Main bearings

Not open for further replies.
Mar 22, 2007
Hello All,
I am new to this forum, but have enjoyed the threads. I have recently split my engine case ( 1972, 750 Combat, 206412). I was first happy to see that it had FAG bearings, which led me to the conclusion the case has been opened before. But my problem is the main bearings are not a tight fit in the engine cases. They are snug fit but I can pull them out. I also see that the outer main bearing cases have wear marks which indicate to me that they have rotated in the case. Could this have been cause, by a previous bearing disaster. I also noted the on one lobe of the camshaft there is a punch mark which may have been caused I suspect by a push rod.
Any suggestions on the bearing case? I suspect some welding and machining.

Thanks in advance Bryan.
Welcome Bryan

You could try Loctite 641 Bearing Fit first: ... tid=018ONA (or equivalent)

Otherwise welding and machining (or finding good replacement cases) is really going to be the only (rather more expensive) solution.

I can't quite see how the cam lobe damage could have been caused by a push rod as the cam follower is between the two parts.

Are the FAG bearings NJ306E ?
Thanks L.A.B,
Loctite, great Idea. I believe they are NJ306E (not in my shed unfortunately), though interestingly I thought the main bearing ( have'nt seen another in the flesh) was complete. The bearing I have the comes apart with the inner ring stuck to the crankshaft. Is that correct?
bryan_woods said:
The bearing I have the comes apart with the inner ring stuck to the crankshaft. Is that correct?

Yes that would be correct, the roller cage should remain captive in the outer race and the separate inner race fits on the crankshaft, although some of the earlier roller bearings had the cage held captive on the inner race.
These earlier bearings had the bearing code *NF* instead of *NJ*306 and one of those types would have originally been fitted to your engine (206412). The first *Superblend* bearings (6/MRJA30) used from eng. number 211891 according to the factory service release sheets.
On initial inspection the bearings look/feel O.K. Should I out of insurance replace them while I have the case split or just clean them up. The Bike has done roughly 20000 miles. Thanks in advance.
G'day Bryan,

I would certainly replace them while you have it apart, especially if they have done 20,000 miles.

I will replace them. By the way how long do these main bearings last? Do I need a special tool to take off and put on the inner ring of the bearing to the crankshaft? Lastly should I get the crank trued? I see Norvil have a bottom end kit. I haven't seen anywhere in Australia where they sell these things cheaply.

There is an extraction tool available for removing inner races from crankshafts, #063970, I don't know whether this is the same for your bike, L.A.B. would know. You may be able to carefully warm the race up with an oxy torch and it could well drop off, but don't have too harsh a flame.

It's quite comon for main bearing to come loose in the housing. I found that loctite didn't fix it. There was enough of an interference fit to squeeze the loctite out, but not enough to hold the bearing when the engine got hot and expanded.
I had a flat ground on the bearing ( I use a ball race on the timing side) and the case was drilled and tapped to take a countersunk screw and washer to hold the bearing in place.
This mod is described in Paul Dunstall's Norton Commando Tuning book.
mike mcmanus said:
There is an extraction tool available for removing inner races from crankshafts, #063970, I don't know whether this is the same for your bike, L.A.B. would know.

Yes the same part number tool is listed for machines that would have had Superblend bearings originally, although from the illustrations shown in the workshop manuals the factory tool appears to be a standard type bearing puller, (but not the general tool box bearing/sprocket two or three legged type puller).
I'm curious to know how you found out the locktite didn't work. Do you hear a noise/shudder etc. I was thinking of using locktite on the flat surface as well a s the circumference.
The crank had end float when the engine was hot, but not when it was cold. That could only be caused by the timing side (caged ball race) bearing coming loose.
Could you use shims to stop the end float? I believe they are available for that sort of problem ( Read in Roy Bacon's twin restoration).
""Could you use shims to stop the end float? I believe they are available for that sort of problem ""

Well, that would only disguise the fact that the bearing was moving. I wanted to fix the problem once and for all.
I looked at the bearing numbers and found the outer case is labeled FAG NU306 and the inner ring on the crankcase is FAG Nj306E. Does this sound right? My other Queries is should I glass bead clean the cases? Or use some other media/Method? Finally I see a few scratches and a Nick on the Connecting Rods, Should I try to polish them out or just replace them.
Sounds like the outer ring of the main bearings were changed but the lazy sods didn't bother with the inner rings. Not good practice as the the bearings are assembled by matching tolerences, as you are going to change them its in the past.

Bead blasting works but beware of it sticking to crud in the bottom of holes ready to release itself when the bike is running, be prepared for lots of cleaning out and washing, walnut shells is a safer alternative and gives a more genuine look.

Yes get rid of the nicks and polish the rods for road, replace for race.
Would this indicate that the engine has allready had a second pair of main bearings (20000 miles)? Any Idea as to why they would have used NU306 instead of NJ306.
Anybodys guess, the letter after the N just signifies the location of the single flange in the SKF numbering, same bearing really.
It would appear that the NU bearing could be offered as a replacement instead of the NJ type by some suppliers?

The main bearing shown for sale on the CNW website appears to be marked as the NU306E.M1 type (the bearing inner isn't shown for some reason?): ... g/1401.jpg .

The NU bearing has no flange on the inner spool, but appears to be identical apart from that, and as kommando suggests the outer parts of the bearings could have been changed, and the inner spools were probably re-used. These bearings (NJ or NU) should also be 306'E' type? not 306.


The CNW website shows the SKF layshaft bearing marked as 2nu03EC on the outer part but NJ203ECP on the inner!: ... og/402.jpg

So maybe the NJ/NU306 main bearing combination of markings could be OK after all?
I noted that the SKF bearing has 11 rollers and the FAG 12 would that imply the FAG's would be better as load dissipated across more rollers? My bearing knowledge minimal to say the least. I would assume all the bearing diameters would be the same for various brands? I was hoping one may be just a little more bigger so it would be a tight fit in the case. But I feel that would be grasping at straws.
Not open for further replies.