2014 FSAE-Australasia

[Kevin Hayward]

Christian,

Main limiting factor was the exhaust manifold. Very close to the gearbox. Could have angled the head up, but theguys were trying to keep machining simple. Hard to run through all the engine design iteration conducted over a number of years. Lots of compromise everywhere with the powertrain. Whenever something looks like we did something a bit odd the reason is usually to do with ease of manufacture, ease of rebuilding and chasing reliability.

Z,

If both wheels can go forward by 150mm some very cool packaging ideas can be investigated. Guys are working on it.none of the things we talk about in the workshop is that simplicity is not easy. Tokyo Denki had a couple of cool ideas that would work well with achieving that.

I think in a few years time we could be seeing a big change in how simple and small these cars become. Enough teams are playing around with the ideas. If a few of the best teams bite and give it a go we could see the end of the mini-f3 cars.

Kev

[mech5496]

Take a look at my post at "beams" tread...driver can be within the WB almost entirely and with reasonably low GC and inertia. I had a similar assembly with a rearward tilted JAWA somewhere...

[Z]

Quick note during mid-Xmas-chaos...
~o0o~

Tim,

Anyone who's ever done any real testing can tell you that trusting laptime alone is a far too simplified way of assessing vehicle performance as a whole.

Unless you think driver skill and weather conditions are legitimate parts of a vehicle's design...

But "anyone who's ever done done any real testing" knows that they MUST ACCOUNT FOR the different levels of "driver skill and weather conditions", otherwise NO POINT DOING THE TESTING!

Fair comparisons of this year's and last year's cars are possible...
~o0o~

Brent (Moke),

So UWA cocked up with their "most innovative suspension design"...
[example compares innovative-cock-up vs polished-faster car] ...
... slight improvement beats innovation.

Yes! I would prefer to see cars with constantly IMPROVING performance score highly in Design, rather than cars that are subjectively assessed (?) to be INNOVATIVE. This because I believe most H. Sapiens are utterly incapable of recognising real innovation. (If they were, then we would see much more of it. More on this later in Geoff's new thread in the "General" section of the Forum...)

However, it is well known that constant polishing suffers from the Law of Diminishing Returns. So occasional "innovative leaps" must be made if you want constant improvement.

I looked back at our 2005/2006 car and the COG was 180mm with smaller wings and, this will make you somewhat happy Z, 60% rear. Didn't win anything at comp.

The lower the CG, the more R% required!

This hinted at (spelt out?) in 2005 post on Gordon Murray's BT55 and MP4/4. More details later.
~o0o~

Harry,

They're coming...
~o0o~

Originally by Paracelsus, about 500 years ago:
"The dose maketh the poison."

Anyone know how much chocolate maketh it poisonous? ... I don't feel well ...

Z

[GTS]

As a few people offline have asked that I respond...

The competition weekend is not important and the results ultimately irrelevant.

What is important and relevant is the 51 weeks leading up to that one weekend. That is when the design and innovation is done, that is when little baby engineers learn what they need to become useful members of the world.

Correct. Glad someone gets the point of why the competition exists, and why universities co-fund it. You're welcome to try the "Dear Prof. University Faculty Head, we need a six-figure sum of money to build a faster car than last year's". Let me know if you do, I could use a good chuckle.

Z; While I appreciate and respect that you like to shit stir and challenge the status quo, your attacks on the creditably and personalities of the volunteers is way out of line.

In the spirit of sportsmanship - which all involved in the competition are encouraged to follow - also wise advice.

GTS,

Details?

No, it'd be a waste of time.

A student on another thread has just asked what sort of differential he should buy. Pat Clarke, a long time DJ, has just now advised the student that regardless of what diff he buys, he will have to "defend his choice to the Design Judges", or words to that effect.

(blah blah hysterical crap later)

That's right - Pat has just advised a student to think for themselves, to research as much leading towards a decision that they will own, and to make the process sufficiently robust to explain to others. Pat has directed the student to any answer other than a formulaic response at best and 'because Pat Clarke said so' at worst. In directing the student to consider their engineering choices, Pat's probably given the student the best advice possible. Asking 'Pat, what might I consider in choosing a differential - here's some stuff I've already researched, I'm a little unsure of these issues, etc' might elicit a different answer.

Let's assume that the intent of the Design Event is to encourage the students to bring cars each year that are better designed than last year's.

Let's not, because your assumption is incorrect. It's not what it (or the competition) is primarily for.

But "anyone who's ever done done any real testing" knows that they MUST ACCOUNT FOR the different levels of "driver skill and weather conditions", otherwise NO POINT DOING THE TESTING!

Fair comparisons of this year's and last year's cars are possible...

This works if vastly underestimating the complexity of this as a field of study and what compounds it. Doing this reliably is significantly beyond undergraduate study for many competing teams. Noise factors in a FSAE complicate matters further, and limited repeatable running creates further variance.

Unsurprisingly, there are reasons why this isn't something the competition's structured around.

I believe most H. Sapiens are utterly incapable of recognising real innovation.

For a keyboard warrior so critical of those encouraging others to challenge and find confidence in their own beliefs/concepts/designs, you're quite keen to promote your own ideologies. With an amazingly low awareness for what these statements can do in a discussion forum (supposedly) dedicated to a constructive, highly-formative, learning activity.

As in you really have no f**king clue.

Some students actually read your bile and take it literally... then when some of us (DJ's or otherwise) attend a university, giving time with departing and incumbent students in this program to help best structure their efforts moving forwards, there's a lot of work to do to restore confidence in thinking independently again... as three of us experienced recently, with this very thread - and your comments - directly quoted.

If you think there's that much merit in your point of view and delivery, volunteer. Apply for judging, join the rules committee, wreck the joint if you see fit. There's always room for professionals with the students best interests at heart, if you fit that bill.

And BTW, GTS, I am considerably your "senior". But I am used to youngsters showing very little respect for their elders these days...

No, you're just older. Respect is earned.

I'm out - this thread (including my own involvement) isn't taking a constructive direction.

Merry Christmas all.

[Pat Clarke]

GTS, read your PMs

Pat

[bob.paasch]

Yes! I would prefer to see cars with constantly IMPROVING performance score highly in Design, rather than cars that are subjectively assessed (?) to be INNOVATIVE. This because I believe most H. Sapiens are utterly incapable of recognising real innovation. (If they were, then we would see much more of it. More on this later in Geoff's new thread in the "General" section of the Forum...)

For a keyboard warrior so critical of those encouraging others to challenge and find confidence in their own beliefs/concepts/designs, you're quite keen to promote your own ideologies. With an amazingly low awareness for what these statements can do in a discussion forum (supposedly) dedicated to a constructive, highly-formative, learning activity.

As in you really have no f**king clue.

Actually, you're both wrong, but GTS is more wrong. H. Sapiens are pretty good at recognizing conventional verses "creative" (or non-conventional) ideas. But within the category of "creative" ideas there is little agreement amongst even "experts" as to what is a "good" idea verses a "bad" idea. So the idea of assessing "innovation" in the FormulaSAE design event and rewarding such (implicitly, the "good" ideas) falls flat on its face.

In the early 1990's I was fortunate to attend a two week workshop on "Creativity in Engineering Design" at Stanford University, led by Design School faculty Bernie Roth, Doug Wilde and Rolf Faste. The workshop remains one of the highlights of my academic career. Rolf and I had a very nice side discussion on evaluating innovative design, and he pointed me to a paper he'd recently written on the subject. I urge forum members, and especially the design judges, to read it themselves: http://http://fastefoundation.org/pu...oved_model.pdf

[Scott Monash]

What a great link, thanks for that Bob.

I think they hit the nail on the head, with their creativity and innovation model.
This framework neatly accounts for the similarity and difference between useful and not useful innovative ideas, and explains why you can't have one without the other.

This is why I think that innovation should not be rewarded for its own sake, particularly at the FSAE level.
Successful innovation should be its own reward.
Take inspiration from the best ideas or solutions around, explore them, understand them and their weaknesses, and then if (and only if) you see potential benefits to innovating, then do so.
Do not think that different is always better, its not.
Do not expect a hand out just because something is different, particularly if you cant demonstrate why it needed to be.
This kind of thinking is probably an artefact of the "copying culture" that the article discusses and the workshop tries to break down, as a key inhibitor to the cooperative design process.
Have a health amount of respect for what has come before, you are not the first engineer to walk the earth.

Teams that state that their primary mission in FSAE is to "be the most innovative" scare the hell out of me.
Completely abandoning convention invites multi-faceted disaster.
In my opinion, the path to success lies in maintaining a delicate and measured tension between convention and innovation.
And not fooling yourself.
We are engineers, let the data be your guide.

The related article on the workshops they ran is even more interesting and relevant:

http://www.haakonfaste.com/fastefoun...curriculum.pdf

If there are any materials from those workshops that you could share with me Bob, I would really appreciate it.
Are the authors still active in design teaching (this was all published some time back)?
Are the workshops still running?

It is also good to see that a lot of what they are talking about has slowly, over time, become best practice in the broader education sphere, not just engineering (even though not that many engineering academics are aware of these developments just yet, the inertia is great due to learned teaching techniques as stated in the article). I am personally working hard to push the flipped classroom teaching model, cooperative learning, peer-assisted teaching, heterogeneous team selection and peer assessment for group work in all of my design subjects and more broadly within my faculty. Its still early days but we are slowly making progress, much of which is driven by student feedback and unit review scores.

Students, if you are lucky enough to experience some of these techniques (applied well) and feel that they are more beneficial to your learning, please make that fact known. It takes take but universities are listening.

For others interested in this field (and it is very relevant to FSAE teams) I would also recommend the work of Felder and Brent (their Cooperative Learning Workshop was the highlight of my academic career), and Prince.

I have a lot more to reply to, with respect to discussions further back in this thread but that is going to take some time so I will leave it till I am back in the office.

Cheers,

Scott

[Z]

GTS,

Respect is earned.

Yes, I was going to mention that earlier.

I start people off at a neutral "respect" level of ZERO, after which their actions move them up or down.

So far I have overlooked your emotional writing and profanity because it is so common amongst your younger generation.

But your complete lack of interest in discussing DETAILS has moved you down a few notches.
~o0o~

Bob,

But within the category of "creative" ideas there is little agreement amongst even "experts" as to what is a "good" idea verses a "bad" idea. So the idea of assessing "innovation" in the FormulaSAE design event and rewarding such (implicitly, the "good" ideas) falls flat on its face.

Agreed! That was EXACTLY MY POINT!

(Errr..., maybe too many "http"s in your link, so http://fastefoundation.org/publicati...oved_model.pdf)

Z

[Z]

WHY MORE R%?
===============

As noted earlier, IMO most cars in FS/FSAE have TOO MUCH REAR-WHEELSPIN coming out of the slow and medium speed corners.

Some might ask, "Isn't that what they're supposed to do, given that the comp is about going as fast as possible, in order to score maximum points?"

My answer is "No, you don't score more points by spinning wheels. Just the opposite!".

So, below are some reasons why you can score MORE POINTS by having LESS WHEELSPIN.

Following reasons are roughly in reverse order of importance.
~~~o0o~~~

REASON 4. FUEL ECONOMY - Mega-wheelspin out of corners = much fuel wastefully injected into the engine for the mostly USELESS purpose of tearing little bits of rubber off the rear-tyres.

A typical 13" tyre rolled, with NO longitudinal-slip, around the 1,200 metre Oz-14 track would do about 730 revolutions. A 10" tyre would do about 850 revs. Of course, some slip always happens whenever the tyre is forcefully pushing the car forwards. On the other hand, the reverse slip from braking reduces the total number of revs. Due to entropy the number of wheel revolutions required for a good "fast lap" will be somewhat higher than the above-quoted numbers. So, maybe a MAXIMUM of 850 revs for the 13" tyres and 950 revs for the 10"s?

I would be interested if any DAQ-Guys could post their numbers for rear-wheel-revs/lap, in Oz-14 AutoX and Enduro. (Think of it as a contest to find the wheelspin champs!) Some simple calcs would then show how much fuel was wasted on unnecessary wheelspin, and how many FE points lost.
~~~o0o~~~

REASON 3. SLIDING MU IS LESS THAN STATIC MU ("Mu" = "Coefficient-of-Friction") - As all you TTCers know, the tyre longitudinal-slip-ratio curves climb steeply to a peak Fx at a lowish "Slip-Ratio" (however you choose to define "SR"), but then usually drop to lower values of Fx at higher SRs. This drop-off can be quite significant, often more than 10%. (Any Tyre-Gurus care to share their numbers, taken from the car's DAQ systems?)

It should be fairly obvious that to get around the track as fast as possible, the car should maximise its longitudinal and lateral accelerations. It follows that the tyres should operate close to their Fx and Fy peaks. So mega-rear-wheel-Slip-Ratio = not good.
~~~o0o~~~

REASON 2. BETTER CONTROL - Auckland's car was the only one I noticed where the driver was able to "keep the boot in" as the rear-tyres lit up. Even so, the driver had to take a cautious line around the bends because of the lesser directional control from the rapidly spinning rears (ie. Fy drops together with Fx, as in point 3 above, so less sideways control). Most of the other cars had the driver rapidly backing off the throttle as soon as the rears broke free, giving jerky forward progress.

I am guessing that some cars (eg. Monash?) had reasonably effective Traction-Control, which did a smoother job of "backing off the throttle" than the other drivers could manage. This TC would also have been effective in preventing "wasted fuel", as above. But note that while turning the power down can save fuel, keep the Mu high, and help with car control, it does not actually make the car faster. TC simply lessens the bad effects of UNUSABLE power.
~~~o0o~~~

REASON 1. THE BIG ONE - To a first approximation (ie. assuming the CG is at ground level, for now) the maximum possible forward acceleration of the car is directly proportional to R%!

So, MORE R% = MORE ACCELERATION out of all those low to medium speed corners!

In equation form the maximum possible forwards force, Fx, capable of accelerating the car is,

Fx(max-from-rear-wheels) = Tyre-Mu x Rear-Fraction x Weight-of-Car,
(where RF = R%/100, and Weight-of-Car = Mass-of-Car x G).

Putting this in terms of "G" units of acceleration, gives,

Maximum-Longitudinal-G = Mu x RF. Simple as that!

So, RF is a FIRST-ORDER performance factor, and should be near the top-of-the-list of any Team's "How Can We Go Faster" discussions.
~o0o~

Looking at this in a bit more detail, we include the "Rear-Load-Transfer" term (ie. second part in brackets below) due to Inertia acting on a CG that is above ground. With CG height H, on a wheelbase L, the above equation becomes,

Max-G = Mu x (RF + (G x H)/L).

Putting this in less self-referential form,

Max-G = RF/(1/Mu - H/L).

Using this last equation, it is worthwhile to take several different levels of Mu, say a highish "peak" value and a lower "wheelspin" value, then for each of these do a surface plot of Max-G (vertical-axis) against different values of RF and H/L (on the two horizontal axes). Two things should be apparent.
1. The lower the car's CG, then the more RF it needs for good acceleration.
2. Cars at the low-CG/high-RF end of the map have relatively less change in Max-G levels at different Mus, compared with cars with high-CG/low-RF. In other words, low-CG/high-RF cars accelerate faster out of the corners even when their rear-tyres "light-up". Conversely, the low-RF cars just sit their smoking their tyres.
~~~o0o~~~

(More coming ...)

Z

[Z]

Finally, not "Why?", but ...

REASON 0. WHEN IS "MORE R%" NEEDED?
===============================

All of the above reasons are only relevant when the car is "traction-limited". That is, when the car has more than enough power to light up its tyres, but not enough traction to make good use of that power. So, when is that?

The calculations required to find the dividing line between "traction-limited" and "power-limited" are extremely simple, namely,

Mechanical-Power = Force * Velocity,
(where F, V = vectors, * = dot/scalar-product, and thus Power = scalar).

In agricultural circles this Power is usually called the "drawbar-power", and the Force is the "drawbar-force". So the Force in above equation is that exerted by the ground to accelerate the car forwards (or to pull a plough). From the previous post, and taking all vectors to be longitudinal, we get,

Power = Mu x RF(dyn) x Mass-of-Car x G x Velocity,
(where RF(dyn) = Rear-Fraction INCLUDING the dynamic-rear-load-transfer term).
~o0o~

Let's do some rough comparisons of different types of racecar to make sense of the above equation. We seek the "VelocityT/PL" at which the car switches from Traction-Limited to Power-Limited.

To simplify things, let's assume that in all examples below,
Mu = 3/2 (ie. 1.5),
RF(dyn) = 2/3 (eg. like a car that has static 50F:50R, but during hard acceleration ends up with 67%R),
so Mu x RF(dyn) = 3/2 x 2/3 = 1,
and G = 10 m/s.s.

With these simplifications, VelocityT/PL = Power/(Mass-of-Car x 10).
~o0o~

EXAMPLE 1. FORMULA-VEE - Cheapest formula racing series ever, so naturally low powered, and using heavy-ish VW-Beetle parts. These cars race on conventional circuits, so spend most of their time between 100 kph (~30 m/s) and a top speed of about 200kph (~60 m/s).
Power = ~50 kW (at MOST!, actually closer to 50 hp),
Mass = ~500 kg,
so, VelocityT/PL = 50,000/(500 x 10) = 10 m/s (36 kph, at most).

So, only traction-limited off the start line (ie. for ~one second), then power-limited everywhere else. Hence they manage fine with about 50-55%R, close to equal sized tyres F&R, and NO NEED for more rear grip.


~o0o~

EXAMPLE 2. FORMULA-5000 - Inexpensive but fast formula racing series from ~1970s, with F1 levels of power, or more, from hotted-up, 5 litre, stock-block V8s, and very few restrictive Rules. Raced on the same circuits as F-Vee and F1, so mostly speeds of 100 kph to 300+ kph (~30 - 90 m/s).
Power = ~420 kW,
Mass = ~700 kg,
so, VelocityT/PL = 420,000/(700 x 10) = 60 m/s (216 kph).

So, traction-limited for about half the lap (ie. out of ALL slow and medium speed corners), and power-limited only on the really high speed stuff. So MASSIVE NEED FOR REAR GRIP! So 60-70%R and monster rear slicks...


~o0o~

EXAMPLE 3. FORMULA-SAE - Student Engineering competition, nominally based on "autocross" racing, so conducted on very twisty tracks to maintain low speeds, and thus better overall safety. Rules mandate track-design such that average speeds are about 50 - 60 kph (~15 m/s) and top speeds below about 100 kph (~30 m/s).
Power = ~60 kW,
Mass = ~300 kg,
so, VelocityT/PL = 60,000/(300 x 10) = 20 m/s (72 kph).
(Note that a lightweight car, 200 kg total, with a cracker of a 60 kW engine, can have Vt/pl = 60,000/2000 = 30 m/s (108 kph)!)

So, TRACTION-LIMITED FOR MAJORITY OF THE LAP. Or, at the very least, TOO MUCH WHEELSPIN coming out of all the slow to medium speed corners. And note that in the two preceding examples, aero-drag lowers Vt/pl because of the much higher speeds of those cars, so power is lost to aero-drag, but in FSAE the lower overall speeds lessen this aero influence.

Bottom line, find photos elsewhere, or just look at your own car, but ...,
... all your cars NEED MORE REAR GRIP!
~o0o~

FINAL CASE STUDY: SWINBURNE (#E17) - The Swinnians were kind enough to explain to this "old caveman who had only mastered fire" how they had managed to "hand-cuff lightning, and put it in a bottle", or some such. And something about "the benefits of learning how to integrate complex electro-mechanical systems ...", and, err..., other marketing buzz-speak ... which I have forgotten.

Nevertheless, they had a headline power figure of 70 kW, and a car (~240 kg) plus driver Mass that could have been under 300 kg total. So, comparing these figures with all the other cars at Oz-14, they SHOULD have had a lightning quick car.

However, by the above reckoning and their quoted figure of 55%R, with 70 kW available they would have been traction-limited around most of the track. So they would have been EVEN FASTER if they had more rear grip from more R%.

BUT (why!!!???) for reasons that are beyond this caveman, they geared their 2 x OVERSIZED Emrax 228 motors so that the notional "70 kW" is only available at the very top speed of the car, about 110 kph (ie. the motors are direct-drive). Even more baffling, is that this gearing puts the motors in a rather INEFFICIENT part of their efficiency map, for most of the car's track time (ie. spends most time at <90%, cf. peak = ~96%). Even worse, the torque available from this gearing CANNOT give good acceleration, maybe only 0.8 G maximum... (I suspect an irrational, ideological belief that "E-Car MUST = direct-drive"?)

Soooo..., speaking only as a caveman ... if Swinnie want to stay with this overall concept, then I suggest two smaller (and lighter!) motors, perhaps the Emrax 207s. But gear these ~4:1 so they operate most of the time at their peak efficiency. This gearing then gives an available thrust force (ie. "drawbar-push") of ~5 kN, implying ~1.7 G acceleration, IF the rear tyres are suitably loaded (ie. ~100%R(dyn)). The ease of re-positioning batteries, etc., should make a low-CG/high-RF car very easy to do.

Furthermore, this "more R%" allows smaller front wheels and tyres (eg. <6"x10"), for less mass. Toss the PushRods&Rockers for less mass again. More R% also means more braking-regen-energy can be extracted from the rear-wheels. Perhaps all of it in Enduro, so front brakes are only for scrutineering and AutoX. So along with the lighter motors, a lighter battery pack can be used. And if the Team has any clever software/EEs, then torque-vectoring means that, in principle (*), the steering-linkage also becomes redundant, with directional control as per ZTR lawn-mowers. (* But conventional steering-linkage must be kept to please the scrutineers, and as back-up...).

So, all up, the concept can head in the direction of Ben Bowlby's DeltaWing racecars, albeit with full-width front-track. The DeltaWings have about 72+%R. Despite all the experts crying "It'll never work!!! Too much R%!!! ...", and so on, the "half-power + half-mass = same-speed + less-fuel" concept has actually worked very well indeed. And with the relative ease of implementing torque-vectoring on E-cars, an even more extreme low-CG/high-R% E-car might work even better (eg. power-on-understeer is easily fixed, etc.).

Then..., maybe..., Delft and Zurich might have to look over their shoulders...

But, of course, Caveman-fired "C"-cars will always be faster. Especially if Enduro is made longer than "a quick trip to the corner store for a pint of milk...".

Well reasoned criticism most welcome, especially from Swinnie.

Z