Chapter 19 - Section 11

Controls

 

This section is for building the linkage between the main wings and the control sticks. Its a short paragraph in the plans, but much work is ahead. I worked on this section a bit out of sequence because I can make precise measurements to the linkage hardware, hopefully reducing any slop in my control system. I worked on this section AFTER I completed Chapter 19, Section 12 and Chapter 14, Section 10.   

 

Marking the Center of the Torque Hot Wire Hole

The first thing I did was to mark the center of the torque entrance (hot wire hole) to the main wing / aileron. I drew a 1.6" diameter circle with a cross hair through its center with my graphics program. I printed it out on semi-thick paper and cut out the circle with a small handle. Then I placed the circular template over the 1.6" diameter hot wire hole as shown.

 

I extended the cross hair onto the wing rib surface. These extension marks will be used to line up my FMN-10 bearing (later) to the center of the torque tube hot wire hole.  

 

 

 

 

Fabricating the FMN-10 Holder

The plan calls for fabricating a phenolic block to support the torque tube at the entrance to the main wing and aileron. Many builders recommended using the FMN-10 bearing as replacement - I decided to follow. However, I need some kind of mounting widget to hold the FMN-10 in place...

 

I made two (2) holders as shown - one for each wing. I countersunk one side of the holder such that the screws will be flush to the surface. The bolts I used are MS24694-95 (they are 1/4-28 instead of the usual 10-32 screws) because the diameter of the mounting holes for the FMN-10 are 1/4" in diameter.

 

I hot glued the holder in place throughout the whole trial fit / alignment process. [Note] I held off floxing and glassing the holder in place until I completed the entire aileron control system. Then... 

 

I ground a flat on each screw head, countersunk the holes and installed with flox. The next day I tested them by tightening the nuts to it. One held up OK and the other did not. So I increased the size of the flat and re-floxed in place. This time it was OK. I also filled the Phillips head with flox when I eventually floxed the holders to the wings for added resistance to rotation.

 

Once the holders were completed, I trial fit the holder in place and traced the outline of the holder on the wing root. Then I removed everything and applied wet flox within the holder outline (on the wing root). Then I pressed the holder against the flox and added a couple drops of 5 minute epoxy to keep the holder in place. Once the flox cured, I glassed the entire holder with 3 layers of BID.

 

I forgot to take some pictures on the holders when glassing them in place, but you can make it out (somewhat) from the picture below...

 

Aileron Torque Tubes

I added the universal joint and the aileron torque tubes using similar methods as I did in Chapter 16 Step 2. I started from the aileron end and worked towards the CS132 (the welded part by Brock). The short tube (CS152) that fit through CS132 came matched drilled - which made it tougher because the orientation of the last tube is about an inch inside the torque tube hot wire hole. However, with the help of a bit of masking tape and careful marking, the problem was solved. I was able to set the aileron at neutral position and CS132 at 90 degrees.

 

BTW, to set CS132 to 90 degrees, I drew a straight line at the face of CS132 joining the centers of the two holes. Then I used a plumb line as a guide to set its orientation. It worked fine.

 

First Trial

With the FMN-10 holder hot glued in place, I trial fit the aileron torque tubes. I got about 12o up and 26o down movements. A quick investigation revealed that the inboard bolt (of course it had to be the inner bolt) of the universal joint was rubbing/bumping against the glass inside the hot wire hole. I had to remove the entire aileron & torque tubes and ground off some of the glass inside with a Dremel. After some frustrating moments, I decided to put a dab of grease (that's the only putty stuff I have at that time) on the head of the bolt to mark the exact rubbing location. Several more times of removing, sanding and !@#$%, I finally freed the aileron upward movement to about 23o - good enough!

 

I had to repeat the same processes on the left wing. This time the obstruction was closer to the hole entrance, so there was less removing, sanding and !@#$%...

 

Mounting the Belcrank Bearing

I bought the CS127 brackets (that hold the Belcrank bearing) from the Cozy Girrrls. They came with all the holes pre-drilled - including the hole for neutral and 20o positions. That was well worth the money because once I slipped an AN3 bolt through the neutral hole, the 20o position hole and mount the Belcrank with an AN4 bolt, I pretty much got the whole assembly together for fitting.

 

With trial and error, I found the optimal assembly position such that the CS129 torque tube will be leveled and 90 degrees to the Belcrank arm. My CS129 turned out to be 8 1/4" long. 

 

Mounting the Belcrank Brackets

Once the torque tubes were made and trial fitted, I hot glued the Belcrank brackets in place. I disassembled the Belcrank assemblies and drilled the inboard holes first. Then I bolted the bracket in place and removed the hot glue. I re-leveled the brackets positions with a miniature bubble level to make sure the top and bottom brackets are parallel. Note the single bolt through each bracket as well as the 3 vertical bolts to keep both brackets squared and aligned. Then I marked the second holes and completed the drilling.  

 

 

 

 

 

Here's a picture of my completed Belcrank assemble and aileron torque tubes. Notice I held down my rivets with masking tape because I want to eventually paint the torque tubes before squeezing in the rivets permanently.

 

 

 

 

 

 

 

 

 

Setting the Aileron Control Positions

I was having some difficulty in setting the aileron positions. Somehow, I did not get the same degree of movement between the ups and the downs.... I adjusted a bit of this and a bit of that, but was not able to get the 20o up and 20o down at the same time. I do not have any slop in the control system and I have very minimal resistance or obstruction in movements. I did not have a clear understanding of the relationship between the aileron angles and Belcrank movements, I decided to try something a bit more precise...maybe I can figure out the problem.

 

I mounted a digital level at the inboard edges of both the ailerons as shown. I set the initial torque tube positions such that both ailerons were at neutral positions & control sticks at 10o. I recorded the digital level readings. Then I raised the left aileron up to the hard stop and recorded the digital level readings again. I repeated the process by reversing the aileron positions. The readings below were my initial readings with the newly mounted digital levels:

 

 

 

 

 

 

 

Aileron Positions Left Deflection Right Deflection
Neutral -12.4 NA -12.4 NA
Left up / Right down +7.6 20.0 -33.6 21.2
Left down / Right up -36.5 24.1 +9.4 21.8

 

As shown above, when I raised the left aileron up by 20o, the right aileron drops by 21.2o. On the other hand, when I raised the right aileron by 21.8o, the left aileron dropped 24.1o. My concern was that the deflections were not uniform from side to side.

 

I decided to keep the left wing setting constant and just make adjustment to the right wing torque tubes. I shortened CS126 (the link between the Belcrank to the firewall) by 1 complete turn (of the MM-4 rod end) and lengthened CS129 (link between Belcrank and Weldment) by 1 turn and recorded the readings as follows:

 

Aileron Positions Left Deflection Right Deflection
Neutral -12.3 NA -12.3 NA
Left up / Right down +7.7 20.0 -33.6 21.3
Left down / Right up -38.3 26.0 +10.8 23.1

 

As shown above , the deflection uniformity got worse. The good news was that - I probably made the adjustment in the wrong direction. So reversing the adjustments by lengthening CS126 by 2 turns and shortening CS129 by 2 turns got the following readings:

 

Aileron Positions Left Deflection Right Deflection
Neutral -12.4 NA -12.3 NA
Left up / Right down +7.7 20.1 -33.6 21.3
Left down / Right up -35.8 23.4 +9.0 21.3

 

Data above indicated that I was adjusting in the appropriate direction. Therefore, I lengthened CA126 by another 1 1/2 turns and shortened CS129 by another 1 1/2 turns. the results were as follows:

 

Aileron Positions Left Deflection Right Deflection
Neutral -12.3 NA -12.3 NA
Left up / Right down +7.7 20.0 -33.6 21.3
Left down / Right up -33.8 21.5 +7.7 20.0

 

Since the minimum adjustment I can make is 1/2 turn on the MM-4 rod ends which amounts to ~2o of deflection correction, I decided to stop at this point. Though I still do not fully understand the geometry of the Belcrank relative to the aileron deflection, the exercise above showed that - Changing the length differential between CS126 & CS129 affects the up and down deflection angle of the aileron....