Project Logs

Collapse

• Prototype #1!

  Chris Gervang • 09/09/2019 at 05:37 • 0 comments

  The prototype I built is based on the CAD Concept #2. I tweaked the design a bit to make it work, but that's fine since the whole idea is to get something to work! 

  The most difficult part of this was finding and attaching strings to the airplane's center of gravity. I didn't end up with perfect CoG, so the plane pitched up all the time without an extra string to pull it down (see the 2nd red connector on the tail pole? That's to correct the CoG!)

  This mistake gave me idea though.. I might be able to purposefully hang off the CoG in 2 directions and remove two strings for free. That would simplify the motor controls a bit, but might add complexity to getting everything calibrated.

  I noticed the plane wobbles a lot when it's bumped unless all the strings are tight. 

  I also noticed the strings fight each other when they are pulled tightly because they are all doing two roles. All can influence altitude plus one other control. I think one improvement I'll try it to make is to hang the plane from one string on top for altitude, and then try to place all the other stings in an optimal spot for whatever they are controlling. 

  I have some other videos too that I can upload. Hope you enjoyed a first look at how this might come together!

• Preparing A Prototype (cont)

  Chris Gervang • 09/09/2019 at 05:20 • 0 comments

  Now we'll take all of these posts and arrange them around the aircraft in different ways. I have a few ideas on the direction the strings will attach, as well as the number of stings. Ideally, we can use fewer posts so the view of the airplane isn't covered up. I'm not sure how to accurately simulate how the strings will tilt the plane.. so I need more options in case 3 posts isn't enough.

  Concept 1: The northern post is attached to the tail and the two posts in front are attached to the wings. This should be able to control altitude, point the airplane up and down (pitch), and tilt left to right (roll).. but not bearing.

  Concept 2: NSEW. Two will be attached to the wings to roll. Two to the tail and head to pitch. Altitude will work by relaxing all the strings, but once again this design cannot control bearing.

  Concept 3: 8 strings. I wanted to see if there was anyway to control bearing with 4 posts, so I came up with an idea to double the number of strings to be able to pull on the airplane from more directions and hopefully get all kinds of motion the other designs could achieve plus bearing control.

• Preparing A Prototype

  Chris Gervang • 09/09/2019 at 04:59 • 0 comments

  I decided to move forward with the puppet-like design in my last sketch. I'll suspend a scale model aircraft with strings, and I'll change the string length to change how the airplane tilts.
  

  While I waited for the parts to arrive I went worked on a CAD prototype to capture some of the ideas I would try once I have all the pieces in hand. The prototype will use KNEX, an aircraft model, and string. I'll try adjusting the string by hand to understand the features of this design, and any motion limitations.

  I came up with a few options for posts so I could adjust the height, and change the string length easily.

  The most basic approach is to tie the string to the connector attached to the adjustable red piece, and thread the string up through the top connector, and tie the other end to a part of the aircraft.

  If I need more height, this option doubles the height.

  I could also add a wheel to reduce the friction.

• New Concept Sketch

  Chris Gervang • 09/09/2019 at 04:38 • 0 comments

  After some brainstorming with a friend, we thought a puppet-like structure would be a great way to improve the design for the following reasons.

  • No need for projectors or mirrors. I can put a flat LCD screen under the plane.
  • The overall assemble is way shorter, so it'll fit on my desk.
  • The control surfaces are all the same, so the programming and manufacturing is simplified.

  The drawback of this design is I may need a lot of strings to get all the control I'm looking for. Also, there might be a lot visual obstruction from the poles, which kinda ruins the magic.

  I believe this concept is easier to prototype, so I will move forward with it!

• Initial Sketches

  Chris Gervang • 09/09/2019 at 04:32 • 0 comments

  At first I explored the idea of tilting and moving the aircraft from the bottom and projecting the ground up into a screen under the airplane. The idea is to attach the airplane to a linear actuator for altitude, a single servo at the top for pitch control, and place it all on a rotating base for bearing control. 

  To see the video of the ground under the airplane, I needed to cut a hole in the middle of it. A normal LCD screen can't have a hole in it, so I'd project my computer on a projector screen. 

  I thought this assembly was going to be very tall, so we have the option of reflecting horizontal projector off a 45º mirror (that also has a hole cut out of it) to get the vertical image.
  

  The benefit of this design is how invisible the mechanisms could look. After hiding everything below the screen, only the pole and a small servo are visible. That leaves a big screen and airplane to look at. 

  The main drawbacks to this design is complexity and a bad screen quality. The mechanical assemblies are going to be hard to make. I choose to not add a second servo for roll to make the wings tilt for turns. This whole design could easily be two feet tall for one foot of altitude. Also, the pole holding the plane would cast a large shadow on the projection screen.

  Overall, this design is a starting point.. but has many drawbacks.

  I also considered a more complex assembly of mirrors to reduce the shadow casted by the pole... I don't see this idea being pursued.

  Back to the drawing board!

View all 5 project logs