IMS failures on automatics?

[norcalconvertible]

Yea yea, I am casually looking for an AUTOMATIC boxster. (or a BMW z4) I know I won't have the same driving experience as a manual, but I haven't driven a manual in over a decade and my wife never has. Auto it is for us.

Are the automatics known to have the same IMS failures as the manuals? Autos are less likely to have had the IMS replaced because of no clutch changes I assume.

So far, I have limited my search to 1997-1999 cars as I have read in several places that those have the less risky bearing.

I am wondering if the IMS happens to be less of an issue on the automatics for some odd reason, allowing me to broaden my search criteria.

[JFP in PA]

Originally Posted by norcalconvertible

Yea yea, I am casually looking for an AUTOMATIC boxster. (or a BMW z4) I know I won't have the same driving experience as a manual, but I haven't driven a manual in over a decade and my wife never has. Auto it is for us.

Are the automatics known to have the same IMS failures as the manuals? Autos are less likely to have had the IMS replaced because of no clutch changes I assume.

So far, I have limited my search to 1997-1999 cars as I have read in several places that those have the less risky bearing.

I am wondering if the IMS happens to be less of an issue on the automatics for some odd reason, allowing me to broaden my search criteria.

IMS failures are common on all M96/97 engined cars, regardless of transmission type. And just as a reference, the dual row IMS was used up until the 2001 model year, but 2001 cars could go either way. And before you ask, no, there is no known way to positively identify which 2001 engines had it and which were single rows; you have to pull the car apart and look.

[tonythetiger]

Originally Posted by JFP in PA

IMS failures are common on all M96/97 engined cars, regardless of transmission type. And just as a reference, the dual row IMS was used up until the 2001 model year, but 2001 cars could go either way. And before you ask, no, there is no known way to positively identify which 2001 engines had it and which were single rows; you have to pull the car apart and look.

All true, except to call the IMS failure "common" feeds into the fearmongering of the IMS issue. I do agree its a concern. FWIW, I am on my 4th 986 and never had an issue, even replacing IMS on 3 of them. Every bearing I removed looked perfect. (2 dual row and a single) I also don't know anyone that has personally suffered a loss.
History and sanity check:
Through the years, Porsche used one of two bearings; a double-row and a single row. Earlier 986 boxsters used the double-row, switching over mid-year to a single row in 2001. The advertised rate of failure for single-row bearings is 8% for cars over 90K miles, but that number seems to come from guess who? The folks who have made tens of millions on a solution.
Dual Row has a statistically lower failure rate, estimated at around 1%. That said, statistics don't mean much to you if your bearing fails, right?

If your research leads you to worry about this, then for your peace of mind replace the bearing or buying a car that the previous owner addressed the issue in. IMHO, the best-advertised solutions out there are over-hyped and ridiculously overpriced. It's a bearing and it is easy to remove and replace the transmission, a little time and common sense or experience, but if you dont work on the car yourself, this is gonna cost you some cash. I chose peace of mind.

[JFP in PA]

Originally Posted by tonythetiger

All true, except to call the IMS failure "common" feeds into the fearmongering of the IMS issue. I do agree its a concern. FWIW, I am on my 4th 986 and never had an issue, even replacing IMS on 3 of them. Every bearing I removed looked perfect. (2 dual row and a single) I also don't know anyone that has personally suffered a loss.
History and sanity check:
Through the years, Porsche used one of two bearings; a double-row and a single row. Earlier 986 boxsters used the double-row, switching over mid-year to a single row in 2001. The advertised rate of failure for single-row bearings is 8% for cars over 90K miles, but that number seems to come from guess who? The folks who have made tens of millions on a solution.
Dual Row has a statistically lower failure rate, estimated at around 1%. That said, statistics don't mean much to you if your bearing fails, right?

If your research leads you to worry about this, then for your peace of mind replace the bearing or buying a car that the previous owner addressed the issue in. IMHO, the best-advertised solutions out there are over-hyped and ridiculously overpriced. It's a bearing and it is easy to remove and replace the transmission, a little time and common sense or experience, but if you dont work on the car yourself, this is gonna cost you some cash. I chose peace of mind.

Porsche’s last published data showed 12.6% failures for single rows, 3%+ for dual rows, and that data is now several years old, so there have probably been more in the intervening time. Porsche also produced a third design (2005-2008), the non serviceable unit because of its size, and the last rate of failures on that design was 1-2%. So, in essence, all versions failed, just a differing rates. To my knowledge, there is no viable correlation between mileage and rates of failure; we saw them die in cars with less than 10K miles, and well over 130K miles.

If you are not going to replace the facotory IMS, it is all an amount of your risk tolerance.

[tonythetiger]

Originally Posted by JFP in PA

Porsche’s last published data showed 12.6% failures for single rows, 3%+ for dual rows, and that data is now several years old, so there have probably been more in the intervening time.

Is this publically available and can you reference? Without doubt, the design is prone to failure, and even at the lower numbers, it's disconcerting. Id like to know more as I have found a knack for swapping these out in my simple home garage. (i am not a mechanic by trade) Folks in the industry (such as yourself?) will have the exposure to failure since they come to you when something goes wrong.
I haven't seen an IMS failure, but I have seen 3 different engines with cracked heads, always passenger side on the 3.2L. (high mileage 140K+) Suspension wear is another common issue on EVERY boxster I've owned or seen. Overall, this is a well-made car and the little things that are small frustrations or maintenance like leaking spark plug tubes, water pumps, cracked coils, window regulators, microswitches (everywhere) the visor covers, convertible top, etc. just come with having an older car and don't detract that much from the enjoyment, in fact the projects tend to be fun.
Catastrophic engine failure is different. I am replacing my friends IMS for a 2002 996 in the coming days just because he has been worrying about it for years. As easy as this repair is, except for the exorbitant price of parts, this favor will at least let him sleep easier. However, I regard the 900 price tag for a bearing and flange opportunistic. The pelican retrofit is priced favorably, but I'd like to see a stronger bearing in that kit to trust its fixed.
do others have thoughts?

[MrBen]

I say buy it drive it and have some fun don't worry about some dang bearing that's been way oversold. All things mechanical have a weak point. Every chain has a weakest link. For what you can buy these cars for and the value fun factor you receive its a bargain. That said I made enough in the market today to buy 2 more. We may not all be in the same boat

[piper6909]

Originally Posted by tonythetiger

... However, I regard the 900 price tag for a bearing and flange opportunistic. The pelican retrofit is priced favorably, but I'd like to see a stronger bearing in that kit to trust its fixed.
do others have thoughts?

I've been curious about this one:

https://www.europeanpartssolution.com/ims-bearing-upgrade-kit

In case it gets censored: europeanpartssolution dot com

I've never tried it, nor have I seen any feedback on it from anyone who has direct experience with it.

[JFP in PA]

Originally Posted by tonythetiger

Is this publically available and can you reference? Without doubt, the design is prone to failure, and even at the lower numbers, it's disconcerting. Id like to know more as I have found a knack for swapping these out in my simple home garage. (i am not a mechanic by trade) Folks in the industry (such as yourself?) will have the exposure to failure since they come to you when something goes wrong.
I haven't seen an IMS failure, but I have seen 3 different engines with cracked heads, always passenger side on the 3.2L. (high mileage 140K+) Suspension wear is another common issue on EVERY boxster I've owned or seen. Overall, this is a well-made car and the little things that are small frustrations or maintenance like leaking spark plug tubes, water pumps, cracked coils, window regulators, microswitches (everywhere) the visor covers, convertible top, etc. just come with having an older car and don't detract that much from the enjoyment, in fact the projects tend to be fun.
Catastrophic engine failure is different. I am replacing my friends IMS for a 2002 996 in the coming days just because he has been worrying about it for years. As easy as this repair is, except for the exorbitant price of parts, this favor will at least let him sleep easier. However, I regard the 900 price tag for a bearing and flange opportunistic. The pelican retrofit is priced favorably, but I'd like to see a stronger bearing in that kit to trust its fixed.
do others have thoughts?

Let’s start with the simple: The Pelican bearing is the same as the factory bearing, from the outset, it was designed to be the low cost alternative, not the best alternative. So you would be replacing the suspected problem bearing with another one just like it.

If memory serves, the article was in a trade press magazine a couple of years ago that was talking about the class action against PCNA over the IMS issue, and its impact on resale values after the legal action closed. If the years have not addled my memory, it was also reported in and article in Excellence. Similar numbers have been posted on other websites as well. Porsche originally released numbers around 10% for the single row at the outset of the legal action in a deposition, but like most mechanical issues, the numbers continued to rise while the leagal action dragged on. Porsche took the quick “corporate” way out by offering a nominal cash settlement to all the co litigants and no admission of guilt, with the lawyers taking most of the $ as usual.

Over the years, we have seen several failures up close and personal; we even had one customer that had one fail while still under warranty, PCNA approved a replacement engine which the dealer installed. Six months later, the replacement engine failed as well. We knew the owner and the car, it was serviced religiously, and the owner was not one to abuse the car. When the owner picked up the car after the second replacement engine, he drove it directly to a dealer for another brand and traded it in.

At the same time, we have had customers put 100K, 150K, and over 200K miles on similar engines with the factory bearings and without issues. So the IMS issue remains a crap shoot proposition: Some engines seem they will never fail, other simply don’t make it. One theory about the whole ball of wax was postulated by a rather serious Porsche after market engine builder with some serious credentials. He commented about the well known RMS leaking issue were very low mileage M96 engines started leaking oil badly, noting that Porsche released a special “go/no go” testing tool the measured the concentricity of the RMS opening in the case, which tested to see if the case opening was actually centered on the crank center line, and which found many were not. PCNA approved new engines for any that failed this test while under warranty, and released a new designed seal that was a lot more forgiving of misalignment. His theory was if the RMS case opening could be off center, was it possible that the IMS opening just above the RMS could also be misaligned. While the PTFE RMS seal could make up for misalignment, the metal on metal IMS flange had no hope of doing this, resulting in weird loading on the IMS bearing in engines with misalignment. The same engine builder also noted that when he spun up IMS shafts on a lathe before pinning the rear gear to prevent it from slipping (it is pressed on, and yes it too is a potential problem point), he noted that he found a lot of run out at the bearing opening on quite a few shafts, which he tossed out rather than reusing. So there is a whole bunch of possible reasons for problems to occur, which leads us back to the crap shoot description; get the wrong combination and you lose, big........... It also plays into another fact: the oil fed solid bearing IMS Solution is by its design much more tolerant of misalignment than either a ball bearing or roller bearing retrofit, which may explain why there has never been a reported failure of a retrofitted IMS Solution.

If you think about it, an engine with case opening misalignment and/or a wobbling shaft could explain why an engine that was pulled after failure and sent back to the factory for rebuild, could fail a second time; the true problem(s) were never repaired, just new parts installed. And it became a problem waiting for a new owner.

It cost Porsche a rather sizable fortune, both in bad press and engineering and parts sourcing to totally redesign the M96/97 into the 9A1 without an IMS shaft; they did not go to that expense because the problem was a little one, or one that was easy for them to fix.

[BFeller]

Originally Posted by JFP in PA

One theory about the whole ball of wax was postulated by a rather serious Porsche after market engine builder with some serious credentials. He commented about the well known RMS leaking issue were very low mileage M96 engines started leaking oil badly, noting that Porsche released a special “go/no go” testing tool the measured the concentricity of the RMS opening in the case, which tested to see if the case opening was actually centered on the crank center line, and which found many were not. PCNA approved new engines for any that failed this test while under warranty, and released a new designed seal that was a lot more forgiving of misalignment. His theory was if the RMS case opening could be off center, was it possible that the IMS opening just above the RMS could also be misaligned. While the PTFE RMS seal could make up for misalignment, the metal on metal IMS flange had no hope of doing this, resulting in weird loading on the IMS bearing in engines with misalignment. The same engine builder also noted that when he spun up IMS shafts on a lathe before pinning the rear gear to prevent it from slipping (it is pressed on, and yes it too is a potential problem point), he noted that he found a lot of run out at the bearing opening on quite a few shafts, which he tossed out rather than reusing. So there is a whole bunch of possible reasons for problems to occur, which leads us back to the crap shoot description; get the wrong combination and you lose, big........... It also plays into another fact: the oil fed solid bearing IMS Solution is by its design much more tolerant of misalignment than either a ball bearing or roller bearing retrofit, which may explain why there has never been a reported failure of a retrofitted IMS Solution.

If you think about it, an engine with case opening misalignment and/or a wobbling shaft could explain why an engine that was pulled after failure and sent back to the factory for rebuild, could fail a second time; the true problem(s) were never repaired, just new parts installed. And it became a problem waiting for a new owner.

Thank you for posting this information.

[azlvr]

Originally Posted by JFP in PA

Let’s start with the simple: The Pelican bearing is the same as the factory bearing, from the outset, it was designed to be the low cost alternative, not the best alternative. So you would be replacing the suspected problem bearing with another one just like it.

If memory serves, the article was in a trade press magazine a couple of years ago that was talking about the class action against PCNA over the IMS issue, and its impact on resale values after the legal action closed. If the years have not addled my memory, it was also reported in and article in Excellence. Similar numbers have been posted on other websites as well. Porsche originally released numbers around 10% for the single row at the outset of the legal action in a deposition, but like most mechanical issues, the numbers continued to rise while the leagal action dragged on. Porsche took the quick “corporate” way out by offering a nominal cash settlement to all the co litigants and no admission of guilt, with the lawyers taking most of the $ as usual.

Over the years, we have seen several failures up close and personal; we even had one customer that had one fail while still under warranty, PCNA approved a replacement engine which the dealer installed. Six months later, the replacement engine failed as well. We knew the owner and the car, it was serviced religiously, and the owner was not one to abuse the car. When the owner picked up the car after the second replacement engine, he drove it directly to a dealer for another brand and traded it in.

At the same time, we have had customers put 100K, 150K, and over 200K miles on similar engines with the factory bearings and without issues. So the IMS issue remains a crap shoot proposition: Some engines seem they will never fail, other simply don’t make it. One theory about the whole ball of wax was postulated by a rather serious Porsche after market engine builder with some serious credentials. He commented about the well known RMS leaking issue were very low mileage M96 engines started leaking oil badly, noting that Porsche released a special “go/no go” testing tool the measured the concentricity of the RMS opening in the case, which tested to see if the case opening was actually centered on the crank center line, and which found many were not. PCNA approved new engines for any that failed this test while under warranty, and released a new designed seal that was a lot more forgiving of misalignment. His theory was if the RMS case opening could be off center, was it possible that the IMS opening just above the RMS could also be misaligned. While the PTFE RMS seal could make up for misalignment, the metal on metal IMS flange had no hope of doing this, resulting in weird loading on the IMS bearing in engines with misalignment. The same engine builder also noted that when he spun up IMS shafts on a lathe before pinning the rear gear to prevent it from slipping (it is pressed on, and yes it too is a potential problem point), he noted that he found a lot of run out at the bearing opening on quite a few shafts, which he tossed out rather than reusing. So there is a whole bunch of possible reasons for problems to occur, which leads us back to the crap shoot description; get the wrong combination and you lose, big........... It also plays into another fact: the oil fed solid bearing IMS Solution is by its design much more tolerant of misalignment than either a ball bearing or roller bearing retrofit, which may explain why there has never been a reported failure of a retrofitted IMS Solution.

If you think about it, an engine with case opening misalignment and/or a wobbling shaft could explain why an engine that was pulled after failure and sent back to the factory for rebuild, could fail a second time; the true problem(s) were never repaired, just new parts installed. And it became a problem waiting for a new owner.

It cost Porsche a rather sizable fortune, both in bad press and engineering and parts sourcing to totally redesign the M96/97 into the 9A1 without an IMS shaft; they did not go to that expense because the problem was a little one, or one that was easy for them to fix.

An explanation that makes complete sense. My 01 3.2 has 141k and runs fantastic, already had the cracked head at 103k and IMS changed at 90k. When the engine is done it's time for a 3.6 or 3.8 with IMS Solution.

[Gilles]

Originally Posted by azlvr

An explanation that makes complete sense. My 01 3.2 has 141k and runs fantastic, already had the cracked head at 103k and IMS changed at 90k. When the engine is done it's time for a 3.6 or 3.8 with IMS Solution.

Azlvr, IMHO, this would be the best choice if you Love your car and plan to keep it for a long time.
.

[Homeoboxter]

Originally Posted by Gilles

Azlvr, IMHO, this would be the best choice if you Love your car and plan to keep it for a long time.
.

Yes, if you are the anxious type and want to sleep well at night after pouring a lot of money into your engine. On the other hand, the main take on from JFP`s post for me is that if a Boxster ran 150k miles without IMS failure then the engine case has no alignment issues and the IMS does not have excessive run out, so putting the IMS solution in it is overkill. Increasing the performance will not affect the load on the IMS bearing.

[JFP in PA]

Originally Posted by Homeoboxter

Yes, if you are the anxious type and want to sleep well at night after pouring a lot of money into your engine. On the other hand, the main take on from JFP`s post for me is that if a Boxster ran 150k miles without IMS failure then the engine case has no alignment issues and the IMS does not have excessive run out, so putting the IMS solution in it is overkill. Increasing the performance will not affect the load on the IMS bearing.

Only flaw in that assessment is that the bearing design itself is also questionable; while misalignment may explain some failures, particularly low mileage ones, the concept of using a sealed bearing is not a very good one all by itself. Viton seals harden with age and oil exposure, allowing hot oil seepage to wash out the grease while not allowing enough oil in to properly cool and lubricate the bearing. While doing retrofits, we found many bearings that showed they were “on the way out”, with hardened and gapped seals, but had not failed yet (no metal in the filter or sump found during pre qualification). The residual oil the came out of the shaft after the bearing was pulled was very thin and stunk to high heaven (An indication it had lost most if not all of its lubrication and heat transfer properties), and the bearing itself was very loose. After being retrofitted with a ceramic hybrid that was open to oil mist lubrication, the engine ran without issues for another 100K miles before the owner traded it in on another car. It would be reasonable to assume that the original bearing’s poor condition was not attributable to misalignment, but rather simple lack of proper lubrication and cooling.

The moral of all this tale is that the IMS failure issue is way more complicated than most think it is, with multiple possible factors contributing to it.

[tonythetiger]

Originally Posted by JFP in PA

Let’s start with the simple: The Pelican bearing is the same as the factory bearing, from the outset, it was designed to be the low cost alternative, not the best alternative. So you would be replacing the suspected problem bearing with another one just like it.

So, this is interesting and who thought more could be discussed on this subject, but then again, here we are. :+)
The pelican is the same as factory bearing, but a single-row, yes? I believe we cannot source the double-row bearing (or single row, but why would you?) directly from Porsche, the part number doesnt show in any of my parts drawings.
RND has the single to double-row upgrade for 550 bucks, but only for cars withe the single-row. What is out there for double row replacements at a similar price point? EPS is a strong candidate but seems dependant on oil pump modifications.

[Gilles]

Originally Posted by tonythetiger

EPS is a strong candidate but seems dependant on oil pump modifications.

Tony, one of the reasons why I don't believe that the idea behind the EPS design is a good one, it's because it will fill up the IMS shaft with oil withe the possibility of creating other unwanted side effects like an unbalanced shaft, unwanted inertia perhaps..? However, I am Not an engineer... so please take my comments like a grain of salt.

On the other hand what I like about the IMS Solution is that there are none moving parts and the oil is feed directly from the oil filter (similar as the design used on the Metzger air cooled engines)
.

[JFP in PA]

Originally Posted by tonythetiger

So, this is interesting and who thought more could be discussed on this subject, but then again, here we are. :+)
The pelican is the same as factory bearing, but a single-row, yes? I believe we cannot source the double-row bearing (or single row, but why would you?) directly from Porsche, the part number doesnt show in any of my parts drawings.
RND has the single to double-row upgrade for 550 bucks, but only for cars withe the single-row. What is out there for double row replacements at a similar price point? EPS is a strong candidate but seems dependant on oil pump modifications.

The Pelican bearing is a steel single row that they sell with a spacer for use in dual row applications. Think about that for a moment: Take out one of the strongest factory bearings and replace it with the most problematic.

LN has a dual row ceramic hybrid that they designed to work in the single row shaft.

I have never cared for the roller bearing approach for several reasons, one of the biggest it the idea of using a pointed punch to wack a precision opening from the oil pump into the IMS shaft; second, I do not like the idea of purposely running a flooded shaft (the IMS Solution uses a plug pressed in behind the bearing specifically to prevent flooding the shaft. That roller bearing kit also uses a oil pump drive shaft with a grove cut in it to allow the oil to pass, the oil pump drive shaft is already a weak point in these engines, and they are weakening it even more. I like to replace the weak OEM shaft with a chrome moly steel aftermarket part that is way stronger than factory.

[Homeoboxter]

Originally Posted by Gilles

Tony, one of the reasons why I don't believe that the idea behind the EPS design is a good one, it's because it will fill up the IMS shaft with oil withe the possibility of creating other unwanted side effects like an unbalanced shaft, unwanted inertia perhaps..? However, I am Not an engineer... so please take my comments like a grain of salt.

On the other hand what I like about the IMS Solution is that there are none moving parts and the oil is feed directly from the oil filter (similar as the design used on the Metzger air cooled engines)
.

It`s a common misconception that the IMSB is running dry and it`s lubricated only by oil mist, therefore some extra oil pressure through the shaft is going to solve all the IMS problems. The IMS is submerged in oil. (Let`s disregard now the seals on the bearing and the original grease pack that probably lasted a few weeks after the first owner started using the car). That`s why everyone who removes the IMS bearing in situ, reports about a quart of oil flowing out from the shaft. So adding an additional oil pressure line into the shaft won`t hurt, but it`s not necessary either. Even EPS claims that on their website, that oil lubrication is completely unnecessary, they just throw it in, because otherwise people would not be convinced it`ll work.

[JFP in PA]

Originally Posted by Homeoboxter

It`s a common misconception that the IMSB is running dry and it`s lubricated only by oil mist, therefore some extra oil pressure through the shaft is going to solve all the IMS problems. The IMS is submerged in oil. (Let`s disregard now the seals on the bearing and the original grease pack that probably lasted a few weeks after the first owner started using the car). That`s why everyone who removes the IMS bearing in situ, reports about a quart of oil flowing out from the shaft. So adding an additional oil pressure line into the shaft won`t hurt, but it`s not necessary either. Even EPS claims that on their website, that oil lubrication is completely unnecessary, they just throw it in, because otherwise people would not be convinced it`ll work.

The IMS bearing is only submerged in oil when the engine is not running and all of the oil has drained into the sump, assuming normal oil fill levels. Once running, only a mist of oil kicked up by the crank surrounds the bearings while the engine runs.

The reason the shaft is full of oil during an extraction is because the seals have hardened and shrunk, allowing the oil to get into the shaft. The fact that the oil found during an extraction is usually rancid is because it has been in there for some time because it cannot drain out when the level in the sump rises after the engine shuts down.

As for oil mist not lubricating the bearing, every LN IMS restrofit (now over some 25,000) except the Solution depends upon the mist to keep the hybrid ceramics lubricated and cooled, And having extracted a few LN bearings over the years, there was no oil to speak of trapped in the shafts.

[piper6909]

Originally Posted by JFP in PA

The IMS bearing is only submerged in oil when the engine is not running and all of the oil has drained into the sump, assuming normal oil fill levels. Once running, only a mist of oil kicked up by the crank surrounds the bearings while the engine runs.

The reason the shaft is full of oil during an extraction is because the seals have hardened and shrunk, allowing the oil to get into the shaft. The fact that the oil found during an extraction is usually rancid is because it has been in there for some time because it cannot drain out when the level in the sump rises after the engine shuts down.

As for oil mist not lubricating the bearing, every LN IMS restrofit (now over some 25,000) except the Solution depends upon the mist to keep the hybrid ceramics lubricated and cooled, And having extracted a few LN bearings over the years, there was no oil to speak of trapped in the shafts.

So, if one just pulls the seals off an OEM bearing it'll get an oil mist, the same mist the ceramic bearing would get, right?

Plus having an open bearing, it would allow any oil that would get into the shaft to drain out. Did I get that right?

[Homeoboxter]

The IMS bearing is only submerged in oil when the engine is not running and all of the oil has drained into the sump, assuming normal oil fill levels.

With all my respect, I don`t understand the logic behind that. Why would that be? At max level on the dipstick, oil level is close to the top of the shaft. Why would the oil level change so drastically when the engine is running? And if it changes, by how much? A liter? That would mean that one third of the shaft is still submerged in oil. When the engine stops, most of the oil that has been circulating will not flow back to the sump but will stay in the oil galleries, otherwise the proper and immediate oil pressure in the crank bearings could not be ensured right after start up. And we are talking about at least a liter of oil that should disappear upon engine start. Where?

Once running, only a mist of oil kicked up by the crank surrounds the bearings while the engine runs.

Has anybody looked at that experimentally? I read this statement everywhere but I still don`t get what it`s based on.

The reason the shaft is full of oil during an extraction is because the seals have hardened and shrunk, allowing the oil to get into the shaft. The fact that the oil found during an extraction is usually rancid is because it has been in there for some time because it cannot drain out when the level in the sump rises after the engine shuts down.

Yes, that`s one reason, the other is that pressure builds up in the shaft due to the rise in temperature, which then forces air through the seals, which then builds up vacuum in the shaft when the engine cools down, and oil seeps back to the shaft through the seals. After a few hundred cycles the shaft will be filled with oil.

And having extracted a few LN bearings over the years, there was no oil to speak of trapped in the shafts.

Yes, those are not sealed bearings so I assume oil just drains off the shaft when you drain the oil from the sump. I agree, removing the seals is a good idea, I still don`t understand why the Germans placed sealed bearings there in the first place and never changed that over the years.