mclarkson

I’m having trouble with my new flying wing and I’m at a loss for a solution.

The problem in a nutshell is this – as it picks up speed, it pitches down. (And as it pitches down, it picks up speed.) Unless I catch it *immediately*, this self-reinforcing behavior quickly gets out of hand and ends with it lawn darting into the ground at high speed.

This is *not* a motor angle problem; it happens in an unpowered glide, as well.

The wing is pretty big – 60” – if that makes any difference.

It’s using a relatively fat airfoil, but I don’t think it’s an airfoil problem; I’m using a version of this same airfoil on three other flying wings in my hangar.


When I first tossed the plane up, I had the CG pretty far forward and it felt very nose-heavy. It required ridonculous amounts of up reflex and all the elevator I had to keep the nose up enough for level flight. I’ve been slowly moving the CG back and the flight characteristics seem to be improving as I do so – though it still requires a lot of up reflex – but it has retained this one fatal habit.

In two sessions, I’ve put it nose-first into the ground, hard, from pretty significant altitude. I’d like to stop that before it’s destroyed beyond all recovery.

I’m hoping there’s a rule of thumb I’m unfamiliar with that diagnoses this problem. Any thoughts?

Here’s some video from the first flights. You can see the problem start to develop several times before it gets away from me and the plane finally augers in.

fhhuber

sweep angle and airfoil choice... the tips are developing more lift with speed, driving the nose down.

mclarkson

Interesting. So ... less sweep would be better? If airfoil choice is a problem what, exactly, is the problem? Too fat?

JetPlaneFlyer

It's the airfoil. Flat bottom 'Clary-Y' type airfoils have a very strong nose down (or negative) pitching moment. You can often 'fudge' the problem by adding lots of reflex on the elevons but that's not a good solution even though on some models it might appear to work ok. When you get to high reflex angles you get problems with flow separation on the elevon which causes unpredictable behavior as well as being very inefficient.

For swept wing you can use lots of twist (washout) but as a rule for a tailless design you should be using a tailless airfoil that has some reflex built into it's camber line to give it a neutral or slightly positive (nose up) pitching moment. Symmetrical airfoils also work ok on swept wings providing efficiency is not your main concern..

One of the MH 6X series airfoils would be good along with a few degrees of twist: http://www.mh-aerotools.de/airfoils/mh60koo.htm

Some good reading:
http://www.mh-aerotools.de/airfoils/nf_1.htm
http://www.mh-aerotools.de/airfoils/flywing1.htm

JetPlaneFlyer

No. less sweep would make the need to use a genuine reflexed airfoil greater. Sweep is generally your friend on flying wings but you need to combine it with twist and a suitable airfoil. To be honest the airfoil you are using is close to the worse imaginable for a flying wing, that's the root of the problem.

Read the links on my last post to Prof Hepperle's web site.

mclarkson

Yeah - that sounds like me!

Thanks for all the info. It's appreciated.

Nutball

So my guess it either a thinner airfoil at the wing tips, you know how most delta wings are thick at the middle and get thinner, or move the CG back and put a gyro on it.

Maybe the thickest part of your airfoil is too far back allowing the shape of the front of the wing to generate too much downward force. Adding some curve to the bottom to help cancel out the pitching moment caused by the upper surface may be the best solution.

I've made successful gliders with the same airfoil as you, but they could do crazy things if the speed and balance were not correct.

soarrich

You're going to think I'm joking, but reverse the elevator and fly it inverted you'll be shocked at how well it will fly!

I built a delta out of blue fanfold, it would tuck just like your wing does when flown with the curved side up, with the flat side up it flew great.

PS I'm looking for pictures of it, I'll post them if I find them.

fhhuber

Yes. It will be better... because it will nose-up with speed. And the elevons will need less reflex.

As strange as that may seem it will work.

For a soaring style wing, sacrificing a bit of aerobatic ability, its fine to have the high lift airfoil at center, but the airfoil should transition to symmetrical or a reflexed airfoil at the tip, being near symmetrical by half distance between center-line and tip. (Elevons can provide the reflex from the symmetrical airfoil.)
That sweep angle is good for this style flying wing.

quorneng

The big Northrop YB49 used a NACA 65-019 section and it does not look much like the one you are using!
http://www.airspace.cz/akademie_lete...ACA-65-019.jpg

mclarkson

So ... here's what I don't understand. (This may sound confrontational, but it's not.)

Why do I have four different wings using that same airfoil that fly just fine? One of them is among my best-flying planes ever.

What about the millions of flying wings out there using something like an Armin wing? Or the Flite Test Versa flying wing? Really simple sorta-airfoils that seem to work just fine.

fhhuber

Obviously SOMETHING is different...

Maybe you built significant washout into the other wings.

Winglets on the tips being angled different could do it.

mclarkson

Indeed, something is obviously different.

soarrich

It's fine to ask someone to justify the response.

I haven't seen your other wings, but from watching this one's videos you may be suffering from control blowback. You seem to have very large elverons with lots of throw, but small servos that are very fast. This leads me to think they don't deliver much push, that would allow the airloads to over power the elverons. I never looked into this as the reason for my delta tucking, my delta was meant to fly with the flat side up, but it may have been the reason it tucked.

Like I said, it's probably a servo issue. I've built three in the last couple of weeks and they fly fine.
Try flying your wing again, but roll it upside down trim it for level flight, then dive it, and see what happens.

mclarkson

First, I've gotta put it back together again.

JetPlaneFlyer

1. The airfoil on this one looks very thick with the maximum thickness and camber placed well back. With a flat bottom the thicker airfoil means more camber which for a flying wing is bad, as is having maximum camber toward the back.
2. Second you only have moderate sweep and you don't use any twist, some of these other designs may use more sweep and/or twist? With lots of sweep and twist you can make a flat bottom airfoil work reasonably ok.
3. Third size matters. Due to Reynolds Number effects what you 'get away with' on a small wing might not work the same on a big one.

I know that tailless airfoils like the MH-60 are not as easy to build but you will be rewarded with a much better flying wing, that's 100% guaranteed. You can simplify the airfoil shape with minimal impact on performance by keeping the same mean camber line and tweaking thickness so that the top-rear part of the airfoil becomes a flat surface, then you can build the wing inverted.

I can draw up a simplified airfoil if you like?

mclarkson

Sure. I'm afraid I only follow about half of what you're saying.

That actually makes a lot of sense, especially given the amount of up reflex I'm currently flying.

mclarkson

I've posted this in a couple of places. So far, I've been told the problem is or may be:

• Ailerons are too small
• Ailerons are too big
• Servos are too small
• Wrong airfoil
• CG is too far forward
• CG is too far backward
• Motor thrust angle is wrong
• Sweep angle is wrong
• Winglets at wrong angle
• No washout

fhhuber

All are possible contributing factors.

The airfoil choice, washout and sweep angle interact with each other. For a different airfoil that's a great sweep angle. Add washout and it would be fine with that airfoil...
It appears really close to the sweep of the Nitroplanes Falcon 120, (BVM Bobcat clone) which can have the booms and horizontal DELETED if you relocate the vertical stabs. That plane has a near symmetrical airfoil. Programming the ailerons for elevon works even with the booms and horiz stab still there and the elevons more pitch authority than the OEM elevator.

Winglet angle is a bit if a trick... You can mount the winglets in a manner that adds some downforce, replacing the washout.

quorneng

mclarkson
Perhaps what you have identified is that flying wings are tricky.

I find conventional airframes hard enough to get 'exactly' right.

JetPlaneFlyer

That's why understanding the aerodynamic principles by reading and digesting the pages on flying wing design i linked to (written by a genuine professor in aerodynamics) is better than blindly accepting all the advice you hear in forums.

Start on this page: http://www.mh-aerotools.de/airfoils/nf_1.htm

JetPlaneFlyer

FWIW, I quickly knocked up this airfoil.. lets call it JPF-01



I checked out it's characteristics in Xfoil and it has very close to zero pitching moment, so it would work well on a swept wing with a shade of twist or with very slightly reflexed elevons.. Or on a 'plank' wing with very slightly reflexed elevons. It's not going to break world records for performance but it should perform far better than the flat bottoms you have been using.

I made it 'mainly' flat bottom and with a thick trailing edge so to be easier for foam construction.

JetPlaneFlyer

Here are the polars for JPF-01 vs some other tailess airfoils and with a 'Simplex12' flat bottom airfoil that's close to what you use. Considering JPF-01 is designed more for easy building than for performance it does ok in these simulations if i say so myself.

The important one as far as stability or the wing is concerned is the Cm(alpha) graph, which shows the pitching moment (degree on nose up or nose down pitching effect). Note that 'JPF-01' has close to zero Cm for the full angle of attack (alpha) range as do the other tailless airfoils. On the other hand if you look at the flat bottom 'Simplex12' you will see how it has VERY strong negative (nose down) pitching moment at lower angles of attack. This explains why when your wing flies fast (so AoA is low) it tries to pitch nose down.

Here's the Cm(alpha) graph:

mclarkson

Thank you, sir.

mclarkson

Maybe I'm exceeding the wing's critical mach number!