Not quite F1, but almost (FSAE project)

Post here information about your own engineering projects, including but not limited to building your own car or designing a virtual car through CAD.
flynfrog
Moderator
User avatar
Joined: Thu Mar 23, 2006 9:31 pm

Re: Not quite F1, but almost (FSAE project)

Post by flynfrog » Sat May 10, 2008 2:45 pm

Jersey Tom wrote:More advanced than 'most actual companies' ? What does that even mean? I'm sure its 'more advanced' than CFD McDonalds does.. but I'd imagine its nowhere near that of a Formula 1 team.

CFD is nice but it isn't hard test data.

How consistent are your drivers? Are they able to put down competitive laps and be within 0.2 sec of each other?

How many tenths faster do you run with aero package than without, in autocross? Is the difference definite given the driver consistency?

Does it make any difference in skidpad times?

How much slower does it make an acceleration run?

That's the only stuff I'd be interested in. Or if not that, what validation and verification have you performed? In a rolling wind tunnel, how well would your CFD match to real life? In the straight ahead position? Does it capture yaw sensitivity? Pitch sensitivity?

What uni are you with? I'll likely be around at competition in Detroit.
Bring enough wheels this time :lol:

The wings advantage is as JT pointed out is greatly dependent on how good your drivers are and what course they decide to setup.


Personally I would like to see them set up more realistic racing courses I don't think there is anybody with an open wheel car in SCCA who wouldn't run a wing if given the chance. The mickey mouse course they use at comp are not really representative of what you see at an autoX

AeroGT3
0
Joined: Wed Mar 29, 2006 10:22 pm

Re: Not quite F1, but almost (FSAE project)

Post by AeroGT3 » Sat May 10, 2008 11:52 pm

Jersey Tom wrote:More advanced than 'most actual companies' ? What does that even mean? I'm sure its 'more advanced' than CFD McDonalds does.. but I'd imagine its nowhere near that of a Formula 1 team.
The optimization, meta-modeling, and automation process is very close to what F1 does. The detail of the models is not, but the design processes we've wrapped around CFD are.

More than most companies means that in talks with Lockheed Martin's conceptual design group, Boeing, General Atomics, and other aerospace companies, we've been ahead of the curve in CFD. Hardly McDonald's :roll:
CFD is nice but it isn't hard test data.
I love the fixation people have with "real" test data. People will take CFD that is 2% error over a wind tunnel test that has a 5% error bar . . . unfortunately.

Our CFD is within 2-3% of the wind tunnel data, and well within the error bar of the experiment - so it may well be, and probably is, more accurate than real world data considering the things that can be done in CFD that can't be done in the tunnel.
How consistent are your drivers? Are they able to put down competitive laps and be within 0.2 sec of each other?
Well, we've got an SCCA national champion and the german national karting champion. They're quite fast, and more than consistent enough.
How many tenths faster do you run with aero package than without, in autocross? Is the difference definite given the driver consistency?
Yep.
Does it make any difference in skidpad times?
Yes but not much.
How much slower does it make an acceleration run?
The change is negligible. We have a literally 3-4 newton drag setting for acceleration. Both wings are under 6 pounds so its a very small penalty.
That's the only stuff I'd be interested in. Or if not that, what validation and verification have you performed?
NATO data, comparison with several SAE published experiments, and our own data. All within the error bar for each experiment. this includes tests run with rolling ground planes.
What uni are you with? I'll likely be around at competition in Detroit.
We are going to UK and West, not Detroit.

flynfrog
Moderator
User avatar
Joined: Thu Mar 23, 2006 9:31 pm

Re: Not quite F1, but almost (FSAE project)

Post by flynfrog » Sun May 11, 2008 3:17 am

AeroGT3 wrote:
Jersey Tom wrote:More advanced than 'most actual companies' ? What does that even mean? I'm sure its 'more advanced' than CFD McDonalds does.. but I'd imagine its nowhere near that of a Formula 1 team.
The optimization, meta-modeling, and automation process is very close to what F1 does. The detail of the models is not, but the design processes we've wrapped around CFD are.

More than most companies means that in talks with Lockheed Martin's conceptual design group, Boeing, General Atomics, and other aerospace companies, we've been ahead of the curve in CFD. Hardly McDonald's :roll:
CFD is nice but it isn't hard test data.
I love the fixation people have with "real" test data. People will take CFD that is 2% error over a wind tunnel test that has a 5% error bar . . . unfortunately.

Our CFD is within 2-3% of the wind tunnel data, and well within the error bar of the experiment - so it may well be, and probably is, more accurate than real world data considering the things that can be done in CFD that can't be done in the tunnel.
How consistent are your drivers? Are they able to put down competitive laps and be within 0.2 sec of each other?
Well, we've got an SCCA national champion and the german national karting champion. They're quite fast, and more than consistent enough.
How many tenths faster do you run with aero package than without, in autocross? Is the difference definite given the driver consistency?
Yep.
Does it make any difference in skidpad times?
Yes but not much.
How much slower does it make an acceleration run?
The change is negligible. We have a literally 3-4 newton drag setting for acceleration. Both wings are under 6 pounds so its a very small penalty.
That's the only stuff I'd be interested in. Or if not that, what validation and verification have you performed?
NATO data, comparison with several SAE published experiments, and our own data. All within the error bar for each experiment. this includes tests run with rolling ground planes.
What uni are you with? I'll likely be around at competition in Detroit.
We are going to UK and West, not Detroit.

I think the real question is: Is the car any faster or not? At the end of the day thats all that matters

AeroGT3
0
Joined: Wed Mar 29, 2006 10:22 pm

Re: Not quite F1, but almost (FSAE project)

Post by AeroGT3 » Sun May 11, 2008 3:40 am

flynfrog wrote:I think the real question is: Is the car any faster or not? At the end of the day thats all that matters
No, the wings are slower and we determined that through track testing. So we left them on. :roll:

flynfrog
Moderator
User avatar
Joined: Thu Mar 23, 2006 9:31 pm

Re: Not quite F1, but almost (FSAE project)

Post by flynfrog » Sun May 11, 2008 4:30 am

AeroGT3 wrote:
flynfrog wrote:I think the real question is: Is the car any faster or not? At the end of the day thats all that matters
No, the wings are slower and we determined that through track testing. So we left them on. :roll:
I think you missed my point you can do all the CFD in the world if it doesn't make the car faster it doesn't really matter

that is also what Tom was trying to say as well you went on to defend you cdf.

Im not attacking wings of FSAE cars I think in a few years it will be tough to win without them. But I have also been to the FSAE comps and know it takes more than a good car on paper to win

Jersey Tom
257
Joined: Mon May 29, 2006 7:49 pm
Location: Huntersville, NC

Re: Not quite F1, but almost (FSAE project)

Post by Jersey Tom » Fri Jun 13, 2008 2:59 pm

My guys did indeed bring enough wheels this year.. though they decided to go bowling for cones in enduro.

I have yet to see laptimes showing any appreciable gains for aero cars in FSAE! There are cars with aero packages that are fast, but you can't say that isn't part tire or part driver. There are cars without any aero that are just as fast, or faster.

Laptimes, and DAQ showing pushrod loads with and without wings. That's the only way I'd be convinced its working.
Grip is a four letter word. All opinions are my own and not those of current or previous employers.

AeroGT3
0
Joined: Wed Mar 29, 2006 10:22 pm

Re: Not quite F1, but almost (FSAE project)

Post by AeroGT3 » Fri Jun 13, 2008 6:59 pm

Jersey Tom wrote:I have yet to see laptimes showing any appreciable gains for aero cars in FSAE! There are cars with aero packages that are fast, but you can't say that isn't part tire or part driver. There are cars without any aero that are just as fast, or faster.
That's a pretty rubbish argument. By that logic I could attribute what a team calls great suspension to chassis stiffness or driver or aero :roll:

There are a number of SAE papers about aero on FSAE cars. Most of them have full car testing, strain gauge data, and laptimes with and without aero.

YOU don't see the aero teams testing with and without aero and so you assert the difference must be zero. Do you really think aero teams don't run their cars without aero to get a baseline, or do so and leave the wings on because it makes their car slower?

Jersey Tom
257
Joined: Mon May 29, 2006 7:49 pm
Location: Huntersville, NC

Re: Not quite F1, but almost (FSAE project)

Post by Jersey Tom » Fri Jun 13, 2008 7:31 pm

A rubbish argument it is not. There's a massive amount of variability in those cars.. the engines, the tires, suspensions, overall weight, drivers.. they're massive. As such, you can't contribute say the fact that UMR won Auto-X to the fact they have wings, particularly when Wisconsin was +.044, which at this level especially is very very close. Or that RIT was -.099 on UTA's finely developed long running aero package.

And you're right. You could look at UWA's fast times and if you didn't know any better, just attribute it to their trick suspension. They also have two prime drivers, one of which does Australian Formula 3, and another has been driving all their cars since 2004. Compare that to some competing teams whose drivers have maybe a day or two total seat time, at that. So which is it.

And I don't put anything past anyone. Very few schools even benchmark competing tire brands, which is the easiest and biggest bolt-on performance modification. I have seen very good looking underbody aero setups that people have admitted to me are purely for show. This was a TOP team mind you that had won in recent years, and had no idea what effect the underbody had, if any. Purely aesthetic.

Maybe hard numbers are out there to laptimes and loads with and without aero. I haven't seen them yet, and I don't believe or trust anything until I see it.. particularly with race teams, and at this level. And I don't recall there ever being a year where aero cars have dominated the dynamic events. Lots of aero teams fall way slower than their non-aero competitors. Until I see faster speeds in general and aero cars pulling away from the rest of the pack, I'm not convinced they're good or bad.
Grip is a four letter word. All opinions are my own and not those of current or previous employers.

RacingManiac
9
Joined: Mon Nov 22, 2004 1:29 am

Re: Not quite F1, but almost (FSAE project)

Post by RacingManiac » Fri Jun 13, 2008 10:13 pm

Agree on all points by JT, now that I've been to 6 of the Detroit and 3 of F. Stud in UK. The judgement on aero is very hard to establish especially when for every uber-fast aero monster there is one just as good without it. And with driver variability as much as FSAE I think it is just as useful to building a "good car" as to have good driver in a mediocre car. I remember Leeds in 2006's F. Student have a driver that's 3 sec faster than anyone at comp, including their 2nd enduro driver(who is a whopping 12 sec a lap slower). 2007 comp was the only comp I've been to where aero seems to have a distintive advangtage in autocross. When I think Oklahoma, Rolla, and UTA were like 1.5 sec clear of the non-aero car in AutoX. But comes endurance day with the enduro track being way tighter the cars like UWA and Graz just smoked everyone(and luckily enough we got a 3rd with the car stuck in 3rd gear). What I will say for 2008 comp that I saw at MIS was that the aero cars are no longer the just sub-500lb car they used to be. I think Missouri Rolla(sorry Missouri S&T) weighs in at ~430 ish, and UTA courtesy of their new Aprilia 550 is now at ~400-ish. So that weight disadvantage is slowly going away. Then again, this year's comp also see the autocross being split with wing and non-wing car going just as fast in both enduro and AutoX, and without the massive open area of the Romeo proving ground I think that's the trend for the next little while in MIS. As long as we don't have tracks like UK's track last year(on the AutoX track there was a 5 or 6 cones slalom, done 2 ways!, plus 3 or 4 other slalom and just as many left-right/right-left gates, you need a single cylinder car to be fast on that thing, which RMIT smoked everyone).

alex1015
0
Joined: Wed Apr 16, 2008 4:38 am

Re: Not quite F1, but almost (FSAE project)

Post by alex1015 » Sat Jun 14, 2008 4:36 am

flynfrog wrote: But I have also been to the FSAE comps and know it takes more than a good car on paper to win
Got that right. A moderately fast car can do damn well as long as its plenty reliable. There have been plenty of damn fast cars I've seen that seem to just spend the entire competition on jacks. In theory they're fast...

EDIT:UTA with the Aprilla weighed in at 395 lbs. Which is not bad considering it was still a steel tube car. At VIR though they couldn't seem to get the Aprilla sorted. I don't know how they did at Detroit though. They may well have had it running well by then. And Missouri S+T with the wings was 451 lbs, I talked with a team member of theirs, like most they ran the Yamaha R6 motor (600cc) they were talking about go dry sump but they kept killing motors cause they got bubbles in the lines. I didn't ask if they had a swirler or not.

slimjim8201
12
User avatar
Joined: Sun Jul 30, 2006 5:02 am

Re: Not quite F1, but almost (FSAE project)

Post by slimjim8201 » Sat Jun 14, 2008 7:04 am

What bothers me about some of the teams that use Aero packages is that they carry over the previous years design and then pawn it off as their own. If I were a judge at competition, I'd immediately know which teams used carry over aero packages (and other parts/packages as well). Unless they could prove to me that their identical looking parts are substantially different, I wouldn't even consider them for big design points.

The point of FSAE is missed often times. It has so little to do with winning and so much more to do with a budding engineer's first semi real world project. Dead-lines, concepts, designs, reports, prototypes, testings, failures, solutions. I'd rather see a well designed and documented approach with new and innovative ideas and solutions fail than the same school win with the same car every year.

It's funny that some advisers don't allow their students to deviate from the beaten path because it reduces the chances of winning competition. It's not like they are going to lose sponsor money by not finishing the race. It only hurts the students.

What a rant! :D

flynfrog
Moderator
User avatar
Joined: Thu Mar 23, 2006 9:31 pm

Re: Not quite F1, but almost (FSAE project)

Post by flynfrog » Sat Jun 14, 2008 11:24 pm

slimjim8201 wrote:What bothers me about some of the teams that use Aero packages is that they carry over the previous years design and then pawn it off as their own. If I were a judge at competition, I'd immediately know which teams used carry over aero packages (and other parts/packages as well). Unless they could prove to me that their identical looking parts are substantially different, I wouldn't even consider them for big design points.

The point of FSAE is missed often times. It has so little to do with winning and so much more to do with a budding engineer's first semi real world project. Dead-lines, concepts, designs, reports, prototypes, testings, failures, solutions. I'd rather see a well designed and documented approach with new and innovative ideas and solutions fail than the same school win with the same car every year.

It's funny that some advisers don't allow their students to deviate from the beaten path because it reduces the chances of winning competition. It's not like they are going to lose sponsor money by not finishing the race. It only hurts the students.

What a rant! :D
I disagree id like to see the design judging taken out of the competition. This would eliminate cars that place top 5 in judging but fail to run any of the dynamic events.

IMO The scoring should include only the dynamic events.

RacingManiac
9
Joined: Mon Nov 22, 2004 1:29 am

Re: Not quite F1, but almost (FSAE project)

Post by RacingManiac » Sun Jun 15, 2008 8:29 am

that would be wrong IMO, because it would take the engineering out of the engineering comp and just turn it into a driving event. As said people can just build a mediocre car that have no "design", or may simply be a copy or a legacy design, and just have people who can drive and the team who may or may not understand why somethings are done, and just out run people. Then you can have people who have done their R&D and such but just don't have the luxury of "hot shoes" and have a poor result. FSAE is a design comp first and foremost, and from an employer or industry people's POV, hiring hot shoe is not what they want...they want engineers....

and at any rate, top 5 design car that don't place well in dynamic don't win overall anyway.....so actual balance is usually restored....

flynfrog
Moderator
User avatar
Joined: Thu Mar 23, 2006 9:31 pm

Re: Not quite F1, but almost (FSAE project)

Post by flynfrog » Sun Jun 15, 2008 2:30 pm

RacingManiac wrote:that would be wrong IMO, because it would take the engineering out of the engineering comp and just turn it into a driving event. As said people can just build a mediocre car that have no "design", or may simply be a copy or a legacy design, and just have people who can drive and the team who may or may not understand why somethings are done, and just out run people. Then you can have people who have done their R&D and such but just don't have the luxury of "hot shoes" and have a poor result. FSAE is a design comp first and foremost, and from an employer or industry people's POV, hiring hot shoe is not what they want...they want engineers....

and at any rate, top 5 design car that don't place well in dynamic don't win overall anyway.....so actual balance is usually restored....
thats true but I feel it makes teams over complicate designs. If you can get a car running a month sooner to me is worth more than a month of R&D but this is not taken into account in the judging. Racing is just as much cut and try as it is design. I have also seen cars take top 5 in design that wouldn't pass braking test. Maybe keep the design judging but have it as a multiplier of your dynamic score.

As an example we got nailed for using all hard line for our fuel lines. A design judge asked why we didn't have a rubber line in there to take shocks from the fuel pump. We told him in tuning we didn't see a need for it. He didn't like the answer even though we were one of the best running highest hp best torque curve cars. You can have the best design Idea in the world if it doesn't work it doesn't work you shouldn't get points for it

slimjim8201
12
User avatar
Joined: Sun Jul 30, 2006 5:02 am

Re: Not quite F1, but almost (FSAE project)

Post by slimjim8201 » Sun Jun 15, 2008 8:57 pm

flynfrog wrote:thats true but I feel it makes teams over complicate designs. If you can get a car running a month sooner to me is worth more than a month of R&D but this is not taken into account in the judging. Racing is just as much cut and try as it is design. I have also seen cars take top 5 in design that wouldn't pass braking test. Maybe keep the design judging but have it as a multiplier of your dynamic score.
It doesn't seem like it at the time (and it didn't for me when I was involved with FSAE), but the whole competition has so little to do with racing and the racing industry. It is a design competition 1000x more than it is a racecar engineering competition. I think the current scoring system is quite appropriate. Without placing well in the static events, a team will not finish in the top ten, but a large percentage of the points are based on the "fun" dynamic events.

This satisfies the boy racer in the budding engineers but still requires them to provide a sound design.

What percentage of FSAE engineers end up in the racing industry? The answer to that question gives a good indication of what the competition is all about...