HI Ian, Just a Thought Bolt up the housing out side the crankcase and check for ovality if its ok then the problem is the crankcase.

Location: TINOPAI NZ

Hi Ian,
Interesting! My bearing catalogues give 1.8955" to 1.8960" for the centre main housing (and the same for the Reliant and A Series bearings). So just half a thou tolerance. I can't envisage how or why it would be made oval, shells do need some crush to aid retension and conduction but that's done on the diameter of the shells. Don't forget that the housings aren't split across the centreline but offset slightly.

Maybe this is the missing factor in why the centre main seems to migrate up the crankcase. But a couple of things occur to me. I've experience of bigger bearing housings (steam turbine sized!) spreading. To some extent they re-roundify themselves when bolted up hard - when the flat butt faces are tensioned together. But these A7 ones are small, why have they spread and surely that would give an undersize in the vertical dimension nipping the crankpin?

I've a few spare housings including a Reliant one; I'll rake them out and do more measurements. I know from previous checks that (mine at least) have a consistant depth - but I didn't check for ovality. Getting a valid nip without adding distortion, mimicking a crankcase - or measuring in a crankcase will be the challenge that I ducked last time! How did you assemble yours for measurement?

Dave

Dave they were bolted up into a crankcase for measurement, then transferred to another case and measured again, the spread was repeated by exactly the same amount in both crankcase's.
I am starting to wonder if you are right, and even though this is only a very small sample, we have stumbled across a missing factor for failure of these engines.

Location: Auckland NZ

Hi Ian,
I reckon the bearing housing should be round and 1.5 thousands smaller than the bearing shell nominal diameter (which is probably an inch standard sizing) say 1-3/4"-.0015". Definitely should not be oval. if it is flogged out it will need reclaiming and resizing to correct spec.
The engine reconditioner should know these measurements and fits if he is competent to do the work.I have seen little notebooks with this info in many workshops!.
Seems strange they are all out of round. I would think this has happened because all of them have been "rubbed down at some point to take up slack in the center main when worn-I have seen .0015" shims under shells and caps and all variety of bodges to remedy wear and misalignment. - metal spray and re-machine ?

Location: Wellington NZ

Steve, the problem was that we did not have any bearing catalogues that gave the shell size Dave has provided above. Quite why all three housings are spread by virtually an identical measurement is the real mystery. At least one of them as you know came from a comparatively low mileage engine with just 10tho us journal and minimal wear.

Location: Auckland NZ

Ian,

An assembled A7 housing measured out consistantly across all the 45 degree axis at 1.892" and ditto for a Reliant housing at 1.895". I haven't any internal mic's so this was with an internal caliper transfered onto an external mic - I wouldn't be too confident with the absolute measurement to a thou or so but the consistancy was there from the 'feel' on the calipers. This does open another query on the definite difference between the two, and the A7 strangely being in the smaller undersize direction?

Try reassembling yours and seeing if the bottom housing will move sideways before tightening. It does fit between shoulders in the crankcase and if these are a bit oversize you could end up with a few thou bigger across the horizontal plane? Installing shells with their overlap across both housings should re-align everything?

If you're getting the correct 1.895" in the vertical plane it would seem to suggest assembly to me and not be a factor in why the bearing migrates up the crankcase?

Dave

Thanks Dave, I do not think that in this case it is an assembly issue, the measurements are consistent in more than one crankcase, and repeatable on each assembly. There is also no radial variation from side to side suggesting a misalignment of top and bottom housing, what ever the cause we will re size and line bore the housing.

Location: Auckland NZ

If I was going to build another 3 bearing engine I'd go the 9 stud crankcase way,Using 2 offside studs and bolting through into the block which must help pull it all together to line bore the centre bearing.