Yeah, I think it could be punched out to 534 cc's, but I don't think it's worth it... But it would look better on the design report...

The design report is an interesting subject.

From what I read it does seem that different judges view the design requirements with various opinions. The design guidelines being for an affordable autocross car for the "average parking lot racer" seems to be taken strictly by some and totally ignored by others. Must be very frustrating for the teams.

However, if I were presenting my design using the Phazer engine, I would stress the advantages of using a stock snowmobile engine that the owner can simply switch to his snow machine for the winter months and then put back in his autocrosser for the summer seasons.

This would lower the cost of the car as it could be sold "less engine" for those in snow areas and make the Yamaha Phazer snowmobile a year round recreation. For those who don't want a snowmobile, the car would be sold with engine. I think the dual cost breakdown should impress the judges and gain some points for intelligent engineering.

If the judges are going by the practicality aspect of the intent of the design competition this should be a major bonus. If the judges are looking for the most complex and expensive ways of doing things, it will not mean much.

I do read of many teams building a good car only to suffer in the events from engine problems due to their own desire to complicate things and try to get more power. Why shoot yourselves down by trying to reach too high? Maybe after the first year of experience with the motor it would be worth looking into raising the performance.

But that is just my view as an active autocrosser. Cars with motor work rarely beat a similar car with stock engine that is sorted out and well driven. Often cars with more power are slower as they require a very gentle driver that can overcome his right foot's desire to spin the wheels.

James

I'm with the University of Michigan Dearborn. We have run CVT's since 2005, initially on the Briggs V-twins and more recently on the aforementioned Yamaha Phazer.

Without giving away results of the hundreds of hours I spent, this engine was quite a nightmare. There are some inherent challenges on this engine. It is very hard to reach stock power on this motor. Sherbrooke turbo'ed it and made 64hp. Ours made significantly less power at competition, and now after a summer of dyno work makes a just shy of that.

The clutches are still limited on operating range, requiring a large final reduction. By hardening our drive sprocket, we were able to use *just* a 52 tooth driven. Sherbrooke used a 75 tooth saw mill looking sprocket.

The electronics side will also be a challenge. University of Maine? used the Phazer for their clean snowmobile, they ended up extending the crankshaft outside of the engine to get usable signaling.

I would highly suggest that you prove out the engine BEFORE committing your frame to major revisions that a CVT will mandate.

Geoff

Hi Geoff,

Wow, sorry to hear the engine has caused such a pain for your team. I have never worked on the Phazer engine so I have no hands-on experience to appreciate what you are dealing with. It seems like such an ideal motor for the competition. But all the engines have their own problems and situations to overcome, eh?

Did you consider using a jackshaft to lower the gearing from the CVT to the diff? Are you using the stock wiring harness and ignition system? As well as the stock F.I. system? I am big on the KISS application guidelines, For me that means changing from what works (stock) to something else, only after the engine is running at peak performance, and then start experimenting.

Here at work (Engineering) I have a motto that applies (to everything actually).

The definition of a "Quick and Easy Job" is anything someone else wants you to do because they don't know how to do it.

I hope you are able to get the engine running well real soon.

BTW, what does the Phazer engine weigh?

James

Geoff those problems sound like a major pain in the ass. I hope we don't run into the same ones. We'll be working on getting our engine running next week... Our sprocket is going to be in the 64 tooth range, which is freaking rediculous, but necessary.


and fwiw: The Phazer weighs 78.5 pounds. The primary weighs around 10 loaded. For reference a KTM 525 weighs 76.0

WOW!!!

78.5 pounds is fantastic. I will keep watching to pick up a wrecked Phazer someday to cannibalize. My two mid '90's Suzuki 600 four engines are 200 pounds each! But, for under $400 each for the complete good running bike they were in my price range. (Extremely limited budget build)

In regards to what JagLite said,

What the judges consider to be important varies largely on the competition. In Michigan, the rule about "weekend autocrosser" doesn't mean jack. They want fancy engineered widgets that cost a lot of money. They want small, lightweight, carbon everything cars that will cost 5 grand to replace a corner on.

And OU has, almost every year, built a car that is, for the most part, for the weekend autocrosser. No fancy carbon tubs, no fancy titanium outboard. And we take a design hit. But you don't have to use this stuff to engineer something well! We perform well on the track, that much is certain. It's a simple design. But it is a well engineered one.

The vast majority of engineers that leave FSAE for industry will never touch a lick of carbon or titanium in their designs.

In California, the judges look more for robustness, for a car that isn't a peaked, about -to-explode car, that is marginal in ordinary areas but has fancy widget a and carbon component B.

In the case of a CVT, it's a good idea. But how easy will this be to use on the track? To repair? To maintain? Good engineering takes all this into account, not just extracting peak numbers. And many fellow FSAE'ers forget this.

Look at industry and motorsport. How many cars, how many teams, run a CVT? Why or why don't they? It's worth looking into. It's a great design exercise, but couldn't you spend your time working on something that will be more useful?

Basically, ask yourself if you're making a design decision just to make things more complex, or if theres a true gain to be had in doing so.

I can clear up a couple of questions Wesley brought up concerning CVTs.

On the track, it does require a different driving style to your traditional sequential car. Staying smooth on the throttle is key to let the CVT operate how it should. You also need to learn how to trust the CVT to work how it should, which takes longer than it sounds.

In terms of repair time, its great. We were out in Toronto this last weekend and the secondary siezed on us. He finished the lap with the secondary locked at a certain engagement and placed a decent time. We took apart and repaired both clutches in under 45 minutes. We could have probably done it much faster, but I had some new members doing it with me.

Maintenance wise, CVTs have been great for us. Very little maintenance was required on our clutches to keep them operating correctly. Main thing to do is just wipe the sheaves down before a run to get the dust/dirt off of them.

When it comes to why no teams in professional motorsports run CVTs, it could have something to do with them being banned in most racing leagues.

Why did the clutches go out?

The FIA banned CVT's when Williams F1 brought a car to silverstone in 1993 and it was much faster than the other cars. So like anything else cool they banned it. Just so you are aware the new KERS system for formula 1 that Torotrak has designed has a CVT inside of it. If it werent for the ban I am very sure that they would be the desired formula transmission.

http://youtube.com/watch?v=x3UpBKXMRto
http://hk.youtube.com/watch?v=6A9V2O5D8mc
http://www.forix.com/8w/altpower-cvts.html

There still hasnt been a team yet that has really gotten one of these setup right yet but just like anything else with some development with a smart group of people and it will cause a shift.

A couple notes for teams out there that want to use one:

A very heavy backshift is very desirable. This will limit the physical vibration of the belt and give you a very responsive feel when off throttle.

Your overall clutch setup should be for throttle response and not like a snowmobile where most of the time you are wide out. Mid throttle response is the name of the game. It is physically impossible to keep the engine spooled at some given rpm because your wheel rpm requirements vary wayyyyy too much. Some people get trapped in this thinking about perfect cvt blah blah blah. You will spend most of your time in and out. Setup for that.
Engagement should be real low. There are parts of the course that you will get dragged down real low no matter how good of a driver you are and how good your setup is. Set the engagement around 2000rpm because by the time the full engagement sets up (i.e. response time) you will end up around 3000-3500. You will get dragged pretty close to that low at times even with a high rpm engine.

That being said you need to have variable upshift control. At low speeds you actually want the cvt to hang for a little so you can make good effect of gearing for accel. At mid speeds you want it faster and at high speed you want it very fast. Multiple angle secondary ramps are the name of the game for this wether they are stepped or progressive radius.

A caged secondary helix is a must. The on/off throttle applications combined with sticky tires are much more faster reacting then a belt in snow. Without a caged secondary you will "float" the helix which will chew it up real quick.

Rollerize everything!!!

Last but not least judge validation. The first question that anyone asks/brings up is the efficiency and is it worth it. The only hard way to compare that is to do a tractive effort diagram for a cvt car vs. a shifter car to show that you get more area under the curve. ANy other explination/argument is useless. YOU NEED TO DATALOG SHIFT RATIO. You need to use a hall or vr sensor on the jackshaft and compare jackshaft rpm vs engine rpm. This combined with final drive will allow you to compare setups with each other. Just because it feels faster doesnt mean it is. The proof is in the data. More area under the curve along with a higher sustained longitudal g will give you the correct setup requirements. If you dont spend considerable time working on this you are shooting in the dark. You may be able to do this in house and show improvements but when it comes to judging they want to see that you actaully did it against data much like anything else in FSAE. Olav Aaen does it on his SSCA races and snowmobiles to get things setup best. You should too.

quote:

Originally posted by Wesley:
Why did the clutches go out?

Clutches went out because a roller was not properly tightened down on the secondary and it dropped off during testing. While the other two rollers could handle the added stress, the retaining pins that connect the helix to the movable pulley could not and they bent. When they bent, the clutch got jammed halfway engaged.

One thing to realize though. Even with this clutch having "failed", it was still capable of running. Driver hadn't even noticed it had happened and just commented on how he had lost a bit of power out of corners.

I'll second what Rob said about the data logging. There is absolutely no way to tune the clutches or defend using them. Tuning a CVT without some type of data acquisition is very difficult because driver feedback can be very misleading (drivers place a lack of changing RPM with a lack of acceleration). When I got asked why the CVT was advantageous to a sequential, I simply showed the design judges out data plots from an acceleration run and indicated the differences between our plot and a sequential car's plot.

That's reasonable, not like we haven't fried a few ourselves.

What kind of weight for engine/transmission combination are you guys seeing?

Aston University in the UK used CVT on their 08 car. Heard they wasn't too happy with the heavy flywheel or something like that in their brake test... may be it worth contacting them if you want to find out more!!

Primary is 5bs. Secondary is 2.7 lbs. Engine with everything I think is about 90lbs. But there is about 10 more lbs to come off due to a huge ass flywheel that needs to be replaced still. UB can get a more exact number next time they have it out and will post it up here.

Yeah, that steel flywheel is getting pulled off this week to do some testing with an aluminum one. I will get you guys an exact weight number for the whole engine/transmission package then.

What are you going to do about the charging system if you are running the aluminum flywheel?

5 and 2.7 lbs? Are you running gaged clutches?

We are planning on running stock yamaha primary and a TEAM secondary... Big, but they will work well and tuning parts are available everywhere for them.

quote:

Originally posted by Kyle Roggenkamp:
5 and 2.7 lbs? Are you running gaged clutches?

We are planning on running stock yamaha primary and a TEAM secondary... Big, but they will work well and tuning parts are available everywhere for them.

Yes, those are gaged clutches.

Rob, we are right now simply testing with the Al Flywheel to see what difference it makes. Our electrical team is looking into still utilizing a stator with smaller magnets or simply a slightly beefier battery. We tested the car the other night on the steel flywheel with no stator connected and managed 3 restarts with 20 minutes of driving in between before the battery did not have enough power to crank the engine over.

Does anyone interested in Honda Silverwing's? It's 600cc parallel twin with CVT, but probably a heavy engine. Several years ago I saw a Japanese Uni used it.

quote:

Originally posted by y.o:
Does anyone interested in Honda Silverwing's? It's 600cc parallel twin with CVT, but probably a heavy engine. Several years ago I saw a Japanese Uni used it.

I just went and lifted the Silverwing that is sitting in our shop. It probably weighs 100-120lbs and that doesn't include the cooling system.