RC electronics is something that has always fascinated me. The sheer ability to control an electric motor with high precision from a great distace at will is something thats stuck with me. As my experience in mechatronic systems has developed, so has the freedom of maneuverability of designs. With the recent rise of the Formula-E class my next challenge was to investigate how well I could replicate not just the design, but the performance of a Formula-E vehicle by incorporating my own ideas and inspirations to make up for funding and scaling issues at the ~1:10 scale.
Inspiration
The original plan was to model something closer to a 70’s indy car both due to simplicity, the lack of safety features, and the decreased focus in downforce since smaller object scale to have greater relative air resistance; allowing a better scaled design. One of my all-time favourite cars was the 1969 Ford Brawner Hawk III, which was an AWD indy champion. Soon this design became unrealistic for the first few versions due to scaling issues but remained something I want to make eventually.
Before designing; to minimize cost and complexity; I started by choosing a universal screw size of M2, bearings that were 8x12x3.5mm, then retrofitted the design to my current electronics, and spliced it in parts such that it fitted my 15cm cubic 3D printer. This meant each part would be a maximum of 15cm which posed challenges of extra tolerancing and weight from screws and decreased stiffness, aesthetic, and laminar airflow.
V1.0
The first design was modelled after 2018 to 2010 models with certain features adjusted for less downforce and airflow directed towards cooling the battery and ESC instead. The major challenges involved the chassis which was essentially a flat plate as it could not be printed as one part due to its size. The first idea was to make the parts interlocking, however due to how brittle the 3D printing material of PLA was this performed quite poorly. Instead I settled for my own version of ‘friction welding’ on PLA which involved pressing a dremel with a piece of filament inserted into it against the pressed ends of the plates such that the friction produced enough heat to fuse the plastic with a similar strength to the 3D printing.
Although a great proof of concept, this was not yet quite what I had envisionsed. It was fun to slide the back around, but the low gripping performance of 3D printed tires forced me to opt for rubber bands until I could find a better option. Next, most likely due to the printing heat and the filament used, the quality of the 3D printed parts was quite low, so I decided a new version was needed. Some flaws included a very low gear ratio to the wheels, a low maximum steering angle, weak steering servos, and unbalanced weight causing left turns to be sharper than right turns. Additionally, the wheels were quite narrow in comparison to the length of the body, so wider wheels would be highly beneficial in steering agility. Some great potential additions would be a rear differential and a front suspension system. With that in mind I went for V2:
V2.0
Improved aesthetics, steering, responsiveness, grip, and aerodynamics.
The process in V2 was much longer and delicate as this time I learned from my mistakes of being sloppy in the first version. Many parts received a generous 80-100% infill not just for the strength but for the extra stiffness as many parts had quite a bit too much play in their positions. The servo was also swapped and rubber tires took grip to a whole new level. zipties were added to restrict the new ackerman steering optimized steering from locking out; in trying to increase the maximum steering angle, I now allowed the steering joints to extend beyond their operating range of angles. The main flaw remainining was power transmission. My gearbox was not performing reliably due to the low tolerances of my 3D printer resulting in poor meshing with both helical and spur gears, not to mention the excessively loud noise.
Although this was starting to look a lot more like what I wanted, the plan is now to add: a differential as shown above with the help of an SLS 3D printer in nylon, front suspension, adjusted lock-proof ackerman steering geometry, a rear differential, a rear wing with an actual airfoil instead of a flat surface, a lower chassis for improved ground effect, and the removal of excessive body parts which only serve the purpose of imitating F1 cars such as the width of the central section, the headrest, and the front wing design. Additionally, now that I have learned how to use CFD, I can iterate my design much more effectively.
The Future
After that I plan on incorporating a belt drive to connect the front and rear wheels in an AWD system with a front differential, a stiffer wishboned front steering mechanism, internal body stiffeners, and excess weight reduction; this will resemble my first image. Additionally I want to test out 2 new ideas for extra weight reduction; generative design and varied infill printing with a bone-like ‘gyroid’ structure which is an ultra light, high surface area, high strength internal structure to fill gaps of parts that dont need to be 100% plastic. It is also on my list of projects to make a splicing software which uses generative design methods and machine learning in order to properly allocate infill density based on specified loads. Recently I have also been reading a lot on compliant mechanisms and have been especially interested in incorporating the unconventional, yet extremely useful method to create a functional suspension. Although the physics is very cumbersome, many design features can be figured out via current softwares such as FEM so as to always keep the stressed joints out of the range of plastic deformation. Additionally, this method is applied much more easily with stronger, more flexible 3D printing materials than Polylactic acid such as the famous thermoplastic Polycarbonate (PC).
I.R.L. …
Since 2019 I have been part of my shcools SAE F1 team. Most of my impact this year has been in helping manufacture parts due to my experience with machineries and wet layups. In the process I am also learning about the car across all subteams and providing some 3D printed parts. Although my time has been limited it is definitely on my list to help design the 2021 SolidWorks model along with doing some FEM testings and to gain experience on dynamometer testing. Being one of the teams drivers is also an eventual goal for me. Working with this team has tremendously helped my further sophisticate my own design and given me the insight no book can, especially regarding the onboard embedded electronic systems and hydraulics.
because the torque-curve corresponding to brushless electric motors is quite good over all rpm’s I decided to make only a 2 speed transmission which can run in reverse, the ratios are 1:2 and 2:1, all that is needed is to replace my finger with a servo and an infrared rpm sensor hooked up to the arduino to shift at optimal motor rpm’s. The transmission also has a neutral gear which may be useful in some situations. I definitely plan on extending this to a 4+ speed transmission with all helical gears both for the challenge and in order for it to work well with a solenoid engine, which will be described next:
Alternative Power Sources…
Although my current BLDC motor works great, its actually a little overpowered. After finishing V3.0 my plan is to actually switch to either a V4 or V6 solenoid engine, with the main reason being that it resembles a cars engine quite well and because its something I’ve never tried and seems pretty cool. The main issue to tackle will be generating enough torque.
Luckily I have my certificate to use my universty machine shop. This way I can work on my craftsmanship and gain real world experience tolerancing. In addition I plan to make parts out of 3D printed PLA to make parts faster and cheaper to manufacture. Because of the bulk of a solenoid, I may need to redesign to match something dating back to the first half of the 20th century. A front engine mounted 4WD F1 car, but with modern aero features to provide much needed downforce at the rear.
So far, almost everything is finished from the chassis, to the front and rear suspension, the semi-automatic transmission, the steering, and electronics housings. All that remains is a nice body to go over the chassis and the front and rear wings. The arduino nano along with some gyros, accelerometers, and ultrasonic proximity sensors will be wired to for brake lights, obstacle avoidance, and gear shifting, but the possibilities are endless with the right sensors and code.