In a previous post I’ve discussed the advantages of a lightweight aircraft: longer flight times, quieter operation, less stress to the components, more responsive control. In addition, having an aircraft below 500 grams (less than around 1.10 pounds) has the advantage that in many countries it does not require registration to fly, except for where specific restrictions apply, of course. I wanted to build such a quadcopter for some time, and I finally came around to finding the time to do it. Hereby I summarize the build, focusing in different ways of saving weight in builds that I’ve experimented with, and the associated savings that they offer. For reference, a picture of the finished frame is below1:
I’ve chosen to build a very simple X frame for this build. The frame is made out of two square unidirectional carbon fiber tubes of 8mm x 8mm outer dimensions and 40cm (15.7 inches) length, crossing over each other and glued with epoxy glue at the center. A few bits of CF tube are added around the crossing point for rigidity.
The overlapping tubes create an asymmetry that should not be significant wrt flight characteristics. In addition, the asymmetry in motor height allows for overlapping props, thus enabling larger prop sizes than the frame would allow.
Power Distribution and Motor Connections
Power and motor leads often account for a significant factor of the AUW, especially if heavy gauge wiring and high current connectors are used. In this build I tried to shorten the lead length as much as possible, and get rid of all the connections, instead choosing to solder leads directly to the PCB. Removing the bullet connectors from the ESC and motor side alone gave a weight gain of 30 grams (1.05 oz). Removal of long leads from the PDB to the ESC gave an additional gain of 40 grams (1.4 oz). In addition, I’m using a lightweight PDB (Micro HKPilot Mega PDB) which doubles as a switching BEC, and soldered the XT60 battery connector directly on the PDB.
In most carbon fiber builds motor mounts are made out of blocks of CNCed aluminum. While this solution offers stiff connections, it also adds up a lot of weight. In a previous post I’ve discussed the design of 3d printed PLA or ABS mounts, but for this build I wanted to go lighter. So I’ve drilled holes directly on the CF tube 19 mm (0.74 inch) apart, and used them to fix the motor. In addition, I’ve used a mix of steel and plastic screws to secure the motor. This combination shaves off a few grams from the build, without negatively affecting the frame vibrations.
For this build I used my CC3D nano flight controller. While obsolete and with few only capabilities compared to modern models, this FC is really lightweight. I chose to mount it directly under the CF tubes with a bit of cyanoacrylate, to avoid misalignment in flight and maintain minimum added weight. I’ve also flashed it with the Dronin firmware which has a nice autotune feature that optimizes PID values for the specific frame. I’ll be trying this in the near future.
With this reduction in weight a small battery should be enough to keep the aircraft on the air for quite some time. I’m using a 1550 mAh 2S battery, with a weight around 150 grams (5.3 oz). Further reduction could be achieved using a pair of 18850s in series.
I’m using plastic 10-inch props, which are overlapping a bit at the edges. These are really lightweight props, at the expense of being a bit flimsy. Still, I prefer them to a pair of heavy CF or APC, as they have a much softer sound profile. I cannot measure the actual RPM, but from the sound of the quadcopter they are spinning at around 2000rpm when floating.
The frame weighs 483 grams (around 1.06 pounds) when measured at my kitchen scale. I don’t expect this figure to be very precise, but it should be in the ballpark. I still have to remove the leads from two of the ESCs, so it should drop a bit more. A preliminary estimation by eCalc gives the hover time at around 15 min, which is quite satisfactory given the small battery. It’s also interesting to note that around 80% of the weight of this quadcopter is drive weight. So with the current hardware (motors, ESCs, battery), I can’t find any more ways to reduce weight on this one. Swapping the 2S cell for a lower capacity 3S should give 1-2 extra minutes of flight, according to eCalc.
This post discussed a sub-500 gram quadcopter build, realized with a carbon fiber body and weight savings in connections, mounts and overal arrangement of components. The build is not difficult for anyone to try, however all the soldering required does mean that you end up with a cable harness that is one piece. In addition, this project has not been tested at all in crashing, but my prediction is that it is too fragile to survive even a mild crash the way e.g. a unibody racing carbon frame would.