Roller bearings and oil viscosity maintenance

[Blake Sobiloff]

A machinist on a motorcycle-specific forum posted the following:

quote:

---

I think there are two basic issues. First, the oil shears down very rapidly with a KLR. Roller bearing cranks are much harder on oil than conventional bearings. It's been quite awhile ago, but testing on a KZ1000 sheared down from 50wt. to 40wt. in only 800 miles. That was with Kaw oil, which was then made by Sunoco IIRC. Synthetics would be better here.

Second, is the ferrious metal debris which gets very high. The filter won't catch it. A magnetic drain plug helps, but won't get it all. Roller bearing cranks generate more of this stuff. I would not run even the best oils over 2000 miles. No matter how good that oil is, the fine metal debris will slowly chew things and wear them out. One cam at over 200.00 will buy a lot of oil.

---

Both suppositions (roller bearings cause greater shear than plain bearings, and roller bearings produce greater wear metals) are new to me. Frankly, I don't see support for the latter in the UOAs over in the motorcycle UOA forum.

However, it is abundantly evident that many motorcycles shear the heck out of oil. I've usually attributed this shearing to the transmission gears in the shared sump design of most motorcycles, and secondarily to the high RPMs possible.

So, are roller bearings really the great shearers of motorcycle oil, not transmission gears? Do they actually spew ferrous debris into the oil at a significantly greater rate than plain bearings, becoming the limiting factor in OCI length?

 

[bruce381]

roller bearing will not throw MORE wear debris when new BUT they can not tolerate ANY wear debris from other sources as they are prone to spalling and pitting which will THEN throw more debris into the oil and it snowballs.

I do agree oil must be CLEAN and change out offten is the best way to do that.

As far as shearing I would think gear sets will shear far more than a roller bearing and a roller would be more than a plian bearing.
bruce

 

[Eddie]

We need some hard data.

 

[Kestas]

How about my years of experience looking at debris in roller bearings? Bruce is right on the money with every word in his post.

 

[Terry]

Kestas and Bruce are right on.

Add clutch dust to the list of particle issues. Material is a mix of metals and other.

 

[wileyE]

How about the engine in question's track record? They've made those things for decades in several variations, they have great longevity and reputation for being bulletproof worldwide. Greater than many of thier plain bearing counter- parts of simular design in many cases. All this getting no different oil maintenance than the plain bearing motors. I've owned 3, 600, tengai, and still have a 650. Also had quite a stint with gs suzuki motors with roller bearing lower ends, again rock solid motors worldwide for a decade. Agree with the above posts though, debri can tear em up, but never have seen it be an issue with these motors? DO not agree with the machinist assumption that synthetics would necessarily be better. If you look at all the roller and needle bearings used in a motor like the klr with transmission shafts, balancer shafts etc, there's a bunch, does 2 more on the end of the crank make that much difference?

just my 02, no hard data

 

[Eddie]

I also have driven and rebuilt 3 Porsche Super vehicles with roller bearing crankshafts without issues. My experience is that they last about as long as the roller & needle bearings in a transmission.

 

[tom slick]

Look at a two stroke's crank design, rollers in the rod and balls on either side of the crank. the rollers are only lubed by the oil in the gas, which is very thin. crank bearings last a surprisingly long time in that application.

if a roller bearging is showing any wear metals at all it is already too late for the bearing.

 

[jsharp]

I'm not sure the machinist the OP mentioned has a good handle on the old Kawasaki Z1 family 900/1000/1100cc engines. I built and raced a number of them in the 70's and 80's but keep in mind this is from memory -

Those things were as bulletproof as you could find at the time. I remember one in particular I took apart ( my street bike ) to modify that had 45,000 hard street miles and yet the crank still met specs for a brand new one on the shelf.

The engines were 100% ball and roller bearing except for the camshafts. The cams rode in plain bearing inserts in the head. The primary drive and transmission were large gears and I'm sure that's where the oil shearing took place. They did tend to throw a lot of metallic junk into the oil but it was from the clutch, not the engine itself and they all came from the factory with a decent sized oil filter and a magnetic drain plug. The oil pressure was very low on these engines, less than 20 psi maximum and effectively nothing measurable at hot idle. The low pressure sensor consisted of a simple switch in the main oil galley that would close when there was no oil flow and open with *any* flow at all turing off the low oil pressure lamp.

If the guy used 50 weight oils in those engines he was wasting power and making for higher oil temperatures, but not much else. We saw as much as a 20 degree increase in oil temperature using a 20W-50 dino vs. a 10W-40 dino. We tried 15W-50 Mobil 1 but saw no real improvement vs. other synthetic oils in 10W-40 and the engines were noisier with the M1. We ended up using Klotz 10W-40 synthetic, the stuff that was ester based, bright red and smelled funny...

 

[carock]

It was really interesting to find out why Porsche had to abandon the needle bearing crank. As they increased the load on the bearings they went to longer needle rollers. THe longer needle rollers would bend as oil passed under them. The bending force destroyed the needle bearings.