Who ya gonna call?

[90GTVert]

I'm sure they're hot, but 8 miles with a very vented CVT cover and maybe 12 miles in a different vented CVT cover?

[Fox]

I say run 'em both with the covers off and see what's really going on.

If you remove the starter Bendix from the 4T you can spin the crank with a drill motor to start it with the cover off if the variator nut is tightened up good. Maybe the belts are riding up too high on the pulley(s) and rubbing?

[90GTVert]

I got both CVTs apart.

The SunL's case and trans showed nothing indicating any rubbing etc... Just belt dust.


I did notice that the belt has rubbed this section of the CVT cover. The belt always shoots chuncks out there, so that may be all it is.


The belt is still plenty wide, pretty much like new.


One side of the belt looks pretty good. The other is chewed up. Hard to say what's from the aftermath and what could have been a cause of failure.




The Venus showed nothing obvious again. The same marks on the starter clutch that would have to be post-failure.


CVT cover looked liek I expected as well.


Most of the belt was in decent shape. It looks like it just snapped to me.


The last belt had glazed on this scooter.


I cleaned everything up and started marking parts to give me an idea what's up... hopefully.

I colored the vari faces with a marker.


I marked the rear pulleys the same way.


I painted both CVT covers inside so I should be able to tell if/where contact is made.


I used a few different methods to mark the new belts, hoping one would leave a mark on something it touches or at least show contact/wear on the belt.

[90GTVert]

I put both transmissions together this morning and took Dave's suggestion to run them with the covers off. I took a vid of both so it would be easier to see what's happening and I don't need to stick my face down there to watch.



The SunL seems to be getting full travel on the variator while on the stand.


Something is making contact with the belt, as you can see by the line.


It lines up with a part of the starter bendix, so I pulled the bendix off to have a look.


It does show evidence of contact. Now, I'm not so sure if this is the cause or not, because without the CVT cover in place the bendix has no support. I used a wire to keep it pretty much in place, but it may sit a mm higher with the cover on. (I start the scoot by holding the cover on, then pulling it straight off once it's started.)


The last belt does show minor evidence of contact there, but it doesn't look like enough to cause a belt snap. I;ll probably knock the little edge off of the bendix before I try on the road, but I don't think this is gonna be the issue.




The Venus isn't even getting full travel on the stand. I've marked the vari before though and it should be getting full travel when on the road.


The lack of travel on the stand makes me really wonder about the SunL when it's on the road.

BTW, I know the Venus is running fat. It needs a smaller pilot, but I'm not sure where to get one for this carb. It doesn't run so rich up top when I'm riding it. Runs quite well when riding, other than really low throttle settings.

[90GTVert]

I reassembled the Venus' transmission and took it for a test ride. I tried to go slow, WOT and anything between to test all scenarios. Before I went, I made a camera mount real quick out of a bolt and an old mirror and took you guys along for a ride. It's kinda boring, and wow does the vid quality suck... but it's a ride anyway.

I think this is gonna be the last handlebar mount that I try unless it's isolated or something.




Guess I was wrong about belt travel. It did have full travel with this setup and the Gates, but then that belt was wider.


Nice travel in the rear pulley. (This is a modded torque driver.)


There is a worn edge on the inside of this belt. It's on the side of my vari washers. I never cut them all the way down to the exact size ofthe drive boss, and I'm seeing that was a mistake. On top of wearing the belt some, I suppose it would stop the pulleys from closing all the way because they can't slide over the washers. So there's some work to be done.


I also noticed that the paint had been smeared on the outer side of the belt. The patch of red was half this size.


On inspection of the CVT cover, I see that the belt is hit a few spots there.

[90GTVert]

I decided not to touch the starter bendix from the SunL, but rather apply a new mark and try with the cover on.

Once again, I made a crappy ride along...


It doesn't look like the belt hits the bendix with the cover in place. I think I'm going to knock the lip off of it anyway.


Just in that little ride, the belt's already showing signs of wear on the sides.




VERY slight evidence of contact on the case and CVT cover.




Good travel on the vari.


Good travel in the rear pulley. Looks like it would be capable of a little more with mods or a different belt.

[Fox]

Wow Brent! You sure are thorough. Great pics too. I guess it's time to break out the Dremel.

[90GTVert]

I figured being thorough with lots of pics would do 2 things. 1) Let you all see what's up in case you have some input. 2) Provide an outline for anyone else having belt troubles. Besides, I can't afford to go through belt after belt and right now I can't trust these things to ride anywhere.

I'm not too sure what's up with the SunL wearing the edges of the belt so fast. I notice that the variator has belt residue on it so it would seem something is up there, just not sure what.

[2strokd]

They both sure do sound good ;D! I see why you like top-end out there. Looks allot like IN country side. My belt hits the same spot on Skeedr,s cover. Great coverage of the whole test :thumb: !
One for the tech section!

[90GTVert]

I'm hoping to actually solve something before it hits the tech section. lol

[2strokd]

:rofl: Oh yes, that would be good :spin: ! Try a Bando, or a Keli after ya fix your vari shims maybe

[90GTVert]

I fixed the vari shims tonight. Worked on the the CVT cover as well to get a little clearance. I'm gonna paint it tomorrow and see if it still rubs. I'm definitely gonna pick up better belts when I can.





The inside of the cover is repainted. Hopefully I'll get it out tomorrow and see no odd wear.

[90GTVert]

I put the transmission back together on the Venus this AM and took it for a spin.

The spacers are no longer touching the belt.


I'm getting better belt travel. I was doing 52MPH sitting up on the test ride. My previous best was 51.5MPH and I believe I was kinda tucked then.


Very minor signs of belt rub. I'll take just a little more off of these spots in the cover.




Once I take care of the spots in the CVT cover, that should cover everything I've found out of the ordinary in the Venus' CVT. Not sure it won't break more belts, but not sure what else I can do either.



While I was out there, before the heat really kicks in, I knocked the lip off of the bendix for the SunL.


I also removed a tiny bit of material from a couple of spots in the case and chamfered the edges of the case. The CVT case and cover meet up where the belt rides, so I figured it can't hurt to chamfer the case and cover. I've got the cover bevelled, but I still need to remove just a tiny bit of material from it as well.


I really don't think these are the causes of my trouble with the SunL's belts, but I'm just not sure what to do about the way it wears the belt's sides on the vari.

[2strokd]

Good, progress! Thats flying on the Venus man! Now ya gotta change the sig to 52MPH.
How tight is the CVT set up in the Sun? Are you running a heavy torque spring and weights?
Im wondering if the extra friction from the tighter set-up would make a difference in belt wear?
I know a TD with a heavy spring will squeeze the heck out of a belt (and my finger, ouch!).

[90GTVert]

When I checked the max speed function of my speedometer on the Venus, it said 215MPH! Now that's flying! Not sure what it is with the newer models of the Sigmas, but it's said as fast as 150MPH once before that. They work find and I never see it flash those speeds, so I guess it must happen for a split second or something. Maybe I really did go that fast... but it was so fast I didn't even realize it! :paranoid:

The SunL is using 6g sliders and a 1K contra. Nothing too extravagant there. :stumped: