CVTs

[gug]

how many teams do run CVTs? our team is looking into one ourselves, but we are curious as to why more teams dont run them. ive seen some canadians who run them, but the website was in french! http://fsae.com/groupee_common/emoticons/icon_confused.gif
any comments on fuel efficiency?
anyway, some interesting research on the topic revealed: David Coulthard (thats probably not how you spell it) said in regards to the Williams '92(approx.) test car with a cvt "i'm very excited about this car... it is very smooth turning into the corners" and other positive things that i cant quite remember now. also, cvts were banned before they even got one race in, not even ground effects and active suspension did that good! http://fsae.com/groupee_common/emoticons/icon_wink.gif

if ive been an idiot, and there is already another topic on cvts, could someone point me towards it? thanks.

"I come from a land down under,
Where beer does flow and men chunder

[woollymoof]

RMIT? did one a couple of years ago and the judges thought it was crap!! But I think CVT's are a very good idea. If you tune the CVT and the engine to get the most out of the CVT then fuel consumption will be very poor compared to gearbox. But what's that worth, 50 pts?

[gug]

well, here are some of my reasons why i think they rock. i'm posting these here so people can comment on how much they agree/disagree, and to share any experiences they have had with cvts.

pro:
-smooth entry/exit to corners, allowing driver to stay right on the limit of traction
-no more missed/badly chosen gears
-obvious tuning advantages, both in the engine and intake/exhaust
-engine stays on peak rmp giving maximum power

con:
-small power loss. the only designs i have seen are friction/belt type, as on snow-mobiles (if anyone out there knows of any non belt type, please tell me!)
-packaging

i admit im being rather positive here, so if anyone wants to add to this list, please post

"I come from a land down under,
Where beer does flow and men chunder"

[Scott Wordley]

Joe Chueng is a guy from the states who regularly posts here.

He runs a CVT in his A-Mod car and apparently it is the about the quickest thing round. Aside from wings, going to a CVT was his other main suggestion when we chatted to him a while back.

Hopefully he'll post his thoughts

Regards,

Scott Wordley & Roan Lyddy Meaney
Monash FSAE Wingmen
http://www-personal.monash.edu.au/~fsae

[woollymoof]

I'm not sure that the losses due to the belt are really any more than that of the chain and gears in a normal box. Belts are pretty efficient, we're talking about 95%. Chains are about 98% I think, not sure what the extra drop in efficiency is due to the gears in the box.

[J. Cheng]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Scott & Roan:
Joe Chueng is a guy from the states who regularly posts here.
<HR></BLOCKQUOTE>
Canada is not part of the States. http://fsae.com/groupee_common/emoticons/icon_wink.gif

There are 3 things I would not give up on my car-my front wing, my rear wing and my CVT.

Imagine you have an 80 horsepower FSAE engine and actually spend well over 90% of your acceleration time at that power peak.

Imagine never having to worry your driver missing a shift again.

Imagine engine compression doesn't induce trailing throttle oversteer anymore.

Of course packaging is a problem, but isn't a significant part of any race car engineering is about solving packaging problem? To make it easier, Arctic Cat is currently making a 660 cc turbo 4-stroke 3-cylinder engine. http://www.arctic-cat.com/snowmobiles/flash_site/index.asp
I believe more 4-stroke engines are on their way, from Arctic Cat as well as other manufacturers.

It might not win the fuel economy test, but I think it's much cooler to have the fastest car at the event. http://fsae.com/groupee_common/emoticons/icon_cool.gif

Joe

[This message was edited by J. Cheng on May 01, 2003 at 11:26 PM.]

[gug]

just out of curiosity, where did you source your CVT from? im guessing that any a-mod would rip a snowmobile cvt to shreds, and i dont know of any high-power handling ones. is it still the typical belt drive? ive heard rumours of different designs, but after some searching i have not seen anything but the typical belt and variable sized pulley.

and could anyone from a team that do not run a cvt please give your reasons not to? all i can see is benefits for a small amount of time devoted to designing and installing it, but still they seem to be reasonably rare in f-sae.

"I come from a land down under,
Where beer does flow and men chunde

[Charlie]

Reasons not to run a CVT? What engine are you running? To run a motorcycle engine you have to cut the transmission off. Not too tasteful an idea. Snowmobile engines son't seem quite up to par with bike-level performance just yet (not in 4 stroke models anyway).

It's one more thing to package, if you need another gear reduction in addition, it gets really complex. Yes, of course these cars are a packaging exercise to begin with, but you can easily be talking about a necessary increase in wheelbase to accomadate the CVT. Not a desirable thing.

Teams have run (and won!) with CVTs in the past, but don't run them anymore. I'm sure they have more concrete reasons than I do, whether they want to share or not, who knows.

If anyone wants to give it a shot, go for it. But I don't think they are a golden egg.

-Charlie Ping
Auburn University FSAE 1999-present

[gug]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Charlie:
What engine are you running? To run a motorcycle engine you have to cut the transmission off.

-Charlie Ping
<HR></BLOCKQUOTE>
we are running a CBR600 F4. after having a bit of a look in the gearbox, we were thinking about fixing it in the gear we wanted, cutting off the rest of the gears to get rid of useless weight, and just having the cvt attached to the usual output.
of course, the potential is there to make the engine a bit shorter if we do cut off the gearbox...
anyway, as you can tell this is still very preliminary stuff, we are planning to put it on the car that will run in dec '04. thanks for the alternative viewpoint though.

to answer my own question earlier, see this website if you want to know about different types of cvts:
types of cvt & reasons to run a cvt (not f-sae specific) (http://cvt.com.sapo.pt/why/why_cvt.htm)

"I come from a land down under,
Where beer does flow and men chunder"

[Marc Jaxa-Rozen]

FWIW, Sherbrooke (the Canadian guys) are running an F4 with their CVT...I'm not sure about packaging problems but their car goes like a bat out of hell http://fsae.com/groupee_common/emoticons/icon_smile.gif

I wouldn't be very surprised to see more CVTs being used in the next few years.

[DoItDoug]

...if you know how to tune them. We ran a CVT back in 1991(?) when we won and it was great, but our tuner graduated and left no follow-up info. It would take a dedicated student or two to do a LOT of testing. It may take the fun out of a sequential shifter, but it would tear up the track. Plus, you wouldn't have to have the greatest driver's in the world either.

Packaging is the other issue. A CVT will add some wieght, but if you can take out the stock transmission gears, you may even out.

I guarentee we start seeing them in Detroit soon...

-DT

Doug Temple
'02 VT FSAE
www.vmotorsports.com (http://www.vmotorsports.com)

[loose_nut]

Sorry, I feel like we're entering this discussion a little late. We at FSAE Sherbrooke run a CVT. It's from CVTech, Bombardiers CVT manufacturers. They gave us a prototype of their latest model. Like I mentionned in a previous post, packaging isn't easy. I had to cram everything together to get a 72 inch wheelbase. The other thing is you're adding weight to the back of the car which makes it difficult for weight distribution at that wheelbase, but we managed. In 99-00 we ran a CBR 600 F2 which was easy to modify, because the cover was perpendicular to our output shaft. This year we have an F4 and had difficulties because the cover is at a 15 degree angle, making it harder to modify. Why do we go to all this trouble? Because in 99 we came in third place in acceleration with a 500+ lbs car with no turbo or wings. We're a very small team with a tiny budget so being a few miles from the Bombardier factory is the best way for us to have an edge. Hope you'll get to see our latest creation in (gulp!) 12 days.

Keep racing
Michel Morin

[J. Cheng]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by gug:
...im guessing that any a-mod would rip a snowmobile cvt to shreds...is it still the typical belt drive? ive heard rumours of different designs....<HR></BLOCKQUOTE>
A typical off-the-shelf OEM snowmobile CVT belt could handle slightly over 200 hp all day long as long as you keep them cool (no problem on an open wheel formula car with no engine bay cover, just make sure there is a rain shield to prevent water from getting onto the friction surface, don't ask me how I found out). Currently, the highest output snowmobile 3-cylinder engines are running over 300 hp using custom CVT and wider non-OEM belts. CVT consists of a drive clutch and a driven clutch. In terms of design, drive clutches are very similar between brands. In fact, some of the them can even use each other's flyweight. Driven clutches are more different. For high performance applications, most people use Polaris or ArcticCat, with ArcticCat being the favorite the last few years. Reason is AC puts their torque-sensing helix on the outside face, allowing adjustment without having to take the whole unit off.

Two books have been written on belt driven CVT. The older one is written by Olav Aaen, who is a mechanical engineer and the co-inventor of the modern day torque-sensing driven clutch. He supplied my engine/drivetrain on my last car and he races a D sports racer himself with a CVT for R&D purpose. You can get his book directly from him, http://aaenperformance.com/ .

The second book is written by Dale Cutler. He is an ArcticCat dealer and tuner. I haven't read his book but it's avilable at http://www.cpcracing.com/ .

Back in '93 when I first started work on my last car, an autox friend who have been snowmobiling for years told me I should look into using a CVT. I was skeptical at first but decided to check it out. I took a winter vacation at Yellow Stone Park, rented one and rode around in the mountain for a week. It was pretty impressive. For such a simple device, it works amazingly well.

In terms of packaging, the latest AMod car from Todd (former FSAE) and George Bowland (smart guys)place the jack shaft higher and more forward (at almost the 1 o'clock position rather than at the 3 o'clock position). This allows a shorter crankshaft to diff c/c distance. I guess I will be copying that feature in my new car. http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

Joe

[gug]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by woollymoof:
Belts are pretty efficient, we're talking about 95%. Chains are about 98% I think, not sure what the extra drop in efficiency is due to the gears in the box.<HR></BLOCKQUOTE>

sorry to ask you after so long, but where did you get these figures from? is the 95% just for general belt drives or cvts in particular? normal belt drives would be efficient, but everything is just so slippery in a cvt to allow the belt to move up and down in the pulley, do you think your figure is correct for cvts?
i do value your opinion though, so if you still want to have a go at guessing cvt efficency (or anyone else), give it a shot. ill give a prize of kudos to anyone who gets the closest when i finally manage to find some solid data. http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

"I come from a land down under,
Where beer does flow and men chunder"

[woollymoof]

I'm going off normal belts. But, I'm not exactly of what type of CVT you're thinking of. If its the two cone idea I'm probably very wrong, but if it is the variable size pulley type I'm not sure that there would be much difference compared to a fixed one. But then this is all speculation, I wouldn't have a clue if I'm right or wrong. I'm awaiting your solid data. http://fsae.com/groupee_common/emoticons/icon_smile.gif

[James Waltman]

gug,
I'm not sure why you think the V-belt type CVT is slippery. The belt doesn't really slip if everything is working properly. Your list of cons includes a small power loss. All things considered they are more efficient because the engine is always operating at peak performance and you don't interrupt that with shifts (especially missed shifts).

Hope this helps:
CVT Mechanism Efficiency Range
Rubber Belts 90-95%
Steel Belts 90-97%
Toroidal Traction 70-94%
Nutating Traction 75-96%
Variable Geometry 85-93%

Source: http://web.mit.edu/klang/www/cvt.pdf

James Waltman
waltmaj@cc.wwu.edu
http://dot.etec.wwu.edu/fsae/
Formula SAE
Vehicle Research Institute at
Western Washington University

[gug]

thanks for that link James, its a real goldmine. i thought that v-belt cvts would have high frictional losses purely on instinct, guess my instinct was wrong! (wouldnt be the first time, ive had some horrible rejections from women due to that damn instinct! im sure the beer had nothing to do with it...)

"I come from a land down under,
Where beer does flow and men chunder"

[O]

Boy, I dunno. Back in the early 90's, our mini baja team tested their CVT's extensivly and found off the shelf CVT's were transmitting about 30% less peak power than a direct drive.

O

[Disco]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by woollymoof:
RMIT? did one a couple of years ago and the judges thought it was crap!! But I think CVT's are a very good idea. If you tune the CVT and the engine to get the most out of the CVT then fuel consumption will be very poor compared to gearbox. But what's that worth, 50 pts?<HR></BLOCKQUOTE>

Sorry not us. RMIT has never run with a CVT. I got no idea what your talking about.

[gug]

living in the sunny land oz, we dont get many snowmobile wreckers over here. also, it seems that ATVs (quads) with cvts dont get wrecked very often either. does anyone have some useful contacts/reliable suppliers that they have sourced cvts from? and some way of making sure that we dont get a dud through mail-order? any help will be appreciated.

"I come from a land down under,
Where beer does flow and men chunder"

[J. Cheng]

This one is on sale, http://www.ddracing.com/category_search_show_items.asp?Cat_ID2=99&Cat_ID=22
Standard OEM Arcticat unit. D&D is a very reputable supplier in the U.S.

Here is another, http://www.aaenperformance.com/RollerClutch.asp
This one represents the latest in snowmobile CVT technology and the price reflects the same. Aaen also carries standard Polaris units at prices similar to the ones from D&D. Aaen is also a very reputable supplier. I have purchased from him since 1995.

Joe

[Daves]

If your car has a CVT, is it true that no gauges/buttons/lights are necessary on the dashboard and steering wheel?

I know that you'd need a tachometer for the noise competition, but many teams had that in their computer which they hooked up to the car. Also, I would like an engine jacket water temp gauge, but is anything else necessary that can't be put in the data aquisition system on a notebook computer?

Also, since snowmobiles come with a CVT and a 600 cc engine, why don't any teams just use the Arctic Cat or Polaris 600 cc engines?

[This message was edited by dave_s on June 13, 2003 at 07:02 PM.]

[Kevin Hall]

They don't come with nice small 4-stroke engines, they are either 2-stroke, or BIG 4-strokers, which are new this year.

Kevin Hall
University of Saskatchewan
'03-'04 Team Director

[MercerFSAE C. Burch]

Speaking of tachometer for the noise test, has anybody ever incorporated a "noise test" setting in their ECU that will keep their engine at the proper RPM's for the test? I'm thinking that it would be something like a pit-speed limiter in an F1 car, but for RPM's instead. If nobody's done it, it might be a neat idea!

Chris,
Mercer University - Drive!

[Daves]

Chris, that's an awesome idea. I've also planned to incorporate into our car for next year. If you read the July 2003 issue of Road & Track, you will notice an article entitled, "Atlantic Voyage" on page 82. On page 85, Douglas Kott writes, "I'll be using 9000 rpm today, short of the 10,000-rpm CART-imposed race limit . . ." If CART has a 10,000 rpm limit, and if Honda and almost every other carmaker has a rev limiter built in, then it is definitely a great idea to make an adjustable rev limiter. Then for the noise test, all a team would have to do is set the limiter at the necessary rpm and hit the gas as hard as possible.

However, since the throttle is mechanical, wouldn't it be easier to just have some kind of mechanical piece to insert to stop the throttle cable from pulling past a certain point?

[MercerFSAE C. Burch]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR> However, since the throttle is mechanical, wouldn't it be easier to just have some kind of mechanical piece to insert to stop the throttle cable from pulling past a certain point? <HR></BLOCKQUOTE>

No, that wouldn't work... Remember that there are all sorts of variables besides throttle position that will affect what speed the engine would turn. Air pressure, humidity, air temperature, fuel temperature, oxygen content of the air, air density... the list is quite long, actually, of the different variables that affect the speed the engine would turn with a constant throttle opening. What might work in Texas on a sunny day with 80 air temperature and 95 percent humidity would most likely not work in Detroit with 49 degree air and 40 percent humidity.

Chris

[Daves]

Besides being innovative and giving the electrical engineers more to do, what other advantages would this adjustable rev limiter have? So far, the only other advantage I can think of is that you can ensure that you are exactly at the appropriate rpm for noise testing, but this is only important if your engine is borerlining on 111 (or the 2004 rulebook level) decibels and raising the rpm the slightest bit will put you over.

Our car overheated in 2003 because of redlining in 1st gear. Theoretically, could we use our own rev limiter to protect our engine by lowering the redline? Should our car have been able to redline for 22 laps (endurance) without overheating? Would a larger radiator have prevented overheating?

What is more likely to have made it overheat?:
1. The engine being at redline too long.
2. The engine "bouncing" off the rev limiter too much.

[Kevin Hall]

Bouncing off the rev limiter builds serious heat, as it is generally an ignition cut-off. The TEC II has quite a few options as to how it can be set. The typical popping and banging is default, but after playing around and switching to 0* advance setting, it just sounds like it runs out of go at rev limit instead. It still builds heat, but not as drastically.

If you are bouncing off of it that much, you should probably be shifting quicker. The rev limiter should only be for protection. Set it low for driver training, at get used to the sound, as you don't have time to look at your tach, and then bump it up simply to keep your engine together at competition.

Any further questions, feel free to ask

Kevin Hall
University of Saskatchewan
'03-'04 Team Director

[woollymoof]

If the engine is sitting of the limiter - assuming this is done by cutting spark - then the engine will be farting and carrying on, this is probably not going to do much for your sound reading.
If, however, this was done by throttle by wire then you're set.

[gug]

woohoo, my topic lives again!

anyway, a question for those in the know. how much does the gears and gear selecting stuff and basically everything past the clutch weigh on the f4? approximate answers are good.

i was having a look inside the engine today, and by sheer quantity of metal, i think that everything past the clutch must weigh about 40% of the engine weight! lets see, 70kg is about right for the f4 total weight? that makes 28 kgs in gears and housing. not only does it weigh alot, most of it is rotating and alot of it is at higher rpm than the output.

so when a cvt is run, you can basically cut off the back of the motor, and run the output straight off the clutch?

dont worry, im not mistaking this for an easy job. you have to worry about new mounting points, oil flow, any support that could get removed from the clutch, and im sure there are a ton of other considerations.

does anyone have those weights? TIA

- if it isnt coming, you need a bigger tool.

[karter]

One of the many dangers when over reving motors is water pump cavitation. If the motor dosen't over heat when dirven below the red line then this is most likley the culprit.

[MikeWaggoner at UW]

"so when a cvt is run, you can basically cut off the back of the motor, and run the output straight off the clutch?"

I don't think you need the clutch with most CVT's. The belt just goes slack at low enough RPM conditions.
I actually wouldn't worry much about the oil. Just plugging it off would work. I think the real work would come from mounting the drivetrain to the engine in a good way w/out the rear of the engine, but it wouldn't be that tough. The CVT belts aren't super precision like gears.
You'd probably need another gearbox or chain in there somewhere to get the RPM's right for the CVT and final drive. Does your school have minibaja? They all use CVT's and you could learn some by looking at them. They're obviously not suitable since they're made for 10 horse Brigg's, but I think the general idea is the same in snowmobiles.

-Mike Waggoner
UW

[gug]

thanks Mike. no minibaja here, as far as i know there is no minibaja in australia.

a question for the canadians, do any of your snowmobiles have cvts that are actuated by stepper motors rather than all those flyweights? ive had the idea ticking over in my head for a while now, i just have to figure out how much force would be required to move the pulleys in and out, and if that would mean a 10hp stepper motor or not. anyone know/have an opinion?

- if it isnt coming, you need a bigger tool.