Rod Bender True Stories - How it Broke and Why
 

Recently, an old friend had a OM603.971 engine failure. I am trying to understand it so I can help him decide what to do. I have read everything I can find on the Diesel Forum. If you, like Jack, have also experienced a 3.5 liter OM603 failure and repair or know about them, please tell us how it happens.

Jack's car is actually a 1992 300SD with 82K miles. It is from Oregon but Jack moved it to California. It as an excellent all original garaged car. Suddenly it started smoking huge amounts of blue smoke and clattering. He took it to Mercedes-Benz dealer. They compression checked each cylinder via the combustion chambers - all good. They suspected a turbo oil leak. They took off the turbo and saw wet oil in the exhaust side and ordered a remanufactured turbo from Germany. When they installed it - no change. Their conclusion is that the car has a defect inside the engine, so it needs a remanufactured engine for $13,000. Jack is has plenty of money which he has not accumulated by letting the MB dealer's service department have their way with him every time he comes to the shop.

I am not a mechanic, but if one wanted to see if the engine was drinking oil from the turbo, would you not simply remove the air pipe between the intake manifold and the compressor and let the turbo run while the intake aspirates normally ? If oil is only leaking out the drive side, the cold engine wont make blue smoke. I have not seen any threads on exploding turbos, so I dont think it would over rev.

When I asked the service advisor(who is paid by commision rather than salary) if they ever removed heads from any engines at their shop, he said no. I suppose that labor charges being equal that the profit margin on an engine is probably a bit more than on a head gasket.

What actually happens with these engines?
Is the bending always caused by hydraulic lock from fluid in the top of the cyclinder?
Which is the offending fluid - fuel, oil, water?
Where does it come from?

There is actually what appears to be a known defect in the head through 3-31-94 shown in the offical MB repair manual, I will try to attach the pdf article. It shows a "known to happen" oil leak into the #1 cyclinder. In fact, in the repair manual it says if the engine is smoking blue and clattering, that is the likely cause.

Has anyone actually ever heard of a 3.5 liter OM603 with only a bad head gasket?

With only a bent rod, can the engine get enough oil up from the bottom to burn blue smoke?

Does this sound like a "typical" 3.5 liter OM603 failure?

I will appreciate your input and I will also try to get someone to open the engine to see what we can see and get back with the results.

Thanks

The cause is not hydraulic lock.There is no known reason why it happens.

I am glad to hear that. I think that this means that there is a good chance that Jack's engine actually does have only a leak at or above the head gasket. Not the dreaded bent rod(s) ---yet.

The only way to tell is to get the head off. You will see if the head has a defect or if the head gasket is blown, or, more likely, one of the rods is bent. Basically these are wonderful cars with crummy engines. If the rod is bent the afflicted cylinder goes oval and then your oil consumption increases drastically and your idle becomes kind of loping - uneven speed, with a regular increase and decrease in rpm. Taking the head off answers all the questions. Jim

Having NO experience with this engine my thinking is that if this came on suddenly you probably have a better than 50/50 chance that is a cylinder head or gasket problem. The reason i say this is having read a number of posts about the "rod bender" my understanding is that the bent rod causes the cylinder to wear to an oval shape and ultimately the worn cylinder and piston rings allow lube oil to pass to the combustion chamber and cause smoking.Logically this would be a gradual process with first a little smoke then increasing to eventually alot of smoke. I would think the bad cylinder would show up on a compression test,but then so would a bad gasket! Boils down to what has already been said ,get the head off and see whats going on. Don

If you back the injector fitting off #1 and the smoking reduces it is likely the head gasket. As mentioned the onset of the bent rod problem is gradual in all the cases I have read as well.
The only exception would be if the cylinder was doctored up for resale. I just am not sure if this is possible. Car from dealer or original owner?
If a person paid the 13k for a new 3.5 engine it could suffer the bent rod symptom at some point as well. So if ever this engine were to be replaced a 3 litre block that does not exhibit the problem is a better replacement in my opinion. In most cases you only need the short block.
Expensive car new to be plagued with this type of problem and never really resolved by mercedes. They also should have sent the turbo out for a seal and bearing change rather than stiff your friend for a totally new unit. Oh yes it would not hurt to quickly eliminate the replacement turbo from any possibility it is bad as well. Reading the archives covers it well.
If head is removed the deck height of all the pistons must be critically checked. This is the best way to find any bending in the rods presence.

I don't know how a bad head gasket would cause oil burning.

I suspect there are a few engineers in Germany that have a good idea why it happens.

I own one of those rod benders with a "newer" MB engine,installed under recall about 3 years ago...before I laid eyes on it,my point in this is the following:there is guy close to my workplace who runs a tow truck business/dealer,something like that,who has in his back yard a complete 93'300sd with a new/newer factory short block,last time I looked at it(tried to buy it for parts),still had the label from MB on top of the engine...the story behind the car was that it was sold about 2 years ago,buyer paid,never picked it up,it has been sitting ever since,started once in a while.Looks good,complete,most likely a parts car by now,after sitting so long...however I belive it will easy roll on a transport truck .If your friend is interested those guys will most likely sell the whole thing to him,or just the engine,they have a shop and diesel mechanics there for the tow trucks...send me mail if your buddy is in the market,insted of selling his car for parts,because,you cannot find those engines anywhere used,in good shape.

On the front of the engine there is a passage. Obviously a pressure oil head feed. Apparently the head gasket can fail in such a way as to allow oil from that area into the combustion chamber of #1.
Since his problem developed almost instantainiously rather than gradual over many miles it is a possibility. Thats the content of the mercedes tech sheet he is mentioning I believe.
What also suprised me is the continious ongoing efforts to refine that head were still going on at that late a date. Or perhaps since the engine was discontinued even is a possibility.
Another thought I have been entertaining reciently is this engine type may actually benifit from the constant use of vegatable oil. Since the flame front or pressure applied on each stroke would be spread over a longer degree frame the rod loading at any instantainious point would be much less. Maybe the block could stay together that way for many hundreds of thousand miles May in fact be the only way to keep these engines together till normally worn out otherwise.
As someone above mentioned. It is reasonable to expect mercedes engineers would know enough to pass an opinion on this. For probable liability issues it is doubtful if we will ever be told though. Just common sense in another way though in my opinion.
He could also pull #s 1&2 injectors and see if number one is a lot dirtier than #2. From their description though the head has to come off. I would still do as much diagnosis with the engine together first as possible. Especially a leak down test.
One more last point. If it was an ebay transaction there is a small chance if recent they can backclaim on this. I would not hold my breath though.

main reason for this happening is the head cracks and let oil from the front oil gallery enter the combustion chamber. A few things you can try. getting the head fixxed (metal added and then milled to surface level. This can only be done on alluminum..) And then have the head retempered.

FYI: WIS service operation numbers
 

AR03.10-P-6111AW repairing connecting rod.

AR01.30-P-5800HX Remove/install cylinder head.

AR01.40-P-9291AW Boring and honing cylinder bores.

AR01.40-P-9273AW Installing cylinder liner.

AR01.40-P-9202AW Measure cylinder bores.

AR01.40-P-9272AW Remove cylinder liner

AR01.40-P-9271AW Widening cylinder bore in crankcase.

RA05.00-6-0103-41X Removal and installation of the guiding rails.

I agree. The typical rod bearing failure is very gradual and the oil consumption gradually increases with no sign of oil smoke in the exhaust until it's consuming one quart every 250 miles.

The current scenario is a sudden onset and severe oil smoke which points to a head gasket issue.

I really appreciate the input from all of you very much. It is unfortunate that Mercedes-Benz has taken the dishonorable decision not to admit to an obvious design and/or fabrication error. Designing the connecting rod would be a challenge because you want it to be light and tough and it is fatigue loaded. But to pretend there is no problem seems wrong.

Now one thing I forgot to mention is that Jack's wife is rather hefty and he takes her everywhere. She weighs about the difference between a W126 SDL and W140 SD. Some people think the extra weight is a factor. I hate to be the one to suggest it but...........could Jack's wife be the rod bender here?

I am a hair's breadth from buying the car myself just to see what is up with it? But please, if you have any more insight, I will aprpeciate the benefit of your experience. I promise I will report back when Jack's car is properly sorted out.

His wifes additional weight percentagewise overall is of little consequence in my opinion. What may have aggrivated this problem could be pushing the car harder than the last owner did or just luck of the draw.
Good old coincidence in action. The 3.5 can really accelerate if you come onto it. Something to see in a good example. Just get it moving a little first. The head cracking with car not showing any signs of overheating is a possibility. If no overheating observed the head gasket gains in favour. These are just my opinions though. Others will and should vary. Please post whatever it is you find out. Again most the 3.5s give a clear signal the rod is bending or has bent over time.

I had one with a broken head gasket, and cracked head because of overheating. (broken hose) rods were fine.
I also saw one that had the 10mm bolts that hold some plate under the crank come loose. the bolts swirled around in the pan until one managed to get through the screen. it went into the oil pump and busted a hole in it. engine lost oil pressure, rod bearings didn't get proper lubrication, engine failed. rods were still fine...

Posts here and other sites before the world wuz round...
 

Indicates that some of the rod bender engines were caused by the EGR valve. Little balls of tar like substance are formed with an active EGR valve and any leakage of oil from the turbo. Also mentioned is any pump back into the intake of unburned diesel fuel. Anyway these small tar balls apparently go/went into the cylinder to be lodged between head and piston resulting in bent rods. Sometimes allowing antifreeze/coolent into the cylinder also due to head lifting/cracking. IF so, how is Ma Benz gonna piss and ***** about a mandated "green" device mandated by the government?

This is all conjecture at this point, but a good precaution would be to inactivate the EGR valve via a plug in the line to that device.

Regards

Run-Em

A guess at the very most.

This theory has been around for awhile. But, to be valid, one would need to explain why the 603.961 engine suffers a small fraction of the connecting rod failures that the 603.971 engine suffers. If it was strictly an EGR issue, the failure rates should be comparable.

I agree with what was said above, I'd yank the head before assuming its a bent rod. Most likely its a head gasket or an issue with the head.

When you compare the pressure at full-throttle/boost on the piston to what it would take to crush a small ball of tar, doesn't make much sense.

I honestly feel like the W123's were the last cars produced by Mercedes Benz that were built to a standard and not a price. It seems like even though these cars retained their reputation, they became more stylish than engineered. :( I can't imagine the old Mercedes Benz company not doing their research and homework and letting an obvious flaw get mass produced like this!:(

????? How did he get it covered under a recall? Is there a recall I don't know about?? The last time these engines were available in the US in a car from the factory was model year 1995.

-Jim

What exactly is different about the engine? Is it just longer stroke? or is the bore larger as well?

And what about that flexible hose in the exhaust manifold of all OM603 engines. That looks like and expansion joint to me. The block and the manifold should have nearly identical thermal expansion. But the head's expansion rate is probably about 10 times larger than the iron parts. So maybe we get all of the issues with the heads squirming around on the blocks in addition to the bonus of longer, possibly more heavily loaded rods.

If MB knows what is going on with these engines it would help if they would tell us now. There could well be ways to adjust the excessive stresses out by de-tuning(via fuel timing), rpm limiting, or something.

Bore and stroke are both slightly larger, and longer, respectively

Nah the W140 is built better, they spent $1B and 8 years of R&D on that one. Lost money on every one they sold until the mid 90's. MB expanded the 603 and it didn't work well, I guess it was a stop gap until the 606 came out. However all the other engines the W140 was fitted with are supurb in all respects.

The 606 was out before the midpoint of the 3.5's life. The turbo 606 powered the W140's outside the USA.

I had a 3.5 603. It had a bad head gasket when I bought it.

We have had long and passionate discussions about the cause of the rod bending.

My favorite machinist and I have had long discussions over st pauli girls and we think the bigger bore and stroke of the 3.5 weakened the block enough so that it flexes too much and eventually causes head gasket failure. The head gasket fails and if you don't get it fixed quickly enough fluid can leak into the cylinder enough to cause a partial hydrolock and just bend the rod a little.

The combustion chamber holds only about 1.5 teaspoons before becoming full and anything from that on up will bend the rod a bit since liquid is not compressible.

A rod that is bent just a little will also twist and cause a slight binding in the cylinder and eventually will cause loss of compression, etc.

Nobody knows for sure what happens and if the benz engineers know (I suspect they do), they are not saying.

The replacement engines with (supposedly) stronger rods have also been known to suffer from the bent rods too, so making the rods stronger is not the solution.

I would only buy a car with the 3.5 motor if I could get it cheaply enough to put in a 3.0 liter block. The 3.5 heads are great.

When running correctly the 3.5 motor is a sweetheart...smooth powerful and very torquey.

Tom W

The problem with the theory is as follows:

1) A "partial" hydrolock can occur if the volume of fluid in the cylinder is about 3 ml greater than the available volume. Any fluid amount less than this and the engine will rotate without restriction. Any fluid amount more than this and the engine will stop rotation.

So, if the fluid is leaking into the cylinder at random rates while sitting.........there must be a situation where too much fluid leaks into the cylinder and the engine fails to rotate.

To my knowledge, none of the 603.970 owners ever reported such behavior on the engine.


2) The owners of the 603.970 never reported any signs of a failed head gasket during the period leading up to the issue. The engine runs perfectly...........no signs of rough idle or smoke until the oil consumption gets very high.

So, for the theory to be plausible, you'll have to find some answers to the aforementioned questions.

I disagree that the logic that there MUST be an example in which it fails to rotate.

The leakage only occurrs as long as there is differential pressure between the coolant system or oil system. As soon as some fluid drains off the pressure equalizes and the leakage stops.

This will not be a large amount of fluid.

Tom W

What does the EGR valve do? I have seen this before and the gist of it seems to be that it doesn't do anything that is important.

Exhaust Gas Recirculator.

It has to do with reducing emissions.

Tom W

If slightly too much liquid enters the cylinder.........the engine will fail to rotate. You can't compress liquid. Your scenario works only for the one specific situation where the perfect amount of fluid is sitting in the cylinder..........not too much (the engine won't rotate)...........not too little (the rod won't bend).

What would the effect be of reducing the combustion chamber until the compression ratio were lets say 60 to one instead of 20? or 80 or 100? all of these possibilities exist up to the point of a full combustion chamber.

Tom W

I understand what you are saying. I just don't agree that that automatically disproves the theory.

Tom W

The engine would simply push the liquid out the exhaust valve on the first (or third) stroke after the driver turned the key. The rod would not bend in such a scenario and the driver might get a bit of a rough start.........but, just for a brief moment.

It can only bend at the point where the combustion chamber is just slightly overfilled.

Are you thinking that the rod will bend under a compression ratio of 80:1? I haven't pondered that one..........but, I think the rod is strong enough to survive that.

If you agree that the available cylinder volume must be just slightly overfilled with liquid to bend the rod, the theory is largely disproved because of the statistically remote possibility for this to occur with no other symptoms.

However, if you believe the rod will bend at 60:1 or 80:1 compression, that's a completely different scenario. I'd need to study it a bit more.

One other point of note to completely dispel this theory:

M/B changed the design of the connecting rods for the later engines. Clearly, if they identified a head gasket leak as the culprit, they would have addressed this problem. The fact that they redesigned the rods tends to point to a fatigue issue in the rods rather than a failure caused by external influences.

Furthermore, a slightly redesigned rod won't dramatically reduce the failure rate caused by your theory. Sure, it's got a bit more capability, but if it can't withstand 80:1 compression, the new rod is not likely to withstand 90:1 compression.

I don't think we've got sufficient data on the new rods to make any conclusions as to their success or not........however.

Several folks have reported bent rods here with the replacement blocks with the "upgraded" rods.

Tom W

That would definitely help your case. Who are these folks.........I'd be curious if any symptoms of head gasket failure would be evident?

Yes, that is my question. At what point does a rod bend? Does it take a 101% full combustion chamber or would 99% do it?

Tom W

If the head gaskets fail due to excessive flex in the block, then a redesigned gasket would not likely solve the problem.

If benz did not know for sure what was causing the problem the upgraded rods may have been an easy to do solution with no assurance of success. We do know that all 3.5 versions suffer the bending rod syndrome, I believe, so benz eventually solved the problem by going back to the trusty 3.0 block with four valve heads.

This supports in my mind the theory that the 3.5 block flexed too much.

If there were nothing wrong with the 3.5 block why would they go back to the 3.0?

Tom W

You'd need to conclude that M/B was completely in error in the conclusion that the rods needed to be stiffened and that they had no clue as to the true nature of the problem. While this is certainly a possibility, I'm not willing to believe it without some data.

There is no question that the 3.5 block is not stiff enough when the proximity of the cylinders is observed. In fact, the 3.0 block tends to violate the typical spacing distances that are considered "normal".

Whether this lack of stiffness translates into a head gasket failure (no data) and that it further cascades into a connecting rod failure (no data) is a real stretch, IMHO.

Who spoke of a rod bearing failure? I did not, but if a rod is bent I would want to replace the rod bearing too.

And as far as data, is there any data to support any of our theories about the cause of rod bending?

Tom W

My error........I edited it to read "connecting rod".

Nope..........all gas.........no data.

But your theory is especially weak because of the lack of any anecdotal evidence to support the head gasket. If the head gasket fails, there are symptoms to support that failure..........you have none.

What do you mean? Nobody has had a head gasket fail?

My car had one.

What anecdotal information, or any other do you have to support any other theory?

Tom W

Do you have any evidence of a head gasket failure that occured prior to the subject connecting rod failure on a 603.970 or 603.971?

Did your vehicle suffer a connecting rod failure? I don't believe that it did.........and, therefore........your head gasket failure is irrelevant to the topic.

You've got a theory with no supporting evidence.........evidence that should be available from those who suffered the rod bearing failure.........if the evidence exists.

We're discussing your theory..........correct?

I could come up with a theory that a rock was kicked up off the roadway and got sucked into the intake and magically went though the turbo and into the combustion chamber and bent the rod. I don't have any evidence of that..........but, I'd like you to believe it because my machinist came up with the idea. Would that be acceptable to you? I think that this rock theory has some merit.........and I'm going to put it forth in the future..........how about that?

could one conclusively say: "350's are junk, stay away from them?"
just wondering.
as I wondered before...: why do people think MB is sooooo good??


because of their great parts department and because people think you are rich when your car has a star on it's hood?

Well IMHO its all academic. The bottom line is that the rods bend and its an expensive problem to fix. No one is sure how to prevent it, so your better off spending the extra effort earning the money to fix it.:D

Its a shame really, I have seen some cherry 350SDL's but even I'm scared to touch them. I guess the trick would be to get one super cheap with a DOA motor and repower it.

I wouldn't say they are junk, but certainly not MB's best effort. Every other MB diesel is supurb, at least the ones we got in the US. The 3L 603 is a fine engine in all respects.