webddd.jpg

History of Control Linkages

Home | News | Aircraft | Store | Gallery | Contact Us | Info

 
 
Or..."Why does my plane fly so badly?"

 
 

I started flying in the days of the die cut plywood control horns, before Dubro gave us the quick link, and when piano wire with a Z-bend on each end was state of the art. One afternoon I crashed the same plane 3 times trying to do an outside loop. Days later - while studying the balsa chips that were once a plane - I realized the problem was not me, but rather the flimsy piano wire control linkage was buckling when it attempted to push the control horn in the down direction. It was funny how the rod was always strong enough to get the plane pointed at the ground, but could never get the nose back up. Thus began my love/hate relationship with control linkages and a decades long quest for a setup that would place the control surface where I wanted it.

Every time a new linkage system came in vogue, I would give it a shot. They all promised to be the greatest system ever. Ultimately, I was dissatisfied with them all. I knew that a lot of my flying problems were directly related to flexing of the control linkages, but I had no alternative but to live with it.

Nyrods were the first big leap forward. The original had a steel rod inside a nylon tube, which rattled louder than the engine as you flew. There was enough play between the rod and the tube that you could watch the rod flex as it went from pushing to pulling. Since the servo arm operated in an arc, there was always friction as it forced the rod against the tube where it entered. In some installations the rod still buckled between the tube exit and the control horn. It worked, but just barely.

The all nylon Nyrod solved some of the problems, but it introduced a lot of its own. It reduced the slop, but was so flimsy that it flexed even more. A slight bind would cause the linkage to buckle and take an irreversible bow shape, which then made buckling even easier. It could not take the heat from the engine or exhausts……in fact, the outside temperature made a noticeable difference in your trim as the nylon expanded and contracted. If you went a couple months without flying, the fuel residue would lock the nyrod tightly in its tube.

About this time the pattern guys started running pull-pull setups. These setups are EXTREMELY difficult to design. I have seen fewer than one in ten pull-pull setups that have the correct geometry. Poor geometry results in the cable either tightening or loosening as the control is deflected. Tightening is very bad, as it causes binding, which in turn places heavy loads on the cables, servo, and control surface. Loosening is the most commonly seen geometry. The cables become slack when the control is deflected. This is only noticeable in certain maneuvers, and most people don’t even know it is happening. Other disadvantages to pull-pull are that it is not appropriate for short runs, like the ailerons. It involves some real work to get control throws up to 45 degrees, and nearly impossible to get 50 degrees or more for today’s 3D setups. One of the biggest detriments is the load it places on the servo. Both cables pull on the servo at the same time, even at neutral, so bearing wear is accelerated. Some giant scale aircraft use a separate bell crank to carry the cable tension, then connect the bell crank to multiple servos with balls links. Does it sound complicated? It is, and expensive too. Not to mention that the cable tension accelerates the wear on the control hinges. Cable stretch can also be a factor in a pull-pull system. Even the wonder fiber, Kevlar, does stretch under load.

Ball links are the latest/greatest linkage. They are tricky to set up with extreme control throws. To get 45 degrees or more of throw, you have to use a 1- inch servo arm. The ball link then bolts to the top of the arm and sticks up at least inch above it. This causes a twisting force on the servo arm. You’ve seen the advertisements for reinforced aluminum arms to control the resulting flex. But, reinforced arms do nothing for your poor servo bearings that are sustaining the huge twisting loads on the shaft, in addition to the regular push/pull loads. A new ball link has no slop, but a used one has a habit of popping off the ball when jarred. The links must be kept short, so they won’t buckle (like my first plane did in the outside loop). Short links do some funny things to the geometry of the setup. Watch a high throw ball link setup as it goes from stop to stop. The rod angle changes significantly in multiple directions as the servo arm rotates one direction and the control horn rotates on a completely different plane. This is a far cry from providing the linear control deflection we want. We want to learn to fly the plane, not learn to compensate for linkage deficiencies.

All linkages have the same fatal disadvantage. That is , they convert the rotation of the servo shaft into a linear push/pull motion, and then reconvert it back to a rotating motion at the control surface. When you really think about it, isn’t all this converting back a forth a big waste? Each linkage end point has to have some play, and therefore slop, which in turn results in a loss of precision at the control surface. The pushing and pulling forces cause deflections of the servo mounting grommets and the mounting structure, not to mention the light balsa control surface. Every deflection, however small, adds to the sloppiness of the installation. The result is an airplane that feels “spongy” in the air, or it “dances” to every upset.

The forces cause friction at each link point. Friction robs the power you thought you were getting from the high torque servos you bought; torque that will never reach the control surface. It places extra loads on the servo bearings and control hinges, accelerating the wear of your high dollar parts.

All linkages become less efficient as the control moves away from center. At extreme control throws, the rod (or cable) is working with very little leverage. When you think about it, a linkage is totally opposite of what we really want. It is most efficient when the control is near neutral, and the loads are the least. The more we deflect the control surface, the more force we need, but the less efficient our system works. This is backwards!

I hope by now you feel some of the frustration I have always had for control linkages. Like most modelers, I want my plane to be perfect. I was schooled as an engineer and fly jets for a living, so I loath built in imperfections.

 

Previous Page

Back to Home Page