New Technology Weakley

Written by Lucas Weakley New Technology As seen in the Summer 2019 issue of Park Pilot

After building a handful of model airplanes, I began to dread making all of the servo linkages. This step usually took the longest and was the most tedious. For the longest time, all of the control surfaces weren’t deflecting symmetrically or linearly. This inconsistency led me to spend more time programming my radio to correct these mechanical imbalances. I later learned the importance of conscious control surface geometry, and now I tailor each linkage to every control surface’s needs. To start, let’s design the ideal control linkage. This geometry will use the entire 90° servo range, symmetrically deflect the control surface, and maintain linear motion through the entire deflection arc. If nothing else, setting up all control surfaces on a model with these guidelines will yield the most consistent performance and latitude when programming things in a radio later. In an ideal control linkage, the servo arm is perpendicular to the mounted surfaces, the control horn connection holes are directly over the pivot point of the control surface (when there is no control surface deflection), and the linkage connecting the servo arm and control horn is parallel to the control surface. This linkage geometry utilizes the most linear range of the servo’s arc by keeping the servo arm and control horn parallel to each other and perpendicular to the control surface.

This illustration shows the movement of an ideal control linkage. Notice how the control surface moves symmetrically with the full servo range. The sign convention for deflections here remains the same in all other images as well.

The linkage geometry can also be modified to get more or less control surface deflection out of the same servo arm throw. This is achieved by changing which holes the connecting wire is linked to. To get the most range, connect the farthest hole from the center of the servo arm to the hole closest to the pivot point on the control horn. To get the least range, do the opposite and connect the closest hole to the center of the servo arm with the farthest hole from the pivot point on the control horn. As long as the servo arm and control horn remain parallel, and the control horn holes are over the pivot point, these changes won’t noticeably affect the symmetry and linearity of the control surface movement. There are some tradeoffs with varying the control surface deflection this way. More deflection means more coarse movement and more torque required by the servo. Less deflection gives finer movement and less torque required from the servo. Most park flyers will have such low flight loads that servo torque isn’t a concern; however, increasing the servo range, as I described earlier, could make trimming and programming the servo more difficult. Servos step through their rotational arc in increments—not as a smooth movement. If the linkage is set for the maximum movement, it amplifies these increments. A click of trim might make an adjustment of a few degrees at the control surface. When the servo is commanded to move, the control surface will jump between degrees as it follows the steps in the servo’s movement. There have been many times when I set up models thinking I might need plus or minus 60° control surface deflection, but after flying the model, I cut the throw in half using dual rates. I would have had finer trim and model control if I had just set up the linkage to give me 30° of deflection using the full servo arm arc instead of using half of it. As I mentioned, it’s important to keep the control horn holes over the pivot point and the servo arm perpendicular to the surface. However, if these attributes are changed, the control surface deflections will become nonlinear, which can be useful if understood. Moving the control horn behind or forward of the pivot point will make half of the deflection arc nonlinear while the other half keeps its original deflection. For instance, if the control horn is pushed back farther down the control surface away from the servo, the upward deflection will remain linear and the downward deflection will exponentially increase with servo arm sweep.

This graph depicts the motion of the ideal control linkage and its variants, which are illustrated on the left.



Moving the control horn forward of the pivot point will sometimes require a new control horn shape. Notice on the graph how half of the range is linear, and half is exponential.



These illustrations show the alternative starting servo arm angles. The servo will still rotate plus or minus 45° from this starting angle, which has a large effect on the control surface, as seen in the graph.

This is especially useful when implemented on the ailerons. The control horn can be moved in such a manner that the ailerons will bias toward more upward deflection and have less downward deflection. This will create roll differential to help counter adverse yaw. Differential is something that can be programmed on a radio, but doing it mechanically allows for simpler aircraft programs and the latitude to use inline flight stabilizers. Another way to create useful nonlinearity is to change the starting angle of the servo arm. Setting up a linkage so that the control surface is at 0° with the servo arm clocked to 45°, for example, will have a large effect on the deflection arc. The control surface will have an exponentially larger deflection in whichever direction the servo arm is clocked away from. This is why it’s a good idea when making ideal control linkages to remake the connecting wire instead of just rotating the servo arm until everything aligns. Sometimes it is not possible to keep the servo arm perfectly perpendicular because of the teeth on the servo shaft. A few degrees off from the desired servo location won’t have an extreme effect, but it might be a good idea to try to make the misalignment symmetric if it occurs on an aileron, flap, or twin rudder servo. Clocking the servo arm to an off angle acts similarly to exponential programming, but only in one direction. This is sometimes useful for elevators or flaps, where these control surfaces primarily move in the nose-up direction and don’t need much authority in the nose-down direction. Good control linkage geometry is important, but conscious linkage geometry is more important! This knowledge has changed the way that I look at control linkages when building models, and now it’s something I look forward to working on! I hope this article has inspired you to play around with your model linkages and see what can be done.


Facebook Twitter Share


Good info


good to now

Good article, illustrations very helpful.

Credit where credit is due! Excellent article on what is otherwise a rather-hard-to-explain topic.

Airplane Servo Adjustments

Good primer sometimes overlooked.

Your comments are well taken, thanks. One thing that is somewhat related to your subject is the importance of establishing control surface linkage that's FREE of binds. The servo rod (plastic, balsa, metal) must be placed in the aircraft so that there is minimal, if any, drag on the servo! Nothing beats a straight line between the servo arm and the control surface horn! If the servo operating the control surface chatters at any transmitter stick position, there's unacceptable drag that should be eliminated. This is of major importance for two main reasons. 1st, the servo will last longer and will require minimal current (watts) to operate. 2nd, the control surface will center properly each time the transmitter control stick for that control surface is centered.

I have seen some of the Depron Foam Aircraft that are being sold by the "few" hobby companies left in the US and many of them may be unacceptable in this area, my personal observation. The majority of those aircraft are made somewhere outside the US and for the most part aren't worth what people are paying for them. I just worked on a friends Night Radian and the servos were not only binding but were defective out of the box. Each control surface (elevator & rudder) had major binding which I corrected.

This is only one example of the poor build quality used in the construction of many of the so called ARF, PNP or BNF, model aircraft - most if not all are made by foreign manufacturers. They all look great but most of them are NOT made as a quality product! If the elevator, aileron(s), or rudder control surfaces don't move freely and smoothly, problems are just around the corner and most times the end result is a complete loss of the aircraft. Just my personal view!

A great article on control linkage! It really got me thinking about how I connect servos and how I can tweak them for better results!


I really appreciate these basic articles. I'm am busy in my practice and am storing up everything for retirement in 15 months!!

Before computer radios you HAD to set up control linkages properly or your plane would fly erratically.
When I got my first computer radio I tried to let the radio do the set up work. I learned pretty fast that you still need to have proper linkages. The computer just allows you to modify these linkages. Things like dual rates, expo, mixing channels together, or slowing down a retract servo are examples of things a computer radio is good at.
I once built a plane that had a sliding servo tray to mix elevator and ailerons. Computer radios make this easy and you can turn it on and off - not so with a mechanical linkage.
The new computer radios can do things I never dreamed of. But you still have to set up the linkages properly.
Great article, nice work!


Good Info

Add your thoughts to the article

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Enter the characters shown in the image.