Swedeboy

Well - my 944 Turbo is finally back on the road again since this morning. Back in April I experienced a slight power loss followed by dramatic drop in oil pressure accelerating out of a corner on track. I quickly switched off the engine and the initial diagnosis was a damaged rod bearing.

Failure analysis of the old engine after opening it turned out that the crankshaft had been turned down and fitted with undersize bearings by a previous owner. Oddly, the crankshaft had broken in half right in the middle of conrod bearing #1. Basically the #1 cilinder then had the ignition and valve timing gear all to itself while the other 3 cilinders were free to do their own thing. Strangely, only the journal of the main bearing between cilinder #1 & #2 broke - making the block unusable. No valvetrain, cilinderhead or other block damage.

The car has seen pretty much dedicated track use for the last 3 seasons. Running full Cup suspension, R tires, big brakes. However rodbearings looked reasonably ok - definitely not blueish.

Anyone care to speculate on the failure of the crankshaft? I can post a few pictures in case of interest.

(Naturally, this could only be considered a good opportunity for an upgrade so a nice slightly used engine was duly purchased. As the engine was sitting out of the car anyway it would have been rude not to have it opened and tweeked just a tad with flowed cilinder head, more aggressive cam, new bearings/belts/gaskets/seals. )

Swedeboy

Anyone?

Konstantin

either the crank was not straight or the block was not straigt. Older 911 with magnesium housing had this Problem. Too much twisting and the crank wa sbroken. You can brake it only if it was to much load on one place and not divide the load to all bearings

Konstantin

Ahmet

I will second Konstantin.

I would love to see some pictures of this though, and wow, quite unusual to break a crank on a 944...
Ahmet

Swedeboy

Attached. The engine was overboosting - but no signs of detonation whatsoever - if anything there was a fair bit of soot. Tuning was by ETG in Germany.

My guess is that the car was running like this at least for a lap or 2. I experienced some slight hesitation and what sounded like a very rough misfire for a second or 2 - which in retrospect could have been cilinders 2, 3 and 4 running slightly out of tune. It then ran fine for another lap or 2. It then lost power accelerating out of a corner - which I assume is because the teeth are missing from the distributor belt in 3 different stretches.

As I mentioned, for the most part only the crank itself and the block were damaged. The bottom of conrod #1 had a bit sheared off the side of the bearing casing. Two pistons (of which one was #1) were slightly scored - most probably from not running straight in the bore. For the rest - no damage to pistons, valves, cilinderhead, etc.

Could conrod #1 have worked itself loose and broken the crank?

Ski

Most, cerianly not all, heavy duty items such as this crack due to a manufacturing defect. Although small at the outset, over years of stress and fatigue, they finally give way. But with what Konstantin said, you might see if that particular cylinder bore is not perfectly straight and caused some improper loading. Tom Charlesworth had one in his shop last year that broke in almost the same spot but I think it was aft of the the #2 main.

Tom M'Guinn

I know many of the 911 books recommend against regrinding crankshafts, because is it difficult to get them re-hardened (nitriding) to factory hardness. DId they do nitriding on your crank when they turned it down a size? I'm not sure that could weaken the core structure (I thought is was a surface hardening process really), but perhaps the regrind contributed to this?

Swedeboy

I bought the car as-is about a year ago and have some history from the last 4 years. It received a new factory shortblock in 1997, but it now sported oversized pistons and crank bearings. It was obvious the oilpan had been removed with the engine remaining in the car sometime in the last number of years.

To ensure it didn't happen again I sourced a good working block elsewhere, however I am curious why this happened. The car has been track-spec for the last 4 years and has seen sole track use for that duration on the Nurburgring.

FSAEracer03

Good Lord! I've seen con rod #4 break on 944/968 bearings about half a dozen times, but I've never seen or even heard of their cranks snapping. That's incredible. I'd agree with Ski in that these won't break unless they are substandard to start with. Now if the people who built the motor somehow had a main bearing in the middle that was too small, and left a gap, I could see how prolonged high-rpm spinning might create a problem, but this is just wrong. Crack propogation I'm sure would take a long time to actually separate a crank, but without strength hardening, I'm sure it's not only possible, but most likely the cause.

Swede, do you mind if I save these pics for future reference??

Swedeboy

No problem. Send me a PM with your email addresss for the hi-res version.

The key thing I want to make sure is that I didn't suffer the dreaded #2 bearing failure or that the fueling/ignition mapping from my old engine isn't to blame. In my understanding that wasn't the case and it was really attributable to sloppy preperation of parts by previous mechanics I won't be using again anyway.

The amount of force to snap the crank might be high, but on the Nurburgring in a well-running 951 with Turbo Cup suspension you do leave the ground several times a lap - even while still on full throttle. Up to now I didn't always lift when this is the case - so could easily have contributed to the failure.

Danno

Hmm, it's certainly not possible to apply enough force to the crank using combustion-pressure alone to crack it. Failures typically start at a stress-riser on the surface of a part, such as a nick or a scratch. It could be possible that when they turned down the journal, they might have nicked it somehow. Then figuring as long as the damage was below the surface of the surrounding journal, everything was fine. Over time with continuous applications of load, the stress-riser that the scratch induced would extend the crack.

That's why it's standard practice to never grind welds, because all those little scratches would be a start for cracks that would eventually fail. If you do grind, you have to take it all the way to polishing to a smooth surface with no surface scratches.

FSAEracer03

Exactly. Crack propogation starts from something small, sometimes even smaller then an eye can see, and spreads due to prolonged fatigue (dynamic or static).

My email address is boostin1BAR951@hotmail.com

Swedeboy

Email bounced. Is your inbox full - the pictures are about 2MB.

FSAEracer03

ahhh, hotmail only allows 2 megs total i think... if you send them <1meg per email i think it should go through. I appreciate it.

hosrom_951

Man, i never saw a cracked/broken crank befor.........im sorry to hear that happened to you...

But as everyone said, since the rod bearings were infact replaced at some time, they didn't install them properly (and they were the wrong size as well?) thus causing the crank to crack due to loads?