What happens when you lengthen the fuse on planes?

I happen to be building the "Paddle", a twin boom speed 400. I'm subbing aluminum tubing for the supplied hardwood dowels. The tubing I have is several inches longer, and I'd hate to whack it off per plans, if leaving it a tad longer would help stability! I still will balance at the point in the plans?

From : Don Stackhouse

Any time you make a small but significant change to something in an airplane design, there will be ripple effects that influence nearly every other parameter in the design.

A whole bunch of things happen when you change tail moment arm length.

First, let's look at static stability. Adding a percentage of tail moment arm has the same effect on pitch and yaw static stability as adding the same percentage to the horizontal and vertical tail areas. It will move the neutral point (the C/G location that results in neutral static stability) further aft. If you leave the C/G the same, you will see and increase in static stability, or you can move the C/G aft a little to keep the static stability the same as before.

Dynamic stability (the ability to damp out oscillations) is linearly proportional to tail area, but proportional to the SQUARE of the tail moment arm. For example, if you doubled the tail area, both the static and dynamic stability would double. However, if you double the tail moment arm, you get twice the static stability, but FOUR TIMES the dynamic stability you had before.

So far, so good. Let's all make our tail moment arms a gazillion inches long, right? Well, not quite. There are some negatives.

A longer tail moment arm results in more whetted area (and therefore skin friction), and more weight, for the tail section of the fuselage. For a pod and boom type of design, the penalty is usually small, but real. To some extent this can be offset by making the tail surfaces smaller as you make the tail longer, but there is an optimum. I have on occasion used total whetted area (on a Q-40 pylon racer, for example) or total tail assembly weight (on a pod-and-boom sailplane design that was extremely weight sensitive) to determine the optimum tail moment and tail area combination for a given amount of stability. There is an optimum, and the design gets worse with respect to those criteria if you have too little or too much of each parameter.

Too much yaw stability, coupled with insufficient dihedral effect, can result in spiral instability (i.e.: the plane tends to steepen the bank angle by itself when held in a sustained turn, sometimes called "overbanking tendency"). This is usually not a big issue for a sport model, but for something that has to do thermalling turns it can be a major problem. This is also somewhat dependent on lift coefficient, so your model might have positive spiral stability (i.e.: tends to roll back to level by itself) at some bank angles and be spirally unstable at other bank angles.

Conversely, too little fin effect and/or too much dihedral effect can result in "dutch roll", a side-to-side, wallowing, falling-leaf type of motion. The inertia of the model about the yaw axis is a major player in this as well, and a longer tailboom will increase the inertia. Usually there is some combination of dihedral/fin area/tail moment that will simulrtaneously result in an absence of dutch roll AND acceptable spiral stability at typical bank angles. However, if the inertia is too high, the zones of combinations that represent dutch roll problems and spiral stability problems can overlap, resulting in a model that always has one problem or the other or both.

Excessively long tail moments can sometimes (but not always) result in poor control response as well. The extremely small tail areas that often go with very long tail moments can also have Reynolds number problems, resulting in poor control authority and quirky handling. At least one fairly popular long-and-small-tailed 2-channel RCHLG I've had the displeasure to test fly had an annoying hysteresis in pitch, and was about as nimble as a school bus. Much of the problem could be attributed to exessively small tail surfaces (with a less than ideal airfoil for that Re), and too much tail moment.

Aeroelasticity (the aerodynamic effects of the way the model's shape changes in flight due to the flight loads imposed on it) are also an issue. If you change the material and/or the length of the tail boom, you are also changing its stiffness and mass characteristics. This opens up the possibility of flutter, tucking, and other aerodynamic problems.

There's another interesting effect that crops up on very lightly loaded models that are capable of very tight turns (i.e.: where the turn radius is relatively small in comparison to the model's length and wingspan). In a turn, the airflow past the model is curved, and in a very tight turn the curvature of the airflow is enough to significantly influence the handling of the model. The direction of the "relative wind" at the tail, in both the pitch and yaw sense, could be ten or fifteen or more degrees different from the relative wind at the wing. The curved airflow is typically trying to push the tail up (and the nose down), and to yaw the airplane towards the outside of the turn. If the model has enough dihedral effect, this yaw can enhance the spiral stability, although the pitch effects will tend to make the model dive a bit towards the outside of the turn and pick up excessive airspeed. You may need to hold some "up" elevator and some into-the-turn rudder deflection to counteract these effects. The amount of elevator deflection required can be surprising; it's not uncommon in a really tight thermal turn with a good, light RCHLG to need twice as much up elevator to hold the wing's angle of attack constant during the turn as it takes to stall the model in level flight!

All of these factors influence the required length of the tail moment. Obviously the designer needs to consider the model's mission profile (including the excpected abilities and flying style of the pilots who will be flying it) and the resulting performance and handling requirements.

Ralph, in answer to your original question about what to do with your model, it depends. All of the effects listed above can become issues when you change a design. If the model was carefully tuned by its original designer to achieve just the right balance of all the factors listed above, then even the smallest alteration is probably going to make something about the model's performance and behavior get worse. When you monkey with a design, you always have the potential of opening Pandora's box. OTOH, if the model wasn't tuned all that precisely to begin with, then there may be some room for yo to do some additional fine-tuning. Check with other builders of that model on how the stock version behaves, and try to find out if the designer is obsessive about tweaking his/her designs (I probably know a few folks who are like that), or if they just get the model to fly reasonably OK and start shipping it. The amount of tuning the designer did is probably a good indicator of how much room there is for you to improve it.

In any case, as soon as you start to change something, you become a designer and test pilot. When that happens, you can be virtually CERTAIN that you will have to go through at least several iterations of changes before you can even begin to hope for improvements. Before you start down that path, you need to ask yourself if you want to develop what is likely to be essentially a new design by the time you're through with it, or if you just want something to fly. Development work can be rewarding for its own sake, if that's what you love. Some of us do, and we encourage others to give it a try, IF that's want they want to do.

However, if you just want something to fly, you are almost certainly better off if you buy a well-designed model kit, and build it with few (if any) modifications.

Don Stackhouse
DJ Aerotech