Perhaps one of the more misunderstood and frequently debated issues in our hobby centers on (pun intended) establishing center of gravity or CG of a model. In particular, the influences of a CG shift forward and aft on flight performance. When one changes the CG, you are causing two changes in the plane’s behavior, one minor and one major.
The minor one is elevator trim. Obviously, if you move the CG forward, it makes the nose want to go down, making the plane want to start to descend, and the airspeed to increase.
Move the CG back, and the nose wants to go up, the plane wants to climb, and the airspeed decreases. Taken far enough, with enough airspeed decrease, and the plane will stall.
The obvious correction for this is a corresponding change in the elevator trim or the incidence angle of the horizontal tail. Note, as the plane gets more nose heavy, it takes increasing amounts of up elevator or nose-up stab incidence to hold the nose up and maintain the desired airspeed. Eventually you will reach the maximum the tail can provide, and the tail will stall. The forward CG limit on full-scale airplanes is often determined by the need for enough elevator authority to pull the nose up in the landing flair at touchdown.
The early Cessna Cardinals had a problem with this. At the forward CG limit, the tail could stall in the landing flair, which resulted in some bent nose wheels. The initial corrective action was an Airworthiness Directive from the FAA that restricted the forward limit of the CG. Later, Cessna developed a mod to the tail (lift-increasing slots) that increased the tail’s pitch authority and allowed the plane to safely use the original forward CG limit.
Note, for reasons we will discuss next, you should NOT use CG changes to adjust the plane’s “hands-off” flying speed. That job belongs to the stab incidence and the elevator deflection.
The major effect of a CG change is to adjust the plane’s amount of static stability, the plane’s tendency to return to the trimmed angle of attack and airspeed when the pitch (or yaw) attitude is disturbed. Move the CG forward and the static stability is increased. If disturbed, the plane wants to more aggressively return to the original attitude. Move the CG aft, and the plane’s desire to return to the original attitude and airspeed decreases.
If you move the CG back far enough, to what is called the “Neutral Point” (“NP”), the plane loses all desire to return to the original attitude. The plane goes where you point it. If a gust or a control input pulls the nose up, the nose stays up, with no tendency to go back down by itself. Shove the nose down and let go of the controls, and the nose stays down.
If you move the CG aft of the Neutral Point, the plane becomes statically unstable. Shove the nose down and let go, and the nose wants to go down even more, perhaps even into an outside loop. Pull the nose up a little, and it wants to keep going up all by itself, possibly into an inside loop. The static stability is “negative”, or “divergent”.
The amount of static stability is a function of the CG location. There is another kind of stability, “dynamic stability”, the ability to damp out oscillations (“porpoising”). It’s mainly a function of tail design, in particular the tail moment arm. There is a connection between the two. If you have a more forward CG, the plane has more static stability, and will more aggressively try to return to the trimmed airspeed if disturbed. If there is not enough dynamic stability to keep this return under control, the plane will “overshoot”, ending up on the other side of the trimmed airspeed. At that point the static stability will make it try to go back the other way to get back to the trimmed airspeed and attitude, where it might overshoot again. Thus begins a cycle of overshoots and corrections, an oscillation or “porpoising”.
If the oscillations get smaller and smaller, eventually damping out, the plane has positive dynamic stability. If the oscillations continue, getting neither larger nor smaller, the dynamic stability is “neutral”. If the oscillations get bigger and bigger, the plane is dynamically unstable.
If the plane oscillates but the oscillations damp out, the dynamic stability is considered “underdamped”. Most airplanes are underdamped a little in pitch, typically needing about two and a half cycles for oscillations to damp out.
If the plane comes back to the trimmed attitude and airspeed slowly, without any overshooting, it is “over-damped”. A few airplanes are very slightly over-damped, such as our Chrysalis, which makes the handling very gentle. However, too much over-damping can make a plane handle sluggishly. I’ve seen models with truly excessive tail moment arms that exhibited this problem.
If the plane has exactly enough dynamic stability to prevent any overshooting and oscillation, but no more than that, it is said to be “critically damped”. For dynamic stability, critical damping is analogous to the “Neutral Point” in static stability.
So how does that involve CG location? If the plane has very weak dynamic stability (tail moment arm too short, and/or tail area too small), a too-far-forward CG may give it more static stability than it can handle. It tries to return to the trimmed airspeed and attitude too aggressively, resulting in excessive overshooting and porpoising. In a case like this, moving the CG aft, weakening the static stability, may slow things down enough to get back to what the plane’s dynamic stability can handle.\
Aerodynamic Center (AC), Mean Aerodynamic Chord (MAC), Center of Gravity (CG), Neutral Point (NP) and Wing Area
The infamous “dive test” is checking for these behaviors. You get the plane in trimmed. level flight at a safe altitude, push the nose over into a steep dive, let go of the controls, and see if the plane pulls out of the dive by itself. This version of the test has been largely discredited and is generally discouraged. It’s very abusive to the airplane, and also does not really measure stability very well. When you push the plane to such extreme airspeeds, you can get into non-linear aerodynamic behavior, aeroelasticity issues (where the aerodynamic loads distort the shape of the plane’s structure enough to alter its aerodynamic behavior), and flutter. Even if you don’t break something, the plane’s behavior may not accurately represent the plane’s behavior at more normal attitudes.
Instead, try something along a similar concept, just not as extreme. Get in trimmed level flight, then pull the nose up about 3-5 degrees and let go. Watch whether the plane tends to come back to the original attitude, and whether it overshoots and oscillates, whether the oscillations get larger or smaller, how many oscillations it takes for them to damp out. Get the plane back in trimmed, level flight, and try the test again, but this time push the nose down 3-5 degrees and let go.
As I mentioned above, most planes will go through about 2 1/2 oscillations before damping out. If the oscillations are excessive, or there are too many, try moving the CG back a little. If the plane takes an excessive amount of time and altitude to return to the trimmed airspeed and attitude, try moving the CG forward a little. Note that in both cases you will need to readjust your elevator trim and/or stab incidence to get the plane back to the original trimmed airspeed after you change the CG.
Visit this liked site Aircraft Center of Gravity Calculator for more information on determining CG of a model.
Thanks to Don Stackhouse of DJ Aerotech for this article edited by Steve Pasierb
