X5 4.4 - Sludge in engine and oil filter housing... help!

[X5 Meister]

This thread has really grown by leaps and bounds, fortunately however, buried within it is some really great information. One of the best posts, if not THE best IMHO, is #180 above by Weasel. I agree with basically everything he is saying, but at the same time I want to throw in a few of my own comments / observations into the mix.

1. BMW recommends 100,000 mile transmission fluid changes, not "lifetime" which actually is a recommendation for other other fluids such as coolant. (Now let's forget for a minute what is the motivation behind this recommendation) The caveat is that this recommendation applies to transmissions that were operated under "normal conditions." What constitutes "normal" I personally have no idea. Is it me driving a car or my grandmother? Further, in the case of a vehicle specifically designed to go off road, tow up to 6,000 lbs, and yet still perform like a sports car, I REALLY have no idea what is "normal!" In any case, the recommended fluid change interval is in fact different for an X5 driven in not-normal conditions (let's put it that way).

2. The multitude of problems that come with trying to define normal / not-normal situations are pretty obvious. So here's just one example. If you bought your car used, who knows how the previous person drove it? Maybe they pulled a trailer every day up until the point of sale? Maybe they thought they were Michael Schumacher (though not the current dude in F1)? Or maybe they drove to Bergdorf's every other day for shopping? All 3 of which I would consider not-normal, for differing reasons. Since you don't know for certain, perhaps an early fluid change is the way to go? And since you really DON'T know, perhaps an early fluid change at ANY mileage is the way to go?

3. In something I read, ZF seems to back point #2, however they state that looking at the fluid should be your guide. Last time I checked, my eyeball did not contain a microscope and built-in fluid analysis equipment to make an educated determination on fluid composition. Few people have JCL's experience in this stuff and so unless the fluid is burnt to a crisp, I don't know any one of us normal people who would know if indeed the fluid needed changing or not without some sort of detailed lab report to go by. Interestingly ZF recommends that the fluid be changed when it smells burned, however "the damage is already done." Nice one.

4. On their website, (see attached screen shot photo) they discuss their LifeGuard Fluid 6 (which is a special fluid blended by Shell for ZF; Shell calls it M1375.4; see photos) as offering extended maintenance intervals to 100,000 km / 62,000 miles. Interesting. This is certainly no 100,000 miles / 162,000 km recommendation. So if an EXTENDED maintenance interval on this fluid (which is after all the OE fluid and the ONLY fluid recommended for this gearbox) is 62,000 miles then perhaps driving in not-normal circumstances would warrant fluid changes at intervals shorter than 62,000 miles?

5. The transmission maintenance recommendations from ZF (see attached fluid recommendations and interval recommendations below) have some different information. Here they indicate that should not-normal (my words) conditions exist, maintenance intervals should be as low as 80,000 km (49,600 miles) or as high as 120,000 km (74,400 miles) or 8 years. So now what does this mean? If you don't have a clue how your transmission was treated should you have the fluid changed at 49,600 miles? How about if you have 25,000 miles and your car is 8 years old? Are they saying that the fluid is now sufficiently broken down to warrant changing?

6. In a letter from ZF circulated to shops (see attached) it discusses "lifetime fill" and leaves it up to the car manufacturers' to decide what that means, however it does address the notion of what I am calling not-normal conditions. In such conditions it calls for 100,000 km or 8 year fluid changes. Yet another different recommendation!

7. Given all the above, the way I understand the recommendations is that the best option is to have a fluid analysis done in order to determine whether or not a fluid change is necessary. A company like Blackstone Labs (no I don't work for them) might be the best option, see attached sample report (thanks to the original poster of it in a different thread).

I don't have any of the answers, just wanted to post some of what I had discovered on this clearly complicated matter.

Confused yet?

[JCL]

Originally Posted by X5 Meister

This thread has really grown by leaps and bounds, fortunately however, buried within it is some really great information....

Meister, good background info.

A couple of comments to muddy the waters (fluid) further:

We keep discussing miles (or km) here when in fact that is not likely to be the best predictor of life. I have experience with heavy duty off highway transmissions in haul trucks and other applications. These trucks (up to 400 ton capacity) have a form of automatic transmission, and mechanical drive, not electric drive. As the trucks got larger, and when we got up to 2000+ hp (they are now up over 3500 hp) the impact of a failure was so extreme that the monitoring of components like transmissions got very precise.

We had traditionally measured component life by hours (since these are low speed applications, up to 60 km/hr). What emerged was that there was a better indicator than hours, or truck load, or grade, and that was the number of transmission shifts. It became a predictor for transmission component exchange intervals. On a long haul road out of a deep mine pit, the trucks would sometimes shift at every cross road, as the effective grade changed. That caused early transmission failure. It turned out that the grade didn't matter as much as the consistency of the grade.

We would have fleets of 20-50 trucks, at each of 10-20 minesites, and so could produce good analytic data (which we can't really do with over the road vehicles, where we have anecdotal reporting at best).

The point of all this is that just like with heavy equipment, the number of transmission shifts is likely to be an indicator of failure. Driving on the highway in 5th or 6th for 100,000 miles is nothing like driving in city traffic for 100,000 miles, and I think that this multiplier is far more than it would be for trailer towing, or spirited driving, which you mention.

Second point is about fluid sampling. I am a big fan of fluid analysis, have worked in a fluid analysis lab, and used it for years. The company I worked for had a very large lab (not Blackstone) That said, it has limited value when applied to cars. That is simply because the value is in trending the results, preferably with a fleet. Point samples are virtually useless. Many car owners do one, or two samples, and use it to justify what they already believe to be true. Other vehicles can't be compared to theirs, due to different duty cycles, driving conditions, fluids, etc. An oil sample can find water, or antifreeze, and that is valuable. It can measure TBN, which if trended can show additive life. It can measure viscosity, but you need to know what the acceptable viscosity is to make that valuable. Unfortunately, what usually happens with oil samples is that people focus on the metals found through spectrometry, which are dissolved in any case. Those results are telling you about your engine (or transmission), not about your oil, in most cases, unless you are paying additionally for large particle analysis with your oil sample.

Edit: I just had a look at the attached sample, and I think that it supports this. There is no detailed information on the results expected for the specific transmission (they don't even note who made the transmission, so how could they?) The universal averages are for an indeterminate population. The viscosity is (surprise) within spec for both measures, yet there is a recommendation to change the fluid based on elements found in the fluid that may have been there since break in. I wonder if that copper came from the clutch plates, or if there is perhaps a cooler in this system that could have provided the copper over the last 83,000 miles? And then with all that inconclusive data, the interpreter advises to change the transmission fluid after 10,000 -15,000 miles. That doesn't seem to be a supported recommendation based on the sample results as reported. This just shows the cautions needed when using fluid analysis, which is in general good science.

[X5 Meister]

Excellent points, and particularly interesting real world examples I must say.

I agree with your thoughts on fluid analysis and that, just like in virtually every science, multiple data points are preferred if one is to make any educated guess on the future based on prior results. I think what Blackstone is in reality saying in their assessment is that they need more data points to see what is really going on. And much like you said, an analysis at another relatively short mileage would help answer (somewhat at least) the question as to when some of the particulates are appearing in the fluid; at break in a long time ago or relatively recently.

What I do find interesting is that ZF seems to be implying that since modern fluids are quite stable and have the ability to hold particles in suspension for longer periods of time, that failures caused by fluid changes shouldn't happen... at least up to a point. The question is just what is that point... 80,000km, 100,000km, 120,000km, 100,000 miles or 8 years?

Your heavy equipment example got me thinking. What would be an interesting thing to look at is what other car manufacturers that use the same transmission and fluid recommend. It won't be a perfect comparison, because those manufacturers (Audi, Bentley, Hyundai, Land Rover, Jaguar, Maserati, etc) will all have different software to operate the transmissions differently from BMW, however what would be fascinating to know is what Maserati, for example, thinks in terms of fluid change intervals on a ZF 6HP26 transmission used under their definition of "normal" conditions in a sports car versus Jaguar who might use it in a luxury saloon, versus Land Rover who uses it in a true SUV, versus Hyundai who uses it in their whatever mobiles?

[chefwong]

I'd like to chime in on my recent drain interval on such a hotly debate. YMMV, and I would think twice about changing fluid if you are in the higher mielage and do not know the previous mechanical history

At 75K, this would probably be my 3rd fluid change - maybe even 4th. Too lazy to look at my records...Fluid is CHEAP to me. I have done 1 drain bolt as well as a a pan drop. At 75K, I just did another drain bolt. I was having some weird noticeable slight lag issues upshifts to 3rd gear. It's not a placebo effect but refreshed fluid did make that go all go away FWIW.

[Weasel]

Now that being your 3rd change by the 75k mile mark will keep you from having the bad side affects of waiting till a higher mileage like 100k for the first fluid change.