Cyclic drift

[dbennettya]

Last flight I noticed that if I put the heli in a stable hover and took my fingers off the sticks the heli would roll to the right and increase speed like I was inputing right aileron. At zero throttle or throttle hold the swash is level. So to troubleshoot the problem I disconnected the main and tail motors. Then after binding I raised the throttle and sure enough the swash raised to increase pitch but with no cyclic input the swash slowly drifts to the right. Both pitch and aileron servos move together so it isn't a servo problem. If I drop the throttle or hit throttle hold the swash levels as it should.

So I pulled my #2 hardly used mCP X and tested it also. It has the same exact issue. I even tried rebinding the TX to the RX and the issue remained. So I added left aileron trim until the swash doesn't drift. This took about half of the trim range to cancel the drift. Now when I setup a stable hover and don't touch the sticks it will hold position for a long time > 10 seconds. It is interesting that both my mCP Xs need the same amount of trim for no drifting of the swash

I just wonder if my new DX-8 aileron channel has drifted/shifted or something. The only things I use the DX-8 on are my mCP Xs. Has anybody else seen this issue?

Thanks, Don

[Ampdraw]

Hi Don,
Not sure how the rx reads tx signal on bind. Subtrim read as zero or does it look for centre pulse?
Perhaps try subtrimming a channel heavily, cycle rx power and see if you still see the drift.
You could also look at the monitor screen on the Dx8 and leave it powered for a while.

Not sure if my mcpx does it. Haven't looked at that aspect...yet!

There was a factory calibration mode that was not documented for the DX7 but it allowed you to adjust all sorts of paramiters.

Edit: Just thought of something....the gyro chip on the board is on a slight angle so that it sits hrozontal in flight when the heli has a list due to the tail. Perhaps this has some effect when on the bench??

Amp

[dbennettya]

Amp,

I'm pretty sure the TX is ok as it doesn't drift while running according to the monitor screen. Zero shows right on according to the monitor with no trim. When set for no drift the monitor shows 3% due to the trim. I don't think subtrim wouldn't allow enough range to cancel the drift.

I'm pretty sure on bind the RX reads all channels for zero so the failsafe will go to safe zero settings.

I thought of the gyro tilt issue and that is why I tried getting into a stable hover, then not touching the sticks to see how long it would hang. Without the trim offset it didn't hang for long, but with the trim it hangs for > 10 seconds.

Maybe others can try this to see if it is trend.

I did notice something strange on the monitor screen. When moving the elevator channel the pitch channel also moves? I verified that it is set for 1 servo swash but it still moves the pitch channel? Is this a bug in my DX-8? I am running the latest firmware.

Thanks, Don

Originally Posted by Ampdraw

Hi Don,
Not sure how the rx reads tx signal on bind. Subtrim read as zero or does it look for centre pulse?
Perhaps try subtrimming a channel heavily, cycle rx power and see if you still see the drift.
You could also look at the monitor screen on the Dx8 and leave it powered for a while.

Not sure if my mcpx does it. Haven't looked at that aspect...yet!

There was a factory calibration mode that was not documented for the DX7 but it allowed you to adjust all sorts of paramiters.

Edit: Just thought of something....the gyro chip on the board is on a slight angle so that it sits hrozontal in flight when the heli has a list due to the tail. Perhaps this has some effect when on the bench??

Amp

[TheQuixote]

Similar weird problem.
At the beginning of every flight helicopter is fine. By the end of the flight the helicopter has a severe backwards drift that takes 3 trim clicks to level out. Put in a new battery and I need to reverse the trim back to normal. At first I thought it was the 3-1 but I have the same behavior on a second helicopter which would seem to point the finger at the TX.

DX6i, monitor looks good and don't see this behavior on the sim so I'm pretty baffled by it.

[dbennettya]

Mine is different as it is an offset that lasts the entire flight. I had to retrim the heli for a verticle takeoff. At least now if I leave it on the ground spinning the main rotor it doesn't want to tip over like it used to.

YMMV, Don

[RedOctobyr]

Originally Posted by TheQuixote

Similar weird problem.
At the beginning of every flight helicopter is fine. By the end of the flight the helicopter has a severe backwards drift that takes 3 trim clicks to level out. Put in a new battery and I need to reverse the trim back to normal. At first I thought it was the 3-1 but I have the same behavior on a second helicopter which would seem to point the finger at the TX.

DX6i, monitor looks good and don't see this behavior on the sim so I'm pretty baffled by it.

That's very interesting, I have developed the same behavior on mine. Fine at the beginning of a battery, but I need to hold forward cyclic by the end, to hover. And if I unplug the battery and plug it back in, it will start flat again, then it will gradually develop its rearward drift again.

Monitor output on my DX6i is also fine, sim is OK, I'm pretty sure it's not the TX. Horizon sent me a new 3-in-1 (which looks used; the antenna wire already has a bunch of bends in it, and there's paint on several parts of the board). But I have not had time to install it yet.

[hydro_pyro]

Level your swashplate using the servo link heads. Adjust until it takes off straight up with a fairly quick punch.

[docz]

I may be crazy but I observed that my chopper was tipping quite a bit on take off last night on rebinding. I had a dim memory of reading somewhere that the binding process could have something to do with this so I did some tests with my DX8.

1) Hold bind button for several seconds on powering on transmitter.... then power off everything and start up as normal.... terrible tipping over even with minimal throttle (right aileron and forward pitch).

2) Hold bind button for 4-5 secs whilst binding then try take off - mild tipping.

3) Hold bind button for just 2 seconds whilst binding (as per manual instructions) stopping as soon as you see 'bind' on the Tx - no tipping??

I have no idea why this would be the case. I would have thought that the binding process on computeriased Tx would just be a link up rather than anything affecting performance.

I know so little about these things that I could understand if in fact I am observing other phenomena and just misinterpreting the cause - would appreciate if someone could confirm if this is a real thing... and if so... why on earth does it happen?

Cheers.

[TheQuixote]

Originally Posted by RedOctobyr

That's very interesting, I have developed the same behavior on mine. Fine at the beginning of a battery, but I need to hold forward cyclic by the end, to hover. And if I unplug the battery and plug it back in, it will start flat again, then it will gradually develop its rearward drift again.

Monitor output on my DX6i is also fine, sim is OK, I'm pretty sure it's not the TX. Horizon sent me a new 3-in-1 (which looks used; the antenna wire already has a bunch of bends in it, and there's paint on several parts of the board). But I have not had time to install it yet.

I thought it was the 3-1 as well, one of the reasons I justified just getting a whole new mCPx instead of a handful of parts. Same problem with the new one.

What I've noticed is it's much more severe depending on the type of flying I'm doing. If I just work on slow piro hovering there is no problem. If start doing some indoor micro FFF which results in a lot hard stick banging it gets really bad.

I'm new to the hobby so it would be nice to know if anyone has had similar problems with different heli or tx combinations. The two possible culprits still seem to be the DX6i and the 3-1. Now that I know some one else has the issue I might start working on this again instead of dealing with it by re-trimming mid flight. Thinking additive software bug on the 3-1 that's triggered by certain conditions.

[hydro_pyro]

Like I said, check the swashplate. If it's not 90 degrees level to the main shaft, it's not going to take off straight. The 3-in-1 doesn't know where the swashplate is at, it only knows to center the servos on startup.

This heli should NEVER require trimmming on the Tx.

[RedOctobyr]

hydro_pyro, not sure who you're replying to? docz with his tipping problem, or TheQuixote any myself, with increasing amounts of cyclic drift during a flight?

I don't think that my cyclic drift has anything to do with my swashplate. Which, FWIW, is level.

Similarly to what you said, TheQuixote, it feels like it may vary by how I'm flying. But it's tough to tell. I tried trimming it out one time, but forgot to remove the trim before the next battery, and then it was trying to tilt forward immediately. That was a mess, so I just keep holding more forward-cyclic as the battery progresses, to keep it from trying to tip backwards.

I don't recall it doing this when it was new. I just noticed it one day, and it's been like that since. Seems to me like one of the gyros is getting confused (a gradual accumulation of small errors?), and unplugging the battery helps temporarily fix it.

[dbennettya]

Ok, for the record both my helis need trim so the cyclic doesn't drift. Since I only have one TX I can't verify that it isn't the TX causing the problem. Rebinding doesn't fix it and rebinding with the trim left in doesn't affect the swash level either. This is very easy to test for, just disconnect both tail and main motor. Then turn on the TX wait 10 seconds and plug in the battery on the heli. Then quickly set it on a flat surface. After the blue light comes on solid move the throttle up to normal flying position ~3/4. Now watch the swash and servos, if they stay absolutely still you've got a winner. On both mine the swash will tilt to the right slowly with both pitch and aileron servos slowly turning. Setting the stick to 0 will reset the swash to level so you can repeat the test while adding trim. Left aileron trim will cancel this out on both of mine and it takes the same amount. The trim has no detrimental effect on the helis operation.

I would like to get a few more folks to try this and report back.

Thanks, Don

[Airworks]

ive noticed the same with mine end of flights , it doesn't want to center well , but oh well , think of it as a flybar


by adding trim to a FBL heli , your actaully giving it a "stick input" not trimming the servos. so it will only be worse "rolling out of level " with any trims added , that goes for subtrims also(as eflite setup has no "built in" unit feature for subtrimming).
Get your mechanical setup correct on center stick, 0 pitch, level swash.

if your pluggin in your heli in your hands and not setting on a flat surface for it to initiate, could solve some issues also by doing that.

dbennettya,,,,what tx you using? dx6i or dx8, could be the tx.

[hydro_pyro]

Originally Posted by RedOctobyr

hydro_pyro, not sure who you're replying to? docz with his tipping problem, or TheQuixote any myself, with increasing amounts of cyclic drift during a flight?

...I don't think that my cyclic drift has anything to do with my swashplate. Which, FWIW, is level...

...unplugging the battery helps temporarily fix it.

Your issue sounds a lot like the voltage-dependent tail gyro drift typical with the Blade mSR. You have an issue with your Rx board. It should NOT drift. HH gyro and flybarless system should both hold "to the death" (until it can't keep up with torque/blowout) Send it in for warranty service.

Originally Posted by dbennettya

Ok, for the record both my helis need trim so the cyclic doesn't drift....

Yours is not normal either. This flybarless system should NEVER need trim if the swashpplate is leveled, and it should NOT "drift." The heli should maintain a heading attitude in any directiono with no wind present.

[Ampdraw]

Hi Don,
I'd test mine but the main is bl and hardwired.

Guys few different issues happening in this thread.
Don's is cyclic drift...similar to tail drift as the voltage varies on a MSR tail for those of you that are familiar with that.

Don, re your dx8 pitch value changing with elev input...that does not seem right. I'd ask in the Spek section of this forum.

Wonder if your cyclic drift is temp or voltage related...could test for both.

Also maybe after the heli has bound, sit a skid on an object so the heli is at a lean and see of the drift occurs when the gyro is trying to hold another angle. May help get closer to somewhere!

Oh...and if the rx sees whatever value signal as centre and zeros to that on bind then you can pretty much exclude the tx as a source of the drift as long as the channel isn't drifting in the monitor.

Amp

[dbennettya]

Amp,

Actually the cyclic doesn't drift after the aileron trim is set to 3% according to the monitor. It is rock solid and flies great with the 3% trim. The trim doesn't affect any flight behavior, it just makes the swash stay where it is put with no drift. Without the trim the TX doesn't drift but both helis swashes slowly lean to the right when you move from zero throttle and sitting on a flat surface or flying in a stable hover. Remember it is rock solid with the trim from start to finish of the flight so it isn't a voltage or temperature thing.

I just retested and the cyclics don't move the pitch channel. The throttle moves both the throttle and pitch channels. Is this right for a 1 servo setup?

I know a 3 axis gyro should not be given any subtrim or trim, but this fixes my drift issue without any negative side effects. I'm going to post on the DX-8 or specturn forum to see if I can get any feedback there.

Thanks, Don

[ss4344]

I know it's an old thread to resurrect (apologies for that), but I have observed some behaviour that might add to the discussion.

I too see a steady collective drift to the right on first power up with my mCPX (v2) that needs some cyclic applied to get off the ground straight up. Interesting thing I have discovered is that if I apply power to the heli and leave it sitting for three minutes without throttling up, then the drift all but disappears.

It is a gradual thing. The sooner that I try and power up, the worse the drift is. Leaving it for longer and longer before power up, the less the drift gets. At about the three minute mark, the drift is all but gone. Ironically, this is about the duration that the stock batteries normally last.

Funny that it took my main motor mosfets failing to pick this up... (Very impressive smoking landing, btw!) I was able to give the little bird some serious maintenance while waiting for replacements to make their way across the pond. Because the blades were not spinning while testing servo operation in all circumstances I was able to clearly pick up the drift early on, and also observe that it went away after a longer period of testing.

I'd be interested to hear from others if they see the same thing.

Cheers,
Steve

[dbennettya]

Sounds like yours may be temperature related, not sure. Have you checked the TX for drift in the monitor?

Good Luck, Don

[ss4344]

Don't think so, Don. The little guy does exactly the same thing whether I bind it to my DX6i or DX4e.

Monitor is stable on the DX6i, and it's nicely calibrated.

Sent from my GT-I9100 using Tapatalk 2

[maarset]

Are we talking V1 or V2 that has cyclic drift?


Sent from my iPhone using Tapatalk