pe reivers

Here is my problem:
Data:
Plane piper cub, span 102", 25lbs, engine 58cc with 23x10 prop. Preparing this plane for glider tow.
Engine thrustline such, that the plane's line of flight shows no pitch, roll or yaw with throttle changes.
wing foil: fancy, blunt nose, NO stall up to 20° incidence and beyond, 14.85% thick@30% chord; camber 3.2%@42%. Incidence to plane datum line +3 degrees. The wing is straight, built like a brick house, and has about 3° V-configuration. Wing halves coupled by 30mm aluminum tube.
CG is at 20% of wing chord.
tailplane foil, flat plate, 8mm thick, radius front/aft, incidence 6 degrees (tangent points to wing neutral point).
Using Duranti's software, the wing airfoil is a very near match in properties (Ca, Cd, Ca/Cd, Cm) to the Eppler E197, even though thickness and camber don't match.

Problem:
1) with low power, like about 30%, the plane flies like a dream, both on rudder and aileron, and shows no stall whatsoever. When speed decreases, it just drops to fly again with applied power. As soon as more power is applied, or speed increases, It is all over the sky, except where you expect it to go to the extend, that all control is lost, and only rudder application will work to correct flight path and regain control again. Reducing elevator throw only results in total loss of control. Raising the wing trailing edge made things worse. VERY nerve wrecking.
2) When the plane gets out of horizontal a bit more than needed, the flight path misalignment will exagerate. Up nose will increase with time, but so will down nose, unless correted by active inputs. (unstable horizontal flight)
3) In landing, the plane is neutral, with very slight down-, but no up pitch.
4) after landing, wings will tend to topple the plane left or right, depending on yaw left or right. Speed needs to bleed down to almost zero, before steering of direction can take place. Three point landings are a piece of cake, and so are wheel landings, if one wheel touches first to avoid bounce.

Your opinions are much appreciated.

crasherboy

My two cents. Your wingspan says 1/4 scale. I am thinking you are over powered ,the scale prop would be 18 inch. Other than that seems that you must have some problem with the engine thrust line ,or some loose linkage in your controls. A full scale pipers speed was only around 75 to 80 mph,so you can see that a 1/4 scale is not intended to fly fast. But it puzzles why you would have problems with your landings,should land like a winner.

Villa

I have had three J3-Cubs with 71" wing span so I cannot be much help. I do not recall having 3 degree positive incidence on the main wing. I'm fairly sure the one I'm currently flying has zero incidence. Have you notice any bending of the horizontal stabilizer in flight? I had to add flying wires to my horizontal stab after I saw severe bending in flight. Mine is a SPAD of my own design so it needed a few imrovements. Are you using wing struts to keep the wings from warping or even folding in flight? SIG used to have a 71" wingspan J3-Cub that had very weak wings that had to have strut.

mjfrederick

My first instinct is to say too much positive incidence.

Rodney

I'd almost bet that the fuse is starting to change shape (bending or oscillating) at the higher speeds due to to weak or flexing structure attaching the tail feathers to the rest of the plane. Can you see it twist if you manually grab the elevators and put a minor torque on the structure?

pe reivers

I'll try to re-arrange the datum line to exit from the stab trailing edge and be parallel to the wing's foil bottom tangent. Then make pictures of the prop down thrust and stab incidence, and take it into cad to truely measure all angles. More info maybe will paint a better picture.

@ crascherboy:
I agree on too much power. I want that for towing. The strange behaviour to lift a wing on the roll-out however points into the direction of some form of plane instability that needs attention.

@ frederick:
If I compare incidence to engine thrust line, it is not all that much. I tried reducing incidence, and it made matters even worse on power, whilst not improving landing behaviour. I needed a lot of elevator up trim though, where the original was a flat plane with the horizontal stab.

@ rodney.
Tested the fuselage. It is a very stable, torsion stiff plywood box construction with lightening holes and stringers. The tail feather area does not have any lightening that could reduce stiffness.
This agrees with the stable in flight vieuws and absolute abcense of any form of flutter. Plane flies the same with, or without the wing struts. The wings are extremely torsion stiff . (sheet covered dense foam, like a brick wall)

da Rock

It appears you have gone to some technical lengths to investigate the model. Duranti's software for example.

You also mention your CG is at 20C. But no mention of how far back your horizontal tail is, nor it's size etc. The size and location of the tail matters quite a bit to an airplane's pitch stability. I would suggest using the application at http://www.geistware.com/rcmodeling/cg_super_calc.htm and find out for sure where a sensible CG location would be. You can use 15% for the Static Margin if you simply want a start CG, or run the application twice with 5 and 20 SMs to get a safe range.

20C really sounds like your CG is "too safe". When the tail has to deal with a CG that's too far forward for it, you wind up with a tail that's used up all it's ability just keeping the CG up. And if you've not changed the elevator throws to keep up with the force requirements, un-funny things happen.

da Rock

You've also mentioned that you're using a 58cc engine. It's worth doing a bit more detailed range checking of the radio when you've got a model that does funny things when the throttle is moved. Have a couple of friends hold the airplane up by the wingtips while the engine is running so the gear and weight of the plane isn't resting on the ground absorbing vibration while you're at a distance.

Gas engines can cause interference from a change in vibration. You won't always see it with the usual range checks.

You're not having an ordinary problem. Extra ordinary tests seem appropriate.

TFF

A couple of thoughts as I know you are very experienced. Great amounts of incidence are not required because the weight is not scaled. A 1/4 cub would weigh 250 lbs; at 25 lbs it does not need as much angle of attack to make the needed lift. I would think at the most 2 deg + wing, measured center of airfoil of wing to center of stab. CG adjusted to mach. with such a big prop there might be a P factor large enough even at idle that causes an aero imbalance. With the CG at 20% it is far forward, but your notes say tail heavy which might be masked by the incidence. I am having a similar problem with a Sig Senior Kadet bashed into a Bellanca CF. With my smaller scale tail. Finding the plane almost unflyable, with my new CG calculations I am off by 3/4 of an in from the place the Kadet would have been if built as designed.

Wilki01

I'm not sure what the recommended prop is for the engine is. I would start with the simplist thing first. Go with small changes one at a time. First you are using a 23" diamiter prop with a pitch of 10 inches. With this prop you are flying on the prop not the wing. Even at low throttle you are moving forward fairly fast. I would drop down to a 6 inch or less for pitch, check your tach top end and keep it within engine specs. With this plane you must fly on the wing and not on the prop. It's wing area, airfoil and wing loading dictate that. With a lower pitched prop you will be able to slow down enough to fly it. Try not to overthink the issues at hand, it's a Cub a very proven design, even with a heavy and over powered engine you can make this a nice flying plane. Just my opinion but I would setup up everything at the recommend setting COG, Thrust whatever and just mess with the prop pitch and or diameter to get it to fly slow first, before you end up crashing it. We all hate to see that happen. Poste some pictures if you can too.

Wilki

pe reivers

I did try different propellers, ranging from 22x10 , 22x10x3, 22x12, 24x8. The plane's behaviour was best with the light load, high rpm 22x10 prop. As far as I could judge, no radio disturbance was involved. Plane speed and maybe prop wash in the large prop sizes definately.
As to elevator arm momentum vs COG:
There is no tendency to drop the nose at lower speeds, and at very low speed 3-point landings are a piece of cake with elevator deflection to spare.

I did a new analysis of angles, this time using the horizontal stab as reference by placing it horizontal:

Wing foil centerline at +1° incidence
prop down thrust a whopping 8°, explaining the complete lack of down pitch with lower power.

The foil has zero lift at -2.8°, making the effective incidence with the tailplane 3.8° . As noted before, decreasing incidence did not improve matters. (I will measure the angle I reduced incidence and post it here)

pe reivers

P.S.
I reduced wing incidence by 2.8 degrees, equal to the foil camber reaction. Like stated before, this did not improve things at all.
However, thinking of how the camber seemed to make the plane look tail heavy, and how yaw would cause very severe roll reaction at speed and with ground handling made me rethink the wing's V-angle and the large difference in AoA this would have on the individual wings in yaw conditions.
So I tested the plane with the ailerons trimmed up about 5 degrees. prop used was 22x10x3.
ROG at about 1/3rd throttle, steady climb at half throttle. Trimmed plane for steady climb angle. When at safe altitude, full throttle was applied in horizontal flight, being prepared for anything. Nothing happened! Plane gained speed, and still nothing happened, except for very slight pitch changes. Left bank stable, right bank stable.
time did not permit more tests.
Landing took more speed than I anticipated so I came in too slow. Having reduced elevater throw, in landing and touching ground I had none left to spare, and the plane nosed over. No damage to plane or prop.

So more questions arise:
Why do I need more elevator in landing with the s-shaped foil, and nothing else changed? The foil pitching moment being so much lower one would expect less elevator input to keep the nose up.
OR
Did the wing's downwash push down on the stab, this no longer being the case with the up trimmed ailerons?
Could the stab working in the downwash have caused the instability?

pe reivers

I think your considerations were spot on. Effective incidence to stab with aileron deflected up could well be about 2 degrees.
Weather now is too bad to do more testing, but my spirits have risen.
I am considering coupling flap input on throttle, keeping a certain amount of S in the foil all the time to reduce piching moment and downwash.

da Rock

How do you feel about the downthrust you measured? You haven't thought to change it, right?

What model Cub do you have? How was the thrust line established? Built in by the ARF mfg?

The wing's downwash doesn't "push down" on the horizontal tail. The horizontal tail experiences an airflow that has been diverted from the line of flight. So it has an AOA that is decreased or increased by the angle of the downwash.

And at higher coefficients of lift the downwash angle will be greater so the tail will be in a higher angled downwash. And the greater the downwash angle, the greater the negative lift out of that component. It's usual for a designer to place the horizontal tail above or below the turbulent downwash to improve it's reliability. When you changed the aileron rigging to emulate a reflexed airfoil, you did a number of things. Things that do quite significant things.

da Rock

TFF, pitch stability is a function of the wing area, wing chord, horizontail tail area and distance aft of that area. Those 4 are the major players. There are formulas that accurately use those 4 to give you a safe and accurate CG location or CG range.

When we alter an existing design with it's established CG range, it's wise to work out the CG range for our new layout. Truth is, when we assemble an ARF, it's wise to double check the suggested CG location. It's quite easy to do that without having to do any calculations at all. Take a yardstick and measure your Bellanca. Plug the measurements into http://www.geistware.com/rcmodeling/cg_super_calc.htm and you'll have a very accurate and dependable location for the CG range of your new layout. The formulas the application uses will sort out where the CG should be for the smaller and less effective tail. It's a good idea to look at how narrow the suggested range is to get a feel for how you should change your elevator throw. If the new range is considerably smaller you probably should expect the elevator to also need a bit more deflection.


Considering nothing beyond the CG location as a function of the wing chord simply ignores too many important factors.

pe reivers

So the downwash did not push down (negative force), but created a negative lift (negative force) right?
The reflexed foil without changing the way the wing was fixed to the plane did following:
decreased camber, and thus decreased downwash, reducing the negative lift of the stab. So I needed to trim the elevator up by about the same amount as the aileron deflection.
decreased incidence angle of the wing relative to the tail plane and fuselage
decreased the aurfoil's Cm
decreased the downthrust of the engine relative to the wing airfoil chord.
Decreased the foil Ca max and thus the lift the wing can generate, so my slowest landing speed goes up a little.
Decreases the AoA for zero lift, thus further decreasing the wing's effective incidence angle.

I do not know what kit this plane is. I bought it used from a guy who did not like the way it flew, who bought it used from... who...., where it had been in glider towing duties using a King 60 engine.
Construction is sound, and I liked the looks of it. Being in need for a new tow plane and needing a flight crate for my 58cc made for a quick deal.

I eyeballed the downthrust, which prooved to be about right. Test flights will finally determine if it needs to be changed. I have a feeling that there will be no large changes, because the engine thrust line is quite low in the planes drag system (aerodynamic neutral point). The same will be done with final location of CoG. The plane will tell me what to do, once I get it to fly well.

I went to the Geistware calculator section, and everything was spot on, except the nose length, which is slightly short in the Piper Cub. My CoG differs only 0.15" from the one calculated for 10% sensitivity.

khodges

My Feedback: (1)

Just my $.02, but even as a towplane, you don't need all that prop. I think a lot of what you're getting when it goes all over the place on application of power at lower speeds is torque reaction. That's a lot of engine and a whole lot of prop for that plane. Based on what I've seen of people using Cubs for towplanes, they do fine with scale power and prop. Full scale Cubs pull gliders with the standard 65hp and a standard 6 foot prop. The Cub flies on its wing, regardless of whether it's pulling something or not, and whatever you are planning to tow (I assume a sailplane or glider, etc) will also have a high lift/weight ratio, so it won't take very much power to pull it anyway.

Don't know what to think about all the calculations, I have enough trouble balancing my checkbook.

pe reivers

Up to now, I have been towing using a Multiplex Biglift, designed for a 15cc glow engine. It is a 220cm span plane. Many fit engines up to 50cc in it, and the plane handles that engine well albeit that the span is a bit small to handle all that torque that the large engines devellop. Directional stability suffers, but the plane never gets unstable or starts to fly uncontrollable like this much larger and heavier Piper Cub did.
The question is not whether the engine is too large, but the question is why does the plane behave that way, because it also happened in a shallow power-off dive. And it misbehaves in the roll-out during landing. The plane's bad manners are speed dependant and yaw dependant.
Keeping the speed way down, and flying on the wing is a way to let the plane survive despite it's flaws.
I try to find out what is wrong with this plane that so seemingly follows all the basic rules.

pe reivers

Here are the data from profili. The first two graphs as compared to Eppler E197, with near identical properties, graph 3 and 4 as compared with a flatbottom design I made for ease of building in combat planes. The planes I built with this foil would curve very tight without snapping out. In longer curves they would retain speed very well. Despite the higher lift and higher Cm, no odd high speed behaviour in this high camber design. So what is wrong with the E197? Any records?
All graphs are at Re 200,000. higher re numbers show no anomalities.

Darryl Usher

Why do you call this a Piper Cub? A Piper Cub climes and turns left in full power and down and right with low power. It has an USA air foil I believe, and a CG 25% and little dihedral . I have flown all real Cubs from 50 hp to 150 hp and they all fly the same. I also have Goldberg J3 with Os 23 that I flown for 15 years.
Darryl Usher

pe reivers

It is an RC plane shaped like a piper cub. As a model airplane, I do not expect it to be a full scale piper cub, but I do want it to fly and handle well.

pe reivers

Another consideration:
Could it be that dehidral is too large? The plane has lots of it.
Regardless of the outcome, I intend to reduce it to nearly nil to make it look more like a Cub. This also would lower the plane's neutral drag line some, so less downthrust is needed.
In the second picture is the amount I took out visualized.
The new dehidral is 0.8 degree. The original model had 7.3 degrees, probably for rudder only steering.

pe reivers

Problem solved! I did extensive tests today
The Dehidral (V-angle) reduction worked wonders, and the plane behaves like is should. It flies very docile on the wing, and can handle any power I choose to apply without the need to compensate. Directional stability is excellent, and vertical instability is gone as well. It flies where I point it.
Aileron defelexing could be undone, and all incidences are back at original. In landing the plane can now slow down to a crawl. This makes 3-point landing a bit more difficult because of loss of elevator authority at very low speed without prop wash. However. tail up wheel landings are quite easy, and the plane can be swerved for braking without one wheel breaking ground.

Thanks everyone for your thoughts and advise.

Darryl Usher

Move CG back!
Darryl Usher

pe reivers

Thanks Darryl,
I am now at 5" from the leading edge on a 15" chord wing, and it probably can be moved back a bit more.
The wing loading is so low, that the slightest tail down movement will cause the plane to climb again. The foil in the wing will not stall (see graphs). For that reason forward speed has to be very very low for a 3-point landing, and elevator has lost authority.
In the last try, the nose stayed up until all speed had bled off, and the plane just dropped the last foot instead of lowering itself to the ground. I will do some stall tests first at safe altitude to see what is in the scope of possibilities.