Printable View
How do you replace Carroll Smith? It's a serious problem. At the time, my memory is that a number of the "racing industry judges" discussed this but we didn't get very far. Others may chime in, but as I understand it, Carroll really elevated the stature of design judging.Quote:
Originally Posted by Big Bird
Yep, Doug, he was a one-off, and I don't think we would ever find another with all those qualities rolled into one package. I had only one encounter with him, in Oz 2002, and he was already quite ill then. He knew his engineering, but he also had a great empathy for the students, and that is what set him apart. It seemed he was there because he sincerely wanted to help the students - not because he wanted the students to know he was helping them.Quote:
Originally Posted by DougMilliken
Dunno, has anyone approached Ross Brawn?? That was just an idea off the top of my head. Are there others around that people know of who might have similar qualities??
This is a good idea.Quote:
Originally Posted by Big Bird
It'd be excellent to add a rubric or template requiring students to demonstrate a PDCA cycle against any set of the competition objectives for whatever their "something special" is. It'd be great; would be good to install a culture away from 'we were creative but we're not too sure why, we didn't invest in a design process covering why and how we best needed to do this' and 'we built this really cool thing but never got around to validating it, because we thought it'd show in lap times and therefore never did any system performance checking along the way, and then the car was completely unexpectedly first-time-ever-in-FSAE late and we didn't get to test pre comp'.
I'm absolutely not advocating that it has to be all-new. I'm simply pointing out that what you've highlighted of your own history -Quote:
Originally Posted by Swiftus
- does not happen much at all. If this is what GFR DE does, great. That's proper grounding for industry work and very much in tune with what the competition hopes to instil in students that partake. This is quite far from what happens often, however. I'd be keen to see the difference appraised formally. A change log, have students justify the difference, etc.Quote:
Originally Posted by Swiftus
We have a few other issues with some teams of 60+ people with a barely-changed package (I don't mean barely-changed concept - I mean a barely-changed design) competing with teams with <10 students and a larger work effort to deliver. Nothing's perfect however we should strive to make the playing field a little more level.
To this end, I'm going to suggest something very contentious: the competition needs a maximum term rule for students. Two years, three years maximum. You register with the SAE and your university, you can only enter a number of times then you're ineligible. This would have good implications through design documentation management, fit within many faculties' learning initiatives, etc. The competition was never intended - and should never be a place for - perpetual students. You know who you are.
I agree, through judges aren't supposed to be casual observers: as an entrant, you're meant to talk though the 80%. We need to find some ways of making this easier to communicate. I really did like the videos and think a better done version of as much could be super useful, with an eye to capping resources going into it and ensuring that teams can use it for other means too (advertising, sponsorship, university presentations, etc).Quote:
Originally Posted by Swiftus
I'm not disinterested in the cars getting faster year-on-year; I hope they do in the sense that between (1) the design criteria required to meet the competition objectives, (2) the knowledge freely available to students to deduce what competitive performance figures are, and (3) the wealth of knowledge and example in prior art in the competition, there is no reason that students should not be able to aim for and realise cars that are reasonably quicker year-on-year.Quote:
Originally Posted by Swiftus
I am completely against, however, the competition being interpreted as being solely about building a quick car, because (1) that's very explicitly not what the competition is about and (2) paying attention to the rest of what it's about lends to an inherently better design, which in turn can allow a greater and more accessible performance envelope and a design process more reflective of the careers students should aspire to. (As in, the people that wrote the rules had a few clues worth following).
This is exactly the point.Quote:
Originally Posted by Swiftus
I'm not dissuading those that prefer to kaizen the windshield or those hoping to replace it with a force field or a beam axle, simply that you'd better have a solid understanding from a decent design process either way.
I'd simply suggest that being hellbent on viewing human nature as a problem best replaced with quantitative metrics won't broadly make for successful or otherwise happy individuals in a professional or personal sense. There are very few roles in life that can equitably be undertaken with complete disengagement from subjective factors - certainly not in engineering and not even in racing (which this isn't).Quote:
Originally Posted by JulianH
(That's the argument that goes a bit like 'write me an equation for the best way to put that bike around that corner. Excellent. Now go speak to Garry McCoy.')
The best way to decrease variance is to increase sample size; more opinions need to be considered in a structured manner. In short some better process on what goes on behind closed doors post-judging could help. I would stress that for the two competitions I've been involved in, the conversations are very positive and couched in terms of the competitor's best interests.
Let's not be too quick to confuse presentation with communication. Communication is something engineers need to deal with every day, and something we should not have students shy away from. Engineering is, in part, an exercise in communication: to take qualitative wants and needs, to abstract and delivery a quantitatively-specified solution, and to use both domains to communicate it's competitive worth.Quote:
Originally Posted by Swiftus
That's part and parcel of engineering. If you don't like it, you're in the wrong industry.
+1Quote:
Originally Posted by DougMilliken
It is judged but this is not structured (FSAE-A) in a manner where judgements from other design areas are considered in context with respect to integration.
We can't replace Carroll Smith and shouldn't seek to. We need a better process.
OK, here I go on another philosophical bent.
My "Reasoning..." thread is all about taking a concept (the FSAE project) and breaking it down into its parts. It is about the design process. The core argument is that you do NOT start with the parts and work backwards.
The FSAE Design Event is about assessing the teams' understandings of design process.
So why do we base our judging process on car parts? We have a composites judge, chassis judge, suspension judge, etc etc
Why not step further back up the process and have judging teams based on the design process:
Problem framing
Concept development and selection
Analysis and validation
Manufacturing and delivery
Performance
It just seems a bit ridiculous to say that this event is so much more than making car parts, but we then judge and allocate scores in categories defined by car parts...
All bents welcome, BB. We are a modern society here at FSAE.com...Quote:
Originally Posted by Big Bird
I'd suggest we judge by system (in aero we certainly do) not by parts alone, and check/judge for a solid design cycle. Effectively we do as you suggest.
FSAE has always been an exercise in giving rope - some student groups will hang themselves with it, some learn how to fish with it, some build Egyptian pyramids. To this end we give a lot of rope in how students go about design management. It similarly amuses me that some students through faculty in the competition put significant effort in sharing CAD for complete system designs with a view to implementation (and some actual implementation - we used to call this "plagiarism")... when what they should be more interested in is in understanding design management processes that lead to what answer they're seeking. Fishing: not fish.
We could specify, for the sake of clarity, some further parameters to this end in the rubric. I've sated above that PDCA cycle evidence should be shown throughout.
I would be a little wary of seeing 'best answers' copied team-to-team. I'm all for information sharing, less for the process becoming formulaic to the point of being a formality. Thoughts?
DESIGN EVENT - WORTH IT, OR NOT?
=================================
I posted some of my views on this in the Oz-14 Competition thread. I expand on this same theme here.
So far there has been some discussion about the "big-picture" of DE and how to improve it, which is good. But the biggest-picture issue to be discussed when trying to improve anything is "Do we need it AT ALL?". This has been barely touched upon. It needs serious consideration.
~o0o~
To start with an analogy, consider the much beloved Push/PullRods&Rockers of FSAE suspensions. For 25+ years FSAE students have started their year with the question "How are we going to improve our PPR&Rs?". It is only recently that significant numbers of Teams have realised that the best way to make such an improvement is to ... TOSS THEM! For example, at Oz-14 five out of twenty+ Teams did just that, switching to Direct-Acting-Spring-Dampers. They placed 1, 2, 3, ... 5, 6 overall.
Note that DASDs are NOT a big advantage in FSAE, nor are PPR&Rs a big disadvantage (eg. 4th place in Oz-14 went to a Team with virtually NO suspension movement at all, so NO need for either type of spring actuation). Rather, the big advantage of abandoning PPR&Rs most probably came from the LESS TIME WASTED on something that has negligible influence on the car's performance.
Specifically, the quantifiable influence of PPR&Rs on competition points is mostly many small negatives, such as a bit more unwanted mass, compliance, friction, cost, time-to-make, etc., together with no real positives that make the car "faster". But the really BIG NEGATIVE is the amount of TIME that Suspension-Guy wastes designing the PPR&Rs so as to optimise MRs, optimise rising-or-falling-rates, optimise load-paths-to-yet-another-frame-node, and so on.
All this wasted time is a BIG DISTRACTION from building a better car.
~o0o~
Similarly, since shortly after I became aware of this competition (ie. late 1990s) I have been convinced that Design Event is a big distraction to any Team trying to build a better car. It is also a distraction to the education of the students.
DE is undoubtedly a huge distraction to the students, as is evident simply by reading these Forums. Too many students see "winning on the track" as rather hard, so instead they aim to do well in DE. They do so because DE is undoubtedly seen as the MOST PRESTIGIOUS of the Events. The students see DE as the epitome of what "real" Engineers do. By comparison, they see "winning on the track" as too dependent on "chance", or "driver skill", or "having better spanner-monkeys". Which, most students think, is NOT real Engineering.
The overarching goal of most FSAE students is to come out of it with something impressive on their Curricula Vitae. To them "Design Finalist" screams "I am a genius, hire me!" On the other hand, "1st in Enduro, 2nd in Fuel, etc.", just mumbles "I are a good spanner-monkey...".
So, sometime during beginning-of-year Concept Meetings, the students decide to build a "mini-F1" car ... with pure carbotanium PPR&Rs! "Ahhh, yes! The DJs WILL ABSOLUTELY LOVE THAT!!!"
Or, perhaps because their Supervisor keeps stressing KISS, they decide to only build a "mini-F3" car ... with merely billet-machined 6061 Rockers. "Err, well, the DJs will LOVE HOW PRACTICAL we are!"
And, yes indeed, the DJs do love that sort of thing, as has been shown historically ... even though such PPR&Rs are UTTERLY USELESS! Some DJs may claim to encourage KISS, but there have certainly been enough DJs encouraging the above sort of bling over the last 25+ years that the majority of Teams have gone that way. Just look at the cars.
In short, the negative "COST" of DE is that it locks the competition into a loop of "build a mini-F1/F3 car, but try to give it a bit more bling than last year". I am sure that many DJs reading this will disagree. But the historical record shows that the above is, in fact, true. Just look at the cars.
IMPORTANTLY, note that it is possible to build a VERY SIMPLE CAR that is MUCH, MUCH FASTER than anything currently out there. But no sign of this car yet. And certainly NO indication, whatsoever, that DE will lead the students to this car.
~o0o~
The above are the negative "COSTS", but what are the positive "BENEFITS" of DE, if any?
1. One main claimed benefit is that the "design-review feedback of your car" helps produce better educated students. As covered at length elsewhere, this feedback mostly DOES NOT HAPPEN AT ALL, or at most is of a very feeble nature. When good feedback of a Team's efforts is given, it is very often OUTSIDE of the DE proper. And often completely outside of any competition, such as the feedback given to questions and build-blogs on this Forum.
DE is NOT necessary for educational feedback.
2. Another possible benefit is that the DE gives the students an opportunity to practice their "job interview" skills. Or, turning the same thing around, it gives the "Industry Experts", who have "volunteered their valuable time to be DJs", an opportunity to spot talented students who they can employ. Either way, exactly the same results can be had by holding a "Careers Fair" alongside the competition.
DE is NOT necessary for "talent spotting".
3. The only real benefit I see (which makes it a bit left-field) is that the "distraction of DE" teaches the students THE IMPORTANCE OF NOT BEING EASILY DISTRACTED! The students who learn this lesson realise that to do well in the competition overall, they must FOCUS ON FAST CAR, and BLOCK-OUT DE DISTRACTIONS. (As an example, when coaching young Rugby League players I used to make sure they were under maximum distractive pressure when executing critical plays. Things like name-calling, shouting, throwing balls at their heads, etc., worked well... :))
Following this line of thinking, DE could be improved by adding more cunningly deceitful distractions. Perhaps a shiny DE trophy for "Most Spectacular CF Wishbone Failure During Brake-Test". "Yeayyy!!! We've won it three years running!"
This sort of change to DE might be fun, but is NOT necessary.
Much more that can be said, but in summary I see NO GOOD REASONS for having DE. The negative costs outweigh any positive benefits.
~o0o~
What might a competition look like without Design Event?
Let's start with,
"A1.1 Formula SAE Competition Objective
... to conceive, design, fabricate, develop and compete with small, formula style, vehicles...
... [that] should have very high performance in terms of acceleration, braking and handling...
... [and also have good] cost, ergonomics, maintainability, manufacturability, and reliability..."
This was put much more succinctly in past versions of the Rules, but the gist is that the car should be,
1. Cheap to build,
2. Cheap to run,
3. Be very fast on track.
All these objectives are easily measured objectively. Namely, with a stop-watch for the main performance factors, with a calender for reliability, and with a competent beancounter for cost, manufacturability, etc. A specific "Design Event" is not required to evaluate any of these.
So, one option is to simply roll DE into the Cost Event. Currently, the "real cost" of the car is only worth something like 40 points (out of 1,000). Make "real cost" (independently audited, if necessary) worth 200 points, with the other current Cost song-and-dances worth 50 points. Presentation can stay as the main "subjectively judged" event at 75 points.
Or maybe,
Presentation = 100 points, which includes a "The Design of Our Car" song-and-dance,
Cost = 200 points, with "real cost" worth at least 150 points,
Dynamic Events = 700 points, with Fuel ECONOMY worth a full 100 points rather than the currently devalued 40 points.
What would students who want a killer-CV do now?
I see broadly three options.
1. First year, low-budget Team. - Build a really cheap car with super-thrifty engine. That wins 300 points for a start, then right-foot-flat-to-the-floor in Dynamic Events.
2. Established, big-budget Team. - Use mega-power-engine, mega-tyres, and mega-aero, to win ALL those 600 "speed" points, plus slick Presentation for another 100, and be "creative" in Cost.
3. Clever Team. - Build "brown-go-kart-with-aero-undertray" and clean up 900 points (ie. Cost + ALL Dynamic Events). Get "Weird-Guy" to do Presentation.
~o0o~
Summing up, a competition with NO Design Event is ALL about meeting the stated objectives of the competition (ie. A1.1). Any students wanting good CVs have to build a car that is CHEAPER & FASTER THAN EVER. And this is possible for a long time to come yet!
In contrast, for the last ~25 years the students have been trying to "impress the DJs" with their bling PPR&Rs, with their MRs that are spot-on 1:1, with their RCs that "don't migrate hardly at all", and with countless other meaningless drivel. The end result is the same-old EXPENSIVE & SLOW cars, year after year after year...
Oh, and we now have a generation of students who, in too many cases, can't solve simple FBDs or make neat little brackets, so are unlikely to make worthwhile contributions to society in the future.
Z
No disagreement on cheaper and faster (I'd add there's more to it than that); DE is simply about explaining how and why it was done. If your design is solid, the event should be a breeze.Quote:
Originally Posted by Z
The competition is not a race competition, it is (not the first time this has been explained to you) about students best delivering a complex project in a group environment.
I'll limit observations to the last 15 years which cover the FSAE-A competitions.Quote:
Originally Posted by Z
Most students I interact with - including my own experiences as a FSAE student - certainly didn't involve trying to break our backs and the university's patience for a year to impress a few people we'd never met with something that vaguely resembled a race car. The vast majority start out with exactly what you hope for in mind, Z - to build it fast - and go from there.
You seem to underestimate the ability of students given responsibility, build resources, money and the like - but very little accountability - to commit to some fairly ill-thought-out choices in design. Many come to remark ruefully on this in later years. The gap between what was and what could have been... has bugger all to do with design event. Very few teams turned up this year at all prepared for it. Same as last year. Same as every year.
Those that asked for a run through prior to comp didn't stuff up because they hadn't put effort into DE preparation, they stuffed up because they'd made some questionable design decisions that was shown up in their design. They didn't trip up on whims, they tripped up on crap decisions that were reflected in the project. Same as last year, same as every year. I've heard some brilliantly comical design decisions; I was even party to a few myself. The student that massively simplified his engine simulations by assuming air to be incompressible - thus indirectly intending the competition's first three-stroke engine - takes some particular beating. I'm sure everyone's got stories to share.
DE exists to examine the thought. Dynamic events exist to examine the result. It's a not correct to suggest the most brilliant designs are immediately the fastest; this cannot be completely judged on track.
None of this has anything to do with any time spent preparing for Design Event. If the event runs to its' intent and the students to theirs, then good work comes through.
I appreciate that you'd prefer good work to show up on track, but again you overstate the ability of students and underestimate the scope of the work they have to do: there are always gaps because putting together a project capable of being reliably judged in a qualitative environment is intentionally beyond project resources. The cars are never perfect or bulletproof reliable. It'd be wonderful if they were and they're not supposed to be. DE offers another chance to shine.
This is a significant reason to need DE.
I'm sure you'd build a highly-competitive FSAE car, however you're not a student.
Not the fault of the DE.Quote:
Originally Posted by Z
I'd disagree completely - your contention doesn't explain why entrants that do well on dynamic events to compete well in DE too. Many students sadly see DE as 'something they have to do'. Which is often reflected in any critical rigour applied to their actual designs; it's easy to get excited about building something, it's hard work to be critical about doing it well. Students entering the competition to do poorly in dynamic events do not meet the competition objectives, and accordingly do poorly.Quote:
Originally Posted by Z
You seem particularly hung up on it being a battle between one or the other. It's completely possible to have good DE and dynamic event performance. Many manage it.
You've a particularly low opinion of DE judges. If you think you can do better, volunteer. That's suggested openly/proactively.
I'd love to have a careers fair at the event, though it's not what DE is for. It's a nice bit to network through.Quote:
Originally Posted by Z
Agreed wholeheartedly - however you're misplaced in suggesting it's for the DE's to lead it there.Quote:
Originally Posted by Z
You've been posting for years at significant detail - it'd take people taking what you suggest that seriously to do it.
Not a new problem in education.
Could part of the answer be as simple as releasing a marking schedule (with comments from section leaders) now-ish, then after the DJ's have a GoToMeeting in October-ish release any changes? This would give students time to see how they will be assessed and prepare for the event.
Judging based on the design processes and systems is a good idea but it is possible to know the process and still have sucky designs. There still needs to be an amount of component level evaluation to assess the successful application of the processes. I like to think of FSAE as a design assignment with a built in bullshit filter that weeds out poor design. We have all seen the great A+ assignment handed in and gone "there is no way that would ever work" but the had a great project log book so here is an A+.
My favourite questions I was asked in design were: "What were your biggest compromises?", "What do you hate, why and how would you change it?", the filpside "What do you love and why?" and "Why is your car better - and not in numbers?". I asked the Auckland guys these each year at their design event practices, and the results are interesting as they often aren't aware of their compromises.
If you dump the design event what checks are there to prevent blind copying of wining designs? A team could see that 5 out of the top 6 teams ran direct acting shocks and decide to just use them since team x does. There would be no need for the team to understand why they might be a better choice or even worse, why they might not be better.
Over a short time innovation will suffer, a few teams will develop a wining design and everyone else will just copy it. Before you know it there will be a paddock of cookie cutter brown go carts and the event will become only a build quality and driver event. The notion of a simple car in the real world has changed and there are a lot of people holding onto the idea that a simple car has no technology from the last 15 years. A basic Hyundai/Mazda/Toyota/etc now comes with ASB, ESP, Air Con, bluetooth, power windows, 50 air bags, keyless entry, etc, so why should FSAE cars not have composite parts, traction control, trick 3d printed parts and the rest. If there is no real points reward for trying something risky or new why bother, just improve what team x uses? FSAE will just become another motorsport event.
Z, you might see being able to justify your design decisions as a distraction from turning a spanner but it is something that will be a distraction throughout their working lives when getting funding or a design approved. Being able to clearly get your ideas across is a very important skill to have and is used in personal and professional life. Being able to make what they design is the difference between a FSAE grad and just another engineering grad. But given less FSAE grads end in up in jobs where they will be making their own design vs jobs where they must get someone else to make it, communication is a part of the job like a doctor needs a good bedside manner.
IMO the DE is a way to explain your top-level decisions and the work done to get into detailed component design and integration into a competitive car. It is therefore (to me at least), something like the engineering discussions we all enjoy so much within these forums. Z, when you are describing your "soft-twist" concept or any other concept for that matter, you are basically presenting in a "DE", with people asking questions and arguing being the DJ's. IMO this is really valuable; do not forget that usually there's a strong correlation between DE points and dynamic performance, at least at European competitions. One main flaw I have noted though, is sometimes the background and knowledge field of the DJ's. If you have an e-powertrain DJ that designs electric motors and you have just picked one off-the-shelf, in my experience you are pretty much screwed; you have nothing to present on his field of interest and he might not be that interested on the rest of the powertrain. Another huge problem is lack of understanding/knowledge of the competition rules, limitations and objectives; to add on that, some of the most reasonable DJs we had (and the ones that usually spend lots of time giving feedback) are ex-fsae team members.
Yup. And it ain't motorsport. Never should be, never will be.Quote:
Originally Posted by Moke
Correct. Check that earlier word: communication. And it gets far more complex than getting someone to build your vision.Quote:
Originally Posted by Moke
Yes, yes, yes, and yes, a strong correlation in FSAE-A too.Quote:
Originally Posted by mech5496
I'd not recommend thinking professional life should be any different - don't ever bet on a perfect audience. Communicate.Quote:
Originally Posted by mech5496