[Robert] once built a quadcopter frame by sawing laminate floor tile. It worked, we’re taking the lack of pictures of this build as evidence of how ugly it was. His latest design used a much better looking material – laser cut plywood – and the finished product is very strong and lightweight, even compared to commercial frames made with glass or carbon fiber and epoxy.
Although the design went smoothly thanks to some Solidworks skills, actually cutting the frame from 3mm birch ply resulted in a few issues. The cheap laser cutter used for cutting include some bottom of the line software called LaserWorksV5. There is a kerf compensation feature, called ‘sew compensation’ in the software’s native Chinglish. The software would always crash whenever it tried to calculate the compensation for circles. [Robert] spent two hours figuring this problem out, and in the end needed to break out a piece of sand paper to get a nice interlocking fit.
The completed frame bolts together without any glue at all, and the best part about it is the weight – only 167 grams. Compare that to a similarly sized glass fiber frame, and [Robert]’s shaved at least 200 grams off his finished build.
What about waterjet cut epoxy infused carbon fiber or fiberglass? Would that give more strength with less weight?
http://www.youtube.com/watch?v=ajV8Xh8XtMw
About 1:00 in.
Sure but not everybody has the possibility to use such an machine. With a cnc-mill you can build also such structures out of crfp or grfp sheets, with special tools (for reduced wear).
I can see that it’s a lot lighter, but … I’m confused about the editorialization in the submission “and the finished product is very strong”
The original article doesn’t mention anything of the sort, it only says “feels pretty stable” which is a far cry away.
I could probably make a frame that was even half the weight of that, but I wouldn’t expect to be able to fly it more than once.
I have a frame, I’m finishing it up tomorrow. Estimated weight will be 150 grams. It is made out of prepreg unidirectional carbon layed up on an aluminum male with a silicone female mold cured at 350f.
I’m not sure I’ve seen a multirotor using a truss system before.
Plenty of products using the same technique. For example http://www.hobbyking.com/hobbyking/store/__16583__HobbyKing_Quadcopter_Frame_V1.html
So it has. Thanks for the link.
Yes, there are similar constructions out there. I have more ideas that I want to try out, this is just a first test. I also had to faimliarize myself with SolidWorks first, before attempting some more complex constructions.
One idea would be a triangular profile for the arms, much like a crane, just upside down.
http://upload.wikimedia.org/wikipedia/commons/2/26/Tower_crane.jpg
This, however, throws problems when it comes to interlocking the parts.
I would recommend another truss between the feet otherwise this part is prone to break.
I built some of such sturctures and often the connections are the weak link, especially in plywood the screw and nut connection used here is precarious, I would recommend wood glue as it increases the total stiffness and durability.
Nevertheless, love to see more truss builds, lightweight builds in its classical way!
I’m tired so excuse the question if it is either been addressed or a non-issue. I’m very curious after looking at this yes the truss system looks nice however what about shear strength of the wood. I understand that it will have flex in it but will this design be more of a disposable frame than anything else because of the stresses put upon it.
Not necessarily, some woods are as good if not better than some metals. Lignum Vitae, the hardest wood in the world, the aft main shaft strut bearings for USS Nautilus (SSN-571), the world’s first nuclear-powered submarine, were composed of this wood.
I’d take carbon fiber any day over wood, unless I need a British police truncheon.
I wonder what the bug with the kerf calculator was, maybe a stack overflow from recursion?
I can’t imagine what they got wrong there. It even fails if you have a document with just one circle and nothing else. Even if the circle was created within the software and not importet as a dxf file.
I have fought with that bug many times in Laserworks. One work-around I have found is to replace all the circles with two (or more) arcs and the kerf/sew compensation usually works.
I intentionally left out the connection piece on the bottom of the landing gear (you can see the slot for it) because I was still unsure of how to go about it and finally wanted to see some results. There will definitely be something to make the landing gear stzrdier.