When to shift -- Acceleration Event

[Daves]

What rpm is ideal for shifting when accelerating as fast as possible?

http://www.letu.edu/formula

Formula Photos & Links (http://what3542.5u.com/dave/fsae/formula_photos.html)

[Daves]

What rpm is ideal for shifting when accelerating as fast as possible?

http://www.letu.edu/formula

Formula Photos & Links (http://what3542.5u.com/dave/fsae/formula_photos.html)

[Scott Wordley]

Multiply your torque curve by your gear ratios and plot against road speed. You should see cross over points which indicate that you are now making more torque (at the wheels) than the previous gear. Start by trying to shift there... but don;t blindly trust theory as its a little more complicated than that, log a few different shift point and see which is quicker.

Regards,

Scott Wordley

http://www-personal.monash.edu.au/~fsae

[RaID]

well according to Chalmers at FSAE-A
dont change gear at all, thats
what go them a 4.05

"Pain is an illusion"
TOOL

[Scott Wordley]

On their quickest run (4.02?) they made one shift, 1st to 2nd obviously

But on a second gear start they also pulled a 4.1. Incredible.

Regards,

Scott Wordley

http://www-personal.monash.edu.au/~fsae

[Max_camber_gain]

How heavy was the chalmers car? and what was the quoted Hp? I am thinking those rims must have hepled them a lot too, that must have been a big reductin in inertia.

I know that wollongong started in 2nd gear, shifted twice and a 3rd time over the line. I think they were shifting around 10500 - 11000 because that is where the torque ended.

cheers

[Daves]

First, congratulations to Chalmers for making such a fast, sexy car. How do you pronounce Chalmers?

There are 2 things I would like to know about the Chalmers car:
1. Overall weight.
2. What size sprockets they used/final drive ratio. All they have to do is specify their car's top speed.

Their car must be light (under 475# or 215 kg) due to the carbon fiber wheels, rear aluminum frame (instead of steel), and general lack of material.

The Mitsubishi sticker on their front windshield looks out of place, and was the duct tape really necessary to cover their front grille?

I appreciate their making their own carbon fiber/aluminum wheels, but why make them look like wagon wheels? Also, are they 13 inch in diameter? Are they easy to replace/duplicate?

http://www.letu.edu/formula

Formula Photos & Links (http://what3542.5u.com/dave/fsae/formula_photos.html)

[Johjes]

dave_s, I´ll try to answer some of your questions:

Put together shall as in What shall we do with mesh as in FEM calculations http://fsae.com/groupee_common/emoticons/icon_smile.gif
Then you get shallmesh = pronounciation of Chalmers

1. Overall weight is 198 kg (wet without driver)
2. Our sprockets (stock R6 gearbox) 16T and 56T gives a final drive ratio of 3.5

[Max_camber_gain]

WOW!!! that is really light guys! No wonder it hammered up the track... Congrats!
That car was so well packaged, low and sexy... I really hope to see you again in either detroit or aussie land soon!

[gug]

wow, thats nothing like what i pronounced Chalmers. lucky you didnt tell me in person, i would have thought you had just come back from one of the wine tasting tours. http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

obviously the photos are not doing the Chalmers wheels justice, they looked awesome. i think it wouldnt be too hard to make something similar to those wheels, but our carbon freak (who had a chat with the Chalmers guys) says that to make them as light would be a work of magic.

- the problem with the world is stupidity. i'm not saying there should be a capital punishment for stupidity, but why don't we just take the safety lables off of everything and let the problem solve itself?

[vinHonda]

Perhaps you can confirm Johjes:

a lotta testing/practice, and nice warm tires got your time down by 2 tenths from FStudent? Do you think u can get it into the 3's?

Vinh

University of Toronto Formula SAE Racing Team
www.fsae.utoronto.ca (http://www.fsae.utoronto.ca)

[Johjes]

For the UK event we focused more on testing (car and components) than driver practice. Now the drivers know the car better.

The tires used in the UK where too hard and never got the temp up. I wasn´t in Australia but I guess they used the new softer BF Goodrich tires and I guess the climate helps rising temps

I don´t know but 3 hundreds isn´t much

[Daves]

Using our 2002 Honda F4i torque curve with a 14,500 rpm redline (Chalmers uses an R6 engine with a 15,500 rpm redline), here's what Chalmer's ideal acceleration run with their 3.5:1 final drive ratio and 198 kg mass would look like. The time is 3.51 seconds, neglecting aerodynamic forces, the tires' ability to gain traction, and many other factors. So 3.51 seconds would be their ideal unless more power was available (kind of like a Carnot cycle efficiency).

With all the forces on the car, you can see why Chalmers only made it to 2nd gear in the FSAE-A acceleration event -- they designed it that way. They could improve by using smaller front and rear sprockets, lowering the overall weight and rotating mass. I like the final drive ratio of 3.5.

http://what3542.5u.com/dave/fsae/displacement.jpg
http://what3542.5u.com/dave/fsae/velocity.jpg
http://what3542.5u.com/dave/fsae/rpm.jpg
http://what3542.5u.com/dave/fsae/acceleration.jpg

http://www.letu.edu/formula

Formula Photos & Links (http://what3542.5u.com/dave/fsae/formula_photos.html)

[Denny Trimble]

Cool plots. I know you said traction isn't considered, but I think 1.5g's forward accel is a bit much for a RWD FSAE car. Also, that's an interesting HP curve you've got (from the 1st gear accel trace).

I see you're using about .250s shift delay times, which matches our manual system performance.

Good work! I've got something similar in excel, with inertia, traction limiting, rotational inertia, and aero drag. I need to add tire rolling resistance, then correlate it with testing data.

University of Washington Formula SAE ('98, '99, '03, '04)

[Andreas]

Hello!

My first posting on this forum...
First I'd like to thank everyone for an awesome competition, we had a great time, the second place is just a bonus. But we couldn't make it here without Pat and the Monash guys help. Scott and his team has help us with EVERYTHING, thank you very very much, or i Swedish: "Dra at helvete"

About our car, in england, our car was 198 kilos wet. Down in australia we were 208 or 209 wet. Some kilos comes from the australian cooling kit and where the other extra kilos comes from we con't have a clue. We don't run BF goodrich and we never have. In australia we ran uniroyal which after all workt really well after tuning the car up for radials. The main difference on the acceleration runs in oz compared to england is shifting and the starts. In england I did the first two runs starting in first and shifted to second. My best run was 4.34. The Toronto guys hinted us to start in second and so we did with our second driver. Too much wheelspin limited our time to 4.33 in just second gear so I think with some more practicing in england we would get down to 4 seconds. We would really like a time below 4 seconds but winning acceleration is good enough

Thanks again for a very good competition

Beanie University of Technology

[Daves]

Denny, thanks for the compliments. What is a "good" limit for forward acceleration? My physics professor, Dr. Forringer (who wrote the C program I used to make the graphs in Excel), says that about 1.0g is the limit. Is this a good assumption?

Andreas confirmed that 1.5g's is too much because of the excessive wheelspin they have had. How tall do we need to gear these cars?

As to the shift delay times, I just guessed at 0.25 seconds. I have not seen any data from our alleged wheel speed sensors, but I'm glad to hear you confirm the number.

That's amazing that you input all those variables into your program! I found the weight of the car and the gear ratio to be confusing enough, but I'd be lost with all those variables. What I want to do is get some data from our last car to insert a fudge factor (call it drag or loss) to see how far off the program is. Then we can add it as a constant to the program and get a good estimate for this year's car. So far it looks like the program is about 13% off when comparing to Chalmer's car.

About the horsepower curve -- it has some big drops in the lower rpms. I don't know what they were due to.

What is the proper way to launch for acceleration? Should we rev the engine to 14,000 rpm and drop the clutch? If so, at about what rpm does the engine start at when the tires actually hook up.

Andreas, can you explain the launching techniques a little more clearly? Are you saying that you have wheelspin problems even when you launch in 2nd gear?

Do you rev the engine before letting out on the clutch to accelerate? Or do you start just above idle?

By the way, I enjoyed Chalmer's 2004 promotional video -- your car runs so smoothly. Are you coming to the states to race at all? I'd like to see Chalmers vs. N.C. State (my two favorite cars).

http://www.letu.edu/formula

Formula Photos & Links (http://what3542.5u.com/dave/fsae/formula_photos.html)

[Denny Trimble]

With the Goodyear tire data, max "mu" is 1.6, and with weight transfer and an iterative calculation to take into account load sensitivity, I'm getting about 1.1g's. But it depends on the weight distribution of the car, cg height, wheelbase, surface, etc.

I'm having trouble correlating my model to the Road and Track speed vs. time data, even with a wide range of drag and rolling resistance values. Once we get wheel speed sensors on our car I'll see if I can do better.

University of Washington Formula SAE ('98, '99, '03, '04)

[Max_camber_gain]

try around 1.2 g's, after around 0.15 seconds (once the tyres have spun up off the line to clear off the dust etc) Also, tyres never develop force without a slip angle... and slip angles need time to develop.

The general proceedure for starting that I have experienced, is drop the clutch around 9000 - 10000, wait till speed increases a bit, then put the pedal to the metal! (then pull on that gear lever like there's no tomorrow... which as we have been shown, is not the best thing to do!)

MCG

[Andreas]

dave_s

Our launch is to rev the engine and more or less drop the clutch. At the right moment we let of the accelerator to get grip. too late and we loose time on too much wheelspin, too early and we are to low in the powerband. Perhaps we can gain som time if we slip on the clutch instead of the wheelspin, but we haven't tried it. I think the wheelspin model is a easy way to get decent times. You can probably go faster by playing with the clutch but i think it's harder to learn it.

Andreas

Beanie University of Technology

[me]

Um, since we test on different surfaces we've had various different methods - all quite unreliable. The best approach so far has been full throttle and dump clutch. The only regulating variable is the rpm at which u dump the clutch.. i think our engine boys 'do' it at 6000 rpm http://fsae.com/groupee_common/emoticons/icon_wink.gif

Gives one or two revs of tyre then hooks up. And yes, the 15,000 rpm engine is the key.