Variator chatter

[szabgab]

Dear all, finally I have gone round to replace the belt and the rollers, upon disassembly I have discovered, both the belt and the rollers are still OK, actually the belt I have measured in-situ completely wrong, and it is barely worn. Anyway, I have cleaned everything, sanded the clutch pads and the bell, cleaned the variator housing, ramps, replaced the rollers, put on the new (Xciting 250i spec) Mitsuboshi belt I have bought, everything is nice and cleaned. The clutch was catching on, chattering away and spinning the wheel ever so slightly before, but now this constant horrible chatter is coming from the variator side at idle completely overshadowing the mellow clutch chatter (that is still there). Also once the engine spins down from acceleration, the chatter gets rather loud for a second it to settle to this constant machine-gun fire mode. What did I do wrong?? Followed everything by the book, put the weights in the correct way, torqued everything to spec, even have taken the shebang apart three times since, trying to make sure, everything is spotless, etc. The belt is spinning the right direction.

I have made a video, sorry it being lousy, but my trouble is clearly audible, the chatter actually consists of three different layers, one is the (now) faint clutch chatter, some metallic noise and the loud chatter.



Any ideas?

Thanks folks!

Gab

[CROSSBOLT]

Hey, how are you? Good to hear from you! Yeah, I have an idea what may be your cause. De dee has a method of making sure the belt does not interfere with the front variation seating properly. I have a different method that does the same thing. His method does not require the rear pulley to be removed but takes patience and must be done in a sequence. What I think happened is the front nut was tightened against the belt and the pulley halves were not touching in the center. When the CVT started working the front pulley parts were loose!

[szabgab]

Hey, how are you? Good to hear from you! Yeah, I have an idea what may be your cause. De dee has a method of making sure the belt does not interfere with the front variation seating properly. I have a different method that does the same thing. His method does not require the rear pulley to be removed but takes patience and must be done in a sequence. What I think happened is the front nut was tightened against the belt and the pulley halves were not touching in the center. When the CVT started working the front pulley parts were loose!

Hey Karl! So nice of you to try to help with all my bloody ailments with this bike. Since we last spoke, the electrics were fried, well basically the battery died every week or so, first I thought it is the cold, turned out to be the alternator cables, where they escape the case were somewhat eaten away from engine movement, one broke of completely. Temporarily I have fixed it with soldering the cable back in (I could just about pull out a piece of the broken cable from the case with some long nose pliers) and RTV'd the hell out of the whole assembly for now (as the other yellow strands started to disintegrate badly). Eventually I will have the to take the alternator cover off and solder on some new cables, but I want to ride and not to f... around with this bike.

Belt wise - I had the same idea, but what I did was to losen the clutch side to the point, the belt was basically hanging between the rear and front pulleys, after that I have made sure, the variator pulley sides are actually touching rather than sitting on the sides of the belt. I still could be wrong, but after running the bike shortly I have tried to torque the variator nut again to make sure it is nice and snug, and it was. But please share your way of doing it, I'm known to be an idiot who tosses up a basic step, like forgetting to put back spacing washers and the likes )) What I did not try to put the variator pulleys on with the belt at first, torque it down and put the clutch side on. What I did instead was to put both sides on, loosening the clutch side, torquing the variator side, torquing the clutch side.

With my old Honda I nearly butchered a clutch, because the nut did not give (I am sorry to say, I sold it like that). I should have done it then, but I have acquired an impact wrench only now... Well at least I can take the variator apart as many times as I want ))

[CROSSBOLT]

This is one way:
Half an inch is too little! That belt needs to have about 1.5 inch of pulley face showing. It is tough but the last time I went through this I had the back pulley off the gearbox, laying on a stout, flat board. Then with a stout bar like a breaker handle or a 3/4 × 1 1/2" stick push the belt into the sheave, hold some pressure with one hand then pinch the belt with the other. Use a string or cable tie where you are pinching the belt to save what you have then do it again until the belt top is about halfway into the pulley. Tie the pinch and put the rear pulley on the bike with the left belt loop around the front shaft. Put the front pulley together with the loads of slack and clearance you worked so hard to get. This way the front pieces will properly bottom out when you torque the nut. Whew! I'm tired just talking about it!

Go to my profile, click "show posts," go to post #73 on page 5, go to that topic and near the bottom there will be EXCELLENT pictures by Snorvey. His photos show the procedure in sequence.

[szabgab]

Oh,  OK,  I will try that method.  BTW,  how much free movement shoulf be on the shafts?  I can most definitely move the clutch side (wheel)  shaft an inch or so side to side without movement in the wheel but also the variator shaft,  which is strange,  as I would have thought,  there should be absolutely no slack,  whatsoever...  Thanks!!

[CROSSBOLT]

There is a LOT of gear backlash in that final drive! It is like a slack chain on a motorcycle. I am assuming the 250 is like the Agility, Yager, DT and People GTi300.

[szabgab]

There is a LOT of gear backlash in that final drive! It is like a slack chain on a motorcycle. I am assuming the 250 is like the Agility, Yager, DT and People GTi300.

Karl, that"s good to know. I guess the engine and final drive will be most likely the same as other 250 and 300cc Kymco engines, basically a revamped Helix power plant...

What about the front shaft? I turned the engine around grabbing the front pulley, and it was relatively easy to turn (the two cylinder 400cc Honda I had was really hard to turn over by hand, but again the generator was bolted to the end of the crankshaft without any reduction, gears, etc) also I could detect free play in it of a half an inch or so...

[CROSSBOLT]

I don't think there should be ANY freely in the front pulley. On the DT the front pulley halves are splined to the crank extention and the Agility 50 was the same. 250 must be the same. Find out if the splines are worn. I hope the extention is OK.

[szabgab]

I don't think there should be ANY freely in the front pulley. On the DT the front pulley halves are splined to the crank extention and the Agility 50 was the same. 250 must be the same. Find out if the splines are worn. I hope the extention is OK.

OH, OK,  that's what I feared...  I will check again,  if it is free play,  or just the crank moves too easily,  but probably now is the time,  to count my loss and sell the bike before it completely falls apart...

[szabgab]

OK,  I have done,  what Karl suggested and mad sure,  the belt is as slack as possible before tightening stuff,  it did not help.  What I have noticed though,  that the crank side of the pulleys do not re-engage as quickly as the clutch side,  making the belt rather slack for a far bit,  which most likely causes the noise upon decel.  I have taken the pulley apart,  and everything seems to be cool,  the rollers are brand new,  but I guess something is somewhat sticky.  The metal shafty thing inside the pulley and it's sleeve look to be fine,  although I didn't measure clearances,  don't ask me why.  So what else? Maybe sliders? The back plate of the roller assembly feels to be loose,  eg having a slight side-to-side movement and I can most definitely crook the plate to the point it binds on the posts.  I'm not sure if this is a good enough test,  as the pulley doing some 2-3000RPMs will behave differently. 

I have done an another video,  if that explains it better



Thanks!

[xsel777]

Perhaps put the old belt back, as that one seems to have a lot of slap. Possibly the edge angle is wrong or it is too long.

Sent from my Mi Max using Tapatalk

[Stig / Major Tom]

Also, please check out CVT belt problems on youtube.
Watching the video it looks like belt slap to me, which is a common problem with scooters.
Belt is touching (slapping) at mid-point of its travel, either above or below, inside the CVT area -- especially prevalent on decel, or over-running, when you roll of the throttle.

Most of the comments on this issue center on improper belt length. A slightly too long belt will have enough slack on decel to flail out and touch. Look for marks on case or on upper side of belt. If you see nothing.....then I can add nothing more.  (have seen people put some white chalk on case above and below the center of belt.....then look to see if it is disturbed after running scooter)

Stig

[szabgab]

Hi,  this is the old belt..  I had the same thought,  maybe a too long belt,  so I switched back to the old one.  Because it did the same,  I videod it and posted here,  as I have ran out of options.  I have cleaned all the old gunk from the ramps,  sliders etc,  could it be,  that the rubbish was cushioning the problem,  like a worn slider? BTW there are no marks on the case as yet,  before me venturing in to clean stuff it was good And the clutch chatter didn't go away either,  although the shoe springs are good,  the bell is rust free and the shoe pads are all within spec...

[Stig / Major Tom]

Eliminating clutch chatter seems to be more art than science, sometimes.
But a different pull-away technique can help. Some scoots hate a gentle take off....preferring a crisp fist full of throttle to quickly lock up the clutch and minimize slippage ( which causes glazing and chatter)
Stig

[CROSSBOLT]

Video is good! Sounds good! But belt action seems a little strange. Just to be sure, measure outside belt length and width. Actually, belt width at several positions around circumference. 5 positions should tell you if is the same for the length. Measure both belts to see what, if any differences. Then the biggie: what should be in there? When I said strange action I meant the front seems to change effective diameter but the rear seems hardly to move.