Functions of trim adjustments
B. Buck, December 3, 2005
The trim of the model is by far the single most important factor (aside from the quality of the engine run) in making the airplane fly well. To fully trim the airplane, people have added more and more adjustable trim features. Everybody know that you are supposed to include all the modern adjustable features in the your killer stunt ship. But it’s often asked what they all do! Without attempting to cover how to trim a stunt plane (a topic upon which we could easily get to Encyclopedia Britannica dimensions), it’s at least worthwhile to know what nominal effects each trim adjustment had. I say “nominal” because one of the most important factors to learn in trimming and even designing airplanes is that a change in one area almost always affects everything else. There are a few examples below, but these are just the tip of a huge iceberg, and the descriptions, while hopefully useful, are an oversimplification. I have divided the adjustments into those that primarily affect the Roll and Yaw axes, and those that effect Pitch. For those who haven’t been studying aeronautics since their pre-natal days, these terms may be unfamiliar or at least not well-defined, so I’ll take a stab at it. The pitch axis of the airplane is the one you control - nose up and down. Positive pitch is nose up. Roll is rotation around the “long” axis of the airplane from the nose to the tail. A positive roll tilts the airplane to the right, so that, for example, in level flight, the right wing is down and the left wing is up. Yaw is a rotation around the “vertical” axis of the airplane, nose to the left or right. A positive yaw is a rotation to the right, so that the nose is aimed to the right of where you are going.
Note that these are all relative to the airplane. Right/Left/Up/Down are all determined as if you were in the pilot’s position - not, importantly, relative to the ground, unless you happen to be flying level. That’s important to remember, and I have seen many, many situations where people became confused about it.
Also, given that they are angles, it begs the question “angles relative to what”. That’s an interesting question, that would take some time to think through and explain, but for the time being let’s say that roll and yaw are relative to the line drawn from the airplane to the pilot. So, there you are, flying happily through the maneuvers. At any instant, you could imagine a line drawn from your hand to the center of gravity of the airplane. In level flight it would be parallel to the ground. In the intersection of the overhead 8, it would be perpendicular to the ground. The roll angle is the angle between this line and the “side-to-side” axis of the airplane, and the yaw angle is the angle between this line and the “long” fore-and-aft axis of the airplane. And let’s say that the pitch angle is the angle of the long axis of the airplane relative to the current direction of flight. They aren’t the same! The nose can be pitched up or down relative to where you are actually going. This is not an entirely satisfactory definition, but it will do for now.
I will also emphasize that virtually none of the information in this article was invented, discovered, or explained by me. It’s an accumulation of the knowledge of others, and I don’t even know where I first heard of some or most of it. So I can’t provide attributions properly. I apologize in advance. There’s no intent to claim it as my own. With the preamble and caveats out of the way, on to the adjustments.
Tip weight: primarily affects the ROLL axis More tip weight than necessary rolls airplane away from you, which, VERY APPROXIMATELY directs the lift vector away from you, and gives you more line tension. Too much and it pulls hard, but the line tension varies a lot, too, and the airplane oscillates in roll (hinging). Adjust to keep roll angle at zero through maneuvers.
Rudder offset (fixed/ground adjustable):
Primarily affects the YAW axis, although yaw and roll are very strongly coupled. Set the *equilibrium* yaw angle, or the yaw angle at which the airplane flies in steady level flight. In my opinion, the goal is to set the equilibrium yaw angle to zero. Rudder offset (in-flight variable - "Rabe Rudder"): pitching the airplane while having a big prop up front, spinning at goodly RPM, causes the airplane to want to yaw "nose-out" on inside maneuvers, and "nose-in" on outside maneuvers. This cannot be fully corrected by a fixed/adjustable rudder. So Al Rabe figured out to hook the rudder to the elevator, so it moves to give more left rudder on insides, and more right rudder on outsides. Usually, it needs to be asymmetrical, moving a lot more right than left (for reasons that might or might not be obvious). Note that the purpose is most emphatically NOT to make the airplane yaw nose-in or nose-out, in fact it’s to prevent the nose from yawing in or out by providing a compensation that counteracts the precessional torques. If you adjust it right, it compensates well for the precessional effects. Al’s worked well. Problem seems to be that unless you are an absolute expert-level trimmer you will not get it adjusted correctly and, given that it's a very powerful feature, it is almost always overdone and causes more issues than it solves. This includes many "name" experts.
Primarily affects the ROLL axis. Bend the flap horn for a ground-fixed "aileron effect" to make roll angle identical with positive and negative accelerations (Gs).
Primarily affects the YAW axis, but cannot be separated from rudder adjustment. I use the leadouts to take out *transient* effects, so that the leadouts and the equilibrium yaw angle are *complementary*. Too far aft, and you a lot of line tension in level flight, but lose it overhead. Too far forward, and the airplane noses in at every control input and loses line tension.
The leadout position is related to the center-of-gravity. For my purposes, the "baseline" leadout position can be calculated using the computer program "LINEIII" (downloadable below). The position calculated this way is the position that corresponds to the "0 yaw angle" ideal. Others use the leadout position to create "opposing forces" so that they have a rather large equilibrium yaw angle, and then overcome the ill effects overhead by forcing the nose back in with a forward leadout position. This creates yaw, roll, and (because the line tension changes when yawing and rolling) pitch transients.
Differential flap area:
This "adjustment" primarily affects the roll axis. It exists to compensate for an interesting observed effect - that sometimes it looks like you need less tip weight to do a square corner than a round corner, and less tip weight to do a round corner than to fly level (all assuming 0 roll angle). Folklore says this is so you can "carry more tip weight", which is vaguely correct if you only see things in "binary". It's really because of aerodynamic asymmetry effect (even if you have equal-span wings) of flying in a circle.
This "adjustment" primarily affects the roll axis. It was originally envisioned as "using the fuselage/engine as tipweight", which, once again, is vaguely, notionally, correct. In fact you are attempting to line up the lateral CG position with the lateral center of pressure. 1/2-3/4" is about right, more leads to less tipweight, but more likelihood of needing a lot of differential flap area. That's because everybody just moved the wing off center - and left the tail right down the middle!
Wing fences/drag tabs/drag vanes - other ways of doing various things. Never seemed to prove useful over the long haul, sometimes useful for specific problems, I don’t think it’s worth going into at this point.
Nose weight/CG: used to adjust the CG of the airplane, which controls the stability, and secondarily, controls pitch rate/lift ratio. More nose weight makes airplane require more control movement and thus control force for a given pitch rate and is more "stable". More stable also means more tendency for corners to "open up" in the wind. The more aft the CG the less control motion it takes for a given pitch rate, and the less "stable" it is.
Used in conjunction with the CG adjustment to create the desired control response for the pilot. More spacing = more sensitive, less spacing = less sensitive. In general set the CG the way the airplane wants, usually as far aft as it can go and still have *slight* positive stability. The maximum allowable aft position depends on the tail volume coefficient. Then set the handle spacing for the control sensitivity you want.
Flap/Elevator movement ratio:
Used to set the pitch rate/lift ratio, which sets the quality of the turn. In some sense this depends on the wing loading - heavier and you want more flap motion than elevator, and lighter and you want less flap motion than elevator. This can be a rather subtle adjustment and for most people just setting it to 1:1 (or whatever the original designer wanted) is probably going to work for you.
Used to adjust the inside/outside turn rate, OR, maximize tracking/stability. Most airplanes of conventional layout need some down elevator at neutral flap to fly best. Partly this is because of aerodynamic asymmetry (thrust line/wing/tail out of line and/or other drag asymmetry) but mostly it seems to be due to the pitch component of gyroscopic precession (which creates a constant nose-up pitch torque, for which the "down elevator" is compensating). Adjust to find the "sweet spot" in tracking. Usually, ANY "up" elevator at neutral flap is severely destabilizing, and a lot of down seems to be pretty well tolerated. Positive stab incidence is just a (permanently built-in) version of the same thing. I would recommend doing this ONLY if you have discovered that a particular design always needs it. It's obviously hard to adjust so leave this to the experts (fools) like myself. If you build an Infinity it wants to be about 1/4 degree.
Used for roughly the same purpose as Flap/elevator neutral, compensating for the same asymmetrical pitch effects. More downthrust raises the thrust line, creating a nose-down pitch torque. Easier to adjust than positive incidence, assuming you don't mind messing up the spinner fit. Once again, build in only if you really know it's necessary or beneficial. I would guess it's less likely to screw anything up than positive incidence. On the topic of these pitch asymmetries - Folklore indicates that "since the airplane has to fly the same upright and inverted everything has to be 0-0". Once again, folklore is vaguely correct, in that it needs to fly the same upright and inverted. But our airplanes are nowhere close to symmetrical in the pitch axis even if the wing and tail are aligned perfectly. So there's nothing magic about 0-0-0 (wing/tail/engine), in fact it's a gross oversimplification. That knowledge helps ONLY to break the closed-minded assumption that if you put it all at 0-0 and it doesn't fly perfectly, it must be "misaligned".
I think this more or less covers most trim adjustments.
Basic (simplistic) trim process
This is a very over-simplified sequence, but if you do it correctly it will get you very close to the optimum trim, assuming the airplane doesn’t have any serious misalignment of the stabilizer to the wing.
Set the CG: The desired starting CG position depends on the tail volume. It’s somewhat of a debatable strategy, but Ted Fancher has a rule, that despite it’s simplicity works out about right assuming you have a relatively conventional airplane.
Compute the ratio of the tail area to the wing area (including flaps of course). That should come out to something like 0.18-0.26 (or 18 to 26%). Then set the CG at about that fraction of the mean chord. So, for example, you have a 25% tail, and an 11” mean chord. The CG should be about 2 3/4” from the leading edge at the point of the mean chord. This is the MAXIMUM aft position. If you are going to make a mistake, or can’t get it quite where it says, err on the side of further forward. And, importantly, don’t leave it in the wrong place to save weight! If you need 3 oz. of nose weight, then so be it. You can’t save weight by compromising the trim. Or, actually, you can, but you’ll lose an awful lot of contests that way. Same with tip weight. Trim is more important than weight, period. That’s so important I will repeat it - Trim is more important than weight!
Plug in all your numbers into LINEIII (program file available below) and make sure you take out the “whip angle” that is in the default settings. Then put the leadouts where it says relative to the CG. (labeled “Leadout Offset”). For a check, a typical 64 ounce 40 -60 airplane on .018 lines, it comes out about 3/4”. So put the centerline of the leadout guide /75 inches behind the balance point. Unless the designers says otherwise, set the flap/elevator ratio to 1:1 and the flap/elevator neutral to 0-0 or a little bit (maybe 1/16) of down elevator at neutral flap, and no downthrust.
Fly it upright and inverted. If the outboard wing is high both ways, add tip weight until it's level. If the outboard wing is high one way and low the other, tweak the flaps to roll the airplane so it's the same upright and inverted. Watch in hard corners. If the airplane rolls away from you in hard corners, take out tip weight. If it doesn't, add tipweight a little at a time, until it does noticeably roll away from you, then take out the last bit you added. If it pulls hard in level flight but severely falls off overhead, or tends to go loose on the first leaf of the clover, move the leadouts forward a bit (1/16 at a time). If it suddenly "comes loose", usually on insides, move the leadouts back a bit until it stops doing that.
That should get you pretty close. If you have an abnormally light airplane watch for a tendency to "swoop" into corners (looks like it rotates around a point behind the airplane) increase the elevator motion relative to the flap A LITTLE BIT. If you have an abnormally heavy airplane you might need to increase the flap motion relative to the elevator. Adjust the handle spacing to get an agreeable control response. Don't set it up for excessively quick response. Once you get it close, start watching for the airplane to “leap” around corners, come out higher than you expect on round loops, or have the control effort suddenly become “light” in square corners. If that happens, try moving the CG forward a little bit at a time until it stops, and then readjust the handle to get the right sensitivity. If the airplane feels heavy on the controls and wants to come out lower than expected on round loops, move the CG aft a little at a time until it quits.
Another CG measure is what happens when the engine quits. If the airplane pitches “up” right when the engine quits or is very “floaty” in the glide, the CG is probably well too far aft. If the airplane pitches “down” as the engine quits, and is very easy to “”whip” in the glide, the CG is probably well to far forward.
That's a far as I think we can go in the hypothetical, but it will be *very close* if everything is straight and reasonable weight and power.
Center of Gravity
by Ted Fancher
(Reprinted from Stuka Stunt Forum)
It occurred to me after wasting more bits and bytes than I could afford on some of the threads started by Godzilla regarding Center Of Gravity, leadout postition and so forth that I had given short shrift to a question which wasn't actually asked but was implicit in the discussion. Specifically, the affects--pluses and minuses, etc.--of moving the CG about with a fixed leadout location. I got hung up on the classic argument of finding the "right" relationship between the two and the need to move them in concert with one another to maintain the "ideal" tangent relationship of the aircraft heading relative to the circle. Here's a little discussion to get the neurons flowing.
It is true that, all other things remaining fixed, moving the CG forward will result in a yaw angle away from the center of the circle. This *will* result in some predictable outward vector of the engines thrust and *will* affect line tension. Whether tension is more or less would depend on the variables we discussed earlier...whether the airspeed is reduced due to less forward vector of the thrust, whether drag from the yawed condition slows the aircraft reducing centrifugal force more than that gained from the outward thrust vector, etc.
Thus, if your powertrain has the capability you could purposely fly with the leadouts aft of the CG a desired amount and, by adapting the powertrain to compensate, the overall line tension would be increased. (Example: The yaw decreases airspeed and thus line tension by "x* amount but the offset thrust vector increases it by *Y* amount which results in a slightly lower line tension of X (-) Y. Now the engine is dialed up or the prop pitch increased to regain the lost airspeed and, thus, the lost tension *X*. The net result is a positive line tension increase of *Y*)
Whether or not this will be good throughout the flight hemisphere is subject to a lot of things like where the resulting revs are on the torque curve and, thus, the ability of the powertrain to overcome the greater drag while maneuvering, etc. With modern engines (particularly with the big four bangers 'zilla champions) this is not likely a problem unless the wing loading is particularly onerous.
Thus, an airplane can be trimmed to the pilot's preference with a more forward CG relative to the leadouts than others would find *ideal*. Al does that all the time and kicked some serious butt during his competitive career while doing so.
All the foregoing assumes, of course, that the CG isn't so far forward to cause the detrimental effects discussed in the thread on forward CGs and lift.
What does this all mean?
Well, it means that thinking about it has turned on a great big light bulb in my fuzzy brain about differences between airplanes trimmed elsewhere in the world and those trimmed for maximum performance here in the states. Here's what I've come up with in the last week or so of thought.
The point has been made numerous times (and I've experienced it firsthand during my World Championship experiences) that flying sites outside the states are generally configured so as to make turbulence a common factor. Lots of trees, hills and buildings to disturb whatever air movement there is.
For the most part, in the states raw turbulence of that sort isn't the issue...particularly where we've tended to hold our major competitions for the last 30+ years; Naval bases, inactive Air Force bases, Municipal airports, the Muncie AMA site etc. In the states our air related problems are usually the result of high winds which brings with it its own peculiar sorts of trim demands. Turbulence, however, is a comparatively minor factor due to the wide open venues of our major comps.
Here's the kicker.
The trim requirements appropriate for dealing with turbulence are not necessarily the same as those appropriate for flying stunt in moderate to high winds. The recent FAI ruling that judges *shall not* move about the circle to place themselves properly for maneuvers but that the pilot should place the maneuvers where the judges can best observe them has made the distinction in trim requirements even greater.
High winds generally make the movement of judges a non-issue. Judges love it when the wind blows hard because there is no question where the maneuvers will be flown. High winds seldom change direction a great deal and therefore it is easy to both pick the proper place to fly maneuvers as a pilot and easy to know where that is going to be for the judge.
Turbulence is a different animal entirely. By definition, turbulence is some form of wind shear, air masses in close proximity whose direction of movement differs to greater or lesser degree. The result is uncertainty on the part of both judge and pilot where exactly the best place to put the maneuvers will be.
The FAI solved this problem neatly by saying the judges ain't going to move after you tell 'em where you intend to do your tricks. Ergo, the judges problem is solved. The pilot's problem has, however, been made seriously more complex. He has declared where he will fly and he must find someway to do so regardless of what changes he may feel during flight due to the turbulent airflow throughout the flight hemisphere. What to do????
Here's the rub...and maybe a look into a crystal ball regarding future trim methods in the US.
When we trim for our usual conditions in the states we (at least those of us on the left coast) do so to maximmize our ability to turn aggressively in conditions ranging from dead calm to high winds. This has led to the trim conditions championed in the other threads. Aft CGs with the leadouts placed so as to keep the ship tangent to the circular flight path, large tails to allow the aft CGs etc. etc (you can read the old threads for more detail than you probably want).
Such a trim set-up loses little when the winds blow. The pitching moments of CG/CL displacement are minimal and the airplanes are plenty stable and flying in high winds is not a significant problem once you get the blamed thing in the air. For our usual US conditions this sort of trim set-up has proven nearly bullet-proof. Witness the dominance of this style of trim since 1982...Walker (Nats titles and team trial victories by the Baker's dozen), Fitzgerald (three nats titles and several team trial successes), and Fancher (four Nats titles and two Team appearances including one victory...in the trials, not the WC). This is a record that defines "dominance".
When we go overseas, however, the dominatrix act seems to lose its whips and chains. I think the site conditions and our common trim set-ups have more than a little to do with this.
Let's revisit the effects of the forward CG and aft leadout position. Quite a common condition for competitive European designs and, in particular with the Beringer style ships.
Offsets of engine and rudder and large amounts of leadout sweep relative to the CG are anathema to an airplane trimmed to fly well in non-turbulent air...even if it's blowing by at 20 knots. If the air's going in a straight line we want our ships set-up to minimize the deleterious effects. These deleterious effects are primarily the problem of the ships winding up in the wind...going faster and faster. When they do so the effects of a large moment between the CG and the CL are exacerbated. The faster the ship goes the greater the Gs required for a given radius (which must stay uniform per the maneuver descriptions). As the ship flies faster and faster any siginificant nose heaviness tries to open up the maneuver and, if too extreme, can make it impossible to fly maneuvers of the size and corner aggressiveness we desire.
If your problem is turbulence, on the other hand, the issue isn't maintaining adequate aerodynamic ability to turn. The problem becomes a mechanical one...simply keeping the airlane on the ends of the lines so that you have the tension necessary to transmit the desired control input to the airplane. In turbulence the airplane is apt to be dancing around, up and down, in and out...an infinite number of possbilities exist when the air is being baffled and flung about buildings, trees and the oragraphy of the site.
This, I think, has been the direction that airplane trim has taken in most of the stunt world outside the US. Airplanes are trimmed to, perhaps, less than their maximum performance potential in an effort to retain what performance they do have while flying in unpredictable conditions.
A classic case of exactly what I'm talking about is the very fine Firecracker design flown a decade or so ago by Brian Eather of Australia (and CF tuned pipe and prop fame).
The Firecracker is almost a chariacature of the type of trim setup I'm talking about. Several degrees of engine offset, lots of rudder offset, and a leadout sweep which can only be termed substantial, I'm guessing 12 to 15 degrees aft of the CG at the point they exit the wingtip. In addition the CG was very far forward. An almost diametrically opposed trim set-up to that used so successfully in the US.
This airplane flew very slow lap times and had substantial tension at all times. In addition it was strange to launch because it had to be angled out 20 or so degrees to accomadate all the offsets. Looking at this ship hanging on the wall of his guest bedroom offers evidence in the form of the wheels being literally worn to a nub from traveling sideways so much on take off and landing.
It is interesting to note that when Brian broght the Firecracker to the US for (I think) our 1985 nats the wind blew hard and straight and, predictably, the Firecracker ran out of corner exactly as we've discussed.
On the other hand, flying that ship in modest winds but at the very turbulent, tree surrounded flying site at Kuringai north of Sydney it handled the turbulence wonderfully.
I think we could well find ourselves leaning in the direction of european style trim set-ups for future world championships. I think we'll compromise our fine tuning for max performance so as to accomadate turbulence while maintaining an acceptable level of pattern...an acceptable level that can be managed even under difficult turbulent conditions. I think part and parcel of the changes will be offsets in leadouts relative to the CG and greater acceptance of engine offset. Not so sure about rudders.
There are a few other considerations that might require a slightly different configuration of the overall appearance of the airplane. I'm going to keep those thoughts to myself for a bit because I think they may well be the key to making these fundamental changes agreeable to the pilot.