I had noticed a bit of vibration from the V2 arms which were cut from 2.0mm carbon. For a 2D shape, 2.0mm is a bit thin. The strength in bending is OK but when twisted the arms deformed quite a significant amount.
Lots of people don’t realise this when building quads and counter the problem by simply making the carbon thicker. This will solve the issue but add lots of unnecessary weight. The best solution is to move the material away from the central plane. The I shape quads with the upper and lower deck are some good examples of how lots of extra rigidity can be added while keeping the carbon thin (1.0-1.5mm). The best solution however, is circular cross section.
When disassembling the quad I also found a fair bit of damage to the carbon itself. When the carbon deforms it doesn’t yield like a metal does but rather fails. The arm below has failed where there is a fairly sharp radius right where it used to mount to the main frame. Some simple FEA analysis would have probably shown this to be a weak area and it’s likely that this was due to a crash where one of the arms pranged into the ground on a downward-side on angle.
There is some delamination here rather than a shear break, probably due to the twisting of the arm as well as the fact that the top and bottom layers are biaxial with some extra resin to make it glossy. Not a big deal, just something I found interesting.
A circular cross section puts the material as far away from the central plane as possible so the torsional rigidity is fantastic. I was concerned that the new rigidity might be a bit too much to absorb crashes so I have opted to join the arms to the body using zip ties. I’m hoping that they will be the weak point and break in the event of a huge crash, they also are cheap and easy for me to replace.
When attaching arms like this previously, I had problems in the flat slipping against the tube so I recessed a small gap in the tube to seat the arms flat. This will cost a bit of the strength of the carbon tube but it’s an acceptable trade off for the practicality of the solution. I opted to not run the cables through the arms themselves for this reason. There wouldn’t be that much aerodynamic advantage and it would have required another large cut in the center of the tube to allow the wires to exit. If this was a from scratch build and not an upgrade, I might have planned this a bit better and fed the wires in from the end.
Some other updates since the V2.0 quad. All of the extras- landing gear, camera, etc have been removed and it’s just the bare essentials now. It’s just much nicer to fly with as little weight as possible. I’ll probably add all of those things back on if I make another version using larger motors. I have also gone back to using three bladed props exclusively. They’re just so much better suited for the motors as is shown by the xcalc chart below:
2 bladed props would have been slightly over the limit rather than slightly under. 2 bladed props also vibrated much more due to their higher speed.
- 10mm carbon tube
- Tapered the tube on the bench sander
- Drilled the motor mount holes by hand
- Hand filed the gaps in the tube to locate the main frame
NA – spare parts on hand