Saturday, May 18, 2013

Making the Frame

For the frame of the quadcopter, I wanted to use a 3D printer because many hobbyists either have one or access to one. This would allow anyone to print out the frame and simply attach the electronics and start flying. So I taught myself how to use SolidWorks and designed a frame to be printed. My design was based off of a design by an MIT student who attempted a similar project. It consists of a central holder for the electronics and places the motors in an X configuration, complete with landing gears.

The quadcopter frame in SolidWorks
The design from MIT

I then printed the part at Stanford's PRL (Product Realization Lab) using their ProJet 3D printer which uses wax support material so it would be easy to melt away after the print. After the print finished, the part looked great but I quickly realized some flaws in the design. The biggest flaw was that it just seemed too heavy. The printing material was heavier than I anticipated and as a result I had serious doubts about its ability to fly. Another problem that was made clear was that it wasn't the most sturdy design. I carelessly dropped it from my bed and one of the arms broke. If this quadcopter was to fly, it would often be dropping from height and would most likely break upon impact and since the frame was a single 3D print it wouldn't be easily fixed.

Fresh out of the printer

Melting off the wax

Quadcopter

So it was back to the drawing board. Armed with the lessons I learned from my previous design, I decided the new version had to have several key features. It had to be lighter, this meant less plastic or at least a lighter plastic. It had to be stronger and more easily fixed. Finally it had to be more easily printed. The first version was a difficult print and may be too hard to do with a hobbyist grade machine. With these parameters in mind, I decided that the best way to do this was use carbon fiber rods for the arms and only 3D print the base that held the electronics and motor holders. This decision had several advantages. First, the carbon fiber meant less plastic and made it much lighter while also stronger. By printing out only the motor holders and electronics base, the prints were simpler, cheaper, and easily replaceable. As an added bonus, the entire frame ended up being cheaper because I only had to use about a dollar of carbon fiber and the print only took 0.66 cubic inches of support and printing material so it was a fraction of the cost of the old frame.

Check out the CAD files at: https://sourceforge.net/projects/nanoquad/files/?source=navbar.

The revised design (the square rods are carbon fiber)
The printed parts. One of the problems was the support material was inside the holes which made it difficult to clean out.

Post support material

The assembled frame

The final quadcopter


2 comments:

  1. Where is the electronics part list? I ordered the PCBs from OSH Park but don't know how to build the rest of the system (motor, where to get the printed parts made, etc)..

    ReplyDelete
    Replies
    1. I just added a bill of materials to my Github (https://github.com/ndanyliw/nanoquad). I found the motors on eBay so I don't know if they are still available. I would just search "7x16mm coreless motors" to try finding some that have the best chance of fitting the frame. The BOM is a bit out of date and some parts have changed between board revisions. I'll try posting an updated one in the coming week. But I sourced almost all my parts from Mouser and all the package information and part numbers can be found in the Eagle files on my Github as well. As for getting the frame made, you probably can get it printed at some place like ponoko.com. All you need then is some carbon fiber rod for the arms which can be found on Amazon (http://amzn.to/1oCt5e5). If you don't have easy access to a 3D printer, I would suggest getting the parts first and measuring them to make sure the dimensions are correct.

      Delete