Cracked crank

[RandalColman]

A few weeks ago I partially dismantled my engine on account of severe camshaft wear, and also to clear the sludge out of the water passages. At the time I preferred not to delve too deeply into the condition of the bottom end but eventually curiosity got the better of me. The crack in the crankshaft was barely visible when it was in place, but on lifting it out it was clear that the crack went at least 90 per cent of the way through the web. It took only light finger pressure to open it up.



I suppose I should count myself lucky to have found the defect at this stage, since it looks as though the crank would have broken within a few hours of further use. The crack was opened up as shown below simply by standing the shaft on its end and pulling up on the no. 1 counterweight.



I have read that it is usually the front flying web that breaks, as here. Mine is 15.5mm thick, while the back one is 16.0mm thick. That would make the front web almost 10% more flexible. Is that typical? I would have thought to look at them that the design thickness should have been equal. This is a late 1951 car, by the way, no. E1-PC-18562.

[paul wilks]

Blooming heck! And I thought ALL Javelins and Jupiters were fitted with oval web cranks by now.
This doesn't look like an oval web to me.

Paul

[Keith Clements]

Yes I was surprised when riding in the Jav for a couple of days in Tassie. Perhaps it was my influence that broke it!

[RandalColman]

Reading what Edmund Nankivell has written about the saga of broken crankshafts it seems that no one could quite put their finger on why they were so vulnerable to flexure in the flying webs (i.e. the ones with the "missing" main bearing), but evolution certainly beefed up crankshafts in this area. Here is an early Subaru:

And BMC were taking no chances with the A-series:

[Nick Webster]

I am currently "inside" a spare engine, known to have been rebuilt relatively recently. Although it rather curiously (for a series III crankcase) carries a D9 xxx number on the front plinth, it also has an "RO" engine plate. I was surprised to find that it is fitted with a pre black sided crank shaft. A salutary reminder that these days it is much too risky to believe anything written on the outside of an engine. This sent me off reviewing crankshaft data and hence back to this article.

Looking at the pictures of the broken crankshaft what occurs to me that it is of the very earliest type without holes in the big end throws. These are the ones notorious for breaking and suggestions are to avoid like the plague. And yet this one, though broken now has apparently survived all these years. Either the car is low mileage or has been driven with the greatest of care?

Nick

[Keith Clements]

Perhaps it is the Coriolis effect that saves them down under.

[BobCulver]

Many Javelins in Oz were very early models with the more reliable shaft not induction hardened.
The occasional car with original crank still turns up here. Fatigue processes are statistically very random. Example is the life of wheel bearings.

I dunno why cranks break in the front half. Torque transmitted is not reckoned to be a major factor. The fundamental weakness of the Javelin is an in sufficiently rigid crankcase(as evident edge wear on the mains and the general thumping even when good).Perhaps the heavy flywheel restrains deflection under firing whereas the front has no restriction.

Bob Culver

[BobCulver]

The observant may note the undercut radii on the Subaru crank. Many cars adopted this. The fillets are rolled, greatly increasing fatigue resistance. And normally safe from slovenly regrinders who have tightened the radii and ruined so many Javelin cranks.The A series crank is a very late one, but all were a bit like that. Would tend to think that very rigid webs would concentrate deflection adjacent to the centre main. Many modern cars have the crank so well suported that the webs are just thin discs. If anyone has stumbled across textbooks on the subject I would be interested in the reference please.

[RandalColman]

I had a bit of a look around on the web and didn't find anything on crankshaft design generally, but this description of the Subaru 5-bearing engine of 1989 is interesting: http://www.surrealmirage.com/subaru/files/SAE_boxer.pdf

Despite the much greater stiffness of the Subaru crankcase they still predicted enough occasional bearing misalignment to increase friction appreciably. Also, the crankshaft vibration damper is stated to be designed to absorb flexural vibrations as well as torsional vibrations.

I would agree with what Bob said about the flywheel inhibiting flexure in the back half of the crankshaft. All the same, my car had a crack in front of the no. 4 crankpin, along with one in front of no.2 crankpin and the major one behind no. 1. Perhaps if you get enough cracks they grow more slowly! The car had 69,000 miles on the clock when I pulled the crank out.

[Forumadmin]

I have added the Subaru engine paper to the Gallery as it contains some very useful discussion items. The paper is now OCR ed and thus text can be quoted.

One of the most difficult aspects of designing
high pressure diecast aluminium crankcase is the
oil passage, because drilling an oil passage is
almost impossible because of oil leaks caused by
cavities within the cast aluminium.
More than 50 cases had been studied before we
could make a satisfactory oil passage with no
machining, some of which are shown in fig. 7.

Another interesting feature of this crankcase
is a water dam which separates the water jacket
into upper and lower sections.
(See fig. 6 * mark)
This works not only for the homogeneized cooling
but is also effective to increase the rigidity of
the entire crankcase and cylinder bore.

For the DOHC 2.0 Lt turbo version, medium
pressure diecast aluminium crankcase is utilized
to secure the higher rigidity necessary with a
closed deck. The sand core with a special
surface treatment is incorporated within this
medium pressure diecasting.

[Forumadmin]

Since I have parts hanging up in my garage from my Subaru turbo engine that suffered from a bearing / conrod failure/ piston holed, I can see a lot of what is being discussed.

The cylinder head gasket is the heat and pressure-
resistant carbon type with a hooked steel
core. (See fig. 13)
The bore grommet with steel wire ring inside
is formed as a part of the heat plate over the
gasket and enables it to withstand the high
pressure and temperature of the combustion gas.
A rubber grommet is used around the high pressure
oil passage for perfect sealing.
Silicon coating is applied on the entire
surface to improve the initial sealing function.

The head bolts are torqued to their yield
points in order to maintain stable axial forces.

[Robin Fairservice]

My crank, which I believe to be an original, has been found to have three cracks in it. My Chassis number is 10100, if I can find an Oval Web shaft, can this be used without modification, or what are the necessary modifications to fit my engine?

[Forumadmin]

I am fairly sure the two are interchangeable, especially if you were fitting new bearings. Read the tech notes on bearings, I suggest.

[BobCulver]

Been covered in Club data but I do not have refs. Basically the radii are greater (unless ground off!) and edge clearance of shells (and bes) needs to be checked. I gather clearance to balance pipe may need mod and possibly elsewhere in crankcase.
Much depends on the particular bearings or substitutes to be used. Ideally the behind the shell grooves, larger oilways etc of the later engines should be replicated, depending in part on the shell type used. Large surface channels alone may suffice if high rpm avoided and all clearances kept small.
As with any Jav rebuild the bes and case should be assembled at full torque (and no more) bare and tunnels checked for size and roundness. Ideally should be rechecked with shells and ground to suit but at least measure the shell thickness and refer to assoc tech notes for intended tunnel sizes etc. The neg clearance sometimes advised for the wm rear main must be avoided if not using wm (and esp in cold BC!). Slight binding due case distortion probably does not matter provided each bearing has clearance. Cranks for some makes were deliberately bored on two centres to quieten crank.
Others will be able to advise more.

Bob Culver NZ