A few answers here. On the Dubo arms, I cant agree with the blanket statement that ball links and dubro arms dont mix. It depends entirely on the application. If it is a 3D set up and a lot is being asked of the servo and linkages, then the little bit of arm flex that can happen isnt good. But, for example, my 1994 TOC model (Godfrey Extra 330S) used Dubro nylon arms and ball links on all surfaces. It worked great, but we werent doing 3D yet (except for Quique) and only had maybe 20 degrees max throw.

On the subject of flutter, there are different things that cause it, not one cause. Depending on the cuase of flutter in a particular model, a measure taken to fix it could work, or it could agrivate it if the guess was wrong. I'll give a great example later in this post.
Some common causes are:
- Mass - (control surface is too heavy and an oscillation builts into a volent flutter that the servo/linkage cant overcome. This is a common cause in many cases. It is greatly aggrivated by using 3D throws, because the servo looses mechanical advantage over the surface in order to get that much throw and it becomes a lot harder for the servo/linkages system to overcome the load of a surface if it wants to flutter.
- Twist - The surface is structurally weak and even though the servo holds it at the attachment point, it buzzes or flutters outward because of low torsional strength.
- Similar to twist, linkage strength is also an issue. If the linkage system is flexable or week the surface can flutter.

Flutter is in general a condition where the control surfaces oscillate at a natural frequency, so it can be excited by engine RPM if it hits the natrual freq, or by aerodynamic factors if airflow hits a natural frequency. Thats why in some cases the flutter we get doesnt happen at full speed, but at a lower throttle setting and in level flight not a dive. Many factory contribute. I believe that any control surface can flutter if the natural frequency is reched, but in some cases the natural frequency that will excite the flutter is well outside of the flight envelope of the aiplane so it can never happen.


Of course if a surface is not aerodynamically prone to flutter (within the airplanes flight envelope), then it wont even if things arent ideal. Some surfaces can sit still even if completely disconnected from the servo. Getting to that point aerodynamically is very difficult is is very elusive. Most models, especialy 3D or aerobatic models with larger surfaces arent there.
The best rules of thumb are to make the control surfaces as light weight and as stiff as possible. Mass is the biggest enemy in a control surface flutter with models. Mass balances are also an option to help is mass is the cause of the flutter.

Good example of looking for the wrong cause and how the fix actually hurts the situation is a mistake I made several years ago when I worked for Midwest. We were developing the Super Stinker. It originally used 2 x aileron servos with slave rods connecting the top and bottom ailerons. It flew well, but I kept hearing a buzz that I was pretty sure was from the ailerons, I was right there. (buzz is generally a slight, very high frequency but low amplitude oscillation of a control surface around neutral that does not grow into a high amplitude or destructive flutter). I used 4-40 rods connected at the trailing edge of the ailerons as slave pushrods and I thought it might be buzzing and causing the ailerons to buzz with it. So, I added airfoiled aluminum over the rods to stiffen them and to make then less aerodynamically prone to buzz than a round rod. I also knew that I was adding mass to the trailing edge of the ailerons by doing this and if I was wrong and the buzz was a mass realted issue I just made it worse.
I took off, made a turn and went to full throttle in level flight, since that was a condition where I was hearing the buzz before. Did it buzz...no. Instead there was a violent explosion and a bilpane came out one side and 4 ailerons the other. Fortunately it was a Pitts, so the rudder generated a ton of roll and I landed the model. Lesson learned, and I went to 4 x servos on ailerons and the issue was solved.

On the Showtime set-up, what you are seeing on the bench isnt flutter, its servo jitter. They are two completely unrelated issues. Flutter is aerodynamically induced. Jitter is a electronic or mechanical issue. It can also be a case of the mass of the surface sets up a natural frequency that starts,the servo oscillating. But in the air the load of the surface will be very different. It is probably an issue of servo wear causing it. I see it all the time, and it usually isnt a problem. When it becomes a problem, it doesnt cause flutter, the model normally just doesnt fly as well becuase the servo has lost precision. Typically if the jitter stops if the damp the surface (touch it wth your fingers) then its fine and the ail loads in flight will damp it out. If it wont stop, there is a servo problem and soemthing needs to be changed (servo pot, or the entire servo)