I don't think it's only the rod because it's predominantly the #1 rod that bends. Sometimes it's #1 and others, but I haven't heard of others but not #1. In my case #1 and #6, at least from what I could measure.

I tend to think it's oil leaking into the combustion chamber pushing weak rods over the edge. The reason #1 is always involved is the poor cylinder head design. Oil probably leaks into #1 from the moment the car is driven off the lot. Over time valve seals start to go causing leaks into other cylinders.

Something else that occurred to me is maybe canted injection means the combustion charge doesn't propagate evenly putting unbalanced pressure on the piston. More pressure on the forward end of the piston. Doesn't answer the why #1 question. Maybe it causes hot spots that cause oil to coke...

Unfortunately we have no data on the propensity of the 2.9l 602 to bend rods.

Sixto
87 300D

If it is due to the head gasket losing its seal form a block that twists too much, It might follow that the end cylinders would fail more often.

Tom W

__________________
[SIGPIC] Diesel loving autocrossing grandpa Architect. 08 Dodge 3/4 ton with Cummins & six speed; I have had about 35 benzes. I have a 39 Studebaker Coupe Express pickup in which I have had installed a 617 turbo and a five speed manual.[SIGPIC]

..I also have a 427 Cobra replica with an aluminum chassis.

I'm late chiming in here but also a very interested party. As far as real compression #'s, I just did a test on my ill 3.5L a few months ago. It ranged from 420 to 450 psi across all 6 cylinders. The suspected #1 and #6 were not in the lower range.

In consulting a few mechanics on the direction to move forward, I had one that thought my head had been removed once already. This agrees with some of the stripped head bolts we found replacing the turbo. There was no listing in the MB service documentation, so it probably was not done by MB. I think the previous owner must have suspected something and maybe had the verdict given to him already. Oil consumption was at a quart per 2000 miles (@130K) when I got the car. It has degraded to a quart per 300 miles, but has now stayed there for about 10K miles. The engine runs perfectly, starts in the cold, no knock or nailing, no coolant loss. The dealer and another mechanic suggested the next step is pull the head to look for ovaling / tipped pistons or possibly a bad head gasket.

I do have a slight external oil leak near the exhaust on #1, between the head and block. This area seems to be a common leaking place. I think there is a high pressure oil passage in that area of the block going up to the head. If the head gasket leaks there I imagine it could put a significant amount of oil into the cylinder.

More misc data... Another person who went through this pulled his engine apart. He did see a minor bend to the rod. With the head off, it showed 2 "tipped" pistons similar to what Sixto saw. He pulled the pistons from the engine and saw that opposite corners of the piston were worn ( top on one side and bottom on the other I think but don't remember at the moment). The cylinder was oval but not that far out.

Chuck

Jack gave me his car. It will be back on the road in a few weeks one way or another. I plan to conduct a reasonably careful examination of engine #235to determine what the heck is going on.

Than you all very much for your thought and discussion thus far. I will appreciate a review my test plan before proceeding.

Background:

#235 began spewing large amounts of blue smoke and the engine began clattering at higher rpm. The onset was instantaneous. One moment it was consuming no oil and running perfectly, the next - blue smoke.

Shop "A" isolated the clatter to cylinder #1 by removing fuel to the #1 injector. They removed the cross-over the air pipe and observed oil. Their diagnostic conlusion - bad turbo seals.

Jack took the car to the MB dealer. He passed along the diagnosis. They performed a compression check (1-6, PSI) 550-400-370-425-390-400. They removed the crosss-over air pipe and observed oil. They concluded that the turbo was leaking and replaced it. They filled the car with oil and ran it at idle for 45 minutes. It clattered when revved, made blue smoke and drank 2 quarts of oil. They said also, "engine still runs very poor and starts runaway from excessive oil" The new diagnosis - broken engine, possibly piston #1 oil ring broken and leaking.

Theorectical discussion:

1) There are weird dynamic forces possible that could could help set up the rod for buckling like vibration and other dynamics but hydraulic forces can be huge, plenty to buckle the rod if loading is slightly eccentric. 603 rods bend under hydraulic lock conditions.

2) No fatigue. Too hard to screw that up in design. Would require sabotage and conspiracy, assuming designs are reviewed in the MB engine department.

3) 603 engines have a well-known propensity for head gasket faliure at the #1 cyclinder adjacent to the chain tower in the long web between the cycinder and a long oil gallery there. See The Mercedes-Benz 601,602,603 Engine Manual, Programmed Repairs Section page: PR01.1014/7 The article does not specifically include the 603.970 and 603.971 engines, however, it says the defect ran through 3-31-94. I assume this means the defective head/gasket was used until then.

4) the combinations of high pressure and thermal mismatch seem to have stumped the MB design people when dsigning Al head engines.

hypothesis:
#235 has the #1 cyclinder oil gallery leak. High pressure(550PSI) in #1 is from oil in the cylinder/combustion chamber. The leak is such that large amounts of oil enter the cylinder during lower pressure segments of the cycle. Compression test pressures are far lower than ignition pressures and the compression test simply measures peak pressures, not leak down. During ignition, combustion gasses are forced through the breach at high pressure causing an atmosphere of suspened oil droplets which are pulled by the turbo compressors suction side into the turbo, thereby supplying the entire intake plenum with a somewhat compression ignitable gas. Only #1 makes blue smoke becasue only #1 has enough excess oil to do so.

This is an optimistic hypothesis because if correct it means I get a nice car that is easily worth $10K for the cost of a head gasket repair.

All of the cylinders except #1 are within 15% of nominal pressure(I think ? what is nominal pressure for the test?) if any rods are bent now 1 will bet that no more than 2 are and that they are #1 and possibly #3.

The plan:

1) Look inside #1 cyclinder if I can get a scope into it somehow.

2) Remove #1 injector, take the guts out of an injector and screw the body into #1 , fuel bypass #1. This is to restrict gas flow so that #1 will still compress and suck a little, while not allowing too much oil build-up

3) Remove cross-over pipe, disconnet blow-by tube from turbo inlet.

4) run engine on 5 cycliners and observe.

If the engine runs dry with no blow-past, and oil squirts out of the #1 injector body then it probably has a broken head or head gasket.

Then either way the head is coming of for further examination.

I will appreciate yor review of this plan. I am especially interested in not damaging the engine any further. I think that what I plan to do will be safe, will not hurt anything.

If you can suggest any further precautions or tests, please do so.

I promise to make a complete report with many photos and measurements etc.

Thank You,
JMF

__________________
JMF

I have never seen or heard of a 615,616,617 engine leaking oil through the valve stem seals. At least not enough to notice. But what about 603? do the Al head engines ever leak/consume oil via the stem seals?

__________________
JMF

Head cracking between the oil supply passage and chamber as mentioned by another poster. Or a head gasket failure sound valid in this sudden onset situation. You will of course check for the rod being bent?
In the more common problem the onset is always gradual. I hope and expect you not to have a bent rod. The rod or events that cause the rods to bend seem to impact very gradually initially.
Probably as soon as oil consumption drops to more than a quart every 3k or even earlier. As this consumption is targeted probably at one cylinder initially That cylinder may be constantly a little overfuelled.
So it makes the problem gradually still increase by constant overpressure. Stresses may be generated well beyond design intentions. You do not to have to hydrolock almost to have the larger than design stresses. Now if also running at elevated temperatures compared to other cylinders is another consideration. Something is going to become problamatic with some abnormal thermal stress thrown in as well.
Very few if any engines are taken apart at this point so the rod may not even be bent yet. .Or the bending may be very gradual. Everyone is aware the fuelling of that cylinder is now going to increase substantially from the base oil. The resultant stresses are going to grow until the efficiency of that cylinder passes a peak. By then the damage has been done anyways.
Probably by the time oil consumption has reached a quart every thousand things are getting very serious in there.
Also mercedes has seemed to almost always had a problem with their #1 diesel cylinders. If you are going to throw a rod in a 616 for example it is good money it will be the #1. The actual cause of this is unknown to me other than the machining tollerance for #1 piston fitting is different. Not as prevelant in the 617 as I am not totally aware but still there I suspect.
Do not know enough reported incidents myself to get an indication of percentages. I do not know or hear anything on the 603 engine throwing rods if they do. Perhaps by beating down one problem years ago mercedes added another?

220D and 240D engines from before the W123 chassis had a throttle plate in the intake manifold that regulated air flow at idle, and just coming off idle. I believe 240Ds continued a version of this on later vehicles too, but it was much improved. Early Diesels took some getting used to as vacuum created by this device had to be overcome before you could transition from essentially idling to normal driving.

Anyway, this feature causes enough vacuum in the intake manifold to cause worn valve seals and guides to allow engine oil to dribble into the intake area and all over the top of the valves. On my old 220D the oil coked up all over the valves and lead to difficulty for the valves to seat, and then, it choked up the air flow passages. Head came off, replaced two seats and cleaned up the valves, replaced guides and seals, and the car was as good as new. Jim

__________________
Own:
1986 Euro 190E 2.3-16 (291,000 miles),
1998 E300D TurboDiesel, 231,000 miles -purchased with 45,000,
1988 300E 5-speed 252,000 miles,
1983 240D 4-speed, purchased w/136,000, now with 222,000 miles.
2009 ML320CDI Bluetec, 89,000 miles

Owned:
1971 220D (250,000 miles plus, sold to father-in-law),
1975 240D (245,000 miles - died of body rot),
1991 350SD (176,560 miles, weakest Benz I have owned),
1999 C230 Sport (45,400 miles),
1982 240D (321,000 miles, put to sleep)

The rods are Fine. Its not the rods fault. This problem is caused by hydrolock. Any rod will bend under that condition. Fix the reason its hydrolocking, and you dont have to build a new rod.

__________________
Have you ever noticed that anybody driving slower than you is an idiot, and anyone going faster than you is a maniac?

As long as they would add one additional commandment for you to keep thy religion to thyself.
George Carlin (Wonder where he is now..)

1981 240d (engine donor 1983 240d) recently rebuilt engine hurray! - No more.. fought a tree and the tree won.

pearl black 1983 240d 4speed (Converted!@$$%) atleast the tranny was rebuilt.

Related perhaps, ... or not.

I pulled the head from a junked '87 SDL today, 196,000miles, saw some interesting things.

The head gasket is of the original design, likely original, #14 head. No evidence on any of the cylinder seals or the passeges of any leaks, pretty much a perfect seal. Also no cracking. In the spare parts rack.

The turbo was a different story. Full of sludge, so was the intake and crossover. The wastegate was not hooked to the vacuum diaphragm, so it was left to be open all of the time, no boost.

Since the crossover dumps into the #1 & #2, it seems that oil or anything else ingested through the crossover would be likley to run down the #1 and #2 runners. This can occur with the engine off also of course.

This was an interesting case since the turbo wasn't able to produce boost and purge itself, in fact with the goo in it, it probably couldn't turn at all (about 1/3 full of what looked like driveway sealer) due to the wide-open wastegate.

Just considering the possibility that this engine (and others) might have enough blow-by oil to create a hydrolock or a partial hydrolock. It could accumulate in the riser from the turbo for some time on a long drive or idle, then when high-rpm comes along with the resulting higher airflow, a significant amount of oil/sludge could then make it over the top into the plenum and the cylinders.

So, could the hydrolock theory have nothing to do with head or head gasket failures? Funny that they manifest themselves on the latest design heads and gaskets, just doesn't make sense that they would be failing more than the earlier design.

Food for thought.

__________________

Gone to the dark side

- Jeff

After consultation with three local R&D forensic automotive engineering firms.

We can settle this issue forever.

Parts needed:
* 50 bad engines
* 15 running engines with certified accurate records, half for tear-down, half for dynamometer testing.
* 15 perfect vehicles for testing
* 10 months
+ 17,000,000.00 USD

YOu have presented some food for thought here.

I just read your post again and must confess that I did not follow it. The one thing I think I am clear on is that the #1 with the elevated compression must have either pooled liquid in it or a lot of carbon build up to get those pressures. I would suspect a bent rod there. The sudden onset points to a bent rod in my mind too, so I would check carefully for that while you have the head off.

Where I got lost in your explanation was where compression pressure vented into the oil galley. This being a return galley? I am unclear how this would find its way into the intake....

Tom W

__________________
[SIGPIC] Diesel loving autocrossing grandpa Architect. 08 Dodge 3/4 ton with Cummins & six speed; I have had about 35 benzes. I have a 39 Studebaker Coupe Express pickup in which I have had installed a 617 turbo and a five speed manual.[SIGPIC]

..I also have a 427 Cobra replica with an aluminum chassis.

The assumption is that there is a leak along the #1 cylinder seal of engine #235(my engine). After the breach and during each ignition when the #1 cylinder pressure can exceed 1000PSI at full throttle, combustion gasses are now forced through the breach. So these 1000PSI pulsations pressurize the oil galley or gallery or whatever it is called - the long cast-in and cross drilled oil channel that runs right next the #1 cylinder seal for about 3 inches.

The oil system, or a portion of it, might well be receiving a 1000PSI pulse every other rotation of the engine. These hydraulic pressure pulses could then be pushing on seals designed to hold peak oil pressure which would be somewhat less than 1000PSI. Furthermore, the loading would be fatiguing

A few possibilities come to mind:

1) Secondary seal breaches are causing oil suspended in now excessive gasses in the cam cover to be forced through the blow-by tube into the turbo inlet. (the crank case , chain tower and cam cover are all one open volume)

or

2) Secondary seal breaches of the oil system are somehow injecting oil into the intake system

3) High pressure exhaust gasses leaking from the primary breach are forcing oil to unanticipated places and/or in larger than anticipated quantities and getting into the cyclinders.

and

4) At some point cyclinder #1 begins to draw oil in and it accumulates faster than it is burned and expelled.

According to research done recently for SwRI, these are the most common failure modes for heavy duty diesel engines:

1) Cylinder head cracking
2) Combustion seal leaks
3) Cylinder block cracking
4) Excessive wear of rod or main bearings
5) Piston pin or piston cracking
6) Crankshaft failure

"Bent rods" dosen't even get honorable mention.

I do not understand how enogh oil can come up from the bottom of the cyclinders and get past pistons to loose 2 quarts of oil in 45 minutes at idle. I do not think enough oil can come up past the piston to burn blue smoke unless the piston has a hole in it or a broken ring or skirt.

I like the idea of oil related hydraulic overload for all of these 3.5 liter 603 rod failures. In the oiling system first, and then if you don’t catch it in time, rotation through TDC oil lock. Typically, diesel engines are designed for infinite life byusing a safety factor of about 1.5, and the peak load safety factor of about 2.0. That means each cycle could experience a 1500PSI load with no fatigue. It also means that the occasional 2000PSI lock should not break anything. Where can I get injectors or glow plugs with built in relieve valves?

If the cylinder seal begins leaking combustion gases first and then only sporadically, the leak would only be a gas leak into the oiling system. This would also mean that the cylinder seal ring would be subject to unusual stresses at each ignition. It would then be subject to unexpected fatigue loading and could work its way loose over time.


If my guessing is correct, then there is clearly a design/manufacturing flaw. The enlarged cylinder bore took thickness away from an already thin wall between the oil channel and the cylinder. When the tolerances stack up for size and location of the head, the oil channel in the head, and its wall thickness, the diameter of the cross drilled holes and their location, the location of the cylinder bore, the size and location of the head gasket and its seal, etc. It seems more than plausible that this is were the design limits were exceeded. When it all stacks up in your favor, the engine will go for 500,000 miles. When the inspector pinches a bit of molding sand in his micrometer and passes a too-thin head, your engine goes for 80,000 miles.

__________________
JMF

There was a 3.5L that failed and was posted in the past year, and it was said that there was no breach in the head gasket (if I remember correctly).

I am going with the ingested sludge theory from the sludge-filled 603 I just tore down, which is plausible, and wondering if the failure mode of the sludge/oil hydrolocking the early 603 was damage to the head and in later engines with the stronger head; damage to the rods.

Unless we find that 100% of engines with bent rods have breached head gaskets around the bad cylinder (compression metal ring around the cylinder). the head gasket leak cannot be the culprit IMO.

__________________

Gone to the dark side

- Jeff

All 603s have blow-by into the turbo inlet. With EGR you have bonus exhaust particles. Intakes always coated with sludge.

Gas leaks from #1 cylinder to the adjacent oil channel might be difficult to detect.

I am still strong on hydraulic lock because it is really about the only way that any internal combustion engines have ever typically gotten bent rods without disconnecting first.

I also like the idea that fixing the heads might have focused the problem onto the rods. But also keep in mind that the larger bore possibly weakened the #1 cylinder seal as well. If only the rods had been redisgned properly we could have broken cranks instead. I still want to buy some of those "pressure relieve" style glow plugs for my 603s. Where can I get them?

__________________
JMF

Thank you, I believe I did follow that explanation. It sounds plausible to me.

Tom W

__________________
[SIGPIC] Diesel loving autocrossing grandpa Architect. 08 Dodge 3/4 ton with Cummins & six speed; I have had about 35 benzes. I have a 39 Studebaker Coupe Express pickup in which I have had installed a 617 turbo and a five speed manual.[SIGPIC]

..I also have a 427 Cobra replica with an aluminum chassis.