I've certainly done low throttle landing mixes so assume the opposite is possible.

KE couples are outside the box of a single declaration.

Center of drag, center of balance, symmetry of lifting body, turbulence variations, lateral balance center and the list goes on.

What we know is that some planes aren't troubled and some are. Effort can sometimes pin down the cause for a specific model.

I'm still of the school of thought that my goal is to obtain either pitch or roll stability in KE... I can manage one with the stick without a problem... but I know my skills won't manage both.

Just keep on flying - and practice the hard parts
Just like playing music -you will at some point - suddenly find that corrections are automatic.
I always wondered why some flyers try to set up perfect hands off results for all attitudes
In real life -with winds n gusts it is more important that the model is easy to correct and be stable and predictable with given inputs
The best setup is teaching the ol grey matter how to deal with needed corrections.
Not saying that getting the model as neutral as practical is a waste of time -but at some point this approach can give away stability .

OK... I did an extensive CFD analysis of this phenomona... I have not run all the flow cases I am interested in, but I will 'splain below....

An aerobatic plane (Say 3D) is set up to be very nearly pitch neutral. Pattern planes are a little more nose heavy as they track and feel a little better overall... (I have flown my Compy SX at a CG that was just a RCH behind neutral actually)

So generally speaking, the CG effect is rather small for pattern/IMAC/3D planes. Trim your plane to the CG that feels/performs the best and don't worry how it affects KE coupling!

I set up my CFD study using a 78 inch span YAK 55. I did the study with the plane in a 35 degree yaw condition with 40 degrees of rudder deflection... similar to about what you would expect flying a KE pass.

I setup the model so that I could adjust the location of the stab on the fuse. I could move the stab to the top of the fuse, run the simulation, move the stab to the bottom, run the simulation, and place the stab in the middle and run the simulation.

I actually fine tuned the center position until I got a net ZERO pitching force.

For each simulation, I am able to determine the pressures and forces on the stab surfaces. Top'n'Bottom, left'n'Right.

With the stab located on top of the fuselage, there was a pronounced Pitch DOWN (towards landing gear).
With the stab located on bottom of the fuelage, there was a pronounced Pitch UP (towards the canopy)
With the stab located in the middle... in just the right spot, there was No Pitch!!

Ever notice on the Giant Scale IMAC/3D aerobatic planes the stab is positioned well below the scale location??? This is done to reduce/null the yaw coupling... it is apparent on pattern planes as well... the stab is located low on the fuselage....


I can't fully explain the exact mechanism that creates this phenomona... I believe it has to do wth the flow that goes around the fuselage at some yaw angle
being captured by the stab... there is a tendency for the flow to wrap around the yawed fuselage and if the stab is high, it captures that flow creating a higher pressure on the bottom surface of the stab (Nose down) and if the stab is low, that flow is captured on the top of the stab (Nose up) and if the stab is located in just the right spot, the flow is evenly split with no net pressure difference.

Interestingly, I did a run with the fuse at zero yaw angle and 40 degrees of rudder and there is a nose up force regardless of the stab location. it is only once the yaw angle is achieved that the other forces dominate the flow.

As a control, I split the model down the middle and ran that case to verify there was no net pitching force... and there wasn't

when I have time I will get better imagery showing the flow and pressures....


There was a post earlier suggesting that the downwash was a dominating factor.....well... it isn't!! lol.. as a pitch neutral plane will fly inverted hands off.... so the wing-tail angles are zero-zero

The yaw coupling is a phenomona that exists at all flight attitudes, upright, inverted, upline, downline .. whatever.... leave the mix on!! (I do)

that downwh
Mithrandir, interesting stuff... i've heard others talk about how tail position effects tuck in KE based on practical experience so i'm sure your CFD tells the truth.

I'm interested in the quote above regarding downwash (the suggestion it had a part to play was mine)... I'm not sure why you say that the fact planes fly inverted without down elevator proves downwash is not a factor??? if anything i'd suggest the opposite was true. i can take a plane that has some small positive stability (according to calculation of static margin and overall flight charecteristics) and that plane may fly inverted without any down elevator (I've had this occur on my own models).. The same plane would climb when inverted ifstatic marginwas moved close to zero.
The only explanation for this that i can think ofis the effect of downwash on the stab.

Steve

I don't disagree that there may be downwash on the stab... but if it was a significant effect, then the Yaw-Pitch coupling would be different in different flight regimes. If the Stab size location CG etc is such that the wing-Tail angles are "Zero-Zero"... and the plane is neutrally stable, upright and inverted will require no "Top Elevator".....

true... if the wing makes lift, it makes downwash and that does apply an induced AOA on the stab... so if in level flight, Rudder/yaw pitches the nose down (towards the earth), then if the plane was inverted, the same effect would cause the nose to still pitch down (Towards the earth)

Plus.. the Yaw-Pitch coupling is apparent in all flight conditions.. uplines, downlines, +/- 45 degree lines, upright and inverted... the pitch is always to the same direction. (Unless the stab is located to the null spot and other aerodynamic force become dominant such as P-Factor)

There are reports that the pitch coupling on the Krill Katana actually changes direction as yaw angle varies... so there must be other factors in that instance that have varying influence thru the yaw range...

additionally... the net pitch forces I measured in the simulation were on the order of .1 to .4 pounds... that isn't very much, but imagine just putting 4 ounces on the tail of a 2 meter plane and the amount of elevator needed to trim it.... that is prolly the same amount as would be needed in a rudder-elev mix....

Looking again at your CFD images of aiflow around the stab. These images are (i assume) looking at the 'low' side of the fuselage i.e. the side that faces the relative wind approaching the fuselage. What do things look like on the leeward side?.. If flow remains attached (that's probably a big 'if') then i'd expect flow direction to be opposite and to (at least partly) counteract any stab height induced tucking effect.?

I will have to bring up the simulation again and interogate that.... I didn't see before that the effect is countered or reversed.......
I base my conclusions on the differential pressures (forces) between the stab upper and lower surfaces....

But the flow on the "Suction" side is certainly chaotic and seperated at the higher beta angles

If I remeber correctly it would start with moving the CG and then I would go in and increase the angle of incidence in the wing and test fly. Keep playing with it until either you corrected the problem or just got tired of messing with it. After every change, one at a time, you would need to start over in the trimming process. Look at the Pheonix series of pattern aircraft, each one had changes made to correct unwanted tendencies. Phoenix 7 was a great airplane. My favorite was the Bridi UFO. Just enjoy the capabilities of the new radios but I recommend getting the setups as close as possible mechanically before playing with the electronics.

could this also be caused by the fact that the rudder usually isnt centered? it is usually larger in the canopy side so the plane can rotate on takeoff. this would cause a small pitch to the belly in knife edge position. this may not be tha main reason but it could be a factor.

I'm sure the assymetric vertical stab/rudder is part of the complex interaction, and i can see why that would cause a rolling couple... But i cant see any reason on the face of it why it would cause a pitch to the belly?

the CFD Simulations I have done so far seem to reveal it is predominantly a Stab height positioning phenomona....

I do want to run some follow on simulations to investigate the influence of the asymetrical vertical....

The practical flying experiments we have done with vertical stab positions +/- 1.5" from the thrust line (all else unchanged) revealed NO difference to pitch coupling on pattern planes (with applied top rudder). But a rudder area distribution change affected pitch (and roll) couple a noticable amount.

Years ago I remember talking wiith Don Lowe about some of his experiments with stab position. Turned out that the stab needed to be placed either right on the bottom of the fuse or on top of the fin to make a significant difference in pitch coupling. So, for those inclined to believe that vertical stab location change of 1/2" in either direction will fix a particular flight attitude, you may be disappointed after all that work.

Any wing will generate "downwash" at its trailing edge. It doesn't matter whether the wing is completely flat like a foamy, flat bottomed or symmetrical. Also doesn't matter if the wing is flying upright or inverted, "downwash" of air is a must for lift generation. Ever wonder why a flat bottomed wing can fly inverted? Some fly inverted well, BTW. I am not too sure that wing downwash is even touching the stab on a long tailed pattern plane whose tail moment is 2.5*WingMAC or greater

The stab position thing is SACRED with some enthusiasts.
Do I buy it ?
nope not for a second
Having setup countless pattern /scale models big n little
I never worried about it simply because this issue was put to bed looooong ago - for me (and Lowe and many others ). As for down wash - well mostly hogwash on a lightly loaded model - you need for the wing to really work to produce much "downwash.

Nice pun! Not disagreeing, may I just trade some figures for your experience?

Downwash depends entirely on Cl. For a tail moment arm 2.65 times the wing MAC, downwash angle (at the stab) is only 0.9 degrees for Cl=0.15 and 1.8 degrees for CL=0.35 (only approximation, but in the ballpark).

I think there was no argument about downwash in knife-edge. But what is it that makes for belly pitch? Rusty Dose once mentioned that the 1980 Z-50S models had a wicked down pitch due to stab position. (He didn't even mention wing position.) I have a "scale" Z-50 ([link=http://richard.ferriere.free.fr/3vues/zlin_50_3v.jpg]3-view[/link]) in the simulator, which not even renders such things as flow around the tail and fin/rudder shape. Yet there's belly pitch and roll out, though, but I manage to set up the Z-50 to no pitch and no roll out in knife edge and in level flight yaw by just setting +2.5 degrees incidence for both wing and stab (no down thrust) as shown in the drawing. Of course, the C/G position has to be tuned both horizontally and vertically. How's that?

I'm still interested to see the results of the CFD analysis and I'm wondering why the streamlines seem to go through the stab.

Belly pitch= the accululated forces acting against gravity MINUS the gravity when in knife edge
I can't think of a more simple explanation..

Downwash angle also depends on aspect ratio too but i assume you took that into account. I'd argue that a 1.8 and even 0.9 deg angle is not at all insignificant, certinly far from 'hogwash'.
What this means is that when flying with wings level (upright or inverted) compared to when it's in KE (wing makes no lift in KE) then the angle of attack the tail sees is different by 0.9-1.8 deg.. The difference is such that the tail has a more positive angle when in KE flight due to absence of downwash, so makes more lift (lift toward the canopy) and so tends to tuck the nose toward the belly... exactly what most models experience.

I'm sure that there are many other factors at play too but i still maintain downwash must be a significant contributor... Sadly there is not much that could be done to reduce downwash on the tail other than making the wings much higher aspect ratio or the fuselage much longer, both of which would have very negative impact in other areas.

Steve

On a decent aerobatic design , One can setup a cruise speed and trim hands off then roll to inverted and add what trim elevator necessary to again acheive level hands off flight.
the minute trim change indicates that typically the DIFFERENCE in AOA for the wing is well under one degree.
remember one degree is 1/4" in one foot.
I would venture that any downwash to the tail is of little or no consequence
I guess someone could calculate the exact angles but I would again venture in this case, the error percentage is close to the calculation.
As speed decreases or wing loading increases, the angle will increase - that's a given. I frankly don't think downwash is any kind of a factor in this situation.

rmh,
You cant really judge AoA by how much down is needed for inverted..I have a model that willfly inverted with no down but that doesnt meanit hasno AoA.. Again downwash plays a part here because it creates a downforce on the tail regerdless of if the plane is upright or inverted... This downforce alone can be enough to produce stable level flight without any furtherelevator input... which is why some model can be stable yet still fly 'hands off' both upright and inverted.. or even in some cases climb when inverted.

1.8 Deg downwash would be similar to packing the TE of the stabiliser 3/16" (assume 6" chord stab)... that would make a huge difference to trim. Even if you take 0.9 Deg thats still 3/32" which would still have avery noticeable impact. In effect this 'virtual stabiliser packing' 'swaps sides' when you fly inverted.. and is removed altogether when you fly KE.. that's got to make a difference?

Well, I like to balance the plane to neutral upright and inverted, hands-off flight. The more subtle effects neglected (i.e. in the simulator), that needs a far rear C/G but still ahead of the NP. That's an effect of the downwash on the stab. A small positive wing incidence (1/32) may be needed to compensate the landing gear drag moment, but basically it's a zero-zero setup.

The Z-50 is set up zero-zero as well, but even though wing and stab are parallel they both have +2.5 degrees incidence. Thrust is parallel to the imaginary centerline. (Which should be defined by the vertical C/G position, shouldn't it?) So it's actually downthrust.

Now imagine knife-edge. If the airplane is not set up quite correctly and still has some belly pitch that may be corrected by outside bank. Isn't that counter-intuitive, as well as the positive incidence?

Not to knife-edge, only to the downwash issue, some figures:

The Integral pattern plane, set up for 100 ft/s (about 70 mph), needs 0.5 degrees decalage if static stability margin is 9.6%, 1.0 degrees if static margin is 30.4% (C/G exactly at 25% MAC). Or an equivalent elevator deflection. In the first case, the stab is still lifting up (Cl 0.03), in the other case it's pushing down (Cl -0.01). At 29% static margin (maybe 27% MAC) stab Cl is zero, but decalage still 1 degree.

How did you measure 1.8 degrees of downwash?
The flow from te of a wing is a moving target-it isn't a straight line consistant flow
It is turbulance that disappears very quickly -in the case of these models .
There is disturbed air aft the wing when lift is being produced -o course .
My example was only meant to point out that in these instances , there is very little of it because the wing is operating at an extremely small AOA

Understood, I just tried to show how small. The values were not measured but calculated, and indeed downwash is not linear so it's tricky and the known calculation methods are all approximations. But we're in the ballpark.

The examples were some I have at hand. The last example, the Integral pattern plane, has indeed 0.14 wing Cl at 100 ft/s (70 mph), so the downwash angle at the stab is about 0.9 degrees what I took from another example. Decalage is rather 0.94 degree in this example for zero stab Cl (if set up for that). Another counter-intuitive thing here?

There's no drive/thrust taken into account in these calculations. The wing downwash is part of a quite wide-ranging flow system, as was also shown in another thread where (shoe?) calculated the change in pressure at some distance (below the wing, IIRC). So even a T-tail is in (nearly) full downwash, even if not in propwash. And it's not in the turbulent wing wake that stems from the boundary layer (except in a deep stall where the stab may be blanketed).

I disagree with this statement.

My CFD simulations are not consistant with this assertion....

EVERY decent 40% (35%, 30% etc) scale aerobatic plane has the stab lowered... this is not done friviously.. it is done to directly affect Yaw-Pitch coupling.
(Yaw-Roll coupling can be nulled with either wing height on the fuse or dihedral)




I disagree with this... and you can prove it to yourself....

Do the following....
Fly along straight and level, Pull to a 45 degree climb. Roll 90 to KE.. let go.. no top rudder.. nothing.... what happens....???
The plane flies a mostly tractory... with a small pitch if it is nose heavy with compensating up trim....

repeat the experiment and then apply rudder... the plane will usually pitch....
Try rudder'ing on an upline or a downline.... the pitch only occurs if there is yaw...




I understand and I agree.... but basically.. if a plane pitches to the gear or the canopy.. it will do so in all flight conditions.. upline, downline, 45 degree...
My comment regarding the downwash is that if the down wash is significant, it seems that the pitching with rudder would always be the same direction with respect to the down wash
It isn't... if it pitches to the gear upright.. it will prolly pitch to the gear inverted...or on a upline or a downline.....

Well- I just have to disagree with you - simply because my findings are based on building and flying a shload of different aerobatic designs .
I found I could put the stab in any practical location (strength rigidity being prime consideration) and trim it out for nice neutral handling.
By the way I have never subscribed to the idea that upright /inverted flight can be perfectly achieved with ONE trim setup
Seen it claimed many times -never seen it- close but somehow the different AOA for the wing just had to be there.
The only pitching problems I found to be a bear to fix, involved biplanes
especially those with stagger.
Straight upline /downline - a different game as the speed is typically no longer a constant (gravity somehow enters into this.)
The UPline problems typically disappear with power and this means a LOT of reserve thrust in order to hold constant speed.
The big downline problem to me has been braking (till electrics ) but then the "whirling brake" inserted a roll couple.
So I take each setup on it's own merit and tweak till I like it -
I have never seen any magic setups or moments /locations for any of it
So far the best setup for being stable enough and easy enough for solid predictable flying has been engine/stab -on a common line-with the wing slightly below this line
This seems to give a good tradeoff with upright flight as the main drag (the wing) is below center of applied thrust .
a perfectly symmetrical biplane setup should be best -but ain't- at least for me
I don't have any math models I just cut n try.

Can't cut and try, just have only math models. (No offense, just plain truth.)

At closer look, the (calculation) experiment turns out subtly different. Yesterday, I didn't consider the fuselage effect which is accounted for in the tool. Today, I re-set the incidence angles so that the fuselage centerline is level. The wing's angles of incidence and attack are the same: 1.64 degrees. The stab's incidence is 0.77 and it's AOA is virtually zero (0.03). Accordingly, the stabs lift (Cl) is virtually zero. That means decalage is 1.64 - 0.77 = 0.87 and the wing downwash at the stab is 0.77 - 0.03 = 0.74 degrees.

Both is noticeably smaller than the values I had yesterday with the fuselage slightly pitched down due to big incidences. That should mean the fuselage produced some down lift and/or down-pitching moment. If you look at the side view that may be plausible, or doesn't it?

Now imagine the airplane in knife-edge flight (the thread topic). Notwithstanding the other effects (fuselage lift, airflow around fuse and tail), the typical low- or mid-winger fuselage shape (especially with the popular cheeks) would produce a belly pitch. It's countered by opposite lift and pitching moment of wing and stab. Decalage required depends on C/G position. Fuse and wing/stab are "braced" (?) giving a directionally stable system.

Do you think that's possible? And considering all other influences, how to cut and try only for this one?