Name that bolt.. most bazaar mode of failure yet!

Ok... Who wants to guess what happened first?? I have more pics to share, but they'll give away the mode of failure completely.. See that bolt? What is it? Where is it from? It was in perfect condition before it's ride up the "conveyor belt" (timing chain) where it met the exhaust cam and thats finished the engine off. Note that bolt isn't broken or stripped.. it simply loosened during normal operation.. This happened at 60 MPH, the car wasn't on the track.. Bazaar!

Only thing I can think of is a rod or main bearing bolt. But it does look like a length of a head bolt. Which ever it is its miserable to have that happen.

Rod bolt... :-) Guess how it ended up there?? It fell out! We never know what we'll see next around here!

What year? How many miles? ------------------- I bought a Miata years ago with a locked engine. Crankshaft wouldn't budge. So I started to take the engine apart. (BTW it had run dry) #2 rod only had one bolt and it was in the process of backing out. I pulled the rods and main caps. Thing still refused to budge, I'm thinking only the seals are left on the crank- this doesn't make sense. Ends up the missing rod bolt was wedged between the block and a counter weight.

One more motivation to get an aftermarket warranty

and I take a quote from my favorite 80's movie.. "its an imperfect world; screws fall out all the time". Just sucks for him it was that one. I wonder if he had any bottom end work done to the car recently by another shop or him self.... thats an odd one to fall out. ya know?

Before shutting down I'm always thinking about making sure the valve temps have a chance to semi stabilize. From the old days of guys managing to suck valves or warp them racing in the mountains I just have the thoughts etched in my brain. Having a 2 mile uphill I just wonder if the torque and the compression is giving me any amount of heat I'm best to cool off first. Half the time I'll either let it idle parked or take a different road at the top of the ridge thats flat to make myself think the engines getting treated right.

Quotes source post: 8021408

Did the bolt snap and come loose or did the nut just come off and allow the bolt to then fly free? As someone already asked, and it is a very insightful question, has someone been inside that engine? Sincerely, Macster.

Quotes source post: 8021497

Oh now.... its not like this happens every day. We also dont know the whole story of the engine or the past work done to it.. It could very well be a DYI project gone wrong... I have seen a lot of weird stuff people bring me on their computers too. But I dont make them do a daily back up becasue of one weird failure I never seen before.. When I do harder I usually find a virus related to a **** site or the customer tried to fix it himself and put in the wrong cooling fan..or they didnt put in any heat sink compound, and it ran for a while then burnt up.....And then blame the POS cheap computer..... when they caused the failure My point?, this failure could have been the result of work done to the car 3 years ago, maybe more maybe less. Wonder what kind of noise it made when it quit.. The engine had to come to a stop VERY fast.. and if a 6 spd car locked up the drive train. I bet it went BANG!

DAMN IT!! NOW I HAVE TO UPDATE MY RODS BOLTS???!!!

The engine was never repaired.. Original owner, 55K miles.. .3.4 from a 2000 996. The rest of it shot out the top of the block when the one remaining rod bolt snapped.. We've produced ONE M96 engine with stock rod bolts and even that one is too much, but at the time nothing else was available. Macster, these engines don't have rod nuts. The rod bolts thread into the connecting rod. What happened to this one is what we see often in tear downs, the bolt stretched which made it start loosening.. Then it loosened enough to fall out during operation. Usually this results in the bolt becoming a piece of shrapnel and flying out of the crankcase, but this one failed differently and the bolt was picked up by the timing chain. A Porsche Dealership that we work with had a new Cayman that had an engine failure, when they took the engine apart the rod bolts could be loosened with your fingers. In mid 2009 Porsche issues a TSB to change the rod bolt tightening procedure to include the bolt torques.. The values INCREASED. Imagine that. Why?? So some clown sitting in a cubicle can fight with you for 3 months about why they aren't going to pay you?? Our experiences with aftermarket warranties and big time engine failures have not been good. Not at all.

Quotes source post: 8021497

Is this on some auto reply algorithm?

Great pictures!

Another engineering oversight on the wonderous M96 motors. I cringe everytime I think about the day my motor rendered itself useless and came apart due to a broken cylinder sleeve on the way to work. This happened at just over 40k miles and with a lot of hard work I was able to convince Porsche to goodwill me a new motor even though the car was years out of warranty. Tic, tic, tic... and now to twist the knife... during the late '90s, early '00s, Porsche was the MOST profitable car maunfacturer in the world!

That is unfortunate. There must have been a defect or someone simply didnt tighten the bolt to spec. I have never seen that type of failure before. Which rod did it come from?

Quotes source post: 8021591

Should I pull the pan on my 15k mile '07 Cayman S 3.4 motor and check the rod bolts? Or just swap in some better ones?

I would say if you that concerned just retorque the rod bolts...

Came from cylinder #5 These are torque to yield bolts (stretch bolts) the only problem is that during operation they stretch farther than the torque values dictate, then they loosen. This isn't our first exposure to this MOF, just the most bazaar considering the bolt was picked up by the timing chain. Generally the bolt is never found, it usually launches into orbit.

Yes, that's definitely a 'WTF' kind of photo. Thanks for sharing.

Jake, do you recommend retorquing the bolts? To what values?

You can't re-torque a torque-to-yield bolt. They're a one time deal.

Quotes source post: 8022076

Yes, but Jake mentions Porsche TSB with new torque values from 2009 on; can they not be then tightened to the new value?

Don't you have to break the engine down completely to access them?

Quotes source post: 8022089

Unless Porsche got the torque value wrong such that bolt wasn't stretched to the proper yield point, I would imagine that the TSB also mandated new rod bolts too. Once a bolt reaches it's yield point, any more torquing after that will weaken it dramatically. I wouldn't trust a bolt that's been torqued past it's yield point, especially one holding my engine's bottom end together.

Jake, Is the rod bolt slightly bent? It looks from the picture that the force of stopping the engine in the chain bent the bolt. Rick 99 996C4 87 944S

What a thread to start off the morning....

It's called excessive deflection and it is a well known failure mode. It's the 'why' and 'how' that is the issue. It could be a single failure point where the product was missed in the QAR process (ie. heat treatment process failure) assembly error, product design issue, product material selection (ie low carbon (iron) steel instead of copper-nickel lets say). Even in the area identified as an 'assembly error' this could be an individual worker failure point or a production engineering issue (which is the worst for all of us) where Porsche based the design loads and set assembly requirements based upon faulty data. The same failure point can also be caused by a CFO interfering in the design process where by engineering request a budget for chromium metals and he tells them to pizz off and come back to him with an earth bound proposal using galvanized metals or something cheaper. The other issue could be owner abuse although it is usually manifested in other parts first. Unless you send these pieces to a metallurgist you won’t know how it failed and if the material is appropriate for the intended use. I’m willing to bet it is either a production engineering issue (material changed) or a product material final finishing issue. The skeptic in me says it was the fault of the Porsche CFO being driven hard by Wankerboy in the 90's to cut costs!

The TSB I referenced was related to new engine builds or very deep internal repairs.. No way to access the rod bolts with the engine assembled.. Even if you could torque them the process would be wasted effort since the bolts are torque to yield. Keep this in mind: The M96 engine in one of my cars has been used for all sorts of test work, I have held it at 7,400 RPM for 4 hours at a time on my chassis dyno, driven in second gear for 5 miles against the rev limiter, made abrupt downshifts that send the engine to a stage II over-rev and have generally tried to kill it. No luck, still going strong. I did that for 30K miles in 2007! From what we've seen its really hard to force one of these engines to fail through abuse, makes no sense but they seem to fail more often and more explosively when just being driven normally. We finished the autopsy of the engine this morning and will be inducting the entire assembly bottom end components onto our "trophy shelf" later this afternoon as more offerings to the Gods of Speed.. Here are a few more carnage pics...

Those are some pretty damn harsh tests there Jake. A bit OT, but do these engines have piston squirters ?

Quotes source post: 8022474

Yes, they do have piston squirters.. We were testing the IMS bearing retrofit during all that abuse, so the environment is never harsh enough.

Yuck, what a disaster.. looks like the engine is toast..

Quotes source post: 8022437

Love quotes like this. And the wife complains about me only taking it out to romp around in the hills.....

Quotes source posts: 8022474, 8022624

I was able to salvage the heads, everything else is destroyed.

Was it from your Millennium? or DE car?

Jake - When you assemble an engine do you use a thread locking compound on the rod bolts? I am wondering if in the case of the rod bolts stretching if thread locker would help in keeping them from coming loose.

So they do. I just checked some pics I have of a 996 engine torn down for an IMS bearing repair and sure enough I can clearly see the rod bolts have no nuts and in one pic I can see the threads in the hole in the rod. Interesting failure mode. Sincerely, Macster.

Quotes source posts: 8022076, 8022089

You can reused bolts. In this case bolts stretching is mentioned as a common thing that happens when torquing the factory standard. So I'd figure new bolts would be a very good idea. When torquing a bolt you only go to the value once. Any subsiquint torquing to the same value gives a higher torque value even though the gauge will read the same on a clicker or bar gauge. Loosening the bot back up then going after the proper torque again is the way to go. I believe in hand tight then then snugged. Then if chasing a proper torque like some heads require there is a stepped approach where lower values are used around the pattern to finally increase to the end value of torque. Sorry I'm being long winded on this but previously I watched a DIY video on here showing the torque wrench being used to repeat the same value on bolts torquing down bolts using the same end value. It's just not the proper way to operate a torque wrench and get the proper value. Just reloosen the bolt and go after the value again. Repeated tightenings like I mentioned gets the bolt over the set value even with just one second shot at the value. If the case is there is a new value as in you are wanting to go higher or lower after the engine has been using one prescribed value I'd say take the bolt out and clean things up or use new bolts. I'm not sure if there is a combination lube and lock tight solution. I'd imagine a lock tight sort of goop would act like a common lube used for attaining a good torque number but eh. Good luck.

Quotes source post: 8022918

The man said to floor it. Can do

Quotes source post: 8022437

You're not the only one who has said this, and it's just one more reason for me to keep beating the living hell out of mine, redline every chance I get, and my car runs like a top. If she goes, hell I've put 30,000 hard and fast miles on her and never thought twice about it.

Quotes source post: 8023120

No, but I also refuse to use stock rod bolts. Generally all my engines use FSI/ LN billet rods with custom aged ARP 2000 bolts. I have 13 years experience with these components in several different applications and a 0% failure rate. I've built ONE engine with stock rod bolts.. I didn't want to do it, but that was back when there were no other choices and all parts had to come from Germany. The bolts feel like spaghetti when they are torqued. I have used a stethoscope on the rod big end when torquing the bolts and slight pops and cracks can be heard while the bolts are being torqued to their yield point. SCARY!

Quotes source post: 8023468

Pac996, I'm afraid I have to disagree with you. Torque-to-yield bolts are designed to be torqued not just until they stretch a little bit, they're designed to stretch until they physically stretch to the point of yield. Once a bolt is stretched to yield, it will no long "spring back" to it's length; it is deformed permanently. However, bolts which are not designed to be torqued to yield (what you may be thinking of), are designed such that the torque value specified for that bolt will not be enough to stretch it to the point of yield. Only in those applications can you re-torque a bolt or stud.