Replaced the roller clutch bearing in the flywheel

[Carl Watts]

Well finally got two bearings out of the USA. Decided to go with the Koyo /Torrington RC-121610-FS which is the stainless steel leaf spring version as the integrated acetyl resin polymer (plastic) spring version on mine got "Weak" and failed.

What I did is make up a puller so you DO NOT have to remove the 3 torx screws and remove the hub itself. I decided although removing it would make the job much easier it could potentially upset the balance of the flywheel, even though the hub could be a precision fit to the flywheel. I just happened to have a bit of alloy about that came out of a partition that had an adjustable foot on it and the holes were on a perfect PCD to recess to go over the 3 torx screws.



The bearing is an extremely tight fit and some M10 x 1.25mm fine thread bolts were needed to extract it.

Also decided to remove just the inner shield from the two roller bearings and move from grease to synthetic oil and modify the internal spacer to catch any surplus oil while rowing. When you stop the oil that has moved out by centrifugal force to the outside groove can drain back down and re-enter the bearing.



Everything is back together and runs as smooth as silk with no slack at the catch.

[woodgeek]

Nice work. Which model do you have? Do you know if the roller clutch bearing is the same across all models -- B through E?

[bisqeet]

i'm going to have to save these - or we could make a wiki...

alternatively i might have gone with the SKF 129 TN9

(SKF is just down the road from me)

[Carl Watts]

Nice work. Which model do you have? Do you know if the roller clutch bearing is the same across all models -- B through E?

Pretty sure they are the same bearing at least from C to D/E anyway.

You have to make up a puller or use a press, you cannot use a HAMMER to do this job. Its a precision bearing you cannot bash it with anything.

[c2jonw]

Pretty sure they are the same bearing at least from C to D/E anyway.

You have to make up a puller or use a press, you cannot use a HAMMER to do this job. Its a precision bearing you cannot bash it with anything.

Yep, Same clutch from B through E and BikeErg. The SkiErg and Dynamic use a 12mm clutch. And Carl is right- these are precision mechanisms that require a very high tolerance sizing of the housing and axle to get them to work properly- and no hammers!! C2JonW

[froque]

Hi Carl

Did you get a picture when you where using the puller to extract/press the bearing? or how it works? I want to build one.

How that internal spacer modification is working? it is pretty difficult to contain oil.

Thanks

[Carl Watts]

Essentially its a nut and a bolt and a tube to pull the bearing into. If you have a lathe you can obviously make the bearing a nice slide fit in the tube to help keep it square going in and out.

The thick alloy washer on the bolt is a slightly smaller diameter than the outside of the bearing, you put the bolt through the bearing and do it up on a thread and pull the bearing out. You could just use a stack of say 4 or 5 washers to do a similar job. You cannot just use one washer, the force required to remove the bearing will just concave the washer. The bearing makes a cracking sound as it comes out when the force overcomes the stiction and the bearing moves and you retighten the nut until it moves again.

Use the same technique to pull the bearing back in, you just need the right length bolts with enough thread or simply change bolts half way through so you don't run out of thread.

The two roller bearings in the flywheel form an almost sealed unit if you run the shields to the outsides. Note that the shield was only removed from one side of the bearing, not both. Centrifugal force will spin the oil to the outside of the bearing anyway, any tiny amount thrown out into the bearing spacer goes through the holes to the outside holding area that then just drains back out when the flywheel stops spinning.

Technically there are no gains to be made here by reducing friction. During the drive the bearings are stationary and only spin during the recovery. As the drag Factor calculation is looking at a decelerating flywheel, whether it decelerates slightly slower makes no difference. It was just an engineering idea that popped into my head and I have fun implementing them.

[froque]

Thank you Carl

I do not have a lathe, but I will start collecting bits and pieces to build my own puller, I will try on an old unbalanced flywheel first.

Was aluminum you material of choice?

I will look into the cycling and biking shops, they might sell a puller for small bearings that could work.