Reducing dutch roll on the FT Dart.

[slembcke]

I got an FT Dart put together a couple weeks ago. I substituted some Emax 1306 3300kv motors w/ 3x3x4.5 Dal props I had laying around, and added a RMRC Dodo flight controller. The AUW is ~250 grams. Before I tuned the flight controller, this plane was *really* hard to fly as it is unstable in the roll axis if I fly it at less than 2/3rds throttle. I've tried sticking a couple quarters in the nose to really push the CoG forward, but it doesn't seem to help. It's definitely not nice and docile like the Mini Arrow is.

I *think* that it's dutch roll, but this is the first time I think I've experienced it to a large degree. The part that makes me unsure is that the frequency is pretty high (like maybe ~5 hz?), and it builds quite quickly until I give it full throttle and greatly increase the airspeed. Does anybody else have issues like this? It's borderline unflyable without the flight controller stabilization turned on. I'll try and pull some example footage from my DVR later tonight.

Bruce Simpson has a video on flying wing aerodynamics and suggests adding wing fences. I'm considering trying that, but it still seems odd that it flies so poorly as is. The bigger motors and FC add maybe 10% to the weight, but in the review video they say "it's a nice gym flyer on 2s". Doesn't seem like it should make that much difference.

 

[Hai-Lee]

I got an FT Dart put together a couple weeks ago. I substituted some Emax 1306 3300kv motors w/ 3x3x4.5 Dal props I had laying around, and added a RMRC Dodo flight controller. The AUW is ~250 grams. Before I tuned the flight controller, this plane was *really* hard to fly as it is unstable in the roll axis if I fly it at less than 2/3rds throttle. I've tried sticking a couple quarters in the nose to really push the CoG forward, but it doesn't seem to help. It's definitely not nice and docile like the Mini Arrow is.

I *think* that it's dutch roll, but this is the first time I think I've experienced it to a large degree. The part that makes me unsure is that the frequency is pretty high (like maybe ~5 hz?), and it builds quite quickly until I give it full throttle and greatly increase the airspeed. Does anybody else have issues like this? It's borderline unflyable without the flight controller stabilization turned on. I'll try and pull some example footage from my DVR later tonight.

Bruce Simpson has a video on flying wing aerodynamics and suggests adding wing fences. I'm considering trying that, but it still seems odd that it flies so poorly as is. The bigger motors and FC add maybe 10% to the weight, but in the review video they say "it's a nice gym flyer on 2s". Doesn't seem like it should make that much difference.

Mind you I have not built or flown a DART but I have designed and flown a number of flying wings of different configurations. So this post is based on things that are my experience elsewhere!

I do not think that you are suffering from "Dutch" roll but are actually having an effect of asymmetrical thrust or at least non-linear thrust from the motor setups. You could easily increase the wing tip plate sizes to provide more directional stability.

Just what i would check for!

Have fun!

 

[thenated0g]

In my experience this kind of wobble can be solved by adding some nose weight.

 

[Hai-Lee]

One other thing I would check is that you added heavier motors and so it might be possible that the motor weight is causing some wing twisting due to the leverage of the extra weight so far forward of the wing. At low speed the motors could effectively be "Bouncing" and changing the wing profile. At full throttle the "Pull" of the props would seriously reduce the effect.

Just something else to check!

Have fun!

 

[Arcfyre]

I would also check to see that your flight stabilization is correcting in the right direction. Meaning if you bank the plane to the left by hand, make sure the right aileron pulses up and not down. I've made this mistake before.

 

[TEAJR66]

Fly it faster.

 

[slembcke]

Here is a 4x time stretched DVR capture of the issue.

The phase of the yaw/roll seem to agree with dutch roll as I understand it at least. Yaw to the left, right wing rises, yaw to the right, left wing rises, repeat.

To be clear, this is with flight stabilization off. The motors are a few grams heavier than the stock ones, and mounted on those plywood pods. The whole thing is quite rigid given it's size, so I don't see how it could be flexing much. I've tried moving the CoG forward up to a full centimeter past the recommended point. That just made it nose heavy and unstable.

I guess I'm going to try increasing the size of the tip fins see if the extra yaw stability helps. Might also try enabling electronic stabilization on only the yaw axis to see what sort of effect that has. I usually leave yaw stabilization really low in iNav since it tends to fight a plane's natural tendency to make coordinated turns.

 

[L Edge]

Are you aware of the Aileron differential function in your radio?

It is used to reduce unwanted yaw characteristics during roll inputs. Suggest you try that and see if it works for you. I first started to use it when doing pattern competition on doing 3 successive roll as well as my scratchbuilt X-47B. That is why it is a built in mixer in a radio. Most pilots are afraid to use it. Make sure your CG is on target before you start.

 

[slembcke]

Sure, I've used that a little. Adverse yaw is maybe one of the initiating factors. In the video above I had just straightened out from a bank when it started wobbling.

The more immediate problem is that with neutral inputs the plane is unstable, and the wobbling increases until I speed up to nearly full throttle or enable electronic stabilization.

 

[Hai-Lee]

It would be easier to see what it happening if you had an external or ground based video of the issue. The supplied video does not show AOA or anything else other that a rolling action.

A better view of the problem may result in a quicker fix!

Have fun!

 

[L Edge]

Sure, I've used that a little. Adverse yaw is maybe one of the initiating factors. In the video above I had just straightened out from a bank when it started wobbling.

The more immediate problem is that with neutral inputs the plane is unstable, and the wobbling increases until I speed up to nearly full throttle or enable electronic stabilization.

A big question for you, how do you fine tune your CG, on the ground or in the air?
Definetly need to see flight video from ground, especially need to see it in both right and left hand turns. That is say 1/2 throttle and full open. That will show what is going on.

 

[slembcke]

I did try adding some wing fences and extended the fins with some tag board taped on to them. That helped significantly, it's still wobbly, but flyable at half throttle now. I'll make some more bigger/more rigid fins next and report back.

It would be easier to see what it happening if you had an external or ground based video of the issue. The supplied video does not show AOA or anything else other that a rolling action.

A better view of the problem may result in a quicker fix!

I hadn't considered that. Hard plane to record since it's so small and fast.

A big question for you, how do you fine tune your CG, on the ground or in the air?

Dumb question, but how do you tune the CG in the air? I usually just put quarters or washers on the nose of my planes to adjust it.

 

[Merv]

Could you have a twisted wing? That's how the Wright Bros controlled their first plane, by twisting the wing.

 

[L Edge]

I did try adding some wing fences and extended the fins with some tag board taped on to them. That helped significantly, it's still wobbly, but flyable at half throttle now. I'll make some more bigger/more rigid fins next and report back.



I hadn't considered that. Hard plane to record since it's so small and fast.



Dumb question, but how do you tune the CG in the air? I usually just put quarters or washers on the nose of my planes to adjust it.

Do you balance your plane lengthwise also?
To fine tune the CG, you fly inverted, if you have to push your elevator forward more than a tad you need to move the battery back until a gentle push is needed.
Then you have to take a look at turns as to what you do next. Does your wing drop?

I do 3D quite a bit and you need an aft CG to fly where I use this method as well. Also, when I did F1 pylon(a football length per sec) we had to have CG (static and dynamic) so plane flew straight and in turns (no wing drop) so thats why I wanted to see external video. It may be your problem. That's where differential can help you because of uneven drag issues.

I scratchbuilt and designed a X-47B with no vertical rudders and used the same method to get it to fly and turn proper.

 

[Hai-Lee]

Sure, I've used that a little. Adverse yaw is maybe one of the initiating factors. In the video above I had just straightened out from a bank when it started wobbling.

The more immediate problem is that with neutral inputs the plane is unstable, and the wobbling increases until I speed up to nearly full throttle or enable electronic stabilization.

OK. This may be a little long but there is a setup catch when setting up the CG on a wing that uses a normal, (not reflexed), wing profile and when it is heavily loaded.

I check and adjust the CG or balance point based upon flight performance for a smooth and controlled stall.

You get the wing flying flat and level, (with a bit of height for safety), and slow the speed trying to keep it level by increasing the elevator input until it is at maximum. If when flying at the stall point the wing starts bouncing its nose up and down in a stall, recover, stall, recover, etc pattern then the wing is nose heavy.

If at stall the wing wanders from side to side with great directional instability then the wing is definitely tail heavy.

Ideally the wing should just settle into a wing level glide that is smooth and controllable, (a minor amount of nose bouncing is acceptable if smooth cannot be obtained, or is too fiddly or difficult to obtain.

Once the balance is obtained any wing diving or nose down tendencies will require you to increase the reflex angle to provide sufficient lift for the wing at its current weight. Conversely any nose up tendency will require a reduction in the reflex amount.

The increase or reduction in reflex angle is normally obtained by elevator trim unless the trims max out in which case the change will need to be done mechanically by pushrod adjustments.

When you build a wing it is the reflex that determines the amount of lift and hence the amount of weight the wing can support, and the CG that determines the wings stability in low speed flight and at stall.

Just what works for me!

Have fun!

 

[metaldirtnskin]

I know this is an old thread but I was just rebuilding my Dart and tinkering with it, so...

Which motor spins in which direction? Since yaw is done with differential thrust (when set up as FT describes), then to yaw to the right, the left motor spins faster. And if the left motor rotates to the right (counterclockwise as viewed from behind), then when it spins faster the torque will twist the airframe harder to the left. So right yaw will produce left "adverse roll". With the left motor spinning to the left, you would get "proverse roll" - left yaw produces a bank to the left.

I'm wondering if the flight controller is just fighting this adverse-roll tendency and producing the ~5Hz oscillations - just a half-baked theory, I guess.

 

[slembcke]

Interesting thought, but the stability problems were with the stabilization turned off. It was fairly flyable with some tuning, but only if you made very gentle maneuvers.

Anyway, I was able to fix my problems by making the tip fins bigger as posted above. I also added some small wing fences, but I suspect they don't do much.