2015 FSAE Michigan

[JT A.]

JT A,

I'm not sure I agree that the rules have been stagnant. They have become larger, more complex, and more prescriptive since I was a student. There also have been small but significant changes to weightings that have seen the ideal concept move a fair bit.

I would also propose that without some stability in the rules (as well as freedom) there is little opportunity for real innovation. To do so takes a lot of time if you want innovative solutions that are performance improving. It is a common mistake to assume that improvement happens gradually at all times. Unexpected rapid changes is what we should be expecting. Spaceframes for an eternity, and then all of a sudden lots of carbon tubs. 13" wheels as standard, then everyone on 10"s. Almost no aero, then nearly all the top teams (yes a rule change helped this, but current regs allow about as much downforce as what they were pre big change). 600cc 4s then singles. The list goes on, but the pattern is the same. The difficult part is trying to predict the next big change, and be ahead of the curve.

If there is stagnation it is, and always has been, in the teams. Hopefully no-one enlightened enough on these forums think that GFR has produced the ultimate incarnation of a FSAE vehicle in the last few years. There are so many good ideas running around some of the lesser known teams that with a bit of development could turn into some great performance. At the moment I think we are on the verge of entering a new style of car build in the next few years, one that will be a little easier for smaller teams to complete when compared to the mini F3 cars we have had at the top since about 2006. I guess time will tell though.

I am always curious as to which team is currently on a five year plan that will produce the new "ultimate" FSAE car. We don't need F1 style shotgun style rule changes derailing these efforts.

Kev

I guess you and I have a different idea of what the competition should be about.

I don't want the competition to evaluate which team had the smartest guy 5 years ago that came up a long term plan, and a good initial design, allowing the next 4 years of students to copy 80% of the design over every year. Allowing them to spend their extra free time training drivers and rehearsing their design speeches.

I would rather see the winner of each competition determined on which team did the best engineering work that year. Who had the smartest engineers that year. Who designed the fastest car that year. Not "who already had the best starting point handed down to them from the year before and managed to not screw it up"

[Kevin Hayward]

JT A,

Why would having a decent long term plan mean that any one year does less engineering?

Why would a decent long term plan involve copying designs every year?

Why would students be involved in the project for only one year?

A decent long term plan would involve setting targets well beyond the capability of one year group, and then chip away at them year after year, with the team developing good methods of information retention and mentoring along the way.

Kev

[Z]

... the rules have been stagnant for way too long! The competition needs a shake up.

JT A.,

I ... agree more with Kevin on this one...

While I agree that the "shake up" suggestions you listed COULD make the competition more interesting, in practice I do NOT think they would make any difference at all to the "Design" efforts of the students. Quite honestly, I think the students would still bring exactly the same "traced" copies of the cars that they bring now.

The "proof" of this, IMO, is that despite the relative constancy in the way the competitions have been run and scored over the last 30 odd years (ie. fairly constant points awarded to fairly constant types of events), the cars are still nowhere near "optimal". That is, their performance is a long way away from where the Laws of Nature allow it, even for very low cost cars. After 30 years of development the cars should be MUCH, MUCH, FASTER.

To spell this out with a specific technical example, the cars started out 30 years ago with a F:R weight distribution of about 50:50. This was an accidental by-product of using bike-engines, with their forward-leaning cylinders and rearward-power-take-off, placing these unmodified bike-engines behind the driver as on real "Formula" racecars, and then using short-as-possible wheelbases to suit the Autocross tracks. The resulting layout with its 50%R weight distribution has hardly changed in 3 decades.

But, as I have pointed out at great length, taking a "big-picture" view of the Mechanics of this problem with its typical mass, power, and racing speeds, or by simply studying the prior art, shows that this 50%R is WRONG (at least for 2WD). The only excuse the students can use to justify this ineffective R% is that "...it allows us to minimise build-time and costs by using an unmodified off-the-shelf bike-engine...". This is hardly a valid excuse after 30 years of competition.

Worse yet is that the more recent E-cars have EXACTLY THE SAME R%! So while these E-cars should be blowing the C-cars off the track in Acceleration and AutoX (and only having to hold back a little in Enduro, maybe..., to conserve batteries), instead they are EQUALLY AS SLOW. Yep, they have abundant power, easily put-down, but they are equally traction-limited out of all the slow corners as the C-cars. THIS IS NOT A DRIFTING COMPETITION!

Bottom line here, for 30 odd years the FSAE students have been encouraged to be "Tracers", as per my previous post's Drawing Office hierarchy. And as popular as those girls might have been in the male-dominated DO (especially the younger ones!), they themselves would happily admit that they know NOTHING about Engineering Design!
~~~o0o~~~

... making the [Rules] more and more restrictive ... takes freedom & creativity away from the students in design, which was one of the best things about this competition ...

This I agree with 100%. Certainly keep the Rules as open and NON-prescriptive as possible.

In some senses I think that the stagnation of the actual "designs" of the cars (for example the constant, though poor choice of, ~50%R) is down to the over-prescriptive Design Scoring Sheet discussed before. This, together with the overly prestigious nature of Design Event, and the fact that DE is worth 150 points, which is quite a lot.

It follows naturally that typical student thinks,
"Well, we are no chance of winning outright. But let's at least try doing really well in Design. Its worth 150 points, and ... a high score there is what we really want on our CV when we go looking for that top job!"

So the typical students go through the Scoring Sheet, item by item, and start "Tracing" each and every one of them. Yep, neatest correct entry wins. NO original thought required...
~~~o0o~~~

So, How To Change Design Event ... To Produce Better Engineers?
================================================
1. Get rid of Design Event.

As noted before, the Dynamic Events + Cost give a straightforward, very objective (err..., if Cost is more realistically done) way of measuring the "fitness of purpose" of the overall Design. And should a Team build a car that is really fast on track, even if (?) they do not understand "why" it is fast, well I reckon they should still score high points for making such a good "guess" of which car to copy. (Seriously, no Team can "accidently" build a really fast car...)

At the very least, DO NOT allow DE to be worth any more points than now. In fact, about 100/1000 points seems better.
~o0o~

2. Let Paul-The-Octopus pick the Design scores.

Hey, his success rate back in the ~2010 Soccer World Cup was something like 85%! That is a lot better than this competition's Design Judges' record of picking winners...

At any rate, having a Design score of, say, 100/1000 points that is down to a roll-of-the-dice, is better than a DE that encourages the students to "... just TRACE last year's car, but do it more neatly this year... with more Matlab graphs, and stuff...".

With 100 points down to a roll-of-the-dice, the Teams that REALLY want to win will make a big effort to build a REALLY fast car (ie. must be 100 dynamic points faster than anyone else!).
~o0o~

3. Get rid of the Scoring Sheet guidelines.

Instead of the prescriptive scoring, get a small number of very GOOD Design Judges to score the Teams any way they see fit. The Teams should have no idea how they will be scored. Only that a small number of people who have track records of producing outstandingly good designs will have a quick look at each Team's car, and a short chat with Team leaders, and, after seeing the car's dynamic performances (or not?), they allocate scores.

For educational purposes these DJs should also give a very short written review to each Team. Just a short list of what, in their opinion, is OK, and what could be improved next year...

Z

(PS. Kevin, "A decent long term plan would involve setting targets well beyond the capability of one year group, and then chip away at them year after year, with the team developing good methods of information retention and mentoring along the way.". Agree 100%!)

[JayNoon]

Gentlemen, having been though Michigan Design Finals, I report my faith in the the design event has been restored.

- I did not feel our score was "prescribed" from a score sheet.
- They went to great lengths to give our whole team feedback following the event.
- I felt that the judges were probing the limits of my knowledge and understanding first, my design choices second. "How do your dampers work?" then "How did you tune them?".
- The actual physical design of any one component on the vehicle mattered very little compared to designer's thought process and understanding of load conditions, material properties, etc.
- To quote a design judge from two years ago "The most impressive students we encounter don't need the physical car to give their design presentation". And I felt this was very true, I hardly looked at the car itself.

Z, As far as picking "winners", that doesn't seem to be the design judges job description. That's what people do at the horse races. The judges are trying to evaluate which team is most knowledgeable, understands fundamental concepts for their systems, and backed up their choices with testing and validation. This took almost three hours. They don't know if the drivers are slow, or something unlucky will happen. I'm not saying this is how I THINK design should be, I'm REPORTING on how I felt at the competition. Also, ETS's car with inboard suspension and roll/pitch independent damping was AWESOME and deserved the win even if it's not super fast.

Kev, I agree with your comments entirely. Our program has been chipping away at big long term goals the for last three years and focusing heavily on documentation and knowledge transfer along the way. Moderate innovations based on resources available, finish the car on schedule, lots of testing time, very reliable. Seems to be working pretty well.

One last note: I would be wary of calling 10" wheels, carbon tubs, and singles "improvements". Our placement in Design AND Endurance should reflect that any platform can still be competitive.

[Z]

Jay,

I spend time here because I see this FS/FSAE competition as one of the very few places in the world today where students who want to learn a bit about "Engineering", can, and do, actually learn something useful. Importantly, IMO, it is primarily the stopwatch and things like DNSs and DNFs that do most of the teaching. That is, it is they that "examine and assess" the state of the students' education.

The reason I harp on so much about the Design Event is because, IMO, it actually works AGAINST the education of the students.

Here are some examples of how this negative education works, taken from your last post. (Note that similar arguments to yours can be found in countless defences of the DE, so please do not take this personally).

1. The Team that had the fastest car in AutoX, and the fastest car in Enduro, and the most Fuel Efficient car, and also won the competition Overall, were assessed by the Design Judges as being BAD DESIGNERS.
Why?
And what message does this send to all the other students about the difference between "good design vs bad design"?
Well, maybe this next one answers that...

2. "- I felt that the judges were probing the limits of my knowledge and understanding first, my design choices second.".
So why do they call it "Design Event", rather than, say, "Knowledge and Understanding Event"?
Taking an example that shows how wrong this approach is, should a person who has fantastically good "K&U", but makes shockingly bad "design choices", be called a good "Design Engineer"?
Should an Engineering company working in a life-critical industry, such as building Jumbo Jets, Big Bridges, etc., employ such a person as a "Design Engineer"?
I wouldn't...

3. "- The actual physical design ... mattered very little compared to designer's thought process and understanding...".
As before, WHY?
The above attitude is similar to "In Maths exams, show all your working, so even if your final answer is utter bollocks, we can still give you some marks if you were sort of on the right track...".
That attitude might be acceptable in primary school, and maybe even in a lot of the useless parts of the real world, but I do not want to fly in a Jumbo Jet, or drive over Big Bridges, that were designed with that sort of "rough enough is good enough" attitude.
These competitions are pretty much your Final-Exam-of-Big-School, with the next stop being the Real-World-of-No-Second-Chances.
Which is why the stopwatch makes such a good examiner, IMO...

4. "Also, ETS's car with inboard suspension and roll/pitch independent damping was AWESOME and deserved the [DE] win even if it's not super fast.".
Again, WHY???
I have no idea of the details of ETS's car, but isn't the hypothetical customer looking for a car that can win races?
Nowhere in the design brief (ie. the introductory section of the Rules) can I find anything about building an "AWESOME" car....

5. Instead, the design brief, specifically "A1.2 Vehicle Design Objectives", asks for a car that has "... very high performance in terms of acceleration, braking and handling... [and is] ...sufficiently durable to successfully complete all the events...".
It seems to me that ETS's car, 1st in Design Event, so apparently best "designed" car, failed to meet that design brief on both emboldened counts. Namely, not very fast in AutoX, and it broke in Enduro!
On the other hand, GFR/Oregon's car met all those requirements, yet got a -45 point flogging in DE...

I am sure that Paul-The-Octopus could have judged it better...

Z

[Kevin Hayward]

Jay,

The reference to 10"s, carbon tubs, and singles wasn't meant to name them as improvements, rather that the change of the "standard" design can happen quite rapidly. The majority of top teams might start running DASD's and beam suspensions. The fact that the majority of top teams run a certain system does not necessarily mean it is the best.

However I would resist the temptation to say that these decisions are made between equal, but different alternatives.

By the way stunning looking car again, and good to see you guys doing well. I have always liked the build quality you guys achieve, and it is almost impossible not to like a team that choose gulf colours. Also nice to see the Aero working for you. How much do you feel it added?

Kev

[JayNoon]

Thank you Kev!
We were targeting a 1.5 second/lap improvement. It's difficult to tell from competition, but I feel we were seeing about that in peak lap time. For the record, roughly an equal points boost was earned by more consistent drivers and fewer cone penalties.

Z, the fastest team won overall! And top-10 endurance results typically match overall results to the letter! So what more do you want?!
(Rhetorical question, your point is already well received.)