One Step Forward, Two Steps Back

As you can probably tell from the title, prototype number two (shown below in the photo), was not quite as successful as my proof-of-concept build.  I learned several important things from the first prototype; the first, and very exciting, was the fact that I was on to something that no one tried before.  If you haven’t read my Proof of Concept article, you can find it here.

First Prototype Lessons / Prototype 2 Planned Improvements

  1. My first design was very over powered.  Prototype 2 could be built larger and stronger.
  2. My first design was built with a simple KFM airfoil.  This one will utilize a better semi-symmetrical airfoil.
  3. The first motor hinge design worked great, was inexpensive, but very labor intensive.  I had a simple idea for a stronger motor hinge that could be easily reproduced.
  4. Lastly, the first plane was beyond my skill level to maiden and beyond my skill level to fully test its capabilities.  This time I will look for a more experienced pilot and a larger flying field.

As you can see from the photo, prototype 2 was beautiful.  It was larger and I found a better, more experienced pilot for its maiden flight.  Unfortunately, if you are observant enough, you may find something very important that I forgot to include in the design.  If you like puzzles, look closely and leave a comment if you find it. Click on the image to see the full size photo.

Build Photos

Construction went well.  Size was good and it was definitely stronger.  I wanted this build to be very clean, so I decided to mount the servos front and center under the canopy.  For it to work, I needed a good set of full sized high torque servos.  Unfortunately, I chose the Tower Pro MG955.  They worked, but were not very precise and had a lot of servo jitter.  The servos were mounted forward in the nose as planned.  They were connected to the motor hinges via Sullivan Cable Type Gold-N-Rods. I don’t know why I chose these, because I’ve always had trouble using them in the past, and this was no exception.  It looked great, but drag from the flexible push rods amplified the jitter, the servos would not center properly, and if the control sticks were moved too quickly, the servos over shot their endpoint and had to return to their intended position.

The new motor hinges were a definite improvement over the first prototypes.  They were constructed from 1/2 inch square wooden dowel and 1/2 inch aluminum channel that can be found at your local Home Depot or Lowe’s store.  The only caution with this design is that only two of the motor’s mounting screws can be used to mount the motors to the aluminum channel.  This probably won’t ever be a problem because the weak point is the hinge bolt at the channel pivot to the wooden dowel.  To strengthen the pivot point and eliminate any slop or play in the hinge, the hole was drilled one size smaller than the bolt and the area was saturated with thin CA glue.  After sanding and running the drill bit though the hole one more time, the fit was tight and as strong as it could be.

To make this plane more crash worthy, I chose to use a more rigid foamboard.  The first prototype was built with Dollar Tree foamboard.  For this model, I chose to use Hobby Lobby foamboard.  In theory, this was a good idea, but in practice, it was a very poor choice.  The wings were so large, and so much of the plane’s wings were behind the center of gravity, the model was extremely heavy for its size and hard to balance.  It might have worked out better if I had used a combination of foamboard.  If I build this prototype again, I would either use the Hobby Lobby foamboard on the underside of the wing, or only use a small section of the Hobby Lobby foamboard glued inside the wing towards the front.  This would have been a good compromise between strength and weight.

Have you figured out what I forgot?  Until I got to the flying field, I hadn’t either.  I forgot to add a set of elevons.  I fully intended to.  While working out the design, I had intended to attach two sets of pushrods to each servo so each servo would control a motor hinge and an elevon.

Things didn’t go so well at the flying field.  The first couple of tosses, the plane just looped in place and crashed to the ground.  After adjusting the motor reflex, I got it airborne but it had no pitch authority.  It would gain altitude, but as speed increased and it leveled off, it would nose down and no amount up up elevator would bring it up.  The roll axis worked as you would expect, but flights were short and landings hard.  After a few short flights it became obvious that the plane was very tail heavy.   To find the center of gravity, I used an online Flying Wing CG Calculator.  To get it correctly balanced, I needed to add approximately a pound of weight to the nose.  As the plane was already too heavy, this was not an option.  A wooden dowel was attached to the front of the model that extended about 16 inches forward and a few ounces of lead was added to the tip.  We got it airborne one last time and the flight was longer, but the weight on the end of the dowel began to oscillate making control impossible.  The plane nosed in at full throttle and was destroyed.

Although the plane looked great, it was incredibly heavy, had inadequate control due to a missing set of elevons, poor quality servos, badly run flexible pushrods, and an improperly set center of gravity.  Stay tuned to my next article for prototype number 3 where I corrected all of these mistakes.


Leave a reply