CVT selection

[oz_olly]

Just want to know from all the CVT teams out there how to get access to CVT clutches. As you can imagine there aren't too many wrecked snow mobiles in Australia. I am interested to know how you guys decided which one to use. I have found loads of info on clutch tuning kits and parts but nothing on purchasing the actual clutches. I would particularly like to hear from Rob Woods or Joe Cheng if they still post here (anyone know how to contact either of them?). Also if any of you guys have drawings of the clutches or know where to find them I would be extremely greatful.

Cheers

Olly

[Parker]

Most Polaris ATV's will have them. if you want to get a new one that would be more suited to your powerband, contact Team Industries. they have TONS of information and I saw a few teams last year using their stuff too.

http://www.team-ind.com/

[rjwoods77]

Depends on a bunch of things but here is a quick rundown.

Choices:

Whats your power level? Under 50hp and a briggs jr dragster clutch will work good for you but only after some additions that make them upshift as well as downshift. Helper springs on the fulcrum arms on the primary and a caged helix on the secondary will make he clutch work both ways. Otherwise they are pnly really suited for acceleration runs.

More than that you will probably need to use a snowmobile clutch. Many different types with many different advantages but you are best off sticking with a polaris p-85 or p-90 clutch due the huge variety of parts and belts available to them. The BRP tra clutch and actic cat clutches has some real interstign features but stick to what is easly available. Get a team industries roller secondary with a caged helix they call a twin trax. The primary is a stcok unit but change the weights and springs. Aaen makes this crazy 4 tower roller primary which is the shit but crazy bucks it is.

The dearborn/mini baja route is to use a comet 700 series clutch(770-780-790) or a CVT TECH IBC powerblock setup. They both work well in baja cars and has brought dearborn some great finishes but the overall performance potential and/or hp limit is in question. They really are kinda "sticky" compared to other cvts. UB mini baja which has been a top 20 team for 5 years has been grenading the helix on the secondary of their 790 for 3 years now. Even the steel rienforced ones but it is due to overrun and the secondary catchiup and smashing the cam buttons off. We call it slap. We have a video of the sound of it happeneing but only really happens at Ohio in the woops.

Power:

Debateable what the max rateds are but the aforementioned is close.

Cost:

Comets are dirt cheap. Call Roy Navarete at Quality Drive Systems.

Call polaris for a polaris and get a free base model. Go look on the mini baja area for polaris parts requests then call team industries.

There are a number of briggs jr. dragster clutches out there and I dont wanna say who we get it from but its not hard to figure out. But they are the most expensive.

Size and Weight:

Polaris primary are heavy(6 lbs and 10lbs) and big (9" and 12-13")

Jr Drag clutches are real light(2.5lbs and 5lbs) and small (5.5" and 7")

Comets are half way in between.

So here is the big deal. So I thought I was done with FSAE due to crazy personal problems. But I cant leave a half finished car alone. But this time around we are going to actually calculate and do all the hard work since everyone is on the same page and working to secure the future of our organization. One of the projects is going to be datalogging the shift ratio of the clutch in order to figure out what is really going on. Simple rpm comparo between the crank and the jackshaft. Without this you will really be tuning in the dark. Keep in mind that when a cvt works great you will smoke fools. When it is a little off it doesnt work for shit. I can help with drivetrain layouts if you are looking for help. You can do some real intersting stuff with a cvt if you are using the right engine.

[oz_olly]

Thanks alot to Rob and Parker.

Some people may think I am absolutley crazy for thinking about attempting this but... I am looking to do some serious modification on our GSXR. One of our strengths is we can tune an engine really well. We came third in the Australian Accel event with a 270kg car plus driver. Put our drivers on a curvy track and they aren't so fast so I want to try and make it easier for them by not even having to consider gear shifts. I have done a fair bit of research to date and CVTs are looking like a pretty good tranny option.
What I am considering doing is lopping the engine off at the split between the upper and lower halves of the crank case. I would then design a transmission casing including dry sump that outputs off the primary to the first clutch then the second clutch comes back into the casing which drive a jackshaft and gear to the torsen. So it would almost be a CVT transaxle. It is going to be a major excercise in packaging but I think it is achieveable. By doing all this i could integrate suspension mounts, engine mounts, caliper mount and rocker and shock mounts. I think if done well it will produce a very nice result.

What is the deal with getting a free base model from polaris? I have priced some Arctic Cat stuff from DD racing it seems within the price level I expected. Is parts availability and tunibility much of an issue with Arctic Cat?

Cheers

Olly

[rjwoods77]

Not really. Jusrt figured cat stuff would be harder for you to come by. They arctually have a really nice reverse helix design for the secondary.

[jsmooz]

Rob did a pretty good overview. However I still disagree with the Jr. Dragster route. I highly oppose this option because those clutches are designed for acceleration only and not constant up/down shifts. I know Rob was working to modify them to make it work but you need a LOT of experience and I'm still not sure it'll work (Buffalo never made it to compeition to prove otherwise).

If you're looking to get something that'll work pretty decent the Comet's will work. We've used them for 2 years and have had no issues. Now we haven't spun them as fast as a 600 (we also ran a Briggs) would so I'm not sure what your application would result in.

The snowmoblie route is pretty good cuz there speeds are similar to what a 600 would output. The Polaris primary is pretty good because of the rollers, and the Artic Cat secondary is also pretty trick (I think a little better then the Polaris). The only think with these is that you'll need to figure out a setup that works for you. We've already done some of the datalogging using different components to see what happens. That can get expensive but really tells you alot.

FYI, CVT's require a different driving style that focuses heavily on smoothness. It, the CVT, really helps the driver focus on driving.


Bring it Rob. We're still waiting to have a good Briggs CVT'd vs Briggs CVT'd comparision. It'll be good to see how the different designs compare.

[rjwoods77]

Originally posted by jsmooz:
Rob did a pretty good overview. However I still disagree with the Jr. Dragster route. I highly oppose this option because those clutches are designed for acceleration only and not constant up/down shifts. I know Rob was working to modify them to make it work but you need a LOT of experience and I'm still not sure it'll work (Buffalo never made it to compeition to prove otherwise).

If you're looking to get something that'll work pretty decent the Comet's will work. We've used them for 2 years and have had no issues. Now we haven't spun them as fast as a 600 (we also ran a Briggs) would so I'm not sure what your application would result in.

The snowmoblie route is pretty good cuz there speeds are similar to what a 600 would output. The Polaris primary is pretty good because of the rollers, and the Artic Cat secondary is also pretty trick (I think a little better then the Polaris). The only think with these is that you'll need to figure out a setup that works for you. We've already done some of the datalogging using different components to see what happens. That can get expensive but really tells you alot.

FYI, CVT's require a different driving style that focuses heavily on smoothness. It, the CVT, really helps the driver focus on driving.


Bring it Rob. We're still waiting to have a good Briggs CVT'd vs Briggs CVT'd comparision. It'll be good to see how the different designs compare.

Hey Jim,

Me too buddy about the briggs vs briggs. Doing baja for a number of years has me familiar witht eh comet like it was an ex-girlfriend. The only thing that keeps a jr drag clutch from backshifitng well is the lack of springs on the fulcrum arms. SO just like the 790 you put springs on them and bingo. The difference is that the jr drag clutch uses these for a off throttle pre tension to keep the roller wieghts on the ramps but the main function of primary springing is actually done with a spring concentric to the hub. Our clutch manufacturer had a severe problem with their drivers "pedalling" the throttle down to track to make their index time because these their rails are way faster than the indexes. He put fulcrum springs on the and eliminated the DOF that was there. Damage went away and the drivers said it felt the clutch was always "connected". So based on this knowledge I felt comfortable making the decision to run that setup. The secondary isnt any different from any other cvt so that isnt a concern. However ours does have a roller unit on it and i think the compony just got done developing a caged helix after I pointed out the off throttle advantages of such a unit(the overrun helix disengagement) that may happen. If you give me our email i will send you the video. You can here it overruning and the helix smashing into the buttons. I always did have a bcakup plan. The center to center distance on the cvt we use is anywhere between 8.5 to 9.25. I have designed room in the car to swallow a huge 790 if I have to. Belt chart on a 790 has a 8.16, 9.41 and a 9.86 so I can always switch to a 790 if I need to but I suspect I wont have to. I may have big eyes but I am not a fool to design myself into a corner on unproven parts.You are absolutely correct that we havent gotten it done and it is all theory but I feel confident that it will work fine. Got a team with actual leadership and goals for the first time in 2 years so I hope it will turn out good. Cars already half done anyway.

[rjwoods77]

Originally posted by oz_olly:
Thanks alot to Rob and Parker.

Some people may think I am absolutley crazy for thinking about attempting this but... I am looking to do some serious modification on our GSXR. One of our strengths is we can tune an engine really well. We came third in the Australian Accel event with a 270kg car plus driver. Put our drivers on a curvy track and they aren't so fast so I want to try and make it easier for them by not even having to consider gear shifts. I have done a fair bit of research to date and CVTs are looking like a pretty good tranny option.
What I am considering doing is lopping the engine off at the split between the upper and lower halves of the crank case. I would then design a transmission casing including dry sump that outputs off the primary to the first clutch then the second clutch comes back into the casing which drive a jackshaft and gear to the torsen. So it would almost be a CVT transaxle. It is going to be a major excercise in packaging but I think it is achieveable. By doing all this i could integrate suspension mounts, engine mounts, caliper mount and rocker and shock mounts. I think if done well it will produce a very nice result.

What is the deal with getting a free base model from polaris? I have priced some Arctic Cat stuff from DD racing it seems within the price level I expected. Is parts availability and tunibility much of an issue with Arctic Cat?

Cheers

Olly


If you are gpoing to go through all that trouble to do all of that then be smart and turn the motor around 180 degrees from the way it normally sits in a typical cal. You wont have the jugs and cylinder head making a "space triangle" between you and the engine. Lean the cylinders along the angle of your firewall. It will necessitate a log manifold with single side entry but big deal. It will be you, firewall, engine and zero lost space. Also with the exhaust ducting out of the back of the car it frees you to do some cool stuff. Make an adapter that goes on the cranksnout on the other side from the stator. Thats where you will hook your polaris clutch up to because anything else wont survive the rpm or power since you are using a big engine. Then run the secondary cluthc above the diff that will now be sitting mear inches from the block. I would use a chain final drive in order to ture you gearing right.

You can do all that and swamp your team resources or make your life easy and use a polaris 500cc engine and cvt out of a polaris quad. Olav aaen is making 47 dynoed hp using a carb. http://aaenperformance.com/Power500.asp

I saw some 12.5:1 compreession pistons for that thing once but i cant remember where. Rememeber RMIT said they had 55hp on their engine and they won the comp if the power level is what you were worried about. Ayway, you'll save a bunch of weight overall on teh engine as well as the small size of the rear end of your car.

[Parker]

Im glad to see that others are using the CVT so well. Our first plan was to go with a hopped up Briggs motor and a CVT, but we got such a killer deal on our Suzuki motor that we switched up the plans.

[oz_olly]

A few weeks ago I was lucky enough to go to Oran park (circuit in australia) with our tech advisor who runs a production celica. The main draw card was the Formula 3 races. After looking at the way they do stressed engines, I became very inspired (stressed rocker cover). I was then reading Racecar Engineering V16N09 the one with the review of the two US comps and the british formula students. There was one team, no name mentioned in the caption but they had a Vanguard Racing sticker on the manifold (anyone know who this is). I noticed that they were running their CVT so that the belt is running in the east west plane. The two concepts combined in one almightly clanking of gears in my head and I think I have an even better concept now. I currently have four options for the implementation of a CVT. My thesis supervisor wants to do a thorough comparison of available configurations.

Here are my four concepts to date:
1. Keep engine standard run it off the end of the crank where the alternator is, then jack shaft to chain drive the diff.
2. As Rob suggested turn the engine around and do the same as above.
3. Try my original CVT-Transaxle idea, engine mounted east west, replace existing transmission with a custom housing for diff and integrate all the required components as I outlined in a previous post.
4. My current preffered option of turning the engine through 90 degrees running one the primary off the alternator end of the crank, use a structural rocker cover to assist in engine mounting which also mounts a jackshaft for the secondary. A toothed belt would then be used to return drive to below the crank (belt drive wheels allow for minor adjustments in overall ratio). The shaft running underneath the crank would then mount to a bevel gear and the diff. Shafts out the side of the crank case. External mounts would be used to integrate the shocks, rockers and brake caliper.

I definitely welcome feedback on any of the above options.

I really like the north south option because I feel it would allow me to lay the driver down more for same wheel base. We seem to have really tall guys in our team each year.

I'm currently avoiding studying for my exams that I am in the middle of so I welcome as much correspondence to procastinate over as possible.

To any of the Melbourne based teams a few of our guys (myself included) will be in Melbourne for two weeks from the 19 Nov till 01 Dec, for some training at Point Cook. If you would like some visitors we would love to invite ourselves. It's too close to the competition to steel your current ideas, but it would be good to have a chat as we are a bit isolated in Canberra.

Cheers

Olly
UNSW@ADFA

[jsmooz]

Olly - That picture in RCE was of our car (Michigan - Dearborn). Hopefully DanG will see this and send you a link to pics of our car he has. (I don't remember the url right now)

Our drivetrain is similar to your design except in the "east/west plane." We run a jacksjaft to a chain drive, to our stock ATV differential (including stock housing), which has an internal bevel. Dan should have pics.

[rjwoods77]

Olly,

Sounds neat but seems way overkill. You wont beat the packaging of turining the engine 180 degrees. You can almost use the engine a seat pan mount that way. You can mount the diff underneath the jugs/cylinder head if you are so interested in cutting wheelbase which seems to be the reason for running under the oil pan in a longitudal application. Stressed engine is cool but it is wasy more work the than standard and maybe wont work as well. Look at the alum sprocket thread. See how that team tied their rear roll hoop supports into the stock engine mounts. That will work great to hang the engine since 180 degreeing it will make those points about the same place of more toward the rear axle. The stock swingarm mount can be used to attach the engine at the firewall. Now that would be slick. Run the cvt inclined toward the rear to jackshaft supports mounted a a bunch of inches below the exhaust manifold bolts and then run a chain to the diff that you would mount between the engine and the rear suspension space frame. Again stressed applications are cool but is way more work than doing what i mentioned. the rear axle centerline would be under and ahead of the jugs/head.

Added this edit. Look at the aluminum sprocket discussion. Take that car, turn the engine 180 and cut 1 foot out of the wheelbase. You will have to find somewhere else to put the gastank but who cares since you only really need about a gallon anyway.You could hook your rear a-arms front joints intot he firewall depending on geometry which would avoid the need to stress member anything. SImple square tube rear subframe would do the trick.

[rjwoods77]

Olly,

I was just thinking. If you did want to make it complicated for complicateds sake then do this with the configuration I described.

Make a stressed oil pan than connectes at the firewall and runs toward the rear of the car that then loops upward to tie into the rear main roll hoop supports. You can easily make an integral diff carrier as well as upper shock mounts and rocker arm mounts depending on your geometry. Should also go the extra step and use this as a dry sump pan so you can get the engine nice and low. So team has already done a structural dry sump pan setup but I cant remember. Again if you want to get complicated with it. I am going to reintroduce my earlier suggestion as well. For all of this trouble you are going to go through to run a cvt why dont you just run a polaris 500 with the cvt that comes with it. You can probably get polaris to tos you the whole thing since they already give the cvts away for free to sae teams. You say you are good at making power so do it with this. Save yourself a ton of weight, complexity and money. Like I said RMIT won on 55hp. I guess my general point is dont look to make things more complicated when you can make them easy. RMIT and Texas AM are doing it very well. At this point running a big 4 cylinder is almost a "just because" proposition. Join the small engine/car trend. Thats where it is all going anyway. The polaris stuff is already sized and works well with itself. I have some junk ones in our shop that our clean snowmobil used if you want pics and sizing info. They turboed theirs witha gt12 and won that year. The sled also did wheelies just about. Made about 80hp on the turbo from what i remember. By the way the engine is actuall made by Fuji. It is a Fuji engine so it has good quality unlike everything else polaris makes with is twice as heavy as their competetors and breaks 4 times as often.

[Dan G]

Here's that shot from the magazine...
http://evilengineering.com/gallery/d/2824-2/IMG_0201.jpg (http://evilengineering.com/gallery/v/UMDracing/FSAE/2006/IMG_0201.jpg.html?g2_imageViewsIndex=1)

There's a few more shots on this page (http://evilengineering.com/gallery/v/UMDracing/FSAE/2006/?g2_page=3), I'm still working on uploading the rest of my archived photos to my new image gallery.