Home page | Links to other sites | SPEEDEX History | SPEEDEX Parts | SPEEDEX’s and their owners | SPEEDEX advertising etc from the ‘50s & ‘60s | Where are they now? | For Sale and Wants | Stupendous 750 | Special Builders Guide

Fitting the centre main bearing without machining

Centre main bearing reliability of the final A7 engines has a poor reputation in comparison to the earlier two main bearing engines. This shouldn’t be so from an engineering point of view, the extra support for the ‘bent wire’ crank should help enormously. So what’s gone wrong? Simply put, the three bearings aren’t in line so the crank is either assembled with the bearings stressed or it’s not assembled correctly at all. I’ve taken down a few three bearing engines over the years and most of them have centre bearing nuts not fully tightened or a rash of shims in between various parts of the centre main housings. Both of these seem to be in response to the crankshaft tightening up as the two bearing nuts are pulled up.

 

Bearings are designed to have a running clearance and essentially be circular and concentric. Shell bearings also rely on the ‘nip’ between the top and bottom housings to tension the shells securely into the housings. Any shims between the top and bottom half housings looses the ‘nip’ and also puts the bearing concentricity and clearances way out. It doesn’t work, neither does not tightening the nuts fully.

 

I believe this misassembly is mostly responsible for the poor three bearing reliability. It can be made to work, Lotus 3 had a three bearing engine as did Chris Gould’s record breaking car of the seventies (albeit with a Reliant pressure fed crank). The trick is getting the three bearings in alignment in the crankcase, and keeping them that way when it’s running.

 

There’s been, like most things A7, a lot written about the centre main. Consensus of opinion to get this right is to machine out a set of undersize shells to accommodate the out of true of the centre main housing. I don’t like this from an engineering standpoint, why machine a precision part to accommodate an error in another component? Shells along with the cams, pistons and rings are the most finely machined parts in any engine. How do you replace the shells without having to machine all the shells again? How do you know it goes together in alignment? And stays that way while its running?

Three bearing engines can easily incorporate an extra block holding down stud. The second centre bearing stud has a block stud fitted with the block being drilled. Note also the cut down tappet guides to give clearance with a modified camshaft and the double valve springs.

Others have made new bearing housings and bored them to give alignment. This does sound the best engineering solution, spare shells will fit straight in. However it is quite a bit of work especially for those without an engineering background or good friends who will do it for them.

 

Of the engines I’ve worked on the centre main recess in the crankcase is ALWAYS high, somewhere between 20 and 25 thou (hence the shims fitted by anonymous builders in the bearing butt half joints). Assembling the bearing and just pulling everything up tight puts a 20 to 25 thou bend in the crankshaft that reverses every revolution it makes—no wonder they crack and break. Side to side alignment seems OK. I can’t account for why its this far out consistently, I can only imagine it was machined true when the crankcase was first machined?

 

I came up with this way of assembling the three bearing engine with no machining and nothing more complex than an accurate shim in the right place. Working out the right size and getting it there is a bit of an exercise but that’s what we enjoy doing isn’t it? You do need a Dial Test Indicator (DTI or ‘clock’ in engineering speak) but more of that further on.

 

Firstly have your crank ground to size after crack testing it. I was lucky, the crank was standard size and still useable without a regrind. I suspect that the more regrinds the crank has experienced the more rebuilds it has had and consequently the more chance it has been run out of alignment strained and possibly inclined to crack. Select your bearing housings. Note that the shells have little tags to stop them rotating, work out which way round to have everything, the tags shouldn’t be together but on opposite sides. Have front and rear marked on the bearing housings and always—always, put them together that way to remove one source of misalignment.

 

Assemble the crankshaft into the front and rear bearings. Get a feel for how tight it is, this is a useful gauge if it starts to tighten up further.

 

Trial assemble the housings to the crankcase. Spin in the top half and push down the bolts and fit the bottom half. Make sure the housings are the right way round. Check that the bearing is, by eye, central to the journal on an axial basis.

 

Remove the bottom half. You need a selection of different shell sizes and insert one that’s 40 thou bigger (thicker) than your journal size (ie if your journal is –60 thou use a –20 thou shell) into the bottom half. Even decent old shells are good enough for this as long as they aren’t completely wiped. Set up the DTI on one of the centre webs, from this point forward the crank hasn’t to be rotated. I find a steel strap bolted across two gauze holes makes a good base for a magnetic stand.

Lift the top bearing housing on the bolts and slide in the shim. A single shim is much easier to handle and more consistent than a sandwich of a few. Assemble the two tensioning sleeves over the bolts with the cut always towards the journal, they’re there to allow the top shell to be spun into position. Tension up the bolts on the tension sleeves. This will pull the top housing into the crankcase and nip the shim in place. Oil up your shell and spin it round the shaft into the top housing. It probably won’t be easy and you may have to rotate it in with the crank to get it round.

 

Remove the tensioning sleeves and fit the bottom housing with your new shell. Before tensioning reassemble the DTI onto the crank web. Pull up slowly, the DTI shouldn’t move indicating that there is some bearing clearance.

 

Turn the crank over, it shouldn’t feel any tighter than when it was on two bearings.

Refit the bottom housing with another matching shell 40 thou bigger than the journal yand slowly tighten the retaining nuts up to full torque to pull everything into place whilst carefully watching the DTI. Any flicker on the DTI suggests that the shell and housing are pushing against the crank and the crankcase is more than 20 thou off set. Continue tightening if you were close to being up to the full value (I use 35 ft.lb). Add the movement from the DTI to half the additional 40 thou shell size ie 20 thou plus the clock reading, and that’s the crankcase off set. If it doesn’t touch with 40 thou bigger shells try 50 thou in the same way until some movement registers on the clock—then add half the 30 thou shell oversize—25 plus the DTI movement.

Cut a shim of the required thickness to the shape shown. This is as wide as the bearing housing and recessed at the ends to clear the bolts. The recesses also hold the shim in place between the bolts. Don’t forget the oil hole in the middle, make it big enough to cope with any misalignment. You also need two tensioning sleeves as per the photograph. I made mine from 5/8” diameter tube (old Imperial conduit), they need to be a close fit over the 7/16” bearing bolts and the same length as the bearing housing. Note that they are cut away at the sides—that’s important and we’ll come to that in a minute.

Other bits and pieces I discovered when I was trying this out;

 

· The bearing position doesn’t vary with bolt tension. Different tensions are not a way of setting the position.

· Tightening the block in place doesn’t vary the position but only tighten the block in place once the bearing is tensioned to ensure the bearing end is seated in the crankcase rather than taken onto the 5/16” block thread.