[Efthreeoh]

Originally Posted by freon480

well it did reset but only to 12k. not that ive gone that far. so we can confirm it is not programmed to work passed a certain mileage.im consuming/leaking a quart every 5k. i usually change when the +1 message first appears. which is around 5k. oil pan/possible rear main seal has been leaking since 90k. ill fix when I do clutch if that goes before engine lol. I noticed it would not tell me oil level right away after starting like it should although prior to oil change I was getting +1qt message immed after starting for a few days as I was procrastinating tossing a quart in to only drain it out for the oil change. im eagerly awaiting thw next oil change now. im very curious to measure oil correctly. the reminder I couldnt care less about since I dont do my changes by that. but if there is an issue with the capability to correctly measure the oil level then im intrigued to find out how or why the two are connected.

BMW suggests that once the OCI will not reset, to start changing the oil at 7,500-mile intervals, which tells you the type of influence the oil quality input has to the algorithm used to calculate OCI. I can't prove that there is an causal relationship between the CBS not using oil quality data and the oil level measuring capability of the system. All I do know is that once my engine reached the mileage where the CBS stopped using the oil quality data to calculate the OCI, its oil consumption behavior totally changed. It could be I'm leaking a bit more oil from the rear seal and pan gasket, or perhaps the oil separator is allowing more oil into the combustion chamber. Observation shows the belly pan is not all that covered with oil, and my exhaust seems to have the same level of carbon at the exhaust tips as it always had.

[mweisdorfer]

Originally Posted by Efthreeoh

That's the other concern I have; if the software is programmed to stop accepting the data from the oil quality section of the sensor, and BMW can't code the warning out, what's to say that the integrity of the computer code is good enough to properly interpret the oil level data the sensor produces.

When the e-dipstick started acting abnormal (compared to what it has behaved like for close to 200K miles) and reporting oil level discrepancies, I really got concerned the whole system is f'd up. I just changed the oil about 3,200 miles ago, and the e-dipstick seems to be acting normal again. I just went down a 1/4 quart after 3,200 miles. That extrapolates out to about 13,000 miles for a quart of oil burned, which is normal for my engine at this stage in its life.

I know it will be expensive, but the sensor is located in the bottom of the oil pan, so it is easily accessible. Can't one just change our the sensor?

[freon480]

Originally Posted by mweisdorfer

I know it will be expensive, but the sensor is located in the bottom of the oil pan, so it is easily accessible. Can't one just change our the sensor?

I believe he has already done so.

[Efthreeoh]

Originally Posted by mweisdorfer

I know it will be expensive, but the sensor is located in the bottom of the oil pan, so it is easily accessible. Can't one just change our the sensor?

I did change it. That is how I discovered the issue. I explained that in the first post.

[freon480]

so yesterday my +1 quart light came on today. im a hundred miles from 215k. I drained oil and had 5 1/2 quarts come out. I didnt change filter though so im sure not all oil came out. I reset cbs w snap on scanner again. this time it set to 7000. I wonder where the glitch is. it seems that after a certain mileage that it will not reset like before but I think we are getting different reset mileage which intrigues me. it seems as it is still computing information and not completely shutting down. I believe my oil level readings are off by 1/2 quart. I felt I had the same discrepancy last oil change as well. maybe next oil change ill change sensor and see what happens.

[Phil325i]

Originally Posted by freon480

I didnt change filter though so im sure not all oil came out.

No offense, but why on earth would you change the oil without changing the filter..?!

[Efthreeoh]

Originally Posted by freon480

so yesterday my +1 quart light came on today. im a hundred miles from 215k. I drained oil and had 5 1/2 quarts come out. I didnt change filter though so im sure not all oil came out. I reset cbs w snap on scanner again. this time it set to 7000. I wonder where the glitch is. it seems that after a certain mileage that it will not reset like before but I think we are getting different reset mileage which intrigues me. it seems as it is still computing information and not completely shutting down. I believe my oil level readings are off by 1/2 quart. I felt I had the same discrepancy last oil change as well. maybe next oil change ill change sensor and see what happens.

There is almost a half quart of oil in the filter housing. Like Phil said, why would you not change the filter? With very long drain intervals it is imperative to change the filter.

Regarding the reset issue you are experiencing, it is the software not the oil sensor that is altering the reset value. I'm not sure if you read this entire thread, but that is exactly how I discovered the issue. I thought it was the sensor, so I changed it and still had the issue that the system would not properly reset. That's when I took it to the BMW dealer and it was discovered that BMW programmed the CBS to stop using the oil quality data from the oil sensor. You'll be wasting your time and money if you try an change the oil sensor; I've been down that road already.

[freon480]

Originally Posted by Phil325i

No offense, but why on earth would you change the oil without changing the filter..?!

Originally Posted by Efthreeoh

There is almost a half quart of oil in the filter housing. Like Phil said, why would you not change the filter? With very long drain intervals it is imperative to change the filter.

Regarding the reset issue you are experiencing, it is the software not the oil sensor that is altering the reset value. I'm not sure if you read this entire thread, but that is exactly how I discovered the issue. I thought it was the sensor, so I changed it and still had the issue that the system would not properly reset. That's when I took it to the BMW dealer and it was discovered that BMW programmed the CBS to stop using the oil quality data from the oil sensor. You'll be wasting your time and money if you try an change the oil sensor; I've been down that road already.

if you recall as i said above i change my oil usually every 3k. my last oil change was 5 weeks ago and i changed filter then. bmw recommends 15k so obv the filter is ok for that long. although i dont change my filter every change i do change it every other which still has me changing it twice for every 15k at least. ive done it like this for the last 115k. and at 215k i think its safe to say my method does no harm. if yall are using your filter for 15k the 7k i use it for is quite sufficient. as far as the cbs and its programming i think what interests me the most now is we have different reset values on the decline. i believe you had 3 resets before it went to 0. im on my 3rd and although i dont change at same intervals as you i still am able to reset to a mileage other than 0 so far but it does decrease everytime mileage wise. i still get a date of 2016 for my next service date though which is funny considering i do between 3-5k a month.

[Efthreeoh]

Originally Posted by freon480

if you recall as i said above i change my oil usually every 3k. my last oil change was 5 weeks ago and i changed filter then. bmw recommends 15k so obv the filter is ok for that long. although i dont change my filter every change i do change it every other which still has me changing it twice for every 15k at least. ive done it like this for the last 115k. and at 215k i think its safe to say my method does no harm. if yall are using your filter for 15k the 7k i use it for is quite sufficient. as far as the cbs and its programming i think what interests me the most now is we have different reset values on the decline. i believe you had 3 resets before it went to 0. im on my 3rd and although i dont change at same intervals as you i still am able to reset to a mileage other than 0 so far but it does decrease everytime mileage wise. i still get a date of 2016 for my next service date though which is funny considering i do between 3-5k a month.

I checked your posts, it wasn't clear in you posts that you change the oil every 3,000 miles and change the filter every other oil change. I'm not sure why you just didn't follow BMW's oil change regimen. What I did get out of your posts is that for several oil changes you didn't reset the CBS and let the CBS go negative by 29,000 miles. I'm relatively certain your change regimen without resetting the CBS has affected the OCI when you finally did reset the CBS.

On the other hand, I followed BMW's oil change regimen from day one for 200,000+ plus miles, so the way my CBS has reacted is most likely the way BMW intended it to once the magical 186,000 mile milestone was reached and use of the oil quality data is suspended from the system. The date interval is independent of the mileage calculation, so it's not unusual that the CBS has a date that seems totally out of alignment with the mileage-based OCI.

All I can tell you is don't waste your money installing a new oil sensor because it will not change the behavior of the system past 186,000 miles, and besides you've never used the CBS in the way it was intended, so it doesn't matter to you anyway. Your oil change regimen is out of the norm; it hardly makes sense to not change the filter when refilling the system with fresh oil. Based on your posts I've determined that your engine consumes oil at a rate higher than mine (about double) and I have 30,000 more miles on my engine. I understand it's just a data set of two (your engine and mine) and obviously we drive our cars differently, but it is an interesting evaluation of the two different oil change regimens.

The takeaway of all this is BMW is a screwy company. If it was GM doing this, most of the E90 Post members would go into full GM-bashing mode and expect something like this to be "typical GM". But for the hallowed BMW to do it, some how they get a pass (just like with the HPFP issue) from the members of this Forum. I can say BMW has lost one customer over this, me.

[alexwhittemore]

Originally Posted by Efthreeoh

The takeaway of all this is BMW is a screwy company. If it was GM doing this, most of the E90 Post members would go into full GM-bashing mode and expect something like this to be "typical GM". But for the hallowed BMW to do it, some how they get a pass (just like with the HPFP issue) from the members of this Forum. I can say BMW has lost one customer over this, me.

Of course, this ignoring the most likely scenario that this is simply a bug. Granted, it's bullshit they just don't seem to care about fixing it, but I'll bet my $5 it's a software bug somewhere that some engineer is telling his manager "give me two days and it'll be fixed" and the manager is saying "it affects .01% of customers and would take an extra $100 of labor per affected customer besides."

[Efthreeoh]

Originally Posted by alexwhittemore

Of course, this ignoring the most likely scenario that this is simply a bug. Granted, it's bullshit they just don't seem to care about fixing it, but I'll bet my $5 it's a software bug somewhere that some engineer is telling his manager "give me two days and it'll be fixed" and the manager is saying "it affects .01% of customers and would take an extra $100 of labor per affected customer besides."

It's not a bug. It was intentionally programmed into the software. Look at the BMW document I posted in this thread. The BMW dealer told me BMW programmed it that way. If it would take two days, they'd have fixed it. BMW NA sent it to BMW AG as a PUMA case. BMW AG could give a rat's ass. Made my decision easy for my next purchase, which will not be a BMW.

[alexwhittemore]

Originally Posted by Efthreeoh

It's not a bug. It was intentionally programmed into the software. Look at the BMW document I posted in this thread. The BMW dealer told me BMW programmed it that way. If it would take two days, they'd have fixed it. BMW NA sent it to BMW AG as a PUMA case. BMW AG could give a rat's ass. Made my decision easy for my next purchase, which will not be a BMW.

That document only proves that the problem is known, not that it's intentional. These are the types of bugs programmers spend 60% of their time trying to test for and 20% of their time trying to fix. It's a miracle of thorough test methodology that, with half a million lines of code in the car, it even runs at all.

Not to suggest that it's okay they don't bother fixing it.

[Pilot a330]

Thread starter Efthreeoh why is it then when I change my oil at 6000 miles the oil change readout does not change and stays at what it was removed at? These sensors are just rubbish and in my opinion they act just like a countdown.

[Efthreeoh]

Originally Posted by alexwhittemore

That document only proves that the problem is known, not that it's intentional. These are the types of bugs programmers spend 60% of their time trying to test for and 20% of their time trying to fix. It's a miracle of thorough test methodology that, with half a million lines of code in the car, it even runs at all.

Not to suggest that it's okay they don't bother fixing it.

Now you are pissing me off. I've been through all this shit with BMW from the dealer level all the way to BMW NA Customer service and beyond. I'm not sure why you think you know more about this issue than me and why you want to continue to have this stupid discussion about software bugs. I've dealt with BMW for almost a year now trying to get them to fix it. It is not a fucking bug in the software. The VP of Service of BMW of Sterling told me BMW NA Engineering told him that BMW AG purposefully programmed the CBS to stop using the oil quality data to determine the OCI once the car has reached 300,000 kilometers (186,000 miles). I specifically asked if it was a software bug. The VP of service said the CBS is purposefully programmed that way; he said it was the dumbest thing he had heard from BMW. You need to realize that BMW of Sterling would not have had that level of technical discussion with a customer (i.e. me) since most BMW customers wouldn't know what the fuck he was talking about. Most BMW customers don't even know the first thing about how the oil level/quality system works. The easy answer would have been, "Oh, it's just a software bug."

If it was a true software bug, BMW would have found it and fixed it. Rather, if you read the document I posted, it clearly states that updated software is installed to stop the CBS from issuing a notification that the car is past due on oil service once the CBS OCI can no longer be reset. BMW spent company software programming resources to develop a software patch to change the behavior of the CBS rather than find your supposed bug in the software. The software patch is not intended to fix the software code that calculates the OCI to work past 300,000 kilometers. The software patch is to prevent the CBS from annoying the customer that the car needs an oil change every time it is started. The document clearly states that once the car reaches 300,000 clicks, the owner is to start changing the oil on a pre-determined mileage schedule. BMW of Sterling told me BMW NA (in accordance with the document) recommended changing the oil every 7,500 miles once this condition is reached. Sterling brought in two East Coast BMW NA service reps to review my car. They even hooked my car up to BMW's NA engineering department network to allow the engineers at BMW NA in New Jersey have a crack at it. Both BMW of Sterling and BMW NA reloaded my car with the latest software updates and the software patch noted in the document (a total of 5-times - they had the car 8 days). None of it worked; the CBS still shows the oil service is over due every time I start the car. What it doesn't now do is tell me it needs oil service every time I turn the engine off (thank God). The Service Manager of Sterling told both BMW reps that this customer (i.e. me) has kept his car in such good shape (at 220,000 miles at the time) that if BMW can't fix it, BMW should give me a new car just for the hassle I've been through.

FWIW, I work for a very successful large systems integration company and I am very familiar with the design, development, and testing processes of software and bug fixes. It makes no sense that a software bug would show up and prevent the CBS from using the oil quality data from the sensor exactly at 300,000 miles. It makes no sense that BMW would use company resources (hell, they might even subcontract the SW development anyway) to develop a SW patch to change the CBS notification rather than just use a bug fix tool like Bugzilla to find the errant code and fix it.

So can we drop the discussion now? Thanks.

[Efthreeoh]

Originally Posted by Pilot a330

Thread starter Efthreeoh why is it then when I change my oil at 6000 miles the oil change readout does not change and stays at what it was removed at? These sensors are just rubbish and in my opinion they act just like a countdown.

The system is not designed that way. The oil quality sensor is only one data point in a multitude of other data that is used to determine the OCI. The other data are things such as cold starts, engine warm up time, road speed, time, miles driven and MPG. BMW has been using a conditioned based oil service interval algorithm since the mid 1980's, so the concept is nothing new. What BMW started with the N52/N54 was add in a oil quality sensor to determine the level of oil contamination by measuring the dielectric of the oil and comparing it to a predetermined level of undesired contamination. However that information is used in conjunction with the other data sets mention above; it is not the sole determination of when to change the oil.

What you are thinking that is if you just keep adding in new fresh oil, the oil quality should improve and the OCI should lengthen. However, BMW has designed a fool-proof system where the oil quality data is stored and the system is programmed to ignore improvement in the oil cleanliness (this is why BMW wants you to add top-off oil in 1-quart increments); it only determines when the oil reaches an undesired contamination level. The OCI is also determined by time (i.e. the low mileage scenario where the car is not driven far enough in a 12-month period to trigger a use-based oil change), which is why the system needs to be reset at every oil change to start the time counting part of the algorithm. Dielectric-measure based oil condition systems have been in use for decades in the machine tool industry and large over-the-road trucks. So use of the oil quality data to determine OCI is nothing new either. BMW decided to incorporate it into the N52/N54 CBS.

[alexwhittemore]

Originally Posted by Efthreeoh

Now you are pissing me off.

Sorry to hear that, I'm not trying to.

The VP of Service of BMW of Sterling told me BMW NA Engineering told him that BMW AG purposefully programmed the CBS to stop using the oil quality data to determine the OCI once the car has reached 300,000 kilometers (186,000 miles).

I didn't read anywhere you saying that you'd been told specifically it was intentional. More to the point, if it were the case, it seems like BMW NA wouldn't have wasted so much time and debugging effort on it, but I guess maybe you mean that after all that effort, AG told them "it's a feature not a bug, fuck off"? Would certainly be shitty.

if you read the document I posted, it clearly states that updated software is installed to stop the CBS from issuing a notification that the car is past due on oil service once the CBS OCI can no longer be reset.

Again, it only shows that they know about it, not that the behavior is intentional. Though apparently there's other information mentioned above suggesting it's intentional that I hadn't seen.

BMW spent company software programming resources to develop a software patch to change the behavior of the CBS rather than find your supposed bug in the software. The software patch is not intended to fix the software code that calculates the OCI to work past 300,000 kilometers. The software patch is to prevent the CBS from annoying the customer that the car needs an oil change every time it is started. The document clearly states that once the car reaches 300,000 clicks, the owner is to start changing the oil on a pre-determined mileage schedule.

Unless you mean something different from what's mentioned in the OCI Note document you linked out front, the procedure in that document isn't a "software patch," and didn't require a programmer's interaction. Enabling and disabling individual aspects of CBS is a configuration function accessible through NCSExpert and certainly the more advanced ISTA/P/ISIS/ICOM/whatever they call that tablet they've got.

It makes no sense that a software bug would show up and prevent the CBS from using the oil quality data from the sensor exactly at 300,000 miles.

Oh I can think of a bunch of reasons, though none of them are "holy crap that's definitely it" likely - buffer overflow, out of memory, state machine logic error...

It makes no sense that BMW would use company resources (hell, they might even subcontract the SW development anyway) to develop a SW patch to change the CBS notification rather than just use a bug fix tool like Bugzilla to find the errant code and fix it.

Again, unless there's something else you haven't mentioned, the fix up front in the thread is a pretty simple workaround. Though I bet you're right - I bet the code was originally written by, say, Continental (or even a subcontractor of theirs), which may be why changing it is deemed more trouble than it's worth.

Again, I'm not TRYING to antagonize you. It's just that in all my (extensive) experience as a software engineer, this sounds for all the world like a bullshit bug that someone isn't being given resources to fix. I'm not saying BMW is above being shitty, but I've also been provided no evidence to suggest WHY it'd be programmed intentionally this way. I can think of some reasons, but none are better than "don't piss off the customer."

And my real point at the end of it all is that it seems a sort of lame reason to swear off BMW as a brand - if it IS a bug, it's not as though Audi or Hyundai or Bentley is going to be less prone to this sort of thing.

Of course, if you're pissed off at how they're handling it, that's a pretty good reason.

[Efthreeoh]

Originally Posted by alexwhittemore

Sorry to hear that, I'm not trying to.



I didn't read anywhere you saying that you'd been told specifically it was intentional. More to the point, if it were the case, it seems like BMW NA wouldn't have wasted so much time and debugging effort on it, but I guess maybe you mean that after all that effort, AG told them "it's a feature not a bug, fuck off"? Would certainly be shitty.



Again, it only shows that they know about it, not that the behavior is intentional. Though apparently there's other information mentioned above suggesting it's intentional that I hadn't seen.



Unless you mean something different from what's mentioned in the OCI Note document you linked out front, the procedure in that document isn't a "software patch," and didn't require a programmer's interaction. Enabling and disabling individual aspects of CBS is a configuration function accessible through NCSExpert and certainly the more advanced ISTA/P/ISIS/ICOM/whatever they call that tablet they've got.



Oh I can think of a bunch of reasons, though none of them are "holy crap that's definitely it" likely - buffer overflow, out of memory, state machine logic error...



Again, unless there's something else you haven't mentioned, the fix up front in the thread is a pretty simple workaround. Though I bet you're right - I bet the code was originally written by, say, Continental (or even a subcontractor of theirs), which may be why changing it is deemed more trouble than it's worth.

Again, I'm not TRYING to antagonize you. It's just that in all my (extensive) experience as a software engineer, this sounds for all the world like a bullshit bug that someone isn't being given resources to fix. I'm not saying BMW is above being shitty, but I've also been provided no evidence to suggest WHY it'd be programmed intentionally this way. I can think of some reasons, but none are better than "don't piss off the customer."

And my real point at the end of it all is that it seems a sort of lame reason to swear off BMW as a brand - if it IS a bug, it's not as though Audi or Hyundai or Bentley is going to be less prone to this sort of thing.

Of course, if you're pissed off at how they're handling it, that's a pretty good reason.

All I can to do relay to this thread what BMW told me; which is that it was an intended software design with no explanation from BMW AG why it is so. I have no idea how BMW programs the software, what language it uses, or processes/standards it follows. I used the term software patch for lack of better terminology since I don't have any detail of how the code is conceived, written, nor executed. If there is functionality that is switchable within the CBS code to enable a mode where the CBS does not report a oil change pas due notification, the fix BMW tried to accomplished didn't work regardless. But the takeaway here is the issue is not a bug, but intentional programing on BMW's part as that is what I was told by several independent sources within BMW. What I don't get is why BMW needs to make the CBS OCI functionality go defunct at 300K kilometers. Now maybe there is a "bug" regarding how the CBS is supposed to behave AFTER 300,000 clicks and that the original code was supposed to eliminate the notification altogether and it doesn't work properly, but that is not the original discussion we were having.

Regardless, the whole thing is stupid to program obsolescence into the CBS OCI functionality. And you are right, BMW has no desire to fix the problem. What I as the consumer gets from all of this is that BMW intentionally programmed the car to have a CBS functionality problem once 186,000 miles rolls around, which if one has concern over resale value, this issue certainly detracts from selling the car at the best possible price. It artificially detracts from the value of the car. And without my discovery of the problem, this would be wholly unexplainable to the prospective purchaser of the used car. For that reason, I have no faith in BMW anymore and will not purchase another new car from them.

[alexwhittemore]

Originally Posted by Efthreeoh

What I don't get is why BMW needs to make the CBS OCI functionality go defunct at 300K kilometers. Now maybe there is a "bug" regarding how the CBS is supposed to behave AFTER 300,000 clicks and that the original code was supposed to eliminate the notification altogether and it doesn't work properly, but that is not the original discussion we were having.

That's actually one possible behavior I'd considered - like, maybe the engineers decided they don't trust the condition sensors on an engine so old. But right, even if that were the case, it'd be easy enough to simply ignore the condition sensor and report service required every 7500mi as if nothing was wrong.

[Efthreeoh]

Originally Posted by alexwhittemore

That's actually one possible behavior I'd considered - like, maybe the engineers decided they don't trust the condition sensors on an engine so old. But right, even if that were the case, it'd be easy enough to simply ignore the condition sensor and report service required every 7500mi as if nothing was wrong.

Exactly. The CBS could have easily just switched to a standard 7,500 mile interval and leave it at that. BMW gains a lot of revenue from service and parts for BMWs well past 186,000 miles, the company has a entire department dedicated to maintenance and preservation of BMW models, and the company quite enjoys a reputation of 2nd and 3rd tier ownership experiences of older, high-mileage BMWs; so none of what they did with the CBS oil life monitor system for the N52 makes any sense.

[Obioban]

Ugh, just stumbled upon this thread for the first time. What a horrific and stupid issue.

Even if BMW considers the 186,000 mile switch over a "feature", I'd argue the electronic dipstick misreading after is absolutely a bug. An inexcusable bug, given that there's no mechanical dip stick to use as a backup.

My wife's car only has 61,000 miles on it currently (and puts on ~15,000 per year), so I won't be dealing with this soon. Still, I'm exceedingly frustrated just reading your post. We spent a LONG time looking for a car specifically optioned to hold up over the long haul-- no turbos, no automatic, no awd, no iDrive. After removing the runflats and swapping in non runflat subframe bushings, I thought I was in the clear on modern BMW stupidity. Seems like not.

It seems like there's a good chance BMW is never going to do anything about this. Perhaps it's time to turn it over to the coding guys. At which point, time for some questions. Starting with, do we know were condition based service information is stored? As in, is it an ECU function, a cluster function, or something else?

Seems like somewhere in the code there's likely a "300,000". If it could be located, should be relatively similar to alter it to, say, 300,000,000.

So frustrating! I don't even use the condition based service (I've been doing 10,000 mile OCI's), so I don't care about losing that. But, a warning chime on every startup and loss of electronic dipstick functionality (on a car with no dip stick) is just inexcusable! UG!!

[hassmaschine]

I've looked and I don't see anything with a 300,000 in it. there's a maximum interval of 30,000 miles and 720 days, unless there's a mistake in the definition and that 30,000 is actually 300,000. it's probably hard coded in the program space which is why BMW can't fix it with a simple data change.

anyone good with IDApro?

[Obioban]

Originally Posted by hassmaschine

I've looked and I don't see anything with a 300,000 in it. there's a maximum interval of 30,000 miles and 720 days, unless there's a mistake in the definition and that 30,000 is actually 300,000. it's probably hard coded in the program space which is why BMW can't fix it with a simple data change.

anyone good with IDApro?

If it's hard coded we should still be able to read it, just not write to it, correct?

Are you seeing said interval in the ECU or cluster?