I consider this to be the most serious issue with the NuVinci N360, well with my kind of usage it is, 2 – 3k miles is not a long life. Can this be fixed and if so how?
Why is it happening?
Well I don’t think it’s a mismatch in metal hardness issue. If an axle slides over a surface and rotates in the opposite direction to that which it wants to roll the surface is going to wear no matter what. So why is the axel sliding? If a ball on an axle is going to pivot on its centre when guided by slots at each end, the ends of the axle must rotate in opposite directions. If the two ends are fixed to each other only one end will rotate in the right direction. Whichever end has the highest load will dictate the direction of rotation.
How could this be fixed?
One way to fix it is to isolate the two ends of the axle by putting a bearing on one end. The potential problem is that this friction is a necessary part of the functioning of the shifting mechanism. It may be that it holds the axles and avoids the need for a large resistance in the shifter to prevent auto shifting. Adding friction to the shifter would put extra strain on the cables.
It seems like too simple a fix. I feel there’s a reason Fallbrook do it the way they do and it’s probably because fixing it will require a complete redesign and a much more complicated mechanism.
What I’m considering doing.
The centre of the axle is 5mm and enlarged to 8 – 8.3mm on each end. Also on each end is a 9.7mm diameter washer and a 9mm OD O-ring. The ends need to retain the planets roller bearing so cannot be completely removed. A suitable bearing such as one with a 5mm ID and 8mm OD has a width of 2 – 2.5mm. The enlarged end of the axle is 4.5mm long giving sufficient room for such a bearing.
![[Image: faoo.jpg]](http://imageshack.us/a/img41/2682/faoo.jpg)
Will such a small bearing take the loads?
A bearing of the required size has a load rating of 300 – 400N. A very rough estimate of the load on one of these bearings in a worst case scenario is about 300N. This assumes a power transfer of 1000W to a wheel rotating at 60rpm (26 inch wheel road speed 4-5 mph) and a traction ring diameter of 84.5mm. The bearing will be on the output side so this is why wheel torque is important. So it looks plausible, all I need is some kind of lathe to turn down the axle ends.
Why is it happening?
Well I don’t think it’s a mismatch in metal hardness issue. If an axle slides over a surface and rotates in the opposite direction to that which it wants to roll the surface is going to wear no matter what. So why is the axel sliding? If a ball on an axle is going to pivot on its centre when guided by slots at each end, the ends of the axle must rotate in opposite directions. If the two ends are fixed to each other only one end will rotate in the right direction. Whichever end has the highest load will dictate the direction of rotation.
How could this be fixed?
One way to fix it is to isolate the two ends of the axle by putting a bearing on one end. The potential problem is that this friction is a necessary part of the functioning of the shifting mechanism. It may be that it holds the axles and avoids the need for a large resistance in the shifter to prevent auto shifting. Adding friction to the shifter would put extra strain on the cables.
It seems like too simple a fix. I feel there’s a reason Fallbrook do it the way they do and it’s probably because fixing it will require a complete redesign and a much more complicated mechanism.
What I’m considering doing.
The centre of the axle is 5mm and enlarged to 8 – 8.3mm on each end. Also on each end is a 9.7mm diameter washer and a 9mm OD O-ring. The ends need to retain the planets roller bearing so cannot be completely removed. A suitable bearing such as one with a 5mm ID and 8mm OD has a width of 2 – 2.5mm. The enlarged end of the axle is 4.5mm long giving sufficient room for such a bearing.
![[Image: faoo.jpg]](http://imageshack.us/a/img41/2682/faoo.jpg)
Will such a small bearing take the loads?
A bearing of the required size has a load rating of 300 – 400N. A very rough estimate of the load on one of these bearings in a worst case scenario is about 300N. This assumes a power transfer of 1000W to a wheel rotating at 60rpm (26 inch wheel road speed 4-5 mph) and a traction ring diameter of 84.5mm. The bearing will be on the output side so this is why wheel torque is important. So it looks plausible, all I need is some kind of lathe to turn down the axle ends.
- Oran