C.G. Elevon control question

[poulsbobill]

I scratchbuilt one of these (beegee clone). I have the cg at about 32%, so it flies at extremely high alpha (by choice for a small space). I have rudder and elevons.

My question: The rudder controls it just fine...the probelm is the elevons wont turn it. As i increase aileron stick it noses higher up with a slight turn and then noses down as i increase the throw. The front left of the craft pitches down in a left turn and vice versa ( almost like a tip stall--but is this possible in such high alpha?). Is this the rearward cg out "dominating" the elevons? or?

I have flown it with a more forward cg and as you might expect with the aspect ratio of the wings it is quite twitchy. Also the elevons wont turn it in a more normal alpha flight... ok, I can turn it extremely gradually -but anything more and i get an immediate dive.

It is laterally balanced a little bottom heavy.

Curious as to a fix but also the science here. Any thoughts?

Thanks,

Bill

[luvF3b]

Hi Bill

It sounds like the problem is related to the differential movement of the ailerons. What I think is happening is that application of aileron pitches the model to higher alpha, where it stalls, and the nose drops. You are getting cross coupling from aileron (roll) to pitch. When the CG is changed, a new elevator neutral will be required to maintain level flight, and with a new elevator neutral the roll to pitch coupling will change, and so a different differential setting will be required.

For the current CG position (which does sound a long way back), I'd try dialing in some reverse differential from your current setting, ie increase the amount of downgoing aileron, but keep the up aileron movement the same.

I'm not convinced that a rearward CG is needed to achieve a high alpha condition. High alpha could be just as easily be obtained with a forward CG and up elevator trim, with trade off of softer elevator control and some dependancy of the flight attitude on speed of the model (ie throttle) - but hey - I fly gliders what would I know about throttle?

John

[poulsbobill]

Hi Bill

It sounds like the problem is related to the differential movement of the ailerons. What I think is happening is that application of aileron pitches the model to higher alpha, where it stalls, and the nose drops. You are getting cross coupling from aileron (roll) to pitch. When the CG is changed, a new elevator neutral will be required to maintain level flight, and with a new elevator neutral the roll to pitch coupling will change, and so a different differential setting will be required.



For the current CG position (which does sound a long way back), I'd try dialing in some reverse differential from your current setting, ie increase the amount of downgoing aileron, but keep the up aileron movement the same..
John,
I have done the exact opposite as when i had used the ailerons to turn in the beginning i would dive. I see that what you are saying is causing the dive during hi alpha. Seems like maybe i went too far...a balance here. ??

I'm not convinced that a rearward CG is needed to achieve a high alpha condition. High alpha could be just as easily be obtained with a forward CG and up elevator trim, with trade off of softer elevator control and some dependancy of the flight attitude on speed of the model (ie throttle) - but hey - I fly gliders what would I know about throttle?

John

You are right of course... But, the area i am flying in ( a driveway) is just so small i am trying to make hi-alpha/hover the norm and all other flying a choice, so hands off stick reverts to minimal movement.

I will try your ideas though...that was the point of this!! :)

Thanks,

Bill

[JetPlaneFlyer]

What you have there is a low aspect ratio 'plank' flying wing. On any flying wing the CG MUST be ahead of the 25% chord point to be stable. The fact that the span reduces slightly toward the rear will mean that the CG needs to be even further forward still. I'd try a CG at between 15 and 20% and fine tune from there. for normal flight both elevons will need to be 'reflexed' a little i.e. a little up elevator will be needed for normel flight.

[Brandano]

Consider also that there will be massive elevon blanking by the wing at high AOA. the plane will still be fliable, thanks to the low AR and the vortexes that form on the upper surface of the wing, but the elevons will most likely lose some effectiveness. I'd add a little extension to the flying surfaces over the edge of the wingtips, and perhaps take advantage of that to add a balance horn.

[Bobber Bob]

HI folks...

JetPlane is right, for figuring the "25%" cg point on a flying wing (non sweptback), you have to "throw away" the rearmost 20% of the total area, and figure your "25%" on what remains.

On a non swept flying wing, the rear most 20% is the "tail" (hortizontal stabelizer), as there really is no such thing as a "tailless" airplane. Some of them are so "short coupled", that the tail becomes part of the wing's structure.

Just the ramblings of an old man, but I hope there may be something useful there.

Bob Reynolds
. "ComeUpHere"

[poulsbobill]

Gentlemen,

Thank you all for the responses. Not what i wanted to hear :mad:

But by popular consensus i moved the cg XXXX (actually forward!!) some and lo and behold you are all right!!!!!! Thank you!!!! :)

The elevons worked more like planes i have had before and the plane was much more responsive...although harder to hover and a much faster flyer. I guess i will need to decide how i want it to fly...amazing how different it flies by just changing the cg 3/8"

I remeasured the original cg and to the elevon line and it was 36% (not the 32% i had said) and i was able to move it to the 29% mark. (that is with the battery all the way forward as i could get it). I hate to add additional weight but this is much better already!! (i now have that excuse to go to 3 cells!!)

I am still trying to get my head wrapped around as to why such a great difference in flight characteristics though.

Bill

[JetPlaneFlyer]

Good that the CG shift worked... BTW; the 15 - 20% CG position I advised was of chord including the elevons (this is the normal way to measure chord) so it sounds like the position you have now is in this range.

I suspect previously due to the CG being too far aft the model was permanently pitched up into a stall and most of the lift was coming from prop thrust. As the wings were not 'flying' in the conventional sence then hardly any wonder the control response was not what you would expect.

Steve

[poulsbobill]

Good that the CG shift worked... BTW; the 15 - 20% CG position I advised was of chord including the elevons (this is the normal way to measure chord) so it sounds like the position you have now is in this range.

I suspect previously due to the CG being too far aft the model was permanently pitched up into a stall and most of the lift was coming from prop thrust. As the wings were not 'flying' in the conventional sence then hardly any wonder the control response was not what you would expect.

Steve
Steve,

Yes i was in a 70 degree high alpha all the time...it is fun that way for a small area. The rudder worked great but the elevons as ailerons do not work as expected. Pitch up and then dive or as i tried a gradual turn...a small turn and then dive. It just seems to me that if the rudder works as intended then the ailerons as such would too. ??

Bill

[BMatthews]

Given that this is a 3D style model you really should have an aggresively rearward CG but as you found out your orginal 32% was a lot over the top on this aspect.

For those that are saying that you should have it even further forward to around 20% I have to say "no". Yes it would be a good way to go if this was a regular flyer. But for a 3D style that will be doing lots of hovering and waterfalls and harrier'ing and other ing'ings you want to keep it around the 25% position give or take a couple of %. Especially given that the best prop hanging manners come with a CG at or just a hair behind the neutral point.

So play with it around that 23 to 27% range and zero on on the best flying point. As you've already seen it pays to keep an open mind and play with the CG to find the perfect position.

[JetPlaneFlyer]

I'll be the first to admit that my experience of 3D flying is zero however if you want the model to fly in 'conventional' mode as any time at all then the CG really needs to be ahead of the Neutral point (if only slightly)... A really good pilot can probably cope with it just about on the NP but I don’t think there is a RC pilot who can genuinely fly with a CG significantly aft of the NP for any length of time.

Some think they can but what they usually mean is that the CG is aft of the 25% chord line and on a tailed aircraft this does not mean it's behind the NP at all. Even if the pilot’s reactions and control response speeds were quick enough to cope (which I don’t believe) it would be impossible to quickly and accurately enough judge pitch angle when the model was overhead and/or at distance/altitude. There was some details of a computer stabilised flying wing build as a University project posted here some time ago. The control reaction time they had to use in the computer system was far quicker than any human could ever hope to achieve.

Even with a CG ahead of the NP there is no reason that you cant pull the nose up into high alpha or a hover, you will just need up elevator to do it.

[poulsbobill]

O.K., great answers/ideas all!! Now to a new problem.

I cant seem to get it to ROG. Now i must say i have only about 30 feet in which to do it. But full throttle (i have a thrust/weight ratio of about 1.1/1) and full up elevators and nothing...just rolls. The angle of the craft sitting is about 10 degrees. I have seen others punch up in just a few feet. What am i missing? Your input is all appreciated!!

Bill

[BMatthews]

JPF, a fair number of the flat foamie pilots are flying with the models at or just a hair behind the NP. It obviously increases the workload slightly but models of that sort don't spend a lot of time flying in a straight line anyway. Being a tiny bit behind the NP only means that it won't recover to level and will try to diverge gently. A bit of pressure when needed on the sticks will take care of that amount of divergence. While it may be technically unstable it's no where near the same sort of issue as trying to fly an arrow backwards.

My own flat foamie 3D effort so far has been trimmed to where it's pretty much spot on the NP by shifting the battery pack. In a dive with the elevator at neutral it neither pulls up nor tries to bunt steeper to any degree at any speed I'm comfy with flying it at. And as I moved it back it gets better and better at hovering. Not that I'm in any danger of being mistaken for a helicopter any time soon.... :rolleyes:

Next plan is to add some velcro even further back after I repair the broken fuselage and see how much behind the NP I can tolerate. But I don't foresee any issues with it being 2 or 3% of the chord behind the NP. At the speed most flat foamies fly at being a hair behind isn't really a big deal since all the responses are so slow. Hence the huge surfaces and rediculous control throws to make them snappy.

Anyhow, that's why I suggested playing with the CG over a range around the 25% point which for a rectangle of this sort would be the nuetral point.

[JetPlaneFlyer]

O.K., great answers/ideas all!! Now to a new problem.

I cant seem to get it to ROG. Now i must say i have only about 30 feet in which to do it. But full throttle (i have a thrust/weight ratio of about 1.1/1) and full up elevators and nothing...just rolls. The angle of the craft sitting is about 10 degrees. I have seen others punch up in just a few feet. What am i missing? Your input is all appreciated!!

Bill

Bill,
Try making the maingear legs longer so it sits at higher alpha.

[JetPlaneFlyer]

JPF, a fair number of the flat foamie pilots are flying with the models at or just a hair behind the NP.

How do they know? A Dive test does not really prove CG is behind the NP. A finned bomb dropped from an aircraft does not pull out of a dive but the CG on a bomb is most certainly well ahead of the NP. All the dive test really shows is that the model does not reach it's trimmed flying speed while in the dive, not hard to imagine on a draggy 3D foamie.
In any case getting away with such a set up on a conventional tailed aircraft that has relatively high inertia in pitch and therefore diverges relatively slowly is a different 'kettle of fish' from a tailless wing that has very low pitch inertia so will get out of shape MUCH more quickly.

Interesting stuff though...

Steve

[Brandano]

IIRC the Wright Flier had a CG set behind the neutral point. It flew, but it wasn't a good flier by any means, and in the later incarnations the CG had to be adjusted to allow common human beings to keep it under control