Embedded Files
A 2024 comp review
This is a summary of the Australasia competition from our team's perspective.
To preface the whole weekend, we went in with nowhere near as much testing as we had wanted, nor as much as we needed. This is common in the world of FSAE, but that's hardly an excuse. It's the biggest opportunity to get the most out of the car and for a small team to outperform a high-budget 50+ member team; Newcastle Uni is a great example of how to make the most of what you've got. We should be too, but the team came out of 2023 with too much excitement at a 4th place that was largely due to lots of bad luck experienced by other teams who would otherwise have us covered on pace, although we did have genuine reliability. We tried to do too much in 2024. We succeeded in getting everything designed and on the car, but at the cost of actually understanding how much of it worked.
It wasn't entirely unintentional though. Our team member count is among the smallest at our competition, and the number of heavily involved members each year can vary from none to maybe 6 or 7. So whenever we can foresee a strong year of involvement from a solid number of people, it's an opportunity to target lots of big changes, where future "quiet" years can then focus on testing and understanding when the general team pace can't facilitate any design changes. Even with this strategy though, what we tried in 2024 was definitely pushing it.
Anyway, on to comp, and static events first.
I was in charge of the cost event since my first full year on the team in 2022, and it has been our strongest static event with a 4th place in 2022, and one of the lowest error counts of all teams in 2023. Given the 2024 was to have less powerful motors and a smaller battery capacity, and it being my final year on the team, I decided we'd go all out on the cost report and target maximum cost reduction and a top three placing. Between the motors and battery, a lot of process optimisation, and some pushing of rules definitions, we shaved just under $8000 (or 19%) off the total cost.
For the cost scenario, we had to replace a purchased ready-built part with a custom alternative as a means of saving money. There was a huge opportunity to make our life easy here, as the inverters we use are over-spec'd by a factor of 3.3, and this hindered us (and still does) enormously in the cost report. We only continue to use them because they're what the team has, and we finally have a documented understanding of how they work and how to tune them for maximum control and efficiency. Our team leader had just completed an honours project on custom inverters that were heavily optimised in cost, mass, and thermal performance, so we already had all the data we needed for our scenario. If the team gets these custom inverters built and working over the next few years, there's enormous benefits to be had for both the cost event and dynamic performance.
I walked into the cost judging room as usual, ready to defend and explain various errors in our report, but we were told there were no mistakes and that the drawings were the best standard of any team at the competition that year. Between that, and good delivery of our cost scenario and on-the-spot assembly costing, we got 2nd place and the first event trophy for the team in 10 years. I didn't exactly enjoy leading the cost event for three years, but this made it all worth it. To think we could have easily placed first if not for the inverters that someone chose 7 years ago...
Business was next, and due to never really having an understanding of what is required to do well, this event has typically been sacrificed by our team. We'd go in, state some arbitrary numbers, collect our 15 out of 70 points, and leave. This year, thanks to finally having a business student (and a good one at that) to lead the event and get us engineers thinking properly, we had a cohesive presentation with an actual business case. It was pretty good too, getting us 7th overall with a score of 85%.
Last was the design event, and we knew this one would let us down largely from the lack of testing. Indeed it did, giving us 14th. We could take solace in the fact that we generally scored well in all other non-testing-related categories, giving us a score of about 70% overall. Less than we'd like but better than we were expecting, although in saying that I suspect the same went for many other teams, as having 13 other teams scoring higher than our 70% suggested some generous judging overall.
Inspections were a breeze on both mechanical and electrical fronts, which thankfully we've made a bit of a habit of. The only issue this year was some Velcro needing to be replaced with glue as a mounting method for a pcb inside the accumulator, at the request of the judges.
Now for the dynamic events.
Skidpan was first, and it was wet. The only testing we had under our belts prior to comp was in wet weather, and we'd seen the wet forecast so had put all our effort into optimising traction control, torque vectoring, and geometry setup for wet running. Unfortunately it didn't stay wet, so our promising 3rd place after round 1 eventually slid down to 7th in the dry, but this was still the team's best result in this event, and we were the fastest car that didn't have a full front wing/rear wing/floor aero package.
My first time as a driver at comp came on my third and final visit, for skidpan (above) and endurance
Next was acceleration, in which our simulations had suggested we should place 3rd with a time of 3.92 seconds for a 60kg driver, so we were all excited for another potential trophy here. After round 1 we were indeed 3rd, with a time of 3.96 seconds (with a 67kg driver). A pleasing result for the effectiveness of our launch and traction control logic, which was assumed to operate perfectly in the accel simulation. In the end we finished 5th, failing to better our first run time. We were expecting to have a 60kg driver in the car for the event, but by the time comp came around our lightest was 67. If we had a sub-60kg driver like the two teams that ended up getting past us, 3rd place would have been in the bag as the margin from 3rd to 5th was less than five hundreths. Disappointing to lose out over an uncontrollable factor that had nothing to do with the car's design.
Below is a comparison of our best acceleration event time from 2024 back to 2022 (top to bottom), showing the results of progressively lighter cars and incremental traction/launch control improvements, with 2022 having no implementation at all. Each year was almost exactly half a second faster than the last.
Autocross was over in under 20 seconds. This was the first time the new battery had been charged to absolute maximum capacity. The first time our driver lifted off the pedal, it sent a tiny negative current back to the battery that was able to sneak past the inverter PI controller that allowed small overshoots. With the battery being full, the BMS registered this as charge coming in to a full battery and shut off. Such a simple mistake that could have been avoided if just one of many small things had been done slightly differently. If the throttle was pushed harder off the start line, if it had been released more gently, if the battery had been taken off a few minutes sooner, if testing had flagged this issue beforehand...
We had expected to end up somewhere between 4th and 6th in autocross. Our simulation suggested 7th-10th but our driver was definitely a strong point so our expectations were higher. This would have given us around 100 points, but instead we got none. Before this, we were on track for an overall top five finish. If we had achieved 5th place, it would have felt better than our 4th place in 2023, as we'd earned this one through decent vehicle performance and generally strong static events, not just through others' bad luck. With this possibility now off the cards, we made a decision to target a top three finish in efficiency at the expense of a very slow endurance time which would sacrifice another position or two overall. We thought 10th and another trophy would be a better result than 8th without a trophy.
We re-ran our endurance simulation and it suggested the best way to score maximum efficiency points (which balance pure efficiency with lap time) was with a 9 kW power limit. This didn't consider the hill at Calder Park though and we didn't want to be flagged for going up it too slowly, so to give us some margin we went with 12. With minimal regard to total finishing time, the goal was purely to drive as smoothly as possible. Given our maximum speed was now limited to somewhere around 40-45 kph, this meant using the brakes about once per lap, and going for minimum distance instead of racing lines in most corners. Toe was set to near-0 all around and tyre pressures set reasonably high.
I was the second driver, and for my first time driving on the main track at any FSAE competition, this was about as calm of an introduction as you could get. Unfortunately it was so calm it was hard to keep focused and I rolled right through the main straight slaloms on one occasion, costing us a 20s penalty. In the end it was still enough to score us 2nd in efficiency, and amazingly still mid-field in the endurance part of the scoring, albeit with several DNS/DNFs from other teams. As predicted, we ended up 10th overall, and went home with two 2nd-place trophies. There were a lot of what-ifs regarding this competition, but it was still a very satisfying end to my time on the team.
Cue the afterparty.
Google Sites
Report abuse