which my heading hold chased badly) and it has been much better since doing that. I think the manual suggests the re-cal after factory service as well, and it really did the trick.

On my 6, I currently have the sensistivity set to 7 for roll and 8 for pitch. I also have the max VSI set at 500, the max turn rate set to 2 deg/sec, and the max angle of bank set to 20 deg.

Turn performance is pretty good...still a little "bumping" of the ailerons in turns, but minimal overshoot. Heading and altitude hold seem pretty good (+/- a few degrees and little wandering in HDG, +/- 20 feet in ALT, only occasional occilations), but I want to play with it more to find the smoothest settings that deliver consistent performance. Altitude changes (the couple I've tried) have been poor -800 fpm to 0 fpm, nose moving up and down with noticible "g" in the butt as it searches for 500 fpm. NAV tracking was good on the couple times I tried it. Still only 4 test hops under the belt, so more work to do.

I'm planning a 2 hour X-C to LA tomorrow, and will play with all the modes, giving each a good chance to be steady or occilate around the desired targets. Will play with all the settings, and see if I can come up with a good combo...then report back here.

I think it'll dial in with practice, and will be worth the effort, based on improvements I've noted in just a few hops.

I will say that Dynon has been very proactive and helpful, and concur with Stein on Mike. Didn't know he was grabbed from Chelton, but he's been a really big help to me!

We'll get there...heck, we're sort of the "Gamma" testers as "near the front end" buyers and flyers!

It would be great to hear more from the beta testers to see what tricks have worked for them. Maybe we could even start a data base of what settings seem to work best for different RV and Rocket models!

Back atcha after the X-C!

Cheers,
Bob
 
Last edited:
Here's the lastest on my autopilot performance problems. Dynon (on their forum) ask me to measure the play in my servo with the autopilot engaged. You do this by going into the test phase which freezes the servos in a position. Then you move the aileron and elevator and measure how much movement you get before the servo slips (a loud bang as the lever arm rachets). On mine, both servos allowed 5 degrees of movement before slippage. That's 5 degrees total of about 2 1/2 each way. I called it in and Lawrence from Dynon called me back with a bunch of other measurements and info to obtain. I'll do that tomorrow or the next day. However, he said 5 degrees is too much for an RV-8 and with that much would have a tough time holding altitude or heading. There's a couple possibilities for this play. An internal problem with the servos; slippage between the lever arm and output shaft (bad), or slop in the aircraft control linkage - unlikely in an RV-8 if built to plans. These are the things I'll be checking in the next couple of days. I'll also be collecting servo movement limits from the setup pages in the EFIS. The worst part of this information gathering is removing the bag bin shelf (not too difficult), and the front cockpit floor (bummer). My servo is mounted under the front seat like a TruTrak, which Lawrence says is just fine. For you other guys experiencing these problems, you might want to check yours too and let us all know what you get. You might want to post on the Dynon forum as well. Look for my post under 'Autopilot Discussion/Torque,sensitivity values to keep all the posts in one easy to find place.

Scott
 
Scott,

Just posted this on the Dynon forum as well...good thread you started there!

I did some more playing with the sensitivity on my AP on a X-C in my RV-6 today. Torque is at 100% (pitch and roll), max VSI is 500 fpm, turn rate is at 2 deg/sec, max AOB is 20 deg.

Started out with the sensitivity I had set after the last test hop (7 roll, 8 pitch). In ALT, I saw more deviations than on my last hop (smooth air on both flights). I saw +/- 60-80 feet, with VSI as much as 650 fpm in each direction. Just for the heck of it, I went to 25, and started working my way down. 25 was way too aggressive, though it held alt more closely (+/- 10-20', but really twitchy). Dropped to 20, then down in steps of 2 (18-16-14). Turns out 18 felt and performed the best today. Little twitching, and altitude +/- 10'...pretty smooth, though it did pitch up and down a little visually...but without any noticible g feel. At 16 and below, the altitude chase began, though less wallowing than at the 8 starting point (+/- 350 fpm, +/- 50-60').

In a commanded descent (rolled ALT window from 11.5 to 8.5, and rolled in a little nose down trim). At sens 18, it chased the VSI just a bit, but settled into 500 fpm +/- 50fpm. It also captured 8500' right on, then chased it +/- 20 feet, then settled down pretty well. 18 seems like the nicest spot on my plane for pitch so far.

In roll, at 7 then 6, heading held +/- 2 degrees, and was pretty smooth. Changing the heading bug for a turn caused a relatively quick roll in to 20 deg AOB, and a bit of pitch chasing...need to work that piece a bit, but overall, it was pretty satisfactory. NAV tracking was overall pretty smooth at those settings as well.

Just some numbers for your quest!

Scott, a question on your control surface tests: How much travel in inches (or fractions) does your 5 degrees (+/- 2.5 degrees) equal. Mine move about 1/8th inch in each direction with light pressure on the corner of an aileron or elevator, and 1/4 to 1/2 inch if I press a bit harder on the center of the surface (with the servos locked in test). Any more pressure and the servo(s) slip. How does this compare to your movement?

I'd be interested to hear what other measurments and readings Dynon is suggesting, so I can duplicate.

Thanks much!

Cheers,
Bob
 
Autopilot update

I was having the same issues as Scott was with my servo's. Unfortunately my plane was stuck at the paint shop so I could only watch and assist Scott with his. Since Scott is now taking his plane to the paint shop, he was kind enough to let me have his newly modified(by Dynon) servo's.
We determined that both of our original servo's(4 total) had excessive amounts of "play" between the servo arm and the Shaft/capstan/disc.
This play was translating to about 4.5 to 5 degree's at the elevator(1/2 in. to 3/4 in. at the trailing edge. In my case with an identical RV8 with a tru trak installed that flys rock solid, a comparable test showed less than 1/2 the "play" of the Dynon servo's.
Dynon recognized that this was a problem and has been diligently working on a solution.
The problem lies in the shear screw. A minute amount of "play" in the screw threads in the capstan/disc allows the servo arm to move very slightly, independent of the capstan/disc/shaft. This is BAD!
The fix is to reinstall the shear screw with "Loctite 271 (red)", torque the screw to 1 in./lb, and let set for 24 hrs before proceeding with the re-installation of the servo. Dynon has extensive instructions on how to do all this if you should determine that you might be having a similar problem. Or you can send your servo's back to them and have them do it for you.
Scott volunteered his new servo's so we could carry on as "gamma" testers.
So, as the snow and cold swirled around us yesterday, precluding us from dropping Scott's plane at the paint shop, it was "swap the servo's day" at my hangar. Today, with 10 inches of unplowed snow on the runway at the paint shop airport preventing the delivery yet again of Scott's plane, we launched in my plane for autopilot testing.
After testing with a digital level and seeing elevator/aileron play similar to the trutrak plane sitting beside it, I was hopeful.
In our brief flight we were able to test most modes of the AP74, and I can finally say that this autopilot performs as well as the Tru Trak.
In pitch, it held altitude to +/- 10-20 in light turbulence.
Climbs and descents to a new altitude were at the selected(500 fpm) vertical speed, within 100-200 feet of target altitude the vertical speed rate decreased to about 100 fpm, then a perfect capture of the target altitude.
Heading mode showed +/- 1-2 degree deviations probably due to the yawing in the light turbulence.
Track mode, it kept the pink cdi needle centered.
Nav mode was excellent also. Picking a new "direct to" waypoint 90 degrees off heading resulted in no more than 20 degrees of intercept of the original "direct to" bearing. It then settled in nicely.

I AM EXTREMELY PLEASED WITH THIS AUTOPILOT NOW. Dynon's support has been very good.:D
 
Great news! It sounds like it now performs like my Beta servos.

jthocker said:
...Nav mode was excellent also. Picking a new "direct to" waypoint 90 degrees off heading resulted in no more than 20 degrees of intercept of the original "direct to" bearing. It then settled in nicely.

I AM EXTREMELY PLEASED WITH THIS AUTOPILOT NOW. Dynon's support has been very good.:D
Click to expand...
Also, Dynon is working on improving its intercept routines. If my head cold goes away, I hope to get out and do some more testing but will probably wait until the cold wave passes now. :(
 
jthocker said:
.
.
The problem lies in the shear screw. A minute amount of "play" in the screw threads in the capstan/disc allows the servo arm to move very slightly, independent of the capstan/disc/shaft. This is BAD!
The fix is to reinstall the shear screw with "Loctite 271 (red)", torque the screw to 1 in./lb, and let set for 24 hrs before proceeding with the re-installation of the servo.
Click to expand...

I did a close study of my pitch servo arm after a few test flights. I found the shear screw loose. but why? Two things #1 the castle nut was loose, not even finger tight. Allowing arm slop, loosening the shear screw. I tightened the nut 1 slot, (firm finger tight).
#2, Loctite 271 used in combo with brass & aluminum has poor locking performance. check their web site , do your own tests. I did reinstall using 271... time will tell.
After tightening every thing up my Dynon AP is performing fine. :cool:
One thing to think about- If the servo arm is sloppy? will the shear screw shear clean and rotate smooth, or will it mush over and drag?
 
Last edited:
A note on dynon servo castle nuts

While the castle nuts should not be loose they must not be torqued above 4.5 in-lb. Overtorquing the arms will increase the force needed to break the shear screw and restrict the freedom of movement of the arm in the event the screw needs to be broken.

This would defeat the safety provided by the shear screw in the event of a servo jam.

Please get in touch with us if you have questions regarding your servo arms.

Thanks,
-Dynon Support
 
Tom
The cheapest screwdriver that can "torque" that low was one from Grainger, for over 200 dollars.
It doesn't take much of an excuse for me to buy a tool, and if I could think of one other place to use it I probably would have gotten it, but I couldn't. So I sent the servo's back to Dynon and let them do it, for the 10 dollars in shipping.
Best regards
 
Jamie,

Enjoyed reading this thread. I think you're close to a fix. My bet is you'll find the worn hinges/rod bearing or structure deformation that is causing this.

At this point, if I was you, I'd start replacing heim bolts and keep pushing/bending on your structure until you find whats wrong. The fact it works well untrimmed screams to me play/slop/deformation in the control chain.

You will get it right.

Bob
 
Anyone else have any updates?

My AP performance seems to have worsened. I'm guessing it's a loose sheer screw.

I've looked for more and more slop in the system. The only slop I see is in the servo itself. The Dynon Support guys think I'm nuts when I say it, but the arm on the servo will move just a little even with the servo locked in on a step. That movement gets magnified at the elevator trailing edges.

I think this is the crux of my problem.
 
Automation is usually as perfect as it gets......if set up correct, its most likely that all inaccurate automation stems from mechanical inaccuracy.

I do this for a living and it is what we find time after time!

Cheers!

DB:cool:
 
Jamie said:
Anyone else have any updates?

My AP performance seems to have worsened. I'm guessing it's a loose sheer screw.

I've looked for more and more slop in the system. The only slop I see is in the servo itself. The Dynon Support guys think I'm nuts when I say it, but the arm on the servo will move just a little even with the servo locked in on a step. That movement gets magnified at the elevator trailing edges.

I think this is the crux of my problem.
Click to expand...

Jaime,

I've been continuing to play with and adjust the settings on my Dynon AP (RV-6 with clipped wings and lengthened fuse...for comparison).

Roll servo seems to still be happy at 6-7 on sensitivity. HDG hold is +/- 1-3 degrees, with little jerkiness or wallowing. NAV tracks pretty solid (seems even smoother than HDG IMO).

Pitch servo is less dialed in. Did a couple X-Cs at 14-15 (sensitivity), and it displayed different "moods" over time. Most of the time it was stable, maybe +/- 20 feet or less, with no noticible feel in the keester during corrections. However, it occasionally starts to oscillate, and then go +/- 50-100 feet, with trim annunciations popping up, and a light, but quite noticible feel to the pitch corrections. Those oscillations seem larger during descents.

Flew a bit yesterday and today, some on AP, and did a level stretch, followed by a descent of 2K on AP during each run. Some minor oscillations when level at 14, and during the descent, it would hover around the preset 500 FPM, but ocasionally oscillate down to -850 FPM, then up to -200 FPM. Changed to 16, then 18, and the oscillations dampened a bit, but were still present (-700 to -300 FPM at the outside). I found that if I used a little pressure on the stick to help stop the oscillation, it would dampen itself, but if left alone, it would continue. Still not as tight as I'd like it to be, and I don't want to be staring at the VSI to babysit the AP.

I too see a little movement in the roll servo arm when its locked in "test". In my test I moved the stick left and right, and saw the movement in the servo arm. Didn't have the baggage bulkhead off when I did this, to watch the pitch servo, but I was able to move the stick fore an aft a bit as well. In each case, "a bit" means maybe 1/2 inch of stick movement (at top of grip), and that translated to about 1/4" or a little more movement at the arm. Not sure if the mechanical advantage at the top of the stick is just moving the servo arm in a null between steps, or if the servo arm is actually moving around the shaft (i.e., possible slop at the shear screw).

I've been wanting to play with the settings before opening up to look more closely at the servos, but have an oil change comin' up, and need to replace a TED antenna, so the floors are comin' up and I'm going to pull the back bulkhead too, to get a good look at the servos. Will report back with what I see.

Here's a Q: IYO, what's the best way to check the arm movement for slop? I'm thinking about disconnecting the control rods from the servo arms, putting the servos in test, and then seeing if the arms wiggle around the shaft, or just move a little between the first ratchets in each direction. I've read that other servos don't budge, but just not sure what would be considered normal or acceptable. What say you (or others)?

Cheers,
Bob
 
Last edited:
Bob:

Your findings with the movement are close to what I measured with my first set of servos (yes, I'm on my second set now). I have not made the measurements on this set of servos since I noticed the worsening performance.

I found that the movement of the arm was related to both rotation of the arm on the shaft and internal gear lashing or slop in the servo itself. The measurement you're observing in the arm is very similar to what I'm seeing.

The way I test it is to engage the AP in test mode, then while engaged I crawl back and get up close and personal with the servo. I then just push and pull on the elevator pushrod and can see the movement of the arm on the servo.

I ran the numbers and I should be able to move my pushrod to the middle hole on the servo arm without getting close to an over center condition, so that's on my list of things to try.
 
Folks,
If you inspect your servo arms, check for looseness in all Axis?s relative to the shaft. If the arm is sloppy it will continue to loosen the shear screw.
Both of my servo arms were a tad too loose.

A note fron Dynon for guidance....

"A note on dynon servo castle nuts

While the castle nuts should not be loose they must not be torqued above 4.5 in-lb. Overtorquing the arms will increase the force needed to break the shear screw and restrict the freedom of movement of the arm in the event the screw needs to be broken."
 
I think what needs to be done is this:

Tighten the castle nut to the next hole. This will take about 40inch on a torque wrench. What you say? Yup, if this arm is loose it needs to be known if it really is the problem. So I say tighten the nut and run another test, if there is no more slop. Than I say go fly it and see what it does. If it has no more oscilations than we know what needs to be done. Sheer pin won't break until 140inch pounds oh well, use two hands on the stick. At least Dynon will know what's going on and a fix can be gotten.
 
You must log in or register to reply here.
Back
Top