Home
Note: Due to the proliferation of COVID-19 cases in the USA, the scope of this project was significantly reduced early during the design process. The project was redefined to meet the needs of the students, sponsors, and staff involved.
PROJECT BACKGROUND
The design for this heavy-lift drone motor originates from an experimental electric motor design from 15 years ago. The San Diego Computer Society designed a novel axial gap motor and fabricated and tested it as an electric bicycle motor. The motor uses a ring of NIB rare-earth magnets and stators hosting ribbon coils for high-torque, low-voltage operation. However, the prototype's structure is made of aluminum and suffers from hysteresis, magnetism, and cooling issues, and the project was forgotten. More than fifteen years later, this project is being revisited to see if its former glory can be restored and put to use in a modern context.
THE PROBLEMS
1. Lack of data from previous iteration
➜ Most data from 15 year-old design is lost or nonexistent, all numbers used are estimates
➜ Cannot run the provided motor from previous iteration because of missing parts and damaged hardware
2. Overheating
➜ Unnecessary heat limits the current passing through wiring and reduces future motor viability
➜ Current stator design has no way to eliminate extra heat
3. Poor materials choices
➜ Eddy currents induced in the aluminum stator resist the rotor
➜ Current materials gather additional heat generated by the coils and the eddy currents
OUR PROCESS
1. Change the materials
➜ Find material with high thermal conductance and low electric conductance
➜ Change stator design to address overheating issues
➜ Run analysis tests to find trends for suggested design changes
➜ Manufacture and assemble
2. Introduce a cooling method
➜ Research possible cooling designs and replacement materials
➜ Run analysis tests to find trends for suggested design changes
➜ Manufacture and assemble added cooling method
3. Optional goals include:
➜ Change the magnet pattern to increase motor effectiveness
➜ Optimize the coil pattern and placement
➜ Optimize the caliper size
➜ Optimize cooling methods
FINAL DESIGN
The final design of this motor resulted in a motor with a rotor 2 stators on each side featuring 6 copper coils each, 2 liquid cooling reservoirs, and a new hall-effect sensor mount. The channel for the rotor between the stators was also increased by 33% to avoid collisions. All new parts are printed from a Glycol-modified Polyethylene Terephthalate, also known as PETG. PETG is a chemically-resistant thermoplastic with a low thermal expansion coefficient (relative to other thermoplastics) which prevents warping over time. It conveniently comes in a 3D-printable form, making it a great solution for rapid prototyping for this project.
OTHER CONSIDERATIONS
1. Public Health, Safety, and Welfare
➜ Heavy-lift drones can create many opportunities to reduce unnecessary dangers for many jobs, especially those in the construction, search and rescue, firefighting, and other career sectors involving significant heights.
➜ Heavy-lift drones carry large loads so crashes could be very dangerous. Machine must be regularly tested and maintained to prevent accidents.
➜ Internal coils will reach dangerous temperatures. User must handle the motor with care and exercise caution at all times.
➜ Liquid coolant mineral oil has the potential to cause eye irritation, but not blindness. Flushing eyes with water for 15 minutes should suffice for most cases.
➜ Liquid coolant mineral oil has the potential to cause harm if consumed. If large quantities are swallowed, seek medical attention immediately.
➜ Liquid coolant mineral oil is generally not toxic to humans. If skin contact results in irritation, seek medical attention.
➜ Carcinogenicity:
- IARC: No component of this product present at levels greater than or equal to 0.1% is identified as probable, possible or confirmed human carcinogen by IARC.
- NTP: No component of this product present at levels greater than or equal to 0.1% is identified as a known or anticipated carcinogen by NTP.
- OSHA: No component of this product present at levels greater than or equal to 0.1% is identified as a carcinogen or potential carcinogen by OSHA.
2. Social and Cultural Impact
➜ Heavy-lift drones will eventually have capacities to carry humans. This could revolutionize single-person transportation for short distances.
3. Environmental Impact
➜ In tandem with green electricity, drones can contribute to a large reduction in CO2 emissions, resulting in healthier environments around the world.
➜ Noise pollution is highly likely, current models averaging ~80 decibels. Propeller geometry can be used to decrease this in future iterations.
➜ Mineral oil has a very small environmental impact, but leaks can disturb environments nonetheless and need to be prevented.
4. Economic Impact
➜ Heavy-lift drones will make some industries more efficient, like delivery or construction. It will also affect many labor-based jobs, but will increase manufacturing jobs.
5. Global Impact
➜ Used with green electricity, drones can contribute to a large reduction in CO2 emissions, resulting in healthier environments around the world.
➜ Heavy-lift drones will eventually be able to autonomously deliver to areas that are unreachable by other means, saving time, money, and maybe lives.
EXECUTIVE SUMMARY
In the world of electric motors, there are two options. The most common is a radial flux motor, where the radial components of the magnetic field turn the rotor. Another configuration, which is seen less often, is an axial flux motor which uses the axial components of the magnetic field to turn the rotor. This project focuses on the use of an axial gap motor in applications for a heavy lift drone motor. While there are advantages and disadvantages of both types of motors, an axial gap motor can typically realize larger output torques than a radial gap for the same voltage inputs. This is due to how the motor is configured geometrically, with the electromagnet coils being located on a longer moment arm than a radial flux motor, which provides more torque. This project focuses on a prototype axial flux motor that was developed by a couple members of the San Diego Computer Society (SDCS).
Another key feature that the SDCS wanted to incorporate in their motor was the use of flat spiral electromagnetic coils (AKA pancake coils) as opposed to traditional helical coils. This pancake design was accomplished using copper ribbon as opposed to traditional copper wire. Copper ribbon has a thin rectangular cross section, whereas copper wire has a circular cross section. Using copper ribbon provides an advantage over wire when it comes to heat dissipation. Oftentimes when a coil is wound with wire, successive wraps are buried inside of each other, thus limiting the amount of heat that can escape from the coil. When copper ribbon is used, every wrap exposes part of itself to the ambient environment, thus increasing heat dissipation and allowing for a cooling fluid to interact with more surface area of the coil.
The initial prototype that the SDCS created had the coils made from copper ribbon, as stated above, and with all other structural elements (stator and rotor) made from aluminum. The stator was created in a caliper-style arrangement, with electromagnets on both sides of the rotor. The rotor was a spoked wheel with rare earth permanent magnets located along the rim. Because the stator was constructed out of aluminum, the permanent magnets of the rotor would induce a current within the stator when it would rotate. This would lead to not only an increased temperature within the aluminum, but it would also create a resistive magnetic force. Due to these issues, the main goal of this project was to add a cooling solution to limit the temperature of the coils, while also changing the stator/rotor material to limit induced eddy currents.
PETG was chosen as the stator material due to manufacturing and ordering constraints caused by the COVID-19 pandemic, however, it is recommended that G-10 be used as the construction material if the design is to be iterated upon again. All pieces of the design were 3D printed on a Prusa i3 MK3. Each phase of the motor was chosen to have 2 coils on each side, all wired in series. It is important to note that during wiring, the polarities of identical coils on opposite sides of the stator must be swapped by wiring them up in reverse to swap the current direction going through them. This is caused by the two sides of the rotor having opposite polarities of permanent magnets.
The cooling reservoir was sealed with an O-ring. A 50A durometer O-ring was used to limit the flexing of the PETG for a more even seal. FEA analysis was done on the screw holes to ensure an even enough distribution of stress on the O-ring to create a watertight seal at even the most troublesome points.
For the prototype, free convection was used by filling the reservoir with water for testing. Mineral oil is recommended for the final prototype due to its insulative and non-solvent nature. It is important to note that the flash point of mineral oil is 112 C, so care must be taken to ensure that hot spots of the coils are free from contact with the atmosphere.
Aluminum coil inserts were manufactured from standard NPT plumbing inserts and threaded into the PETG stator. These metal inserts were chosen to increase conduction and allow heat to move towards the cooling fluid without creating extreme hot spots in contact with the thermoplastic material. Coils were wound and inserted into these inserts with 24 gage copper wire and then JB weld epoxy was used as a water-tight sealant.
The overall success of the motor was limited due to social constraints, limited manufacturing techniques, and a lack of documentation for the high powered motor controller. However, the team was able to verify the continuity and wiring for each of the stators, energize all of the coils and create a non-trivial magnetic field from each coil set. The team’s biggest accomplishment was verifying the results from the FEA analysis the team built early on in the project. This will help predict performance calculations for the copper ribbon coil wraps and provide a basic idea as to motor performance. The team was also able to create a solid design that could be easily adapted to a G-10 stator with square cross sectional area ribbon cable. The PETG reservoirs could be re-used with the G-10 stator and provide free convection cooling.