RE: Balancing a canard or tandem wing biplane for initial flight.
For forgiving handling, a canard should be balanced slightly nose-heavy by about the same amount as would be indicated for a conventional tractor layout. A good way to check this is to make up a profile fuselage small cardboard hand-launched glider with the same relative wing and stab areas and tail moment arm, and test for stable glide with various CG locations, courtesy of paper clip nose weights. A six inch wingspan is adequate for the little glider, and flat-plate non-cambered airfoils are fine, as far as CG is concerned. This little trick should get you within a couple of percent of wing chord of the best CG location, although test flying of the full-size version may indicate that some fine tuning of CG is needed. Always start a bit on the nose-heavy side.
Posted on: 10/31/2009 4:03 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=9218020

RE: can you compensate for weight?
Assuming that you have the CG right, you can compensate for overweight by adding engine power, although an overweight airplane will still be more of a handful no matter how powerful its engine. My calculations indicate that 10% overweight dictates 25% more power, not 10%, in order to get recovery from incipient stall with the same altitude loss. The accurate rule appears to be that available power must be proportional to weight to the 2.5 power. A good example is the relatively low powered early Spitfire fighters that weighed about half as much as the later versions. Pilots rated the earliest models much friendlier to fly than the much higher powered later versions, even though power:weight ratio of the late models was higher. If weight is doubled, I think that power must be increased by a factor of up to 5.65, in order to retain the ability to recover from a dangerously low airspeed situation, with the same altitude loss.
Posted on: 10/31/2009 3:44 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=9217993

RE: Where to mount landing gear?
Experience has shown that positioning a tail-dragger's landing gear so that the wheels contact the ground 15 degrees ahead of the CG will result in frequent nose-overs on a grass surface. I like to use 30 degrees for grass, and 20 degrees for a hard surface. A simple way to check whether your tail-dragger's landing gear is located properly is to simply lift the tail with the airplane sitting on the ground. The angle to which you can incline the fuselage downward before the airplane tries to nose over is a close approximation of the angle between the wheels' contact point and the CG. Don't forget that the vertical location of the CG is important, not just the balance point shown on the wing, which is not the true spacial CG. A low CG location will permit moving the wheels further back. A higher CG location requires locating the wheels further forward, to avoid nose-overs.
Posted on: 10/31/2009 3:31 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=9217974

RE: Full flying elevator question.
An all-moving stabilizer/elevator (stabivator) combination looks attractive on paper, since it can be built lighter than a conventional stab/elevator combination. Elimination of the hinge line also reduces drag. I tried one on a 125 MPH model many years ago.I hinged it at about 20 percent of the mean chord behind the leading edge. Getting a strong and stiff enough hinge seemed to wipe out much of the potential weight benefit. The airplane handled well, but I could ascertain no advantage. It also got into a scary low-speed flutter once, probably because it was not counterbalanced. Counterbalancing would have added far more weight than that of the counterbalancers on conventional elevators . I didn't repeat the exercise. Piper used an all-moving tail on the Aztec and Comanche. I recall hearing of alleged fatal accidents that may have been caused by a nearly fully deflected trim tab causing the center of pressure to move ahead of the pivot axis. The tail might suddenly take the bit in its mouth, slam to full deflection and tear loose from the airplane, but I don't have much factual information on this. Maybe someone out there has the dope on late 1950s accidents with early flying tails on light aircraft. An all-moving stab develops a much lower maximum lift coefficient than a conventional stab/elevator combination, which acts like a heavily cambered section when the elevators are considerably deflected. Cessna was forced to use an inverted slot to get sufficient downforce from the stab to force the wing into maximum lift angle of attack. The slot probably adds more drag than that of a well-faired hingeline.
Posted on: 10/26/2009 5:37 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=9203875

RE: Shrouded Prop? does it work?
A shrouded propeller may nearly double static thrust per horsepower, which is great for VTOL airplanes, but the thrust advantage rapidly deteriorates to nearly zero as speed increases. In general, at normal cruise speed, the shroud's additional weight and drag negate any advantage.
Posted on: 10/26/2009 4:11 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=9203662

RE: pusher vs. tractor
After testing with a glider of same configuration, I designed a canard pusher competition pattern model a few years ago, but never built it, since the performance advantage seemed minimal, if any, and in even a minor crash, the rear-mounted engine would destroy everything in front of it. I planned on using a rear intake Webra .61 engine, whose rotation was designed to be easily reversed, removing the need for left-hand props. The canard glider flew well, and would snap roll and spin just like any normal tractor layout. I am inclined to disbelieve the canard-lovers who like to claim that canards cannot stall or spin. I think that if the CG is located far enough aft for good aerobatic performance, and with enough elevator authority to force the wing to develop its maximum lift, canards behave pretty much like tractors. One negative aspect of canard pushers that is seldom mentioned: torque is far more troublesome, since torque cancellation by the spiral propwash interacting with the wing and tail feathers is not present. Pushers also would be pretty much incapable of hovering without torque-rolling, since very little propwash flows over the wings and full-span ailerons.
Posted on: 10/23/2009 4:07 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=9196521

RE: naca foil datas
Bear in mind that the lift and drag coefficients of airfoils vary considerably with Reynolds number. I have a lot of old wind tunnel-derived information on various airfoils that does not state the Reynolds number that the airfoils were tested at, and thus the information is useful as a rough guide only. For example, the NACA 0015 airfoil, very commonly used for aerobatic models, shows peak lift coefficient ranging all the way from 0.83 at Reynolds number 43,000, up to 1.55 (87 percent higher) at Reynolds number 3,300,000.
Posted on: 10/18/2009 11:36 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=9184191

RE: how to calculate dihedral angle?
If your goal is to optimize the dihedral angle to obtain no yaw/roll coupling, a low wing layout normally requires about one degree or so of dihedral, and a shoulder wing layout a roughly equal amount of anhedral. This is because of the influence that the fuselage exerts on the wing when the airplane is yawed. For example, in a yaw to the right, the air pressure is higher on the left side of the fuselage, which tends to produce a downforce on the left wing, and an equal upforce on the right wing, causing a low wing airplane to tend to roll left. The result is adverse roll, which can be nullified by the correct dihedral, which depends on the depth of the fuselage and the wing shape. In a high wing layout, the rolling tendency is in the opposite direction, a proverse roll, in the same direction as the yaw, requiring anhedral to remove. Getting the dihedral/anhedral exactly right is difficult. Dihedral is difficult to change after the airplane is built, unless it has strut-braced wings. Some competition aerobatic pilots, after flight testing, slightly bend the joiner tube between the panels of a two-piece wing, until all yaw/roll coupling is removed.
Posted on: 10/13/2009 4:31 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=9170326

RE: washout on Hershey Bar wings?
The full-scale original Cherokee 140 had little or no washout, as I recall. Washout is normally considered to be more important with tapered wings, but I have found that adding washout to constant chord wings on R/C trainers really helps controllability near stall. Although the tips of constant chord wings are less prone to stall than those of tapered wings, their relatively wide tips exert considerably more yawing moment near stall, which frequently overcomes the yaw resistance of the vertical tail and fuselage side area. I would be strongly inclined to incorporate some wing washout in order to get friendlier handling at near-stall airspeed, and to use a fair bit of aileron differential as well, unless the vertical tail area is relatively larger than usual. Mixing a bit of rudder with aileron command will also augment user-friendly behavior.
Posted on: 9/15/2009 3:44 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=9097980

RE: Shear Webbing
I, too, have mulled over this knotty question, doing a fair bit of inconclusive stress analysis. Several years ago, I decided that the only way to resolve it was to build 15 to 20 webbed specimen spars and test them to destruction. The test spars had hard balsa flanges (top and bottom longitudinal members), with various designs of balsa webs. Length was about 20 times depth. The commonly used balsa shear webs with grain oriented vertically failed miserably in the webs, at about 25% of the calculated load. (Very distressing, but model design surely keeps one humble, if nothing else.) The load was applied in the middle of the spar, and each end supported, since the ideal uniformly distributed load would be much trickier to achieve experimentally. Web failure always initiated near the ends of the spars, where the shear deflection was greatest, proceeding inward. With web grain at 45 degrees, the webs were about twice as strong for the same weight. The compression flanges of the spars were restrained from buckling by a continuous slot into which the flange was fitted. Best arrangement proved to be ply balsa webs, with grain direction crisscrossed, as in normal plywood, and 45 degrees to the longitudinal axis of the spar. For the same weight, this was about four times as strong as single ply balsa webs with grain oriented vertically, but I will have to consult my notes for more definitive figures
Posted on: 8/31/2009 11:41 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=9060686

RE: Best low rpm/high efficiency prop choice
Best propeller efficiency is attained at a blade angle of 40 degrees or so at 75% out from the hub, but efficiency of a prop with pitch roughly equal to diameter is only a couple of percent lower. Going down to a pitch of half the diameter cuts efficiency by about 15-20 percent, relative to that of a prop with pitch equal to diameter. If you really need to maximize efficiency, it might be worthwhile to gear the prop down quite a lot, and use a rather high pitch. Multi-blade props can be smaller in diameter than two blade props, and are usually slightly less efficient. They are normally only used to reduce landing gear length, and to cut noise.
Posted on: 7/19/2009 9:44 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8947295

RE: force on a controll surface
This site may be useful: http://www.fayettevillercclub.com/id26.html I like to use about 70 percent of top speed when calculation required servo torque. High aileron and elevator travel is very useful when fighting wind gust during landing approaches, but the same travel may produce excessive roll and pitch rates at top speed. Undersizing a servo results in momentary servo stall at less than full control surface deflection, which may prevent wing breakage upon panic application of full elevator in attempted recovery from a steep dive.
Posted on: 6/22/2009 9:49 AM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8871000

RE: 360 deg. turn by jets banked almost 90 deg....How?
According to my calculations, keeping the nose level without top rudder during an 85 degree banked turn would require pulling 11.47G. 88 degrees of bank would dictate pulling 28.65G, which would be briefly possible for a high powered model with a low wing loading, before the resulting rather large induced drag slowed the airplane too much to allow the wing to develop the needed lift.
Posted on: 5/14/2009 10:32 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8769968

RE: Determining a Biplane's CG question.
For bipes with unequal wing sizes, I would first find the MAC (mean aerodynamic chord) for the each wing. If the lower wing, for example, has 40 percent of the total wing area, the MAC of the combination will be on a line drawn between the MACs of the two wings, at 40% of the distance separating the centerline of the upper wing from the lower wing. Incidentally, wind tunnel tests conducted many years ago indicated that the MAC of an unstaggered, equal winged biplane was at around 23% of the chord behind the leading edge, in contrast to the 25% figure usually used for monoplanes. I don't know of any explanation for this. Always remember that the CG should be a bit behind the MAC normal configurations. With a normal tail moment, and horizontal tail area of 20% of total wing area, I have found that the CG can be at about 40% aft of the equivalent monoplane leading edge location, for best aerobatic line holding, but a little further forward enhances pitch stability while compromising line holding a little.
Posted on: 4/16/2009 1:29 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8687061

RE: Measuring Total Plane Drag
Towing a drag chute could get a bit complicated, since the drag of a parachute varies quite a bit with shape, but the parachute drag could be calibrated in a relatively small wind tunnel, or perhaps by towing behind a car on a towline long enough to get it out of most of the wind wash generated by the car. Most university engineering departments have wind tunnels, but they usually have rather small test sections. Only a few of the larger universities commonly have wind tunnels with test sections larger than about two feet square, and they are very costly to operate. I did some engineering work to fix an underperforming wind tunnel recently which had a test section of about three feet square, and consumed about 300 HP to produce about 150 mph wind, generating quite an electric bill in the process.
Posted on: 4/15/2009 6:27 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8684845

RE: Measuring Total Plane Drag
I made a math mistake in preceding post, and inadvertently submitted while checking calculations. Correct dive height to attain 115.5 mph terminal velocity, neglecting air resistance would be about 445 feet, and probably about 600 feet with air resistance. Correct dive height needed to attain 181.4 mph without air resistance would be about 1100 feet, and probably about 1500 feet with air resistance factored in. Both appear do-able if the airplane hangs together and doesn't turn itself into a dangerous missile.
Posted on: 4/11/2009 9:40 AM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8670263

RE: Measuring Total Plane Drag
Measuring static thrust is easy - I just use a fish scale tied to the tail. Thrust in flight is normally lower than static thrust, and needs a wind tunnel for accurate measurement. Consider a fairly clean airplane that can achieve 100 mph (146.7 feet/sec.) straight and level, with a 2 hp engine. Note that one hp = 550 foot-pounds per second. At 70% propeller efficiency, thrust = (2x550x0.7)/146.7 = 5.25 pounds. Getting back to my proposal for measuring drag by finding terminal velocity in a vertical dive, with a dead engine, when drag equals weight: If the airplane weighs 7 pounds, it would clearly go faster in a vertical dive than at full power, straight and level. Its actual speed in vertical dive, neglecting the drag of the stationary prop, would be the square root of (7/5.25), multiplied by the maximum powered speed, straight and level of 100 mph, which equals 115.5 mph. If the airplane weight is increased from 7 to 10 pounds, vertical dive terminal velocity would increase to 138.1 mph. If 10 pound airplane is extremely clean, and capable of 120 mph straight and level, thrust at 70% prop efficiency would be 4.375 pounds. Vertical dive terminal velocity would increase to 181.4 mph, neglecting the increase expected from the higher Reynolds number. The height needed to attain this high a terminal velocity would be probably be over 5,000 feet. Height needed to attain a terminal velocity of 115.5 mph would be about 2,000 feet, do-able with good pilot eyesight.
Posted on: 4/11/2009 9:23 AM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8670239

RE: Measuring Total Plane Drag
CrateCruncher, you are correct. Above stall, induced drag is inversely proportional to the square of airspeed, and is relatively unimportant at several times stall airspeed. For example, at five times stall airspeed, induced drag is only about 1/25th as great as at near-stall airspeed. Here is a NASA blurb on induced drag that may be useful: http://www.grc.nasa.gov/WWW/K-12/airplane/induced.html Measuring drag is quite a tough nut to crack, unless you can get ahold of a wind tunnel, and test a reduced scale model of your airplane. Another way to measure drag would be to use onboard airspeed telemetry, kill the engine, dive your airplane straight down from high altitude until it reaches its maximum airspeed. This may produce a dangerously high airspeed and cause the airplane to break up from flutter, turning it into an unguided missile. The drag of the stationary propeller can be calculated reasonably accurately; a windmilling propeller produces tremendous drag that is also difficult to estimate. At maximum speed, vertically down, drag is equal to weight.
Posted on: 4/4/2009 10:11 AM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8647609

RE: Winglets on Sig Cap 231 EX?
I tried rather large winglets, actually more like tip plates on a competition fun-fly model. I had to give them nearly as much area below the wing as above, to reduce their dihedral effect. Made them of epoxy laminated balsa, with hardwood hard points for mounting by small nylon screws so that they would break away without damage on ground contact. The reason for my tip plates was to reduce induced drag. A slightly longer wingspan would reduce induced drag much more efficiently, without nearly as much increase in parasite drag, but a longer wing would also reduce the roll rate. With the tip plates in place, the model loops considerable tighter and faster, and also handles nicer, with very friendly behaviour much deeper into stall, due to its greatly increased yaw resistance. As you would expect, the bird also knife edges far better. Also tried ugly detachable vertical airfoils at about half span on wing of a piped Webra 120 powered, frantic flying 9 pound aerobatic monoplane with coupled flaperons, trying to make it knife edge as well my bipes, with mixed success.
Posted on: 3/25/2009 9:54 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8617004

RE: CG locations other than 30%mac
Full scale aircraft often have forward limit of CG well ahead of the 25% chord aerodynamic center of the wing, forcing the horizontal tail to develop significant download. One example is the all-moving tail on the Cessna Cardinal, with a slotted inverted airfoil to increase maximum download capability. The Cessna 172 manual calls for forward CG limit of only 15.6% of wing chord, and aft limit of 36.5%; a range of over a foot. At the aft limit, the tail would be developing uplift, increasing the efficiency of the airplane by reducing the necessary wing lift coefficient. I tested a well-known model design computer program by throwing a proven canard design at it, just to see whether the software knew how to handle larger than normal horizontal stab area. It said that the canard would have a hopelessly high wing loading, and also be so tail heavy as to be unflyable. It thought that the forward-mounted canard tail was the wing, and the rear-located wing was the tail. It clearly didn't know how to handle a canard, which told me that it was only able to handle very conventional layouts, and unable to calculate correct CG layout for stabs larger or smaller than run-of-the-mill airplanes.
Posted on: 3/21/2009 11:40 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8602309

RE: P51d mustang aerofoil section
I used an airfoil that was quite similar to scale Mustang wing airfoil on a sport pattern model, at 12% thickness, maximum thickness at 50% chord, and found that it tended to get into what seemed like incipient stall at much lower lift coefficients than the NACA 0012 airfoil used previously. The airplane tended to wander off track in maneuvers when the wing was forced to a lift coefficient over about 0.5, but was just fine for large radius maneuvers, and seemed faster than the 0012 airfoiled wing. I think that the Mustang's so-called laminar flow airfoil is better suited to much higher reynolds numbers and Mach numbers over about 0.7. I heard that a laminar flow airfoil was tried on a full-scale P-47, with negligible performance improvement.
Posted on: 3/21/2009 11:20 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8602247

RE: CG locations other than 30%mac
One thing that is certain about experimental aerodynamics, it keeps you humble. I have found that when I am having a hard time getting my head around some concept, it often helps to consider an extreme case. Canards (tail first airplanes) are a good extreme case of grossly shrinking the wing and blowing up the horizontal stab of a normal tractor configuration airplane. Canards need to have their CG far aft of the front-mounted stab. A canard that had its CG in the usual location for a tractor (rear stab) airplane would be so nose heavy as to be virtually unflyable. In my book, any normal tractor configuration airplane that has its CG behind the aerodynamic center of the wing forces the stab to develop uplift; otherwise the airplane would be unstable in pitch. Flying wing airplanes have no stabs, and need their CG to be no farther rearward than about 25% of the wing chord, and 20% would be safer and friendlier. Adding a rear mounted stab permits the CG to be considerably farther aft, since the added stab area moves the aerodynamic center of the airplane considerably rearward. When faced with some unusual planform, I simply whomp up a simple, cardboard, profile fuselage hand-launched glider with about a six inch wingspan and simple flat plate airfoils, and glide test it. I give it equal wing and stab incidence. I add or subtract paper clip nose weights until I get a nice flat glide without stalling or diving, and note the CG position. I don't know why highly successful competition free flight models had such large stabs, but such configurations were arrived at by much patient experimentation by highly experienced and talented people, so I will keep close-mouthed on this.
Posted on: 3/20/2009 4:04 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8598019

RE: CG locations other than 30%mac
For best aerobatic line holding, I have just one flight test for CG location. I trim the airplane to fly level upright, then roll it over on its back. If it needs more than just about 5-10 percent of full down elevator to hold it level while inverted, it is nose heavy. If it climbs, it is tail heavy. If the airplane tends to snap roll with application of a lot of elevator, the elevator travel needs to be reduced. I like to be able to obtain at least a half loop with full elevator application before the airplane starts to wander badly off track. If it performs more than one full loop before wandering, its elevator travel is too low. If it wanders before doing a half loop, elevator travel is too high. Always remember that elevator travel must be reduced when you move CG rearward, and increased when you move CG forward. Best CG location is very heavily dependent on the horizontal tail area and tail moment arm. We used to trim freeflight models that had very large horizontal stabs of up to 50 percent of wing area, at 80 to 100 percent of wing chord behind the leading edge, in order to permit the tail to carry its share of the load and reduce tendency to loop under power. Small horizontal tails dictate locating the CG as far forward as 20 percent chord. For non-aerobatic airplanes, correct CG location depends very much on personal preference, in my book. Basic trainers need to have the CG well forward, in order to obtain good stability in pitch, with the airplane lowering its nose automatically when its airspeed gets a bit low, to prevent inadvertent stall/snap roll.
Posted on: 3/18/2009 4:52 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8591427

RE: Prop efficiency, 2 versus 3 blades
One blade props are usually a bit more efficient, and two blade props are slightly. usually one or two percent more efficient than three blade units. Full scale airplanes usually go to three or more blades in order to absorb more power with a given diameter, thus keeping the landing gear from becoming inconveniently long and heavy, and keeping tip speed below about Mach 0.85, where efficiency starts to suffer because of shock wave generation at the highest velocity part of the airflow around the airfoil. However, wind tunnel tests revealed that model-sized propellers of under 24 inch diameter started losing efficiency at tip speeds over Mach 0.6 or so. For a given horsepower, three or more blades also tend to be quieter. Control line speed flyers commonly used props with pitch/diameter ratio of up to about 1.4, such as a 9-13 for .60 engines, obtaining optimum efficiency for model propellers. Propellers with pitch/diameter ratios of 1.0 or more are best for efficiency, possibly attaining up to about 85% in model sizes. At a pitch/diameter ratio of 0.5, for example, a 12-6, peak efficiency is probably down to about 70%.
Posted on: 3/5/2009 11:00 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=8548781

RE: Wingtip fences
Wing tip plates increase yaw resistance, and thus can permit maintaining control deeper into stall, as well as increasing yaw resistance. I experimented a lot with three different designs of tip plates on my airplane design pictured. The tip plates definitely increased loop tightness, handling, and knife edge. Made from 3/16" balsa plywood, attached by 4-40 nylon screws so they break away upon striking the ground. The airplane has 60" span, coupled flaperons with boost tabs, piped Webra .50, weight 3.1 pounds. Goes like a scalded cat, and a blast to fly. Tip plates need roughly equal area above and below the wing, or they will cause a dihedral/anhedral effect. Tip plates reduce wing drag at high lift coefficients, permitting a slightly shorter wingspan which increases roll rate.
Posted on: 11/26/2007 2:03 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=6676057

RE: Wingtip fences
Adding tip plates will reduce induced drag at lift higher lift coefficients quite considerably, but the reduction in drag would be greater if an area equal to the tip plate was simply added to the wingtip. Clipping the wings and adding tip plates would probably be a bad idea, in general. Experiments with NASA winglets revealed that the best angle in relation to the wing long axis was straight out. In other words, an slight increase in span, roughly equal the the height of the winglet will yield a larger induced drag reduction than that of the winglets. Here is an excellent old NACA paper on wingtip plates: http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930091335_1993091335.pdf Hopefully, it comes through clickable.
Posted on: 11/25/2007 3:59 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=6671551

RE: Wing rock, dutch roll?
Your airplane has a lot of lateral area far ahead of the CG, which can destroy yaw stability. I would try taping on some temporary cardboard extensions to the vertical tail fins as a quick check. The CG may be a bit too far back as well, causing the nose to pitch up, and turbulent air from the wing and horizontal canard tail to blanket the vertical tail. Increasing the height of the vertical tail as well as its area might be a good idea, to get the vertical tail into cleaner air when the airplane is at high angles of attack.
Posted on: 11/22/2007 7:57 AM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=6657910

RE: Inboard vs. Outboard Ailerons
Wind tunnel tests that were conducted many years ago indicated that the optimum length of ailerons was about 60% of the span, extending to within about a half chord of the wingtip of a 40% tapered wing. Such ailerons were found to produce the the minimum adverse yaw and drag when deflected. Extending the aileron all the way to the wingtip was found to produce slightly less adverse yaw, but considerably more drag, in return for only a small increase in roll authority. If wing flaps are not used, I like ailerons that are essentially full-span, since the aileron portion swept by the propwash can provide a lot of roll control, by deflecting the propwash at very low airspeed combined with high power. Such ailerons provide very solid roll control extending into deep stall. When I built an airplane with flaps and half-span ailerons, I had to stay farther away from stall on touchdown, or a wingtip would drop. Aileron-induced yaw was also noticeably increased.
Posted on: 11/14/2007 9:12 AM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=6623532

RE: I covered my control surfaces with light ply did I mess up?
You can also mount mass dampers (also known as mass balancers) internally, but this will probably require a blister on the wing or stab surface to accommodate the needed control surface throw. The P-51, for example, has two blisters on the wing, as I recall, to make room for internally mounted aileron mass dampers. I usually try to mount them inside a small aerodynamic balancer area ahead of the hinge line, at the tip of the aileron or elevator. Locating the dampers farther inboard is better, since concentrating the balance weight at the extreme tip can cause the aileron or elevator to twist, and excite a higher harmonic frequency of flutter that can be very destructive. If the balance weight is located at the tip, the weight should be sufficient to statically balance only about 25% of the aileron or elevator unbalance, to avoid exciting a higher harmonic. I had one airplane self-destruct in high speed flight simply because it had long, flexible ailerons with too much balance weight at the tip. After rebuilding the mess, I reduced the tip balancer to about 25% of the total unbalance, and added an intermediate balancer mounted on a stalk at about 40% from the root of the full-span ailerons. The flutter never recurred.
Posted on: 10/23/2007 12:47 AM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=6526480

RE: I covered my control surfaces with light ply did I mess up?
When an airplane self-destructs in the air when it was not pulling excessive "G" load, it is probably flutter. This kind of flutter is very sneaky, and can occur in an airplane that has flown many hours with no sign of trouble. It needs an initial flex of the part of the structure concerned, after which the flutter thing takes the bit in its mouth, and the oscillation goes through just a few, maybe half a dozen cycles at most, of very rapidly increasing amplitude until something breaks, detuning the system. Unusually heavy control surfaces are poison, flutter-wise. I used 1/64 ply covered control surfaces on some UAVs designed and built for the military, and found that mass dampers were absolutely essential - otherwise the control surfaces would rip off in flight. Many WW2 fighters used fabric covered control surfaces on otherwise all-metal airplanes, simply to reduce the likelyhood of flutter. Although a heavy control surface can be more or less tamed by a mass damper, the additional mass near the trailing edge of the wing is very bad, and can force wing into flutter. Ailerons are the usual suspect when wing flutter rears its ugly head, but wings without ailerons are quite capable of getting into violent flutter. The ideal wing would have its center of mass located near its chord-wise center of lift, somewhere around 25% of the chord behind the leading edge. This is highly impractical, except for helicopter rotor blades. The further rearward the CG of the wing or stab ventures, the higher the likelyhood of flutter.
Posted on: 10/21/2007 4:42 PM by Author "Rotaryphile" in the forum "Aerodynamics"
http://www.rcuniverse.com/forum/fb.asp?m=6519793