A think some pictures of rods are in order....

__________________
1996 s10 zr2 with a 6.5 turbo diesel
1996 c2500 suburban with a 6.5 turbo diesel
1981 300sd with a turbo 617
Monte carlo figure 8 car

I too had read originally that there were no .97x engines outside of the US, then other postings contradict. I'm not sure. However, since the use of the car in the US through its early years (when rods were bending) was limited by the 55mph NMSL, and the same engine in Germany might be run at max. output for extended periods, I doubt that high-rpm and high-power made the US ones fail. It is possible that the stop-and-start useage was lugging the engine (a common component of scuffing) and causing some of the early damage, but if the engine was sold worldwide then there certainly was city use of engines in other countries, which it seems would have caused the same failures there.

As far as failed head gaskets and hydro-locking, I feel that is also a vaild theory although difficult to prove or disprove with existing data. My '91 had the head replaced, years before I bought it, had no unusual carbon nor oil buildup in the intake & cylinders (was running on commercial biodiesel FWIW). It is possible that its gasket or head problems were caught before it manifested itself into further damage.

Josha: Search Rod Bender, you'll find photos and data. Also, I mention the scuffing because I have two 603.960 engines apart now, one with about 200,000 miles, one with 247,000 miles, and the crosshatching is all visible, no scuffing. The difference between them and the 3.5l being bore, stroke, rod angle, wrist pin location.

__________________

Gone to the dark side

- Jeff

Interesting to note that the USA version has a different timing spec on the 603.970 than other markets: 14+/- .5 All .971 engines have the same 14*ATDC spec, and earlier versions all are timed at 15+/-1 ATDC.
(REF: FSM RA 07.1.1044-1100/3)

So, the timing was actually advanced 1* for the USA engine (and other engines starting 1989/1990) with the tolerance tightened up. Didn't seem to have any negative effect on the US market 602s, but I wonder why the timing advance was specified. Emissions likely.

__________________

Gone to the dark side

- Jeff

Lot's of questions so I'll try and find all I can 'from the book.'
If we look at it in the engineering matrix, it gets a little clearer than mud, hopefully.

1. State the problem. 3.5L engines in USA bend rods.

2. Redefine the problem. Do USA 3.5L bend rods more than other regions?

3. Identify the constraints/specs. They tend to bend before 75,000 miles or not at all.

4. Identify solutions. Redesign head gasket, Set max RPM lower.

5. Select most viable alternative. -see four-

6. Readdress problem. Are rods still bending?

1. The main reason the US versions bend more rods is the 3.5L was made primarily for the North American market. The majority of them sold in USA, in S class cars. Comparatively, not very many went to other regions before the rod bending was addressed.

2. The 3.5L 603.970 (350SDL W126) debuted in the USA market 03/90 and that specific production run was until 06/91. NOW let's look at the 'outside USA' production...3.5L 603.971 (300 SD W140) ran 07/91 - 05/93....keep these numbers in mind.

3. What happens BEFORE 75,000 miles?....Are the first 16 months on the road within the first 75,000 miles? Sure, unless assembly lines are making used cars...keep this number in mind, and include 'warranty coverage.'

4. The company actually DID issue a TSB about rod bending and went through the motions of a recall without actually issuing a formal recall. IN THE 75K (or less) range, they DID identify; the head gasket tended to erode in BOTH 3.0L and 3.5L near the #1 cyl. The gasket was redesigned with more material in the front. In the 3.5L, sufficient oil had been found in the cylinder(s) to cause a rod to be bent on compression, in some test engines - some, not all. A few didn't have any oil in the cylinders, they just bent rods. For THIS problem, the solution was to reduce the max regulated RPM, this is why the 3.5L had lower peak horse power than the 3.0L but higher torque and better 0-60 times for similar weight cars. Documents are closely guarded still but there were even bent/broken/thrown rods in 3.5Ls shown to be from oil starvation...from windage tray screws vibrating loose and being pulled into the oil pump - there's a post about all this somewhere on another forum but I don't remember where, someone even found the TSB number. I'm sure someone will find it before I do. Back to RPM, injection timing changes with RPM when the timing advance gear works correctly. Stands to reason that at high RPM.. with full advance.. on a longer stroke.. the rod is under exponentially MORE stress [or is it strain? ] than a shorter stroke... other things equal. BEFORE 75,000 miles MB USA replaced many engines that failed and corrected many that had not failed. They even replaced some failed engines out of warranty for original owners. What they did to correct the problem does acknowledge the problem but it still wasn't 'officially' a recall. Did it really need to be?

5. Repairs, before or after failure, included both solutions. These were being done before the release of the OM603.971 model, outside the N. American market. Makes sense that the factory WOULD implement the new changes on the assembly line. Whoa - I think we got it!

6. "Rods still bending?" No. "Would we keep looking into why?" How much do you want your Christmas bonus?

--Now if somebody can find that TSB number....

__________________
1987 300D (230,000 mi on a #14 head-watching the temp gauge and keeping the ghost in the machine)
Raleigh NC - Home of deep fried sushi!

YES, especially after the trap oxidizer debacle.

__________________
1987 300D (230,000 mi on a #14 head-watching the temp gauge and keeping the ghost in the machine)
Raleigh NC - Home of deep fried sushi!

The trap-ox was though never on the inclined-injection engines, but both timing specs were. Also, the glow-plug was shortened to help improve mixing in the prechamber at the same time.

It stands to reason that several things would change to improve emissions, max. RPM is one of them as a diesel's emissions get worse as RPMs increase, a simple factor of not being able to speed up the flame front plus the port-effect of the injection pulse getting to the nozzle.

Also, the timing device is mentioned. Unless I read it incorrectly, I believe that the advance curve of the timing device ends around 2200rpm, or at least below 3000rpm meaning that max. advance would be reached long before the max. RPM of the engine, and due to mechanical/accoustic/hydraulic/combustion constants would actually be more retarded at higher RPMs.

I'm not sure whether you're saying that these statements are directly from a book somewhere, or derived from data in some books, but am very interested to learn more about the source(s).

There are certainly documentable contributing factors to the rod bending, but many/most of these factors also were present in the 602.96 and 603.96 engines which did not tend to bend rods, ... we're missing the trail to why these factors didn't affect the many .96 engines but did the .97s.

__________________

Gone to the dark side

- Jeff

Not to say the trap-ox bent rods, BUT MB did not want to go through another massive payment to the EPA for yet again for diesel emissions. The trap-ox recall cost a heap of $ in parts and emission fines.

They did really play up the 603 series as a big leap in emission improvement.

Yes, MB issued a technical service bulletin for rod bending (internal notice to all dealers to fix a problem.) It was not officially released to the public (TSBs usually aren't) but somebody on one of the Yahoo diesel groups had a scanned copy of it. That was ~5 years ago and it may have been removed for 'copyright.' The production dates I have are on a DVD I got at the factory. I have tried to post parts of it on you tube but it's copyright protected and in PAL format. The EPC also has production run dates but you really have to dig to find them. At that time, I was in Germany seeing family, and an in-law who owns a garage. I tried to get some 350SDL parts from him for a customer. I then learned from the factory the 3.5L was built for the US/Canada market - specifically the 603.970. SO...if we look at when the .971 and .972 came out (after the gasket redesign and RPM limitation)...the issue(s) were already corrected.

The .96X 3.0L have almost identical have almost identical parts to the .97X but IIRC the strokes and bores are not the same. That changes the load on the rods, perhaps to a point they can't take when the RPM is highest. Also, with smaller bores, a little gasket erosion would not allow a dangerous amount of oil in the cylinder. The bore difference was actually documented as a suspect cause. When we look at when and where the 'rod benders' came out, and when they were corrected, it makes sense as to N. American models bending more rods.

Truth is Mercedes or any car maker, is not going to voluntarily tell the public the whole story about any defects. A guy I used to know during high school is now a detective. We were shooting the bull one day a few summers ago. He compared detective work to counter-intelligence, "If you put the little pieces together you can figure out anything." I am convinced the factory was looking at multiple variables at the same time, and they suspected more than one cause. Sometimes we don't really need to see the words in a TSB to know what they were by the actions of the company. They replaced engines with bent rods (some out of warranty) and corrected/adjusted engines that hadn't yet bent them.

__________________
1987 300D (230,000 mi on a #14 head-watching the temp gauge and keeping the ghost in the machine)
Raleigh NC - Home of deep fried sushi!

I feel like i got bashed for bringing up rpm...

__________________
1996 s10 zr2 with a 6.5 turbo diesel
1996 c2500 suburban with a 6.5 turbo diesel
1981 300sd with a turbo 617
Monte carlo figure 8 car

I dont think so Josha.
They are just being forth right like you !!
Just remember its only the internet & sticks & stones........... !!
At least you have not been a cry baby !!

__________________
Grumpy Old Diesel Owners Club group

I no longer question authority, I annoy authority. More effect, less effort....

1967 230-6 auto parts car. rust bucket.
1980 300D now parts car 800k miles
1984 300D 500k miles
1987 250td 160k miles English import
2001 jeep turbo diesel 130k miles
1998 jeep tdi ~ followed me home. Needs a turbo.
1968 Ford F750 truck. 6-354 diesel conversion.
Other toys ~J.D.,Cat & GM ~ mainly earth moving

I have to agree with this. having seen a lot of failure mode effect analysis carried out both by computer and on dynos and road cars, I have to wonder if MB would actually run that close to the edge on rod strength, considering the economic and marketing risk.

Just to refresh your memory: The so-called "rod bender" car cost 55,000 in 1990!!!!!

so you're probably talking the equivalent of a six-figure car nowadays! Given the expense per unit produced (EPUP) and expense per unit sold (EPUS) and the warranty cost of ONE rod post production, PLUS the reputation for over-engineering, even with the absurd cost-cutting and greed by Juergen Schrempp in those days, it would astound me if MB consciously chose to risk so much on on a such a tiny cost savings.

__________________
Strelnik
Invest in America: Buy a Congressman!

1950 170SD
1951 Citroen 11BN
1953 Citroen 11BNF limo
1953 220a project
1959 180D
1960 190D
1960 Borgward Isabella TS 2dr
1983 240D daily driver
1983 380SL
1990 350SDL daily driver alt
3 x Citroen DS21M, down from 5
3 x Citroen 2CV, down from 6

Not at all...what the factory did, besides changing the head gasket, was lower the max limited RPM. That supports your idea. If we 'go with what we know' everything else is speculation. 1. 603 series was kind of rushed into production because they were in a 'we have the fastest diesel car' fight with Peugeot. Not all the R&D testing went into them that went into other engines. 2. The 3.5L was sold in USA/Canada 16 months before any other market as the .970. The others were .971 and 2 mostly in G models (W463) in South America, Africa on the open market and to NATO military commands. This was well into the time the factory knew about rod bending and the corrective changes were already in place and working. Turns out they are good for 4X4 use because they make good torque for their size. BUT, rod stress and strain do not go up in a linear way like RPM. Under some conditions they go up exponentially with RPM.

Regarding fuel...nowadays Cetane (promotes combustion under compression) and Octane (inhibits combustion under compression) are no longer true enumerations of the content chemicals. They are an 'equivalent index' of them. --see photo--

In Germany and other places, they DO NOT USE the same numbering/labeling system as in USA for either one....The square yellow label on the pumps show the formula "R+M/2 method" IE: the average of 'research' and 'motor' octane. That's always going to be lower than research octane only since research number assumes complete combustion and motor number is an actual 'run it in a motor on a test rig & listen for knock' test. What we see is 'regular = 91 research, midgrade = 95 research' *not shown or dispensed at this pump "super+plus" (premium) = 98 research' They do actually correspond to 87, 89 and 93 octane in US markets using R+M/2 and not just R.

As for diesel fuel...according to this at the time these these engines were being built, cetane was actually higher in the US market than in Europe. I have not found any confirmation or denial of this linked info as of now. EU minimum cetane of 38 in 1994 and 40 in 2000. US- minimum 40, testing in the 42-45 range. **NOTE** as of summer 2010 cetane rating is not required labeling on diesel dispensing pumps in Germany.

translation of photo
(gas1)
"This fuel complies with German Industrial Standard EN 228 regular sulfur free 91 research octane."
(gas2) same except "Mid-grade sulfur free 95 research octane" *again note, 'super' in Germany = 'mid-grade' in USA. and 'super+plus' in Germany = 'premium' in USA
(diesel)
"This fuel complies with German Industrial Standard 51 628 diesel fuel sulfur free."
(rectangle)
"Contains up to 7% biodiesel" I'm told by my bro-in-law it's about 2% during winter and about 5% during summer.


IMO...? (that and $7 gets a whatever designer coffee is fashionable these days) I think some bent rods were from oil, some from load on the rods at high RPM. I do not believe cetane or timing were factors because they both stayed the same AFTER the rod bending stopped.

Attached Thumbnails

 

__________________
1987 300D (230,000 mi on a #14 head-watching the temp gauge and keeping the ghost in the machine)
Raleigh NC - Home of deep fried sushi!

I agree that is a bit cloudy. If cetane and timing were that much of an issue, we should see them both do the same thing.

The .97 is essentially a bored and stroked .96. Same gasket could have sealing issues with a larger bore. Longer strokes will put more lateral load on the rod, especially @ higher RPM. I know the differences are not big but they still need to be factored in.

__________________
1987 300D (230,000 mi on a #14 head-watching the temp gauge and keeping the ghost in the machine)
Raleigh NC - Home of deep fried sushi!

IMO is the flavor of this discussion since no documents have been referenced, linked, copied, even exist.

The RPM argument doesn't make any sense to me, since the '90/'91 OM603.970 was already at a reduced max. RPM vs the 3.0L OM603.96, one of the reasons it only made 134hp vs the 148hp 3l engines. The later OM603.971 RPM was raised back up to within 100rpm of the 3.0 though, which increased its HP again. This along with advanced timing by 1* didn't have any more problems. The head was changed again, as was the head gasket, after the introduction of the .971 though. Does this support one poster's hydrolocking theory? I'd be interested in whether the head bolts were also changed in this case.

I still see this as only opinions and theory, with no conclusive answer, no test data, nothing at all. Anyone who calls this one is getting sand in his ears. Whether it was hydrolockiing from leaking gaskets, irregular combustioin from cetane, lugging from US market low-speed operation, sulfur reduction from regular to LSD to ULSD, weak rods, a rush to market, sunspots, we'll never know unless some real data emerges.

My experience with Mercedes-Benz Engineering was that nothing would begin production until the proper FEA was complete in critical areas, and FMEAs complete & analyzed, and any critical failures rectified. This experience goes through the introduction of the 60x series so I don't buy the rush to market theory, ... at all. I saw examples of the opposite, where an introduction was delayed pending engineering changes.

I do enjoy the theories and discussion however.

__________________

Gone to the dark side

- Jeff

Yes, the documents exist. I have some old microfische from a now closed factory dealer. Covering engines OM603.96/97 turbo. Dated 11/92 document numbers 07 102 2034 03, 07 102 2035 03, 07 102 2036 03 and 07 102 2071 03. Here's the problem- my local library got rid of the fische readers and the only other way to copy them, in my entire county, is a local company that charges $3.00 per page. 180 pages per sheet x four sheets = not happening without the $. Here's the rub! The factory never issued a formal recall. They DID do all the contact of purchasers and corrections that made it look like a recall. They played their cards close and did not have to release lots of information the way the would have under a recall.

The same information might be found in the W126 FSM CDs, I don't know, I don't have one. What remains as cold hard fact are the factory changed the head gaskets and RPM settings in .970s (3/90-6/91) and that was all they changed as of 11/92. These changes are shown in the microfische. They were in effect before the .971 (7/91-5/93) and .972 (9/91-7/93) had stopped production. The rod bending stopped after these changes. The fuel (according to ASTM and DIN) contradicts the idea that US fuel was causing the damage. At the time they were built, German diesel was 38 min. cetane and American fuel was 40 min. cetane.

The rush to market was actually a point of pride with the factory. This is detailed in the documentary "Fascination: The History of the Mercedes Benz Brand" (c) 2007 Producer- Wolf Nagel. Historical consultant DaimlerChrysler Archive- Dr. Harry Niemann. {paraphrased from narration} "With the introduction of the six cylinder diesel family OM603, Mercedes Benz made a quantum leap in emission reduction while simultaneously making gains in power and torque up to 15% more than the venerable five cylinder family. This was even done in a period of just less than two years before its January 1985 debut." I bought this two disc doc. set at the factory museum in Stuttgart in PAL format. When I find a way to work around the PAL and copyright protection, I will post it.

I am honestly not trying to be king of the playground but if I have any information and people are looking for it, I do NOT believe in withholding it.

__________________
1987 300D (230,000 mi on a #14 head-watching the temp gauge and keeping the ghost in the machine)
Raleigh NC - Home of deep fried sushi!

Some direct quotes of the documents in question would certainly help your case, regardless of whether they can actually be produced.

As for the RPM change, my '91 SDL had the lower RPM settings, as did all .970 engines. I think that you need to check your source on that. Further, the later .971 engined had a higher operating RPM, both of which tend to contradict your theory that lower RPM reduced failures.

The head gasket was superceded several times according to GSXR's data, as was the head, the final head revision (22) being after the .970 ended production and toward the middle of the .971 production from the same source.

I've been able to view some of my Mercedes-Benze Microfische using a loupe (sp?), it can be enlarged at a printing shop as we used to enlarge printed-circuit film into blueprints. I don't know how many shops still have these Brown cameras though.

I wonder if any radiologists have equipment for enlarging images? Seems that X-rays used to be stored on microfische the same way we archived obsolete blueprints before document scanners.

__________________

Gone to the dark side

- Jeff