We created the quadcopter frame’s components using the makerBot Thing-o-matic 3D printer which uses ABS plastic for printing. ABS plastic has a good strength-to-weight ratio; it has a tensile strength of 30 MPa and a density of only 1.02 g/cm3.With these measurements, the estimated quadcopter weight is around 150 g. The actual weight of the quadcopter frame turned out to be 176 g which is a bit more than the expected weight but according to the motor specs, each motor should be able to provide 270g of thrust using the 8 inch props that were brought. This totals up to 1.08 kg of available thrust. Comparing this to our final quadcopter weight of 628g, our design has enough power and is relatively light.
The base plate supports the weight of the motors and is very vulnerable to bending due to the extra leverage created by the arms. Because of its spacious location, and its sturdiness, the plate is the piece upon which the battery, Styrofoam guard, and electronics are mounted. Being the central hub of our copter, the plate had major responsibilities in terms of stability and protection – during a fall most of the force experienced by the arms will be transferred to this plate and the vulnerable electronics housed on top. Once more addressing our primary goal, we carried out several iterations of the base plate design to find the best balance between strength and weight. The simulations were conducted with the plate constrained on all sides with an 80 N load (approximately 8kg) applied on the surface of the plate to simulate a 3 meter drop.
The initial design of this base plate was very simplistic. It had the slots to tight fit with the arms, and a hole in the middle to minimize weight. The problem with this design, as the simulation showed, is that the weight from the arms can bend the plate by up to 21 mm, causing the frame to completely collapse on itself.
Learning from the failures of the first design, we added cross beams to the center of the plate in place of a hole. This added more weight to the system, but added much-needed structural strength. With this new design the maximum displacement of the plate under stress is less than 0.5 mm.
Although the 2nd design’s structure provided the strength we needed, two improvements can still be made. The first improvement is to lighten the weight. The blue areas shown in the von-mises diagram are areas which can be removed without effecting too much of the plate’s structural integrity. Thus holes were cut in those areas to reduce weight. The second improvement to the design was to make the tight fit arm slots bigger so they fit more securely with the arms.
The arms of the quadcopter are attached to the base plate and hold the mounts for the motors and fan blades. They were designed to support the load of the entire craft during flight and landing. In the event of a fall, the arms have the highest chance of hitting the ground first. Because it is crucial that the arms do not break down in a flight, we went through several iterations of arm design to maximize their strength. The simulations for the arm was conducted with the arm fixed to the base plate while a 130 N Load (approximately 13 kg, equivalent to a 5 meter drop) was applied upwards from the tip of the arm.
In the initial design, the arms were not intended to be used to support any weight when landing the craft. The feet feature on the arm was added later on to simplify the design. In the early stages of drafting the size and length of an arm was constrained to be 10 cm, which is the maximum printing size of the 3D printer. This constraint was later revised when we discovered that bigger parts could be printer by by orienting the part diagonally from corner to corner.
From the second designs onwards, the arms were oriented diagonally in order to maximize the print size. The arm shape was designed with feet for landing and a slot on the tip for the motor mount. Since the arm supports the weight of the motor 4 inches away from the base, the force acting on the arm will gradually increase as you move towards the base plate. To compensate for this, the width of the arms is thicker as you move towards the base plate. The results of this part’s simulation were very good, with a maximum deflection of 0.003 mm and a maximum stress concentration of 196 kPa.
The von Mises simulation analysis of the second design showed large areas of low stress concentration in the middle. In order to reduce the weight, a large outline hole was cut in the low stress areas; however, after running a simulation, this new design showed major areas of stress concentrations and displacements. Without any support inside, the arm yields a max stress concentration of 297 kPa and a maximum deflection of 60 mm.
After a few more attempts, the holes were sized and positioned until the best combination between strength and weight was found. In the final revision of the arm, the maximum deflection is only 4.4 mm and the maximum stress concentration is 540 kPa.
The motor mount also went through several iterations. The initial design was a simple plate that attached to the slot in the arms. However, this caused some issues because the arm would not stretch far enough for the motor and the fan blades to be at the correct height. In second design, a block was used instead of a plate to fix the dimensioning issues. This concept was later revised by the addition of screws because the original type of fitting was not secure enough. This part is not meant to be disassembled often so using screws does not hinder modularity.
Battery and Styrofoam Mount
Originally, the battery and Styrofoam mount were two different parts. The battery mount was part of an extrusion from the base plate and the levels of the structure were supported by pegs. The Styrofoam mount was in its own separate level from the rest. The major issue with this design is the number of parts required. This not only increased overall weight but also the complexity of assembly. In the second design, the battery mount and the Styrofoam mount were combined into one part. By removing the pegs required in the inter layers, the assembly complexity was reduced and the overall weight was lowered.