Life, the Universe, and our curious obsession with engines

[Big Bird]

Over the past couple of years I've probably confused the hell out of a lot of people on these forums, more often than not with my own personal philosophies on vehicle design. I've probably seemed a little abrasive towards some members, which certainly has not been my intention (Daniniowa and Garlic are first to mind – sorry gents!). I just think sometimes by challenging the established axioms we can all further our own knowledge a little.

So anyway, rather than keep hijacking other people's threads, I thought I'd open my own whereby you can all have a dig back at me. It's a bit of a thesis – so you'll all have plenty of ammo http://fsae.com/groupee_common/emoticons/icon_smile.gif

My personal philosophy is similar to Z's, in that we tend to overcomplicate our designs, and don't pay enough attention to the stuff that really matters. Where I might stray from Z's philosophy a little is that I think our obsession with engine performance is a little unfounded.

Does more power always make us faster?
I started thinking about this in response to a comment made on another thread, when someone wrote that you can never have too much power. I replied that I disagreed, mainly because in principle I think that such outright statements don't really address the issues of balance we face when we design a car. EVERY benefit comes at some cost – whether it be track speed at another part of the course, risk of failure, development time, financial cost, etc etc. It is our task as engineers to find the right balance, and to be aware of the compromises that relate to our designs.

In relation to the power argument, my assertion is that engine performance comes at the expense of weight. The more power/torque we have, the greater the loads we are placing on our drivetrain and cooling systems, and therefore the more material (read weight) we require to construct them (for a given safety factor). If we accept the premise of tyre load sensitivity, (i.e. that a tyre's coefficient of friction decreases with increasing vertical load), then this increase in weight has reduced our lateral acceleration capability – our cornering speed. (There is a graph in the first chapter of Milliken to visualize the concept of tyre load sensitivity, page 29 – or else if you have ever stiffened up the front / softened the rear to cure oversteer then you are aware of the effect). Effectively more power equals less cornering speed.

So a hypothetical argument. Imagine one team builds a car with a gun CBR motor, puts out 90hp. Everything built as light as it can to suit the power, the thing is a rocketship in a straight line and a damn nice little car. Another team detunes their CBR to 45hp and builds a car of the same geometry. Of course, they can downsize cooling system, driveshafts, CV's, chain and sprockets, diff and diff mounts, maybe rear tyre widths, even braking system to an extent. Weight savings? Maybe 5-10kgs? The thing will be a plug in a straight line, but if tuned properly the lesser weight should make it quicker around corners, (not necessarily much!). And somewhere in the middle there is a combination of straights and corners whereby the two cars will be equivalently as fast as each other. One problem, two solutions.....

Of course the above is a trivial example, but it's a philosophical argument to show that even with the same engine we can have less power and sometimes go faster because of it. (Remember I'm challenging the assertion that more power/torque is always better). A more significant effect would be seen when the second team takes the argument further and replaces their 45hp CBR with an engine that was designed for that power. And suddenly we have saved 30-40kg, and we get the situation where the "slow"� car is starting to hold its own on tracks with longer straights.

I'm acting a bit as the devil's advocate here. But sometimes these mental exercises open up trains of thought that can end up working for you.

Our curious obsession with engine performance
So how much does our engine performance really help us? I get the impression that some engine junkies think that if their engine is 20% more powerful they'll be 20% quicker around the track. A more educated guess might be that if the track is 20% straights, then you will be 20% x 20% = 4% quicker around the track. When you work out the figures, it is not even that.

For the sake of simplicity, we'll consider that the car has constant acceleration. (If you have a reasonably flat torque curve within your operating rev range, it's not a bad assumption). The car's equation of motion down a straight can be calculated through the formula:
x = u.t + 1/2a.t^2
where "x" is the distance, "u" is initial speed (corner exit speed), "a" is the acceleration and "t" is the time taken. From the above we can see that for a given distance:
"¢ There is a term (u.t) that is completely independent of acceleration
"¢ The term containing acceleration is attached to a time squared function – meaning any time saving will be related to the square root of the acceleration.

When we first started doing some research on the first RMIT single car in 2001/2002, we estimated the typical 600/4 might be around 20% better off on power-weight ratio, (given our conservative weight target for the first car). So imagining worst case enduro/autocross scenario for an underpowered car whereby you are leaving a low speed corner (say, 36kmh = 10m/s) and accelerating down a long straight (75 metres max according to FSAE rules, which we'll reduce to 60 metres given braking distance and corner entry and exit transients). We'll estimate slow car's acceleration at 10m/s^2 (just over 1g), and therefore the 20% more powerful car at 12m/s^2.

The equations for the two cars become:
60 = 10t + 5t^2, which solves as t = 2.605sec (slow car)
60 = 10t + 6t^2, which solves as t = 2.437sec (fast car)

The car with 20% more power/weight has in the order of a 6.5% advantage along the straights. For an Oz or FStudent track with only around 15% full throttle time, the effect is in the order of 1% time based over the whole track. If Rotor's throttle pot figures are correct, then maybe for the US track the time advantage stretches out to around 2-2.5%.

Remember the above is worst case, assuming all the straights are long with a stop corner before them, and not taking into account that a lighter car should be able to get out of a corner quicker, thereby diminishing the effect of the 1/2a.t^2 term even further.

Effectively with the first RMIT single we only had to be around 0.2% quicker around the rest of the track for our single to be competitive, and we "bought"� ourselves a 40kg weight saving to do it with.

The guts of it all? The contribution of 20% better engine performance may be as little as 1-2% towards lap times – and that is before it is balanced against the negative effect on lap times due to the extra weight. And if your engine development team is only delivering a couple of horsepower here and there, you have to wonder whether it is all worthwhile.

As a real-life validity check of the estimates, in 2003 the best 600/4's had around 20% power to weight advantage over the first RMIT single. In the Acceleration event, the "u.t" term in the equation above disappears, so the time taken should be directly related to the square root of the acceleration, i.e around 10% quicker for a car of around 20% better power to weight. UQ were doing around 4.0/4.1 seconds, whereas RMIT was doing around 4.5 seconds. So the estimates seem pretty valid.

The question your team honestly has to ask itself is, are we really close enough to the Cornells, UoW's and UWA's of this world to be concerned about one percenters? Especially when we pay for it with trade-offs elsewhere on the track? Maybe, if your team is consistently finishing the whole event, and you are continually scoring over 800 points, then that last little bit of straight line speed might be worth it, might not. But if you are scoring less than 800 points each comp then I think you have much more pressing concerns than outright engine performance. Get the thing running smoothly and make sure your hot start works – and direct the bulk of your resources to more important things.

What should we be concentrating on? I'm going to agree with Kevin Heyward here, in that I think the true key to track speed in FSAE is in the handling – dampers and tyres. If you can build up your corner speed, it helps you everywhere on the track, even the straights (see u.t term in the straight line equation above).

If engine performance is so unimportant, why do pro race teams spend so much time and money on it?
I'm pre-empting the above question. FSAE is a completely different game to actual wheel-to-wheel racing, and that is due to the fact that we don't have to pass our opponents outright on the track – rather if one car is lapping faster than another, the slower is flagged out of the way. This is a big difference. Under standard track racing conditions, most passing is done under brakes and into corners, which means that to be competitive you need to keep up with the opposition down the straights. Higher corner speed at the expense of straight line speed doesn't work too well in a wheel to wheel situation – your competitive advantage under cornering is lost when you get baulked by the cars that blew by you down the straight.

If anyone is interested I can relay the story of Aprilia's 500GP entry of a few years back as an example – but this post is dragging on enough already!

In FSAE, we are a lot more open to experimentation with where we can chase track speed - and can more easily trade straight line speed for cornering.

Intent of the competition
The good people who designed the rules for this event were well aware that this competition needs to be cheap and safe for it to survive the test of time. Everyone in professional motorsport knows that power is expensive, and to design a competition whereby power is the deciding factor would be disastrous at a student level. The regulations are, quite cleverly in my opinion, framed to make us think further than outright engine performance. Think about the 20mm restrictor, short straights, slaloms and lots of corners, extra points for fuel economy, speeds not high enough for significant wind drag, lack of wheel to wheel racing as mentioned above. Everything is there for us to be a bit more adventurous in our designs – and yet most still cling tight to the premise that the first thing to consider is to have a greater engine performance to weight ratio than the opposition. I think it is a wasted opportunity to be a little innovative, that's all.

Now I've made a right nong of myself – fire away ladies and gents http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Kamil S]

I liked your post. I think it was well thought.
Not neccessarily that I disagree with the current rules, but I would love to see less restrictions on the engine and its systems. I understand that the FSAE committee might be wishing to see students with a limited power range design the best chassis/suspension out of a vehicle (so to speak), but it is really wasting the innovation of what the more thermal/fluids influenced people need to do.

Throttle body design, intake air cooling, and flow paths are the first topics that come to my head when it comes to engine systems innovation. We deserve a chance to show off too! http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

Anyways, I may have steered a little off subject here, but to go back to your point, yes; more power does not neccessarily mean faster cars. However, if the rules change, it would definitely mean better usage of that power.

[Denny Trimble]

Geoff,
I think I see one of your marbles back in the corner...

Just kidding, great post, and I agree for the most part. However, I'd argue that drivetrain stresses come primarily from tire loading. Even a 45HP single can be geared down to produce the same torque at launch as a 75HP 4-cylinder, therefore requiring the same drivetrain strength/mass. Smaller tire diameters help reduce the weight of these parts, however. Anybody seen those 16" LeCont's on any cars other than Delft?

[Kevin Hayward]

Geoff,

Excellent post ... again. (Although you mis-spelt Hayward)

Engine power, at least for us, is not the main issue with our powertrain development. A big part of why we stay with the CBR is the knowledge we have accumulated on working with the engine. Reliability of a FSAE car has to be number one priority. To go out and out break on the finish line attitude is paramount to wasting your sponsors and Uni's money and resources. At least from our perspective.

I do want to pose a question to you Geoff. Mainly because I know that whatever your answer is it will be well thought out. Why is it that the skidpan event rankings are not just the list of entered cars from the lightest to the heaviest? (At least for the cars with the same tyres)

Given equivalent abilities to setup cars (which I assume exist amongst the top group of teams) the event should come down to the mass of the driver and vehicle as well as the COG height.

Cheers,

Kev

[jjusb]

it is really wasting the innovation of what the more thermal/fluids influenced people need to do.

Throttle body design, intake air cooling, and flow paths are the first topics that come to my head when it comes to engine systems innovation. We deserve a chance to show off too!

Not very much true..... in fact when you have a big restriction like we all do (20mm), innovation in fluid dynamics, flow paths, etc can help you a lot. Then again, a couple more HP won't necessarily win you the competition as Geoff said, but you can certainly "show off"

[Big Bird]

Denny, good point. I wasn't thinking about take-off, my head was spinning around an endurance track as I wrote that. Those extra 860 posts or so have certainly sharpened your eye for such anomalies http://fsae.com/groupee_common/emoticons/icon_smile.gif

Kev, hugest apologies. Never mind, when you become a household name (as predicted in that Racetech article all those months back), we'll all be spelling Haywood right.

You've raised a damn good point, obviously with reference to a discussion here about 12 months back. The skidpan event has characteristics all of it's own as compared to cornering on the autocross/endurance track. I can think of three reasons off the top of my head why RMIT at least has traditionally failed to set the world on fire with skidpan:
1) The driver was, for two events at least, an aging, under-weight, over-rated hack with abysmal driving skills, questionable personal hygiene, and yellow feathers
2) Tyre temperatures
3) Vehicle track width
I'll expand on the second and third points. Firstly, we designed the 2003 car with a narrow track, trading some weight transfer and therefore cornering grip so that the car would be more maneuvreable through slaloms and changes of direction. Going back to the Tailem Bend comp, from memory the beanie boys from Chalmers won both skid pad and acceleration (or at least did really well in both). Wow, you'd think, fastest car in both a straight line and in corners - it will cane everyone on the main track. But it didn't. It had a very wide track, great for grip and the skid pad, but it did look rather awkward through the twisty stuff. (Sorry gents, not trying to bag your car - I loved that thing, and you beat us after all! Just using it as an example of relative trade-offs.)
The narrow track was a design decision we made early given we place most emphasis on endurance/autocross. Wider track means you are laterally shifting your CofG more through a given change of direction. If you design purely for grip, you may find the increase in available grip is overwhelmed by the force you need to generate to overcome the extra "amplitude". Being aware of costs and benefits....
Also, a light car doesn't heat up it's tyres as quickly as something heavier. I now get the feeling this effect is quite significant in skid pad. Before Tailem Bend we were consistently doing 5.1's and some 5.0's around a dusty carpark at uni, on old brick compound Dunlops. We figured we'd be right at the pointy end on new tyres in a stinking hot SA desert. Nope. Ended up at 5.5, midfield, scratching our heads. Even changing to stickier Goodyears for FStudent, which helped in the open track stuff, still had us at 5.5 in skidpad. Tyres not even warm. Once again, a design decision that pays off in one event, with trade-offs in another.

Hope that helps Kev. And give Alma a big hug for me. Will we see her at Werribee this year?

And just a quick clarification that I didn't include in my post above. My primary argument isn't meant to be a singles versus fours (versus twins versus V8's or anything else for that matter). I would rather put the position that whatever you have, just get it running reliably and leave it alone. Our obsession with engine performance figures just isn't justified.

Cheers all, my fingers are tired.

[Nick McNaughton]

Geoff,
Just a couple of things with your assumptions -
A 600/4 won't pull >1g all the way out of a slow corner down a straight, it might average ~0.6G. A single is unlikely to better that given similar gearing.

The 100% throttle generalisations aren't so accurate for a formula car. I can only speak for a 600/4, but I imagine it's the same for a single. The engine might be making all the torque it can at 50% throttle at mid-revs, because airflow is limited by the engine's VE rather than the restriction of the throttle. MAP logging will show this pretty clearly. In short, there's a large portion of the rev range that maximum power for that RPM comes at something less than 100% throttle. If drivers were perfect, they'd be at 100% throttle everywhere they could use more power, but from what we've seen, they can be power limited and still not ask for more. This tends to be more pronounced with frequent corners.

What all this means is that time at 100% throttle doesn't mean as much on these tracks as it might for real race cars, because you can be power limited at throttle positions below 100%.

Other than that, from our perspective the challenge is in fuel economy. There's more points in lap speed than fuel economy when comparing 600/4's, but a single really screws it up for everyone. We have to be significantly faster than you guys to account for the hit we take in fuel economy, and the acceleration event alone might not be enough.

Cheers,
Nick

[Kevin Hayward]

Geoff,

Ahh you have picked an immediate hole in my rushed post. I can't believe I left such a gaping one as track dimension. By the way in the tailem bend comp Georgia Tech won skidpan. Also I think there was an error in the scoring because the average skidpan time for Chalmers was quicker than any of their individual times. I'm guessing on the error ... but it wouldn't be the first spreadsheet mistake from that comp.

The mention of temperature is one of the things that we have been very interested in ... and have tested a lot about. The interesting thing is that given the tyre temperature is a significant problem on skidpan why are the teams running the skinnier Hoosiers not winning? The 7.5" Hoosiers (actually a tread width of 8.25") we run at the rear of our car should be abysmal for skidpan if tyre temperature is the dominant effect.

The driver issue that you mentioned (although being too harsh on yourself) is quite a valid one ... even on skidpan. We have seen that driver error can be quite significant in terms of throttle control and steering corrections around the "theoretical" skidpan. Unfortunately the driver issue is probably the biggest gap in performance of any of the top running cars. Some of the driver pairings in Autocross show this very clearly. Seeing gaps of above 5 seconds between number one and two drivers account for much more of a performance difference than I have ever seen in any of our tuning attempts. I mention this as a fact not a complaint. You use the advantages you have. If that includes a talented and/or experienced driver then good luck to you.

Once again I think we face an incredible balance between different vehicle effects. However we have had a lot of clues in testing that tyre load sensitivity is probably not the main contributor to vehicle performance. This has been pretty much empirical evidence and I don't want to share too much of our results for obvious reasons. However the biggest gains (apart from driver training) we have made in our testing involve analysing tyre temps (dynamically), damper work (not just Kinetics either) and suspension geometry. Before anybody drills me on the damper work having an affect on the dynamic loading of the tyres and hence its effect is somewhat due to tyre load sensitivity I'll concede the point and counter that the dampers also affect the dynamic position of the tyres and the amount of disturbance one corner will transmit to the others.

I do not mean to shoot down the lightweight, low COG car at all. At least on paper it is the superior vehicle. However I would argue that suspension component stiffness, chassis stiffness, kinematics and damper properties have a bigger impact on performance than we tend to give credit for. Some of these issues result in added vehicle weight. I will also put in that the engine/drivetrain package has a much bigger effect on cornering performance than we (at UWA) had previously acknowledged.

I pretty much asked the question because I think that simple analysis between basic vehicle properties is almost fruitless at best and misleading at worst. Which I think is along the lines of how you masters of RMIT approach the design problem. Taking the simple approach of power, weight and COG height misses out a lot of other issues that affect vehicle performance. Pretty much all of those arguments should be prefaced with the line "With all other things being equal". That being said I have been nothing but impressed with the RMIT take on the FSAE vehicle design. The vehicle is so much more than a single taking advantage of its inherrent lightweight. It makes me wonder where we would have been if we had started out on our first car with a single instead of a Honda 4.

Anyway I hope this thread keeps going. Philosophy of design is so much more interesting than the stupid detail work you have to do to actually make the car happen.

Cheers,

Kev

p.s. Geoff if you mis-spell Hayward again I may have to add you to my list of future victims as I sit in the corner of my padded in room in the asylum they set aside for FSAE students who spend too long reading Forum posts.

[Kevin Hayward]

Sorry in my last post when I put the words "theoretcial" skidpan it was supposed to read "theoretical steady-state" skidpan.

I do not propose that the skidpan is only "theoretical". In fact I have seen a few of them set up in real life.

Kev

[ben]

Great posts from Geoff and Kevin. I broadly agree with the points made.

Adding to the point about driver ability. I agree completely with Geoff's argument about autocross/endurance provided that we're assuming a driver with the ability to exploit the theoretical performance of the car. This is emphatically true for RMIT but I wouldn't mind betting a relatively weaker driver would be quicker around the lap in a 600/4 with a lot of power than a nimble 450/1 simply because he'll straighten the thing up early and gain on the straights

Geoff – Interesting that you should almost bring up MotoGP. Since the 4-stroke formula has come in the emphasis has moved from carrying corner speed to being able to pick the bike up and get on the throttle early due to the increase in power relative to the grip. This is why you hear people talking about riders with 250cc riding styles (i.e. carrying lots of entry speed) struggling in the MotoGP class. Digressing further, Dany Pedrosa picks the bike up very early even in 250 and this is one of the many reasons why he should make the transition fairly well next year.

Digressing further still, maybe we should called MRA and convince them to give us all copies of LTS and have a virtual-FSAE competition in the style of the Matlab programming contests? We issue everyone with a track map and see who can get the fastest lap. Maybe that would settle the debate...

Ben

[BeaverGuy]

Geoff,

In general I agree with you. As an engine guy I belive getting the last bit of power is not an effective use of resources. But we all like to know that we have more power than the next guy http://fsae.com/groupee_common/emoticons/icon_smile.gif. However, at the same time I would say that many of the things that create a reliable, easy to drive powertrain are those things that get you that last little bit of power. A flow balanced intake will will be more consistent cycle to cycle making the performance of the engine more consitent. An adequate throttle and restictor, they really should be considered toghther, can get you 70HP with out much of a problem. A well designed combination will give a little more while also giving better throttle control and thus drivability. There are numerous other examples too.

[MikeWaggoner at UW]

There's the points issue also. Doing something neat with the engine gets you design points, and this thread is very calculating about winning the competition.

[Big Bird]

Glad to see this post is generating the sort of discussion it was intended to. Cheers gents.

Nick, the figures quoted were what we estimated to be worst case when we started designing our car. If a 600/4 is pulling 0.6g, (say 6m/s^2) and a single say 0.5g (same 20% difference), the above equations solve out to around a 5.9% difference along the straights, less again. Obviously, referring back to the acceleration equation, the lesser "1/2at^2" figure means the "u.t" term is eating away relatively more of the distance travelled.

Throttle time ratios? To be honest, with our first hand calcs we just got some dimensions for the 2001 track and worked it out from there. We found there were 220 metres along 4 straights, and 320 metres of corners. We assumed you'd only get full benefit of the extra power in a straight line, allowed an average of 25 metres of corner entry and exit transients and braking for each straight, and by distance we were looking at around 20% of the track where power makes a difference. We figured that was a safe estimate - any dispute of transient and braking distances on the straights could be countered by the argument that our estimate was by distance, and time fraction would be decreased since you travel faster on a straight. Rough, but it gives an order of magnitude of the overall effect.

As for fuel economy, I think acceleration event and the economy event pretty well cancel each other out. It is just that the economy event comes at the end of the weekend, so any hit taken in acceleration is already there before you take the final day's event.

Kev, sorry about that. I'll never spell Heywood wrong again. As for the padded cell, if you could see out from under that hat you'd find I'm sitting opposite you.....

You've made a lot of other good points which I'll respond to - but my fingers are now worn down to stubs. I'll let some others have a go on this thread before I use up all the electrons myself.

Cheers all

[Jeff The Pyro]

Very interesting posts from everyone. Big bird brought up a very good point about the differences between time trials and wheel-to-wheel racing... but i think you're missing one of the most important points (it sounded like it was on the tip of your tongue... you just didnt really say it outright), which is that in wheel to wheel racing you have to account for another very unpredictable variable... the other cars. In time trials you can find a perfect rythm... a perfect line through every turn... and once you figure it out you just repeat it over and over. in wheel to wheel... other cars will get in your way which means you cant necessarily pick the perfect line through a corner, and if you have to adjust mid-corner and pick a different line, all bets are off that the lighter weight will improve your corner exit speed, which is why its nice to have that extra horsepower as a buffer to help get you back up to speed as soon as possible. if two cars get in each others way through a corner and both come out of it with relatively low speed, the car with the most power will immediately pull ahead in the strait section.
on a similar note, I'd have to say i agree with ben about the driver ability issue... all these numbers floating around are assuming that car is being driven right at its limits 100% of the time... A less experienced driver may very well not be able to pick the perfect line through every corner and take advantage of the extra corner entry/exit speed... or may not have the confidence to drive a car through the corners as well as they could, but it doesnt take all that much talent to push the accelerator to the floor in a strait section (Assuming there is enough traction available) and thus the extra power, though its benefits may be small, will nearly always be realized in a real life. Of course at the same time, if you have too much power you're more likely to get into the gas too heavy too early out of a turn and lose control. being able to run at full throttle more of the time (because of less power) lends itself to consistency, which in the long run can average out to be faster, especially with an inexperienced driver.

Aside from that, i think to a lot of people view this competition as an opportunity to take an aspect of automotive performance that interests them and really dive into it, and given the rather divine aura that seems to surround horsepower... i dont think many teams are likely to clip the wings of their engine programs, even if you could convince them that it would help them score better in competition.

im kinda tired... just finished finals today... i have a feeling i might have repeated myself a lot in that post or came across as babbling or something... but you get the idea.

[Denny Trimble]

Jeff,
Good points. On the topic of designing a car around a driver who's less than optimal, that rubs me the wrong way. Having fast drivers has a little to do with luck (who shows up), and a lot to do with execution (how well you train who you've got). It takes organization and dedication to train drivers. For one thing, you need a car that runs most of the time. And for another, you need people willing to give up a night of preparation and a full weekend day every week from July to May, just for the "driver training" portion of the program. At least, that's how we have run it since '99. Add those hours to the design / build / school / work / girlfriend equation, and it starts to get ugly.

We lost several points this year based on execution / resource management, but driving wasn't one of them. Well, at least not in Autocross and Endurance... we'll have to practice the others more.

But, I guess I'm rambling at this point, too.

[ben]

[QUOTE]Originally posted by Denny Trimble:
It takes organization and dedication to train drivers. QUOTE]

True, but I could 'train' in an F1 car for the rest of my life and never be as quick as an F1 driver. Inherent natural ability will determine the level you can reach, the amount of training surely just affects how soon you reach that level?

Ben

[Denny Trimble]

Yes, natural ability matters (luck - who shows up). But, except for a few teams who have drivers with years of kart racing experience, most of us don't know how to drive racecars well when we show up in FSAE. I sure didn't.

What I'm getting at is that I think it's worthwhile to lose 5 seconds a lap by training your drivers, even if it takes all year. You might still be 2 seconds behind other teams, but it's a damn shame to work all year on the car chasing vehicle performance details in the 0.5 second range, and have a driver who can't get it within 5 seconds a lap of that performance http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Dr Claw]

True, but I could 'train' in an F1 car for the rest of my life and never be as quick as an F1 driver. Inherent natural ability will determine the level you can reach, the amount of training surely just affects how soon you reach that level?

well, that goes for 98% of every other FSAE driver. none of us are really supposed to be faster then an F1 driver, so the more mediocre training we mediocre drivers do, the more spectacular we look when we drive at competition. otherwise, we'd be in F3000, or sombody's F1 driver programs (or at least a karting champion/contender) if our natural ability was that advanced.

[MikeWaggoner at UW]

Regarding mashing it on a straight...

We've got a driver that's pretty fast and 240 lbs (Denny), and a driver that is ok at 145 lbs. On our ADL plots, you can see the 145 lb guy going like hell on the straights, but losing time everywhere else. Denny is faster by ~.75 seconds on a 30 second lap, and way more consistent.

The car is set up for Denny, however (corner weights, ride heights, roll bars, brake bias). I sometimes gripe (some would say whine) about this (I'm about 165), and Denny keeps threatening to put ballast in the car when I drive, to see if it helps my lap times.

[Dr Claw]

that'd be a wild study, you know that? make it repeatable and you'd get hired to balast danika patric's car to shut up robby gordon and all the people that think she has such a terrible advantage. just think of the possibilities, not only is that sort of thing maybe thesis worthy (at least an SAE paper), but balasting danika patrick sounds all.....fun http://fsae.com/groupee_common/emoticons/icon_smile.gif

back to engines and "performance", why dont more teams run a dry sump? you get to chase ENGINE performance, and deliver a chassis perk that can be turned into CHASSIS performance if utilized. everyone is chasing HP and torque at a drain to a teams overall budget (maybe hurting chassis side??), but you can play to both sides of The Force with a dry sump.

I am personaly all for crazy engine programs. its not only an engine guy's only real opertunity to pour all sorts of crazy thigns into an engine, but its also a suspension guy's chance to have a REAL engine behind his suspension.

[Travis Garrison]

Dr Claw,

I think a lot of teams see a dry sump as a major reliability issue, and at least a few lbs of added weight...making it a lot less attractive if resources are already spread thin.

-Travis Garrison
UW FSAE

[Dr Claw]

to that, thin resourced teams wouldnt have an extravagant engine program http://fsae.com/groupee_common/emoticons/icon_smile.gif

what i was getting at, was: for a team that does spend extravagantly on an engine, in sheer terms of 'doing something drastic to the engine', why dont we see more dry sumps? not to sell the idea of the usefullness of a dry sump to the crowd here (because it has it's own point by point pros and cons that dont need to be hashed here) but im just wondering with all the spending on an engine, it'd seem that more teams would impliment one.

I can understand the reliability issue, but that is supposed to be only an initial problem. progressive designs should get better with time, but that initial hurdle is often a baptism by fire in most cases. That can also be sead about running a turbo, or getting headwork or other 'crazy engine stuff' as well though. with teams running all sorts of various features for 4-5 years now and no DS, it seems lopsided.

that's just one of 'those' observations or hypothetical thoughts though, im not pleading for teams to run drysumps http://fsae.com/groupee_common/emoticons/icon_smile.gif. how many teams do you think spend heavily on their engines though? Lawrence Tech has, with exception for '05...any others care to own up?

[Michael Jones]

Reliability is a problem (hence why we didn't have dry sump until this year) but it's probably the right choice overall.

And I'm (somewhat reluctantly) in agreement with Geoff's analysis - as I suspect Jay O'Connell is, from his design reviews of the last couple of years.

It remains to be seen whether the smaller, nimbler single is absolutely better as an overall design standard for this event, but it's certainly the closest contender for full-scale paradigm shift I've seen. Had RMIT's second driver had comparable times to their first and had they finished endurance, things would've been very close indeed.

I suspect if it is, it'll take a couple of more cracks at it to make it happen - high-power 600/4s will still be effective for a while and have their own benefits (I suspect they're a lot more forgiving for variations in driver weight and style, for example...) However, if Geoff's right, the 600/4s will have to adapt very quickly or be consistently second-tier.

[John Bucknell]

Originally posted by MikeWaggoner at UW:
Regarding mashing it on a straight...


I like to call this 'straight line skilz'.

Anyhow, this is the best thread I have ever read http://fsae.com/groupee_common/emoticons/icon_smile.gif Picture what would happen if Penn State built a single...

[MikeWaggoner at UW]

Regarding skidpad and tire width:

What width rims are you running with your tires? Anyone know how much of an effect there is? The only advice I've gotten is use the widest rim possible for a given tire.

[VFR750R]

Very cool topic.
In Nascar, it doesn't matter how much HP you have unless you can put it down. At 160mph corner exits the car that allows the driver to get on the gas sooner gets down the straight sooner and it can look like one car has a crapload more power then another as it walks another down the straights but it comes down to chassis dynamics. I also believe in a balance of engine and chassis, but when many top FSAE cars are traction limited at average course speeds i wouldn't be surprised to see teams starting to compromise skidpad performance for corner exit traction. ie. antisquat, rear weight bias. And until we fully understand tires for FSAE and what gives the best corner grip it's hard to make assumptions on what car design is best light or power (not to say they are a direct tradeoff).
I'd like to add how amazing it is how close so many cars were this year with drastically different approaches. It's crazy how every teams cars share little more then an engine configuration(and sometimes not even that) and 6" slicks, yet lap times are dead even.

[John Bucknell]

Is any team willing to share exactly how traction-limited they were? Like seconds per lap with steering wheel straight and traction control active/substantial wheel speed differential? I doubt it, but you never know http://fsae.com/groupee_common/emoticons/icon_smile.gif

This may be the secret reason Denny is so fast - Mike W maybe you should try the ballast...

[Denny Trimble]

Originally posted by MikeWaggoner at UW:

The car is set up for Denny, however (corner weights, ride heights, roll bars, brake bias). I sometimes gripe (some would say whine) about this (I'm about 165), and Denny keeps threatening to put ballast in the car when I drive, to see if it helps my lap times.

A small point, I haven't sat in the 2005 car when it's been on scales yet, or when ride height has been measured. I think the "dead weight" (Mr. Quann or Mr. Soden) in the car for alignments was around 180-200 lbs. And I haven't heard any requests to change brake bias or ARB settings...

I don't think a heavier car/driver combo helps, except in warming up the tires.

My thinking on the traction limited corner exit scenario John is talking about, is this:

A light car/driver with a certain weight distribution is limited in its forward acceleration rate by the dynamic weight on the rear tires and their coefficient of friction, until the engine can't put out enough power to overcome this. The acceleration rate will be (CF * Dynamic Rear Weight) / Vehicle Mass.

I would argue that a heavier car/driver with the same weight distribution will still be limited to the same acceleration rate. I think the higher CG height from a heavy driver is outweighed by the further forward weight bias on the car, and the Dynamic Rear Weight / Vehicle Mass ratio probably stays the same as the light car. (Yes I'm assuming CF is constant between the two dynamic rear weights).

So, if I become power limited sooner, I'll actually be slower than a lighter driver (while pumping more heat into the cooling system and burning more fuel, but scoring major throttle position plot points http://fsae.com/groupee_common/emoticons/icon_smile.gif )

[Eddie Martin]

This is a good thread.

On the topic of drivers, sometimes you get drivers with talent and sometimes you don't, this has to be taken into to account when designing a vehicle. If you look at the whole "for the weekend auto crosser"� thing this would lead to the assumption that the driving standards will vary. That is not to say that teams shouldn't have very experienced drivers in the seat at the time of competition, driver training is a vital part of the each campaign, its just that after the equal amounts of training different people will circulate at different speeds.

We are very lucky to have very good drivers at Wollongong who have had lots of training and experience but also have buckets of natural talent that has been honed in fsae. I've driven fsae cars quite a bit but I can't go as fast the quick guys with more natural talent. On the same day, same tyres etc. I'm about 1 to 1.5 secs slower on a 30 second track than UOW's fast guys Woody and Dangerous. No matter how much training I do I can't equal their times. We have had Glenn Seton and Craig Lowdnes (two famous touring car drivers from Australia) drive our 02 car and they are pretty much the same speed as our quick guys and the same gap back to the average guys, maybe a tenth or two quicker. The biggest difference I did see, where I think they picked up their time, is braking. They were on the absolute limit of braking every time. Just an observation but I think fsae cars are normally driven at 95% because hitting a cone or spinning is so costly in our competition. Missing an apex in a v8 supercar by 100mm doesn't cost you 2 seconds a lap. http://fsae.com/groupee_common/emoticons/icon_smile.gif

Having driven an rmit single (thanks geoff and steve), an average 600/4, some very good NA 600/4's and an extremely powerful turbo 600/4 you can notice the difference on the straights in terms of acceleration. You really hang on tight with a turbo 600/4. http://fsae.com/groupee_common/emoticons/icon_smile.gif Being an average driver at the best of times the extra power does help make up for the car being "talent limited"� while I'm at the wheel. http://fsae.com/groupee_common/emoticons/icon_biggrin.gif I doubt I'd be able to make maximum use of the lower weight of the vehicle to gain the extra corner speed needed to make up for having less power. But with a more talented driver you can obviously gain time in the corners with the lighter vehicle.

I think the track layout really affects these decisions as well. Australian tracks are normally a lot tighter with less big straights. Australian and American fsae cars are becoming more and more different as time goes by due to the type of courses at the different competitions and also the practice areas that teams have available. These differences will of course change the philosophies of teams at the concept design stage.

The discussion has mostly talked about the dynamic implications of different design paths but you really have to consider how these relate to the one of the main aims of the competition scoring 1000 points. Just like in the automotive industry cost, marketing, reliability etc as well as performance and handling matter. What ever is going to score more points should be the aim, even if that means sacrificing or comprising in one particular area to gain more in another. The singles have shown to have a big advantage in fuel economy not being an engine guy I assume this is down to the vehicle's lighter weight, smaller engine capacity and with most singles 10 inch wheels, small sprockets etc so less rotating mass. Rmit proved how fast a single can go in the autox and enduro. Delft proved you could do a 4.1 second acceleration run with a single. Lighter cars should do better in Skid Pan but as discussed before it is a complex event. Statics normally depend on the team.

I think at the heart of the competition though is a set of rules that make allowance for change / difference and in fact encourages it. Its great to walk up the pit lane and see 4WD, Turbos, Wings, Singles, Twins, 600/4s, simple and complex approaches and different packaging solutions.

[jonno]

Great reading discussing a lot of the theories behind our (RMIT & Delft influenced) switch to a smaller, lighter formula student car.

denny, you questioned which teams use the 16" LeConts: we've got sets for this year and in our discussions with LeCont, they mentioned four European teams now used their tyres...

One of our suppliers mentioned investigating the Tyrrell 6-wheeler 10" Avon tyres, apparently they are only a 15" o.d. Not sure of how they would perform (ie don't know weights etc) but it would be interesting to see how they compare. One of the reasons we didn't enter the Tire test consortium testing was that the LeCont was unlikely to be evaluated.

I agree with the driver point - if your two drivers have a difference of 5 seconds per lap then that massively affects the result much more than any one component (or even collection of components).

I would like to extend a point just made by Eddie Martin in his penultimate paragraph. He pointed out that Delft did a 4.1s accelleration. They had wiring issues I believe in the main enduro, but they went out on the track later on the Sunday, and I have never seen a Formula Student car go like that one went. The driver was possibly on the edge of his skill, and knocking cones over did not affect them as it was a just for fun demo run but due to the fact that at 125kg, there is no weight to shift, that Delft car went through the slaloms as fast as most cars in a straight line. Couple that dynamic ability with one of the best accelerating cars, and you have a car with a serious chance of winning. As Geoff mentioned, the lighter cars do poorly in the skidpan, but there is no other event they cannot win outright. I think the tracks would have to open up a lot more before a clear advantage swings back to a 600/4

[Z]

Boy, I go fishing for a day and I'm 40 posts off the pace! Nice thread you started here Geoff. These comments are mainly related to Geoff's first post.

Regarding Engines
=================
I think this competition can be won EASILY with a 40hp engine. I have capitalised the word "EASILY" as a reference to my post on the "First year cars" thread - more on this below. The main thing the engine needs is a smooth "electric motor" delivery of its power. This issue is driveline related (CVT, etc.). Also important is reliability and fuel economy, of course.

The reason I proposed the bespoke 80hp engine on the "Single cylinder" thread has more to do with human nature than outright performance. Building a 40hp car is going to have at least half the team wondering "What if?". This could lead to internal friction (team, not engine http://fsae.com/groupee_common/emoticons/icon_smile.gif) and a failed program. So I reckon build a lightweight, low CG, compact, driveable, and cheap engine with close to the maximum power available through the restrictor, and everyone is happy and ready to work on the important stuff. (Cheap is very important to me, and I reckon the one-eighth-of-a-V8-small-block would be quite cheap, and possibly even a money earner if you sold it to other teams. Also there would be a lot of polishing that could be done to keep any future year's engine guys happy.) So, now to the important stuff...

What makes a car fast?
======================
It depends on the type of track. I started to answer this on the "First year cars" thread by suggesting a computer simulation, although I may have complicated the answer too much. Geoff gave the basis of the simulation with his acceleration in a straight line equation and examples. The next step is to add a corner at each end of the straight. Pick a radius (R) and an assumed maximum cornering G (Ac) and you have steady corner speed (V) from Ac=V.V/R.

There are many types of track that can be modelled with this simple "simulation". At one end of the spectrum is the skid-pad - all corner and no straight. Obviously cornering G's are all important here for a "fast" car, and acceleration and braking G's and top speed are irrelevant. At the other extreme is the very long straight where top speed rules and acc/brk/cornering G's are irrelevant - land speed record cars. For a shorter straight (say a dragstrip) only acceleration is important. And so on. You assume a certain type of track, then find what is most important to make the car "fast". I would say that on FSAE type Autocross and Endurance tracks it is mostly about cornering G's. Also important is the "easy driveability" necessary to allow average drivers to achieve those G's. The question becomes "How much extra cornering G's do we need to overcome the lack of acceleration from the 40hp motor?". I would say not much at all!

Just as a BTW, in desert racing the fastest car is often the one with the "softest" suspension that allows it to go at a reasonable speed in a straight line from one side of the desert to the other - acceleration, braking, cornering, and top speed are all irrelevant!

So, what makes a car fast around corners?
=========================================
There are many details here, any of which can slow the car down if they are wrong. But the simple answer is either grippier tyres or aero downforce.

1. Grippier tyres. At absolute best you might expect to get a 100% gain on your opposition here - say if everyone had Cf=1 tyres and you showed up with some super sticky Cf=2 tyres. But, in general, FSAE teams don't make their own tyres, and whatever tyres you buy are probably available to the other teams. There is some "unfair advantage" able to be gained with tyres, but not a lot (unless you have some brilliant Chem. Engineers!). Note that grippy tyres are unimportant for LSR cars and in some types of desert races.

2. Aero downforce. A column of air of cross-section 1 square metre reaching from ground to space weighs ten tonnes. The available plan view area of a typical FSAE car is about 3 square metres. So there is the potential for about 30 tonnes of downforce! That is a lot of potentially available "unfair advantage". You only have to tap into 1% of this (~300kg) to double the cornering G's of the car.

Bottom line.
============
It is a general principle of ecology that a species only needs a very slight "fitness" advantage over its competitors to totally dominate them - there are no even fights in nature. The same applies in motorsport. Look at the remarkably identical cars in, say, F1, then guess the chances of a Minardi ever winning. It is certainly not 10% (ie. 2 Minardis out of 20 cars). The Minardi is just very slightly different to its opposition, yet it has no hope of winning. Compare the 3 litre N/A 1980's F1 cars with their turboed competition. It took a few years for the turbo cars to iron out the bugs, but when they did they blew the opposition away. Same when any racecars started to use aero downforce. A small gain in one critical area translates to a huge gain on the track. These disparities in performance are why the rule makers spend so much time tweaking the rules. An even competition (which is supposedly more interesting to the spectators) is an unstable situation, hence needs constant negative feedback from the regulators.

Anyway, it is my opinion that the first FSAE team to get aero downforce right will blow the opposition away. The current aero approach used by some teams of front and rear wings is (IMO) the least efficient way to do it. This approach is seen on so many "real" formula cars today simply because it doesn't work well! That is why the regulators still allow it (oh, and also because it provides good billboard space). The historically proven best approach to downforce is ground effects. Ground effects is so good that even the very inefficient versions of it that are used today have to be very closely regulated.

I reckon a good ground effects car could win FSAE with only 20hp. It would have only one steady speed around the track with (almost) no acceleration or braking. Since 20hp is enough to push most passenger cars along at 100kph it should be enough for the smaller FSAE car. This type of car would do poorly in Acceleration, and there would be little to brag about during the static events, so maybe a "shinier" motor could be added just for show.

My previous suggestions for a "brown go-cart" always included a reference to an "aero undertray". The brown go-cart is just a delivery system for the aero "unfair advantage". So start with a lot of aero downforce, apply that to the tyres to increase their cornering grip, find some horsepower to get everything up to speed, then connect all the parts with a minimum inertial mass system - ie. a brown go-cart - because low mass is most important when you have an aero car.

And what if your "aero guys" don't deliver? Then you've still got a RMIT/Delft/lightweight-single type car, which is still competitive, though not totally dominant.

Z

[TheShagg]

I would like to make the point, that when you give the car 20% more power, you do not have to make 20% in weight additions. For instance, you do not change the weight of the driver, the steering wheel, etc.

Now, also theoretically, if you scale up the weight, you shouldn't lose anything in the corners, assuming you scale up the tires as well.

[Z]

Originally posted by TheShagg:
Now, also theoretically, if you scale up the weight, you shouldn't lose anything in the corners, assuming you scale up the tires as well.

True for a non-aero car, even considering tyre-load-sensitivity.

But when you go aero then minimum mass becomes much more important.

Roughly speaking, any Horizontal-acceleration = Aero-downforce+gravitational-weight/inertial-mass. So reduce the two "weight/mass" terms, which are in effect equal, and you increase horizontal acceleration, ie. higher corner speeds.

Z

[Steve Fox]

Geoff & Kevin,

I will echo some of the previous replies to this post: This is a GREAT post!

The motorsports design judges have said amongst ourselves (Carroll included) for many years now that this is a handling competition, not a horsepower competition. We have also been subtly conveying this to any team that will listen during post judging design critiques.

My personal opinion on the subject (again, this is just a pet theory of mine) is that most red blooded adolescent males (and a lot of adult males too) see large horsepower numbers as being directly related to the size of their p____. (You all know how the saying goes!) They are both built for impressing someone else, and both are a lot of fun to play with. http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

Geoff, your team did not come through my design que, but I sincerely hope that you presented the engineering analysis to your design judges that I saw you post in this thread! That is exactly the kind of engineering documnetation that we look for to back up good car design. http://fsae.com/groupee_common/emoticons/icon_wink.gif

FSAE 2005 will be remembered as an interesting year due to the variation in engine selection. RMIT had a lot of good execution of a great concept. The Tokyo Denki car sticks out in my mind as impressive (design wise) as well. I hope that a lot of you had (took) the opportunity to check out their single cylinder car as well!

Kevin, I know how to properly spell your last name, (I just leave it off when addressing you, and I can't miss!) http://fsae.com/groupee_common/emoticons/icon_wink.gif

I fear John is onto something as well, if Penn State built a single...

[BryanH]

Z, you may well tell you wife you go fishing but I reckon you just laid back in your tinny and smoked one huge bag of hooch. I have never read such irrational crap in my whole life.

While looking at data from Shanghai GP track (a high grip track) I noticed something very pertinent to aero discussion. Data from F1 car shows readings of 1.8g on several 80 to 85 km/hr corners and 3.5g on 200+ km/hr corners. The Rotor/RMIT car was pulling 1.8G at Silverdome. If $50 Million can't get more G's around a 80km/hr corner I don't think Z can.....

If your Grandma had nuts she'd be your Grandpa
MotoGP is on now, seeya Bryan H.

[Z]

Bryan,

The answer to your post is in its second last line.

If a team with $50 million wanted to build a car that goes fast around 80kph corners then they would NOT build an F1 car designed primarily for 200+kph corners. And they wouldn't shackle themselves with the F1 rules that ban decent ground effects. And they wouldn't be racing on a high speed track in the first place.

I would love to hear some rational criticisms of my ideas, but this "Oh, but they don't do that in F1, and everyone knows that F1 is best..." attitude is puerile rubbish.

Just to spell it out a bit more clearly. Would a $50M F1 car beat a typical FSAE car on a typical FSAE track? Of course not! The F1 car would knock over almost every cone because it isn't designed to go around sharp corners! (Watch them at the hairpin at Monaco.)

Z

PS. It troubles me that young people are so narrow minded and willing to accept the status quo. Where are the free-thinkers, the adventurers, the trail-blazers? Is it because Bryan is scared of the unknown that he clings so tightly to the shirt-tails of F1? Is anyone out there brave enough to explore the dark places?

[drivetrainUW-Platt]

was the Tokyo Denki car the light blue frame with the transverse engine shaft drive to an atv diff, if so a very nice package and clean car....seemed to accelerate nicely in the practice track...not that that is a really good test of acceleration, but I liked the drivetrain, all sealed from the factory differential....oh and their car separated in two in front of the engine

[Storbeck]

I'm surprised that nobody has pointed out that for most "real" race cars more power means you can overcome more drag from higher downforce and thus go around corners faster while still going as fast on the straights. I suppose the same thing could be said about Formula Sae cars.

It seems to me like the question about more power and more weight vs. less power and less weight comes down to a question about tires, one that has always been confusing to me.

Tires are weight sensitive, if you put less load on them they have a higher "coeficient of friction"

So why can f1 cars, in slow corners where aero download is low, pull just as much latteral acceleration as a super light fsae car?

Because they have bigger tires. So it's not really a question of how much weight you have, but more like the ratio of how much weight to how much tire.

So how come we don't use bigger tires on our formula sae cars, they all have skinny little tires?

Because they can't can't get the bigger tires up to operating temp, warm skinny tires must have more traction than cold big tires. So it's not how much weight to how much tire you have, it's how much weight to how much tire, as long as you can get that tire warm.

So why are we so worried about making our cars light, why not just have heavy cars with bigger tires?

Because you eventually will get going fast enough that you are no longer traction limited, so you need power to accelerate, and power is limited by the rules.

So it seems to this line of logic that a heavy car with big tires will be able to corner just as well as a lighter car with smaller tires, it's just a matter of getting the right size tires for the car, that is the biggest that can be brought to opperating temp. Then the only difference between a heavy powerful car and a lighter less powerful car would be the power to weight ratio, and this would only matter when the car is going fast enough to be not traction limited, which according to the info above, is not very often.

What if you have aero? You can use bigger tires because the download at higher speeds will warm them up, so you may bennefit even at low speeds when there is not significant download. You also need more power because of the greater aerodynamic drag. Question for Z: you keep saying that low weight is very critical for a car with large amounts of aero download, based on my above rambling, I can't figure out why, would you explain?

Another factor is the physical size of the car, if two cars have the same power to weight ratio, and the same "tire to weight" ratio, then they should go the exact same speed if they take the same line, but a physically smaller car can take a better line. This is the only real advantage to a smaller lighter fsae car that I can identify based on the rambling above. And this has way more to do with the smaller physical size of a smaller engine allowing a smaller car for packaging reasones, there is no reason a heavy car can't have a short wheelbase and narrow track.

One might also note that due to weight sensitivity of tires a car with the same "weight to tire" ratio but a lower center of gravity should corner faster. Also one with a wider track width. It seems that the "cg height to track width" ratio would be as important a measure of performance as "weight to tire"

So it seems to me like nobody should give half a shit how much a car weighs, how much power it's got, or how high the cg is, what we should care about is the power to weight ratio, the "weight to tire" ratio (pretty hard to put a number on that one), and the "cg height to track width" ratio. And we should be concerned about the overall physical size of the car. I suppose this matters more for small tracks so it might be called the "car size to track size" ratio.

I have sketchy understanding about the stuff I've been talking about, so I will not attempt to talk about things like moments of inertia about the various axis', and transient response of different size cars.


There is also one big factor that all of this "intelectual masterbation" ignores. The speed of the typical FSAE car has way more to do with developementk, tuning, and driver training. If two cars with all of the above ratios equal are built by two different schools, almost definitely the simpler of the two will be better developed, tuned, and have more time available for driver training, so the simpler one will win. That is, in my opinion, the REAL advantage to a single cylinder car.

Of course this depends on the size and expierience of the team that builds the car, so I suppose that would be the "complication to resources" ratio.

That was long, if anybody made it throught the whole thing, tell me if and where I went wrong.

Edit: just a few more thoughts

Why does Cornel win?
"car size to track size"-pretty average
"weight to tire" - thier car is not super light or super heavy, and I think they use the same hoosiers as everybody else
"power to weight" - I bet it's really good
"complication to resourses" - I bet this is how they win

How about RMIT? I bet they win the same way.

my $.02

Andy

[John Bucknell]

Okay, since Z brought it up - let's talk about effective undertrays. I think I'm qualified to talk about this subject since my job requires some modicum of aero wherewithal, but by no means do I claim to be an expert on race car aero.

The short answer is I think they'd be disallowed on the grounds of '3.2.1 Ground Clearance' - Ground clearance must be sufficient to prevent any portion of the car (other than tires) from touching the ground during track events. So therefore any deformable skirt that prevents lateral airflow ingress under the car (or even forward ingres) would get the big heave ho. At FSAE speeds, you'd really have to contemplate underbody sealing as a number one priority to be effective should that be your only aero device. Even then, the best underbody pressure you can hope to have is stagnation pressure equal to the forward velocity of the car - as the wake punched by the body is the only source of low pressure air anywhere near the car. Figure 1.5 m^2 hole in the air at 28 m/s (100 kph), gives stagnation pressure of ~0.472 kPa times 3m^2 underfloor area equals 144 kg of downforce for a perfect system - or barely enough to add 50% more download on the lightest car close to it's maximum velocity (ie nowhere near enough download to maintain that velocity around the tightest corners). It is the low drag characteristic of ground effects that makes them it popular - multi-element wings have substantially higher lift coefficients....

The long answer is - I don't think FSAE auto-x/endurance courses will ever be fast enough to ignore corner exit horsepower, even with all the passive aero in the world (which does have significant drag). Rolling resistance is also pretty killer on slicks, especially with downforce.

PS - took me long enough to write this Storbeck snuck a post in, maybe I answered a question of his already.

[Big Bird]

Thanks everyone for your excellent contributions to this thread. It pleases me no end that we can share our thoughts and views in such a diplomatic and respectful manner – even when such opinions are sometimes different to our own. I've learnt as much through reading and posting on these forums as through any of my background reading and class activities. The forums are a damn fine resource and I think we are all the better for having them.

Eddie, you bring up a very good point about differing tracks influencing different design solutions. With a thread running elsewhere pushing the idea of some kind of "World Championship"� featuring top finishing teams from each event, in such a case the SAE worldwide would have to start considering consistency of track design at the different events. Otherwise we could end up with the visiting teams showing up with vehicles less suited to the final event. Just a thought http://fsae.com/groupee_common/emoticons/icon_smile.gif

I also agree completely that we sometimes get so caught up in our cars that we forget the ultimate goal is to score as many points as we can out of 1000. My whole intention with this topic was to open up discussion that to design well, we must think a lot more holistically than we sometimes do (and in fact a lot more holistically than our education leads us – but that is a whole new topic!!). One way to open our minds to the bigger picture is to challenge the axioms that are presented to us – and when I read a post on another thread saying that "more power is always better"�, it gave the perfect opportunity to raise the matter. Cheers Eddie, and I also think Storbeck was on the right track with his comments that simpler designs can offer advantages in other areas such as development and driver training. (S, that was a cool post – I'm still trying to think up answers for all those questions now!).

Thank you Steve Fox for your input too. It is really nice to know that we have some of the official side of the FSAE community keeping an eye on here as well, we all appreciate it. Unfortunately I couldn't make it to the US with RMIT this year since I've changed uni (for my postgrad studies) – and the other issue of my rather depleted financial state. Too much FSAE, not enough time at work!

Ben, I'm always happy to discuss the bike GP's! I'm scratching together a few notes on my opinions on driver/rider psychology, will post them up here soon in regard to yours and other posts on driver training. That will be thesis number 4....

John Bucknell, cheers for all the good info both here and on the other thread, ("Why didn't you finish endurance"�). I'm not much of an engine head, so it has all been very enlightening.

Z, bloody good to hear from you sir. I was wondering when you'd jump on board, I expected you to be first into the mix! I'm amused by your comments that 40hp could be enough to succeed in this comp, I'd been having similar thoughts myself. The car we used at FStudent last year was built as much to prove that power wasn't important as anything else. It wasn't too light, and it was a bit complicated for a single, but it was a fairly honest little car. After the event, there was a lot of talk about how much better we could go with a lighter car, tuning it up a bit more – and in fact Rotor, Drummy, Dan and all the 04 crew (sorry lads, can't mention you all here!) were building such a beast which was the US car this year. (A damn fine little car, I must say). But secretly in the back of my mind I was wondering how much less we could get away with. First thought was a 250 single (WR again) – kart steering http://fsae.com/groupee_common/emoticons/icon_smile.gif - maybe even scrap the fuel injection and just run the standard carb (with restrictor of course) to really annoy the tech-heads. A real quick-build, anti-technology special (the brown go-kart!!). With a bit of development time I reckon you could surprise a lot of people, and leave a lot of spec-sheet engineers crying into their textbooks. Of course, there is that issue you mention of convincing enough like minds to come along for the ride (we had a hard enough time convincing the doubters about the 450!!).

I like the idea of doing hand calc track simulations too – gives the designer a much deeper feeling for what is going on. Some simple calculations, and we can start to get a feel for how much we need to improve cornering grip, how valuable our braking and engine power is, etc. Programs like ADAMS and the like are great, but they'll just pop out answers for as long as you keep asking them questions – and they'll never stop to ask your opinion if the figures they come up with are reasonable. Much of a modern engineering education gives the student the impression that all the answers can be found with the twiddle of a mouse and the stroke of a few keys – with a negative effect that many can't deal with the fine art of estimating.

As for the rest of your comments Mr Z, I'm having a good think about them – all seems to make sense to me.

Dr. Claw, I assume a dry sump appeals for the lesser friction drag? Have you estimated how much that may benefit you. 1hp? 2hp? So what would that be, maybe a 2-3% increase in engine performance if all works out fine? When you relate that to the above track time calcs, the net effect may only be in the order of 0.1-0.2%. Is your uni within 0.1-0.2% of Cornell on track? Sorry, not meaning to be abrasive, but sometimes we have to step back and think of true costs and benefits to the team and our final competition effort. If I was in such a position I'd be allocating resources to more important issues – designing a decent pedal tray, putting together a good wiring loom, routing cables properly, lock-wiring everything properly, ensuring there are no fluid leaks, etc etc. The devil is in the details.......

TheShagg – sorry, you've completely lost me mate. Nowhere in my original post did I say that a 20% increase in engine performance meant a 20% increase in weight. In fact, I argued that halving the power output of a given 600/4 engine might save you 5-10kg only, a much lesser effect. My point was that as designers, we must decide what gives a better "payback"� – the extra power or the lesser weight. The "answer"� to that is not an absolute one way or the other, and is primarily determined in the layout of the individual track. And of course team resources. And driver skill. And.......

I apologize to TheShagg and others if my first post was wordy, but I wanted to explain in some sort of logical manner where I was coming from. Many of my one and two liners on previous threads have been misunderstood, because I haven't fully expanded on my arguments. So if it was too wordy, my main arguments were:
* That engine performance has trade-offs, and the one I focused on was weight
* That even if two cars have as much as 20% difference in their engine performance-to-weight ratios, the difference may be less than 1-2% on track (and that is worst case scenario, where the less powerful car is no quicker on the twisty bits).
* Given that most of our engine development programs are giving a lot less than 20% gains, we would most likely be better off focusing on other things. (For proof of that, we only need to look at the number of cars that don't make it to the finish line each year – and I won't try to sound high-and-mighty on this front since I've been involved in teams that either haven't made it or have only just finished in times gone by too).
* FSAE gives us an opportunity to be a bit adventurous in our designs – yet many still cling to the premise of power first, deal with the other stuff later.

Cheers all, better do some real work for a few minutes http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Chris Davin]

A few things I want to add:

-I read a saying somewhere, "Horsepower sells engines. Torque wins races." I think it is true in Formula SAE. The original analysis in this thread assumed constant acceleration (flat torque curve) on a straight-away. If that is true, then corner exit power will be much less than the peak power allowed by the restrictor. This is not the way you want to operate your engine. You want to be able to put down maximum power at corner exit. Peak power numbers are pretty meaningless. Speed you already have when you come onto a straight-away is best. Speed you get immediately at corner exit is second-best, and speed you get at the end of the straight-away is the least-valuable of the three - hence the saying. There is far more to engine performance than just peak power but that's the number that the center of most discussions.

-I think Andy might be onto something. But, the problem is that the entire question of vehicle design rests on the behavior of the tires. It has been said in these forums many times by many people that the tires are the most important part of the car (and I said it too, above). But, our design choices can be only as good as our understanding of the way the tires behave. I think it will get interesting after the tire testing data gets released.

[ben]

[QUOTE]Originally posted by Big Bird:
First thought was a 250 single (WR again) – kart steering http://fsae.com/groupee_common/emoticons/icon_smile.gif - maybe even scrap the fuel injection and just run the standard carb (with restrictor of course) to really annoy the tech-heads. A real quick-build, anti-technology special (the brown go-kart!!). [/QUOTE

Having had our arses kicked by RMIT at FStudent last year we ended up discussing a 450cc single car with no ECU and a carb/magneto setup. If you dropped to 250 and given a compression release you could probably use a kick starter and junk the starter motor as well :-)

Ben

N.B. This all happened in a pub so that may of course affect the validity of the idea.

[Big Bird]

I've been trying really, really hard not to use the words "torque" or "power" without making my posts unbearably obtuse - and have tried to use the words "engine performance" or similar. This was so we wouldn't get into arguments over whether torque was better than power and the like.

I have the utmost respect for those that have a good understanding of engine theory, you've done a lot better than me. I have used the constant acceleration assumption to pop out some figures that might give some perspective of how much our engine affects us. In real life, the values will differ - but I doubt that a different "acceleration curve" would stray us too far time-wise from the flat acceleration assumption.

And no, I will never, ever, ever, ever, ever, ever, ever tell a Cornell guy what works and what doesn't!! http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Big Bird]

Ben, you British lads do all your best work in a pub http://fsae.com/groupee_common/emoticons/icon_smile.gif I get the impression the ideal job for you lads would involve a hotel with an autoclave in it.

Hmmm, maybe a hand-start next to the gearshift......

FYI Ben, Tokyo Denki ran a car here in Oz in 2002 with a CR450 in it with standard carb - and a crude flat washer with a hole in it for the restrictor!!! Got to around 7-8000rpm before the thing would just start spitting and popping like an idiot. Did we laugh? No. They were around 2-3 seconds a lap quicker than us that year, doing about a 34 second lap. Wollongong were only around a second and a bit quicker than Tokyo Denki from memory. Food for thought....

[Eddie Martin]

I think a car can win with 50 very driveable horsepower but all things being equal it is a lot easier to win with 80 hp. If you make any mistakes or things go wrong not having the horsepower as a back up, for a handling or driver deficiency, will hurt. If you can make an extra 20 or 30 hp without too many comprises to your team or vehicle package and you have the human resources to do so i think it is worth doing, especially if you have people on the team who really want to develop the powertrain. Winning with 20 hp is a bit of a stretch of the imagination i think.

Z on your theory of natural selection lets look at the facts for a minute. The red car from Ithaca normally has 10 to 15 hp on its nearest rivals and their results are mighty impressive year in year out. This may just be a coincidence because they are an excellent team but I don't think the extra power has hurt them.
Most corners and the skid pan are taken at about 40 km/h, so all your aero calculations should be based around these kinds of speeds. To get an effective aero package working at such a slow speed is challenging.

Geoff you bring up a good point about the devil in the detail. A perfect suspension package and awesome powertrain are nothing without a solid everything else. If one component fails the event is over. If it fails and the car is still working fine it probably shouldn't have been there in the first place. http://fsae.com/groupee_common/emoticons/icon_smile.gif

In terms of tracks take a walk around the 04 USA Autox track and then the Werribee course. If you based your designs on these two examples at the concept design stage you would come up with extremely different final products. You also have to look at where teams do private testing. I don't know whether it is just this part of new south wales but there are no areas of any reasonable size to do (Pontiac style) testing on. So we have to take this into account when designing a vehicle. No point designing the vehicle for 70 + km/h corners when the vehicle will never see anything more than a 45 km/h corner till it gets to the competition, even then it may not go much faster than that.

It will be interesting to see if Delft go lighter than 127 kg this year at Formula Student.

[ben]

Originally posted by Big Bird:
Ben, you British lads do all your best work in a pub http://fsae.com/groupee_common/emoticons/icon_smile.gif I get the impression the ideal job for you lads would involve a hotel with an autoclave in it.

Yeah, and just think how much better that work would be if the beer wasn't warm right? http://fsae.com/groupee_common/emoticons/icon_smile.gif

Ben

[Mountainman]

Years ago we had the RPI team down to the hinterlands of NC to discuss FSAE with us. They had campaigned a KTM single with a turbo (aerocharger) on it in the late 90's.. If i remmber right, they were running about 65 hp, and 78 lb-ft. Basically could pop on the course and drive every corner in 3rd gear.

Always tried to get our old Wolfpack Motorsports team to do a single. But we had zero cash. Ended up with a stock CBR f1 motor in our 2000 car. Rearend was a grease lubed VW Rabbit GTI differential. Good frame and suspension design. Ended up with Rookie of the year.

All in a car that we built for less than $5000.

With the choices in engines now, I don't see why more people aren't going for a torquier motor. HP ain't king on a tight track.

-MM

[fsae_alum]

Originally posted by ben:
[QUOTE]Originally posted by Big Bird:
First thought was a 250 single (WR again) – kart steering http://fsae.com/groupee_common/emoticons/icon_smile.gif - maybe even scrap the fuel injection and just run the standard carb (with restrictor of course) to really annoy the tech-heads. A real quick-build, anti-technology special (the brown go-kart!!). [/QUOTE

Having had our arses kicked by RMIT at FStudent last year we ended up discussing a 450cc single car with no ECU and a carb/magneto setup. If you dropped to 250 and given a compression release you could probably use a kick starter and junk the starter motor as well :-)


I tried to convince the following year team of this before I graduated. I said..."why build a brake and gas pedal when you can buy a shifter kart pedal right off the shelf? Same thing holds true for a shift lever. Beyond that you could use a shifter kart seat and even kart paddle shifters (from NZ I believe) and kart data acq systems too. Has anybody looked at using the Kelgate 6 piston enduro kart brakes? They should work for FSAE since they have no problem making an enduro kart stop on a dime from 100mph+ (even with a 450cc engine as one endure kart has out in CA). What about using the Cannondale 450 single? It's got user programmable EFI, dry sump, is TINY, is geared for a top speed of about 50 mph and has a VERY simple intake and exhaust (unlike most singles, it's in through the front and out through the back!) Using all of these off the shelf components speeds up the design and fab cycle and helps to ensure that you have enough time for testing.

Here's the trade-off...what works best for the design judges may not work best for the endurance event or even for the team itself. You have to ask yourself what do you want to win? Design or Endurance or both? IMO, if you want to win design, it's almost as if you need to design, analyze, and build close to 100% of the car from scratch. The down side is that it chews up tons of precious driver training and car testing time.

When we did our car in 02, we used a number of off the shelf components (that were unfortunately heavy) in the hopes of being able to leverage that to the judges. After all, this car is suppossed to be geared towards a weekend autocrosser right? Not many weekend autocrossers have access to machine shops on the weekend. Likewise, I have never met a racer that would prefer waiting on having something custom made or shipped overnight when they can go to a parts store and pick up the part they need within a couple of hours. As such, the majority of our driveline was from a Miata. We used the Torsen diff (adjustable TBR via fluids), stub axles, modified main axles (VERY HEAVY), bearings, seals, and clips from a Miata. If we (or a theoretical customer) had a failure, we (they) could always go to a Mazda dealer , a parts store, or a salvage yard and get replacement parts within a couple of hours.

We applied the same thinking to our engine. We used the F4i engine, gauge cluster, ECU, fuel pump, and main wiring harness combined with a Power Commander. No sense in re-inventing the wheel. We only spent 2 days of dyno testing and the car would start up on the 2nd try everytime and just purr like a kitten with no hunting.

All of this made our lives MUCH easier and we knew we were buying components that we're already proven. We tried to sell our design as "customer friendly". Unfortunately, the design judges didn't like it very much. That was shock. http://fsae.com/groupee_common/emoticons/icon_wink.gif

What this means is that your team has to ask themselves what they, your university, and your sponsors want out of the competition and weigh that with the resources and time available to decide what route to go. There is probably an optimal design for every event of the competition and the team that can design the car that can adapt (read pit tuning between events)well to each of those events is going to do real good (ala Cornell). You don't have to win every event (or ANY event for that matter)...just do real good in them all to win the overall comp. Is that what you want though?

Just my .02

[Z]

Storbeck,

Your explanations regarding power/weight/tyre-size/etc. are quite correct (IMHO). I think an important point that you brought up is the issue of car width for a particular racetrack. This is something that can be usefully studied with some fairly simple simulations (pen and paper, or computer if you must). Essentially, most normal racecars would be hopeless on a FSAE track because they are too wide and too long. Perhaps the most critical part of the track is the slalom, with the wider/longer cars having to do much more turning to avoid the cones. The question becomes "How narrow can we make the car before it falls over?

You asked "why is low weight important for aero cars"? I briefly answered this, but here it is in more detail. Roughly speaking (with lots of simplifying assumptions);

Horizontal-acceleration = Tyre-Cf x (Aero-downforce + (car-grav.-mass x G))/car-inertial-mass

IE. From F=ma, we get a=F/m, with;
Fhorizontal = Cf x Fvertical
Fvertical = aero-downforce + car-gravitational-weight
m = car-mass

Using SI units, G=~10m/s/s, car mass including driver, and looking at extremes of weight and Cf range;

Example 1:
==========
Non-aero car with best tyres.
Cf = 2
Car mass = N (any figure will do here)
Aero-downforce = 0

Horizontal-acceleration = 2 x (0 + N x 10)/N = 20m/s/s = 2.0 G

Example 2:
==========
Heavy aero car with same tyres as above.
Cf = 2
Car mass = 400kg
Aero-downforce = 2kN (~200kg)

Hor.acc = 2 x (2000 + 400 x 10)/400 = 30m/s/s = 3.0G

Example 3:
==========
Light aero car with same tyres.
Cf = 2
Car mass = 200kg
Aero-downforce = 2kN

Hor.acc = 2 x (2000 + 200 x 10)/200 = 40m/s/s = 4.0G

Example 4:
==========
Light aero car with el-cheapo tyres.
Cf = 1
Car mass = 200kg
Aero-downforce = 2kN

Hor.acc = 1 x (2000 + 200 x 10)/200 = 20m/s/s = 2.0G

So, a good amount of downforce gives the heavy car +50% cornering G's, and it gives the light car +100% cornering G's. Or the light car can have crappy tyres with half the Cf of the non-aero car and still match it in the corners. This is a very crude analysis but I hope it explains what I was getting at.
----------------------------------------------------------

John Bucknell,

Regarding skirts, yes they would be very helpful, but they are not essential.

You say "the best underbody pressure you can have is stagnation pressure...". I am not sure what you are getting at here, but if you mean that the best pressure coefficient (Cp) that can be expected is -1, then I have to disagree (stag. pressure Cp = +1). Most wings comfortably manage a max. negative Cp =~-9 near their leading edge. This is not hard as it only requires a local velocity 3 x free stream. The trick is to get a similar level of Cp over an as large as possible area of the undertray. This does NOT require the undertray to act like a normal aeroplane wing - that's why it is called "ground effects". (I'll leave the details for another post http://fsae.com/groupee_common/emoticons/icon_wink.gif).
--------------------------------------------------------------

Eddie Martin,

You say "To get an effective aero package working at such slow speed (40kph) is challenging". Hmmm? FSAE Rule 1.1; "...so that the knowledge, creativity, and imagination of the students are CHALLENGED."

No offense intended to anyone (hope these help http://fsae.com/groupee_common/emoticons/icon_smile.gif http://fsae.com/groupee_common/emoticons/icon_smile.gif http://fsae.com/groupee_common/emoticons/icon_smile.gif) but my guess why there are so few ground effects cars in FSAE is that there is no one to copy. GE's have been pretty thoroughly exorcized from most forms of motorsport - they are too good! Likewise there are no books on the subject (no detailed tech. stuff, just coffee table histories) because there is no money to be made designing GE vehicles.

Sorry guys, but you are pretty much alone here. But then again, so where the Wright brothers, and they were only bicycle makers without the benefit of a University education.
-------------------------------------------------------------

Geoff,

If you ever get tired of engineering I reckon you should join the diplomatic corps. A few weeks of Big Bird in the middle east and they'd all be hugging and making up. Could be a Nobel peace prize in it... http://fsae.com/groupee_common/emoticons/icon_wink.gif

Z

[Storbeck]

This thread is really interesting, it's great that we have at least two design judges here, adding to the disscussion, but a few things still seam unclear to me

Big Bird made a convincing arguement that improvements in cornering ability will have a more significant reduction in laptime than improvements in power, he even put some numbers on it. This isn't an earth shaking new concept. It has been discussed before.

However, the proof that a lighter car will corner better seems to be taken for granted. Even though at first glance this would seem to be disproven easily by looking at the results of the skidpad competition.

The concept of the weight sensitivity of tires seems sketchy as proof that a lighter car will pull more lateral acceleration, if that's the case, why aren't we all running 23.0 X 12.0-13 C3000 R25A Hoosiers. THey would of course add a lot of rotating mass to slow our acceleration in a straight line...


Z- Why is low weight so important for a car with high aero download?


Not trying to be a pest, this is something that I have spent a lot of time thinking about and haven't been able to come up with good answers on my own.

Andy

Edit - ahh, Z posted while I was posting, sounds like we need another ratio, something like "aero download to weight"

[Z]

Andy (Storbeck),

And I edited while you were editing... Usually takes me a few goes to get it to make some sort of sense!

I agree with your car-weight vs cornering G's arguement. As long as you aren't melting the tyres, or peeling them off their rims, and as long as you are comparing different weight cars with the same tyre compound at the same temperature and same tyreprint pressure, and so with the same Cf, then you should get very similar cornering G's. It just becomes a "second order" issue - are the cambers the same, is the wheel load distribution similar, is the handling balanced, etc. This was my Example 1 above.

(Third edit!) But the heavier car, with its bigger tyres, bigger wheel bearings/axles, stronger suspension, more powerful engine, etc., etc., is going to COST MORE!

Z

PS. (Edit) Re: the aero question. It is to do with the car having both a gravitational mass (which gives a vertical force that can be added to by the aero downforce), and also having an inertial mass (which resists the horizontal acceleration, as per Newton II). I've just read an interesting book by an Indian engineer ("Origin of Inertia" by A.Ghosh) who modifies Newton's Law of Universal Gravitation so that it includes velocity and acceleration terms (as per "Mach's Principle"), and thus accounts for the "inertial" mass of Newton's Laws of Motion, and more. But now I'm rambling...

[John Bucknell]

If anyone wishes to read more about underbody venturis, Ian Bamsey's 'Anatomy & Development of the Sports Prototype Racing Car' has a whole chapter devoted to it.

Based upon the values quoted in the above book, I'm still skeptical it would work at FSAE velocities... http://fsae.com/groupee_common/emoticons/icon_smile.gif

[fsae_alum]

Originally posted by John Bucknell:
If anyone wishes to read more about underbody venturis, Ian Bamsey's 'Anatomy & Development of the Sports Prototype Racing Car' has a whole chapter devoted to it.

Based upon the values quoted in the above book, I'm still skeptical it would work at FSAE velocities... http://fsae.com/groupee_common/emoticons/icon_smile.gif

John, having thoroughly read that book over and over again, I'll agree with you on both of those points!

Another really good resource on gound effects is http://www.mulsannescorner.com

[Eddie Martin]

Z, it is good to have lots of different ideas on the forum and I think the discussion is richer for that.

In terms of my comment about the aero package there are lots of things to consider. Most teams are under resourced (human) as it is so any "extra"� components take away from the essentials. Even if they build a "brown go-kart"� they still need to understand exactly why it is better than a "conventional"� fsae car or any other concept. A lot of time each year is taken up with students learning / researching and basically getting their heads around the whole concept of the formula sae competition. Most students would never have designed or built anything as complicated before so you have to take this into account when you have lots of new members on the team.

The vehicle doesn't need an under tray or any other type of aerodynamic device to operate or perform successfully at the competition so you really have to way up all the pros and cons of a system and how it will effect the team before you do it. In my opinion to get an aero package to work effectively you have to design the car around it and in some cases comprise other systems to do that. This is a big ask at the concept design stage when most students are just getting into the idea behind competition. You then have to look at whether you have the man power to let one or two students go away from the essential components and design the aerodynamic parts.

I have looked at how you would integrate an under tray into a car and it presents a lot of headaches and problems, especially trying to not to affect other systems too much. Yes i know that is the point of the competition but getting a vehicle to the competition thats well tuned with experienced drivers is enough of a problem as it is most of the time. Would you mount it sprung or unsprung? If you did mount it unsprung you have to make it stiff enough to handle to the forces generated, eg. more mass and obviously its all unsprung. If you mount it sprung you have to transfer that down load to the tyres without comprising the suspension. As the vehicles get narrower and narrower you also have to look at whether can generate enough download to make the system pay off with the smaller area of the vehicle and the more crowded space in which to package the undertray.

You have to ask yourself will the aerodynamic devices produce enough benefit on the track (in terms of vehicle lap time) to make up for the extra time, weight, human resources, management time and coordination effort, money, materials and vehicle testing kms the system will take to develop?

Formula SAE is mostly about getting good people and then getting them to work together in the same direction to achieve a goal. It's a management challenge. The car is just the fun bit at the end that is used to get students interested in the first place.

Looking at your calculations, Z, I see you have used 200 kg of downforce for your calcs. Can you generate 200kg of down force at 40 km/h? I'd be very impressed if you could.

[Scott Wordley]

Very interesting reading.

Setting aside some of the side issues such as driver skill, fuel economy, resources needed, cost, development time, static events etc and just concentrating on the theoretical performance potential of the car, here are some of my thoughts:

To my mind this discussion has been dancing around the broader and more general question of whether the FSAE dynamic events (on a points weighted average) are "power-limited"� or "traction-limited"�.

What does this mean?

Consider that, on a very fundamental level, we use an engine to generate energy to accelerate the car and driver, principally in the forward direction. Due to the nature of thermodynamics it is relatively easy to specify a braking system more than capable of matching this acceleration (in a power sense) in the reverse direction.

Acceleration in any direction however, is achieved through forces exerted at the contact patch, or traction between the tire and the road. Basic stuff right.

At any given time a racecar will be either "power limited"� (through insufficient engine or brake power) or "traction limited"� (through insufficient traction in the relevant direction at each of the contact patches).

Judging from the majority of posts in this thread, and in my personal opinion, most people would probably agree that it is in fact a "traction-limited"� category of racing, indicating that for the majority of the "time"� a FSAE car is traction limited. If power gains and traction gains are possible (as we know they are) and equally easy to affect (likely) then the greatest performance (time and points) gains are to be found in increasing traction. This statement does not deny the fact that increased horsepower will result in faster times and more points, all other things being equal, but indicates more significant gains can be made on the traction side of the equation.

What data do we have to support this?
Percentage of time at throttle openings greater than 90% is a good start, as people have mentioned. This will obviously vary with competition, track, and driver but I'd be surprised if anyone had measured over 30% for autocross or enduro.

What I like about this philosophy is that it can accept and explain the performance improvements seen as a result quite different and seemingly conflicting design choices. So lets look at the ways in which you can address the problem of being traction limited for the majority of the time....

The most obvious starting point is tires. Every car has them. If you can find tires which maintain a higher average coefficient of friction over the course of the dynamic events your car will be faster. We've recently done some testing on several different brands and were shocked at the spread of results. If the tire testing consortium returns similar results, many people will be surprised, not to mention disappointed. There are serious gains to be made here and while such data is not easy to come by, it is well worth the effort.

Control of the tire is the next step, which comes down to good suspension and steering geometry and the control of its movement ie chassis, springs, shocks, bars. Most teams do an ok job of laying out the fundamentals but we've always found at least 5% - 10% improvement through solid testing and development.

If you want more traction in the forward direction why not drive more wheels? Depending on how well you control the distribution of torque to all 4 wheels you may have problems with maintaining your levels of grip in cornering, but you'll murder them if its wet. In this application you'll also wear a weight penalty which will increase your % of time at full throttle but not necessarily make you any faster. A tough choice to justify, but worth trying if your courses are very "square"�, point and shoot affairs. Perhaps there could be an instance where straight front wheel drive would be superior to rear wheel drive (although I can't think of one right now).

If you identify a lack of corning grip as your major problem on the tight, twisty FSAE circuits then logically (and operating within the established 600/4 paradigm) you would turn your attention to the generation of aerodynamic down force to increase the normal force acting on the tires and the grip that they can produce (specifically cornering, with braking and acceleration as a bonus). Due to the low speeds involved, in our experience we have found that you pretty much want to make as much down force as possible within the constraints of the rules. As powered ground effects are banned, you would explore the possibilities presented by normal ground effects and wings. Most people would find (as we did) that the small foot print and necessity of locating the driver and engine as low as possible generally precludes the use of an effective aerodynamic undertray (ie. it doesn't generate enough downforce for its additional weight). We have run aero undertrays in the past but never been able to measure (or notice) their contribution, so we don't anymore. The data I have seen suggests CL of 1.4ish are possible for a pure diffuser, well sealed. Rule constraints (no touching the ground) make CLs this high unlikely (as John has suggested), however I am most keen to seen what Z has in mind (email me if you like). I would be delighted to be proven wrong.

Which leaves you with wings. In my opinion you are foolish to run wings anything less than the maximum allowable span, and to a lesser extent, to rigidly mount them to your chassis given the option of "moveable"� or unsprung mounted aerodynamics. Unless your suspension is ridiculously over stiff, chassis mounted wings are practically an admission that your wings don't work. Like Eddie suggests 40kmh is your cornering speed so you probably want to be generating at least 5-10% of your car and driver mass in downforce at this speed. Z also illustrated the effect of car and driver mass on lateral acceleration with aero download, the lighter everything is, the higher your downforce to weight ratio (which is the critical ratio) and the higher your cornering accelerations. Due to the lack of long straights seen in Australia, I'm pretty confident that we have yet to hit the point where our lap times are drag limited. Wings neatly convert periodically unusable engine power into increased grip, higher cornering speeds and corner exit speeds and allow later, harder breaking. For drivers inexperienced with aero cars they also require more time and practice for people to explore the available (and constantly changing) traction envelope and develop a "speed sensitive grip factor"� into their expectations. Like 4WD they also come with a weight penalty and also a drag penalty. Justification of their use is therefore as easy as noting the difference in lap times with and without them, or as complicated as undertaking a detailed series of CFD and Wind Tunnel testing coupled with extensive On-Track data logging and lap time simulation. I agree with Eddie that wings are not and probably never will be a necessity for FSAE as finishing the car early and completing the event will always ensure a finish in the top 50% of teams and often higher(something we hope to do this year). Also I will never argue that "wings are always a benefit in FSAE"� as I've seen many cases in this competition where I'm pretty certain they are not giving the team any benefits. Instead I would argue that "wings CAN be a benefit in FSAE"� when done correctly and well tuned. Enough about wings.

Yet another way to attack the problem of being "traction limited"� is to say, "We have too much power, so lets sacrifice some and see what happens"�. While moving in a totally different direction, this response seeks to address the same fundamental issue of being overpowered for the available levels of traction. Downsizing to a single cylinder engine saves a significant chunk of weight and if done well can be advantageous for all the reason that Geoff others have outlined above.

You could argue that downsizing the engine to a single and adding a CVT for no net weight gain when compared to a 600/4 might be a better solution. Same problem, yet another potential answer.

On the issue of weight (it seems to crop up a lot), can someone please provide any actual track data (times?) or anecdotal evidence on the effect of gradually adding weight to these cars. Anything for acceleration, skidpan and autocross/enduro type tracks? The comments from the Washington boys, other teams and our own experiences show that good drivers can easily easily produce lower lap times with significant weight disadvantages. What i want to know is what is the "weight sensitivity" of these cars in performance terms for a control driver. Such data could potentially be used to justify adding a meagre 10kg to the car inorder to finish it 2 months earlier and generate more driver training time and ensure less potential for breakage. Just a thought.

Getting back to the subject at hand, of course the solutions outlined above aren't mutually exclusive...

Say you took a 160kg single car and bolted on some neat wings, your still only 170kg, say 230kg with driver. Then assuming you can still get 40kg of downforce at 40km/h that's getting close to 20% of downforce to mass ratio which is a LOT.

Is it wrong that I've already started planning for 2007?

Hahaha.... I'm going to die building these stupid little cars.

[Z]

Eddie,

I accept that adding "aero" amounts to extra work and less resources per department. However, if it catches on then "aero" will become the "essential task" and everything else will be an annoying hindrance (at least until the organizers ban the aero, which would eventually happen). Actually I think that some branches of motorsports, like F1, are too aero obsessed, and they could do themselves a favour by fixing the suspension so that the wheels stay on the ground over the kerbs, but that's another matter...

One thing I like about aero, especially an undertray, is that it is so cheap to manufacture. At its crudest it is a few off-cuts of plywood and some wood glue. At its most sophisticated, a 2-D carbon/honeycomb molding (ie. a sculptured but mainly flat panel, which is easier than a 3-D chassis tub). No moving parts, no bearings, no oil supply, etc. The hard part is determining the right shape, but I think a couple of bright guys running a parallel program could do this (I can expand on a suggested approach to this R&D).

As for integrating an undertray onto an FSAE car, everything that I've posted about the "brown go-cart" applies. And a lot more - the devil is in the details! Briefly, beam axles front and rear with the lightweight undertray attached directly to the beams (possibly at two points on the rear beam, and only one central point on the front beam, so as not to restrict four wheel "twist" mode). Or separate left and right undertrays, rubber diaphragm sealed to the fuselage (for lower driver and engine), with the undertrays still attached direct to beam axles. A narrow, streamlined fuselage and engine cover to feed smooth air to a low mounted rear wing (or undertray "flap") which is close coupled to, and hence driving, the undertray. As I have said before, if the undertray is a dud, then ditch it and you still have a neat, small, fast FSAE car.

Regarding rough aero calcs: With local flow velocity 3 x free stream the Cp = 1-9 = -8 (I made a slight mistake in previous post where I suggesting this Cp = -9). This is an easy local Cp to get on wings high in the air, and even easier when the wing is close to the ground. A venturi tube can comfortably get a Cp =-20. So at a car speed of 10m/s (36kph), and assuming an even more modest Cp=-4;

Pressure = Cp x (rho/2) x V^2 = -4 x 0.6 x 10 x 10 = -240Pa (= ~24kg/sq.metre)

So for 3 square metres that is 72kg (720N) downforce. This will allow a lightweight car to go a bit faster than normal around corners, which then gives more downforce...

At twice the speed (V=20m/s, 72kph), pressure x 4, so now Pressure = -960pa = ~96kg/sq.m = ~1% atmospheric pressure. At 3 sq.m this is now 288kg downforce - check those wheel bearings!!!

The trick, of course, is to get as large as possible -ve Cp over as much area as possible. The above figure for Cp=-4 is an estimate for the average over the undertray, but I reckon it is quite achievable. There are a lot of different ways of doing this, but I hope this answers your questions for now.

Criticisms of wild assumptions are welcome! http://fsae.com/groupee_common/emoticons/icon_smile.gif

Z

[Z]

Scott,

You slipped your post in while I was typing the above.

Ground effects are relatively unexplored territory. Just because so-and-so did this-and-that twenty years ago (and were then banned) doesn't mean that their approach was the best possible way of doing it. Imagine if the Wright brothers built their Flyer, flew it, and then for some reason heavier-than-air flying was banned. A hundred years later flying is legal again and you want to have a go. Do you copy the design of the Flyer? It had a total of twelve half-wings, and the frame had to be twisted for control, and a largish engine with two propellors...

Just say you saw one of the modern "ultralights" - simple single wing, small engine, weight shift control. Wouldn't that be easier and better?

I think there are huge improvements possible in GE's, it just requires a lot of people playing around with them - as happened with ultralights.

I can expand on this later but gotta go now.


Z

[CMURacing - Prometheus]

re: ground effects and other forms of racing

yes, they're banned (or highly controlled, as in f1), but how many times have we all heard steve matchett commenting on the fact that modern f1 cars generate 70+% of their downforce solely from that heavily regulated underbody?

[Denny Trimble]

Originally posted by Scott Wordley:

Hahaha.... I'm going to die building these stupid little cars.

Scott,
Awesome post, thanks for the contribution. And yes, I think I will die building these (or similar) stupid little cars.

[John Bucknell]

I would rather die trying to ride a MotoGP bike...

But yes, great posts. I still think Z's estimates of Cp are way high (or is that low?). You might b e able to achieve a velocity ratio of 3 over a short distance, but then you have to diffuse it back out again over a relatively long distance so the static pressure rises up to ambient (or a little less given the wake behind the car and an evacuating wing).

The only F1 data I've seen shows a peak decel of just touching 5g from about 320 kph. Assuming a mu of ~1.5, downforce would have to be a little under 2.5 times mass. At one-third the speed, doesn't that make it 1/8th the downforce? So a pretty trimmed out, high downforce F1 car is only able to generate maybe 30% of it's mass in downfore at 100 kph? Can we double that within FSAE rules? Can one of you whiz kids with your CFD modelling test these assumptions? http://fsae.com/groupee_common/emoticons/icon_smile.gif

[rjwoods77]

Hey John,

MotoGP bikes eat shit.(compared to this) You need one of these. Such a cool design this thing is. I really dig the front mount carbs and the radiator.

http://venus.13x.com/roadracingworld/issues/apr00/tularis.htm

[Scott Wordley]

Again, I agree with John's comments on undertray pressure coefficients and the need to diffuse the flow back up to near ambient pressure, which results in average coefficients much closer to 1 than to 4. Z, is there some new technique that you have in mind to increase this? If so i'd like to hear it. I have a few interesting ideas in mind to help pull more flow under the front wing but im going to try them out first.

I also think 3m^2 for the undertray of a "small" car is excessive, I would halve that figure unless you plan on going back past the rear tires (which is still technically legal, we have done it in the past). Using a rear wing to help pump your diffuser rules this out though (as it needs to be forward of the rear of the rear tires) and further cuts into your undertray package space.

So if we take your calculated 72 kg of down force at 36km/h and divide it by 4 (more realistic average pressure coefficient) and halve it again (more realistic undertray area) we are left with a mere 9kg of downforce or 4% of a light car and driver combo. We have PHYSICALLY MEASURED more than 3 times this amount at the same speed with our current wing package.

I understand your point that we should not limit our ideas and investigations to what is accepted. There is nothing i hate more than people who think we should look to current formula 1 to provide all the answers. They have a drastically different set of rules and requirements.

Personally I think you need to be aware of current technological advances but more importantly you need a solid understanding of what people have done in the past. You then need to evaluate all the information at hand CRITICALLY.

I think it is dangerous to assume that there is a huge potential for further development and improvement of undertrays simply because they were banned early on in the piece. The physical phenomena of ground effect is well understood, and the major limitations have been identified (diffuser angle, ride height, sealing, 3D effects). Unless you have something new to add, i personally cant see much potential for average pressure coefficients of 4 while maintaining a "sculptured but mainly flat panel" as you put it. To my mind there are much more gains to be made from a return to Lotus 49B style high mounted wings. And i have tunnel data to prove it.

So Z i'm interested to hear where you find 3M^2 for an undertray and how you plan on getting Cps of 4. As you say, the devil is always in the detail, so details please.

Have you got any sketches of this brown go kart, I'd love to see it!

Cheers

Scott

[Z]

Originally posted by John Bucknell:
I still think Z's estimates of Cp are way high (or is that low?). You might be able to achieve a velocity ratio of 3 over a short distance, but then you have to diffuse it back out again over a relatively long distance so the static pressure rises up to ambient

John,

This is a big subject, so the following is only a small bit of it...

Your description above basically applies to a single element wing, or to something like a standard venturi tube (as opposed to a "compound" or "double" venturi). Many GE cars used this type of single element shape, most probably because that was all they needed at the time - the GE's were giving them more than enough downforce to cope with. If they tried for even more downforce then flow attachment in the diffuser was a problem, but there were also many other problems to fix such as chassis/suspension strength, porpoising, aero balance, etc.

If an aeroplane designer wants a high-lift wing they don't use a single element, which can only get up to maybe Cl =~1.5 before it stalls (ie. flow separates). For high-lift wings you use multi-elements, or "slots and flaps". Back pre-WW1 (~1910?) Handley Page did pioneering work on these slots and flaps. They took a standard single element profile and carved it up so that it eventually looked like a venetian blind. I think they ended up with about ten mini-wings nose-to-tail. The more "slots" they added to the original wing the higher angle of attack they could run and the higher maximum Cl they got. Ultimately they got about Cl=~4 at an angle of attack of about 45 degrees. The slots prevent flow separation and allow a more aggressive overall wing shape.

Anyway, Cl=4, which is a common value for high-lift slots-and-flaps wings, amounts to pretty much the same thing as my "estimated" average Cp=-4 (since the positive pressure side Cp=~0). Furthermore, this Cl=4 is for free-stream wings, and the effect of a nearby ground plane makes it even easier to achieve high Cl's. And even more furthermore, there are a lot of other "loopholes" that can be exploited because of the nearby groundplane.

Z

[Z]

Scott,

More simultaneous posting! My above post should answer some of your questions.

3 square metres? Well a rough upper estimate. 4sq.m would be too much, 2sq.m quite easy. Extending the undertray behind the rear wheels would certainly help (and an in-front-of-rear-wheels-wing could still be used to drive it - see below http://fsae.com/groupee_common/emoticons/icon_wink.gif). Not sure of the legality here, though? I also can't see how that rear wing rule is supposed to help with safety? With the front wing permitted a long way in front of the front wheel, then an aero centre of downforce in front of the CG is likely, which is very unsafe!

You say "The physical phenomena of ground effect is well understood...". I must disagree here. There was an article in Racecar Engineering ("Aerostatics" - March and April 2002) that proposed a somewhat different from standard explanation for SOME types of GE. Shortly after that article came out one esteemed expert suggested something along the lines of "highly speculative... very doubtful..." etc. Yet the aerostatic mechanism pops up all over the place. For example see RE October 2004 "Aerobytes" column by Simon McBeath (although Simon explains the mechanism incorrectly).

Anyway, I reckon that if there is no fundamental law of nature preventing it (eg. Newton's Laws, or the Laws of Thermodynamics), then it must be possible! Maybe I'm just a born optimist! http://fsae.com/groupee_common/emoticons/icon_smile.gif

Z

PS. I can send you a copy of that "Aerostatics" article if you want.

[John Bucknell]

Rob,

Actually I've met Rob Tuluie (at a SoS race at Nelson Ledges lo these many years ago). Isn't he an engineer at Renault F1 now? Anyhow, yes a Tul-da or Tul-aris would kill me faster than a MotoGP four-stroke http://fsae.com/groupee_common/emoticons/icon_smile.gif

Z,

A cascade as you describe operating in ground effect? Not the same as an undertray, more like two really low aspect ratio wings on either side of the central chassis. Lots of details to chase for sure. Winggrids (http://www.winggrid.ch/) purpose to do the something similar, but can be non-functional if the spacing and chord are off. The more I study these things, the less I realize we know how exactly aero works (Edit: Doh! I seem to be agreeing with Z!) http://fsae.com/groupee_common/emoticons/icon_smile.gif

Okay, now for real I have to pack for Indy...

[Scott Wordley]

Those wing grids are COOL. There goes my afternoon.

By the sounds of things that "simple" diffuser of yours is getting a tad complicated me thinks. Compound diffusers heh? Ive thought about trying a concentric (half obviosuly) diffuser type undertray to generate higher angles of attack (and therefore lift) but you run into the same problem as with bi-plane wings, with the low pressure zone that you generate on the outer diffuser partially negating the pressure differential achieved on the inner diffuser.

What happened to the Brown go kart philosophy?

With this level of complexity it seems to me like it would be easier to design a very narrow cockpit and run two huge (think 2.5m chord), narrow (think 300mm at narrowest, 500mm at widest) multielement wings (in ground effect of course)down each side of driver and engine. Youd have endplates at the outer width of the wheels and grow the wing out side ways infront of the front wheels, between the front and rear wheels and above the rear wheels to meet the endplates. Working it neatly around the front wheels would be fun. Getting your centre of pressure right even more so. Did i mention that it should probably be unsprung as well? And god only knows where your going to put the radiator....

but if you figured it out you'd have the worlds first "race wing", as apposed to a "race car".

Scott

PS: Also please send me that article Z, and i agree with you, Simon tends to have his good and bad days like the rest of us. I loved his Competition Car Downforce book though.

scott.wordleyNOSPAM@eng.monash.edu.au
leave out the NOSPAM bit obviously

[Z]

Yes, interesting site about "Winggrids". Good to see that the aeronautical community is catching up with the birds! I haven't read through it all yet but I especially liked the line - "History is full of unsuccessful attempts... because nature is a bitch...".

Anyone interested in this sort of stuff might also like the book "Birdflight as the Basis of Aviation" by Otto Lilienthal (first published 1889, a reprint with introduction by brother/co-worker Gustav Lilienthal is available through SAE library). Otto made about 2000 glider flights during the 1890's. Unfortunately, his last one in 1896 was fatal, but his efforts did inspire the Wright brothers. The book is a great example of how to investigate something as slippery as flight. The Lilienthal's "muse" was the stork. There is a detailed line drawing of it in the book, complete with all dimensions, and with its characteristic "winggrid" tip feathers.

Regarding aero undertrays. Because "nature is a bitch..." I would caution against taking aeroplane or bird type wing concepts and applying them directly to an undertray. The main differences are that the undertray is longer than it is wide, so lateral flows, inwards from the sides, are more pronounced than on the bird wings. Also the very close proximity of the ground makes a big difference - eg. wings at high angles of attack benefit from a leading edge slot (or a "slat" in front of the mainplane) because this helps the air whip around the LE. A GE wing doesn't need this slat because the ground does the same job.


Scott,

The multi-element wing down each side might work - only lots of testing will tell http://fsae.com/groupee_common/emoticons/icon_smile.gif. You might try "simplificating" it by making an essentially flat undertray (full width and length of the car?) with some largish louvres along the sides of the fuselage, these only reaching out to the inside edge of the wheels. Since there would be significant inward side flows under the flat outer section of undertray between the wheels, the louvres might best be swept back at an angle. Flow visualisation techniques would help - say, a transparent (eg. perspex/lexan) test undertray with wool tufts stuck to the underside.

Email is on the way (soon).

Z

[Mechanicaldan]

OK, I have a new screenname as I have moved to Louisiana, so "Daniniowa" was no longer relavent.

Geoff this is a great post you started. FYI: I am an engine guy.

Let's see if I can quickly output some of my thoughts now while being able to look back after gaining a few years of experience.

1. I feel the driver is the biggest factor to improve upon (with an average car) given that 675 points are available from dynamic events. Unfortuneatly, unless a team has already developed a reliable car to be used for training, they are out of luck. It's almost unfortuneate that the driver is such a big part of how well teams do, because it would sure help to answer many of these questions if there was more of a controlled environment in which to test and compare the different cars. Given that the drivers are such a factor, at least for all the teams outside of the top 10, I feel that making the cars simply easier to drive by less experienced drivers is important. I watched one of the CVT cars cruising around the track at FSAE 05, and he looked as fast as some of the other drivers that were struggling to downshift and upshift at each corner.

2. Tires. We keep discussing the many different aspects of this topic as no one knows all the answers, and even when we do, we will still be forced into a compromise. I do feel that optimizing (whatever that becomes) is the key to making the car faster. More unestablished teams would do better with a better understanding of keeping the tires in contact with the pavement. I know very little about this topic, especially in regards to suspension geometry, but I know that it is important. Risse shocks are the worst performance alternative, but the best cost alternative. Just put them on a shock dyno sometime.

3. If indeed the cars are traction limited, then why are there not more attempts being made to implament TRACTION CONTROL and ANTI-LOCK BRAKES? Yes, you are required to lock the brakes for brake test, but I think an ON/OFF switch would take care of that. I know that there are small ABS parts available from multiple motorcycles. Another question that needs to be asked is up to what speed are the cars traction limited? Iowa State has used the AEM ECU for 2 years now. I bought it because it did have the ability to use traction control. In 2 years, there has never been enough time or effort put into getting it working, but I know that it is worth the effort.

4. Thrust at the wheels. Give me torque, and I will adjust the final drive gearing to adjust the thrust, which directly translates to acceleration. Basically you can pick how much acceleration you want. I feel that the maximum speed any FSAE car should be geared for is between 70 and 80 MPH.

5. Kart brakes. I stumbled upon kart brakes too late. I need to research them more, but I see the potential.

6. Ah, singles vs. fours... My first choice would be the Cannondale if going the single route, especially with ATK supporting it now. It's all there. Small, powerful, fuel injected, electric start.

7. Design competiton. My current thought is that to do well in the event, you have to bring documented evidence of design process, and know what you are talking about. This means that starting at the beginning, there needs to be a paper trail documenting each decision as to WHY a part ended up the way it did. There are LOTS of correct answers, be it cost, time, manufacturing, availability, innovation, whatever.

As far as more power being the key to being faster, it all depends on the conditions. I think that the fastest car is going to be the one that can adapt to the given conditions the fastest, and why I really feel the effort should be made toward traction control and ABS.

It's all about balancing on that outer edge of the traction circle, and staying there with each tire.

[Craig Garfield]

We run novice autox novice schools in the SCCA DC region, and the first thing we tell novices is the most important thing to going fast is the driver. There is no substitute for seat time and proper techniques. This is more important than anything you can do to the car. A good driver can drive any car regardless of setup faster than a less experianced driver. And some people are just much more talented than others. The second thing is tires. Getting a sticky tire to work at peak levels then becomes a function of the suspension.I believe that FSAE teams that have a driver with a lot of seat time in any car be it the team car or not will have a significant advantage in the enduro and autocross.