Lubrication/Tribology expert advice needed ref ball bearings
Discussion
There is a new idea inside the post 1996 Porsche engines and gearboxes - namely to fit grease filled ball bearings with seals each side.
I don't understand this because the grease washes out and although some oil finds its way back in - there is no flow and it is neither cooled nor refreshed, but trapped inside with any debris that grinds off the micro surface, becoming a mush of eventually poor quality oil.
I contacted one of the bearing manufacturers and talked to the person involved in the original specification of one of these bearings and they explained that they advised not to use seals but Porsche insisted.
It is interesting that the 2 most common bearings to suffer occasional premature failure are indeed of this new design - namely the intermediate shaft ball bearing and the gearbox pinion bearing.
Can anyone expert in this field please explain why anyone would prefer to seal the bearings rather than allow the debris to flow away and be caught by the magnetic pick up, and eventuall be flushed away during and oil change - thus renewing fresh oil and also keeping the oil in the bearing cooler.
The oil trapped inside the intermediate shaft is always very sludgy (as is that inside the bearing).
It beats me - but then - as other have pointed out - I am a mere independent individual - not a Porsche empoyee - who doubtless knows the reason - if so - what is it please?
Baz
996 intermediate shaft ball bearings can fail but that they are grease filled and fitted with typical dust seals -
I don't understand this because the grease washes out and although some oil finds its way back in - there is no flow and it is neither cooled nor refreshed, but trapped inside with any debris that grinds off the micro surface, becoming a mush of eventually poor quality oil.
I contacted one of the bearing manufacturers and talked to the person involved in the original specification of one of these bearings and they explained that they advised not to use seals but Porsche insisted.
It is interesting that the 2 most common bearings to suffer occasional premature failure are indeed of this new design - namely the intermediate shaft ball bearing and the gearbox pinion bearing.
Can anyone expert in this field please explain why anyone would prefer to seal the bearings rather than allow the debris to flow away and be caught by the magnetic pick up, and eventuall be flushed away during and oil change - thus renewing fresh oil and also keeping the oil in the bearing cooler.
The oil trapped inside the intermediate shaft is always very sludgy (as is that inside the bearing).
It beats me - but then - as other have pointed out - I am a mere independent individual - not a Porsche empoyee - who doubtless knows the reason - if so - what is it please?
Baz
996 intermediate shaft ball bearings can fail but that they are grease filled and fitted with typical dust seals -
Possible reasons would be the lack of a suitable oil feed - as you know, a single race roller bearing is fairly robust and needs little more than an oil mist, but that may not be available in this location ?
Another possibility would be to protect against swarf ingress if that was a known problem during development. A sealed bearing is a bit more expensive than an open bearing, which would indicate that there is some arcane reason why they did it that way.
Another possibility would be to protect against swarf ingress if that was a known problem during development. A sealed bearing is a bit more expensive than an open bearing, which would indicate that there is some arcane reason why they did it that way.
But the grease flows away (and is is a known poor lubricant for a high speed ball bearing) and there is an even worse oil feed with 2 seals blocking the way in than if they were not there.
Similarly there is not only a good supply of oil in both cases - but swarf would tend to be thrown away by the centrifugal forces at the rotating bearing moving cage and be caught by the magnet - not flow inwards.
I don't doubt someone has an explanation - but I just have this doubt that they are right - although there is always something new I can learn - but if so - why are failing?
Baz
Similarly there is not only a good supply of oil in both cases - but swarf would tend to be thrown away by the centrifugal forces at the rotating bearing moving cage and be caught by the magnet - not flow inwards.
I don't doubt someone has an explanation - but I just have this doubt that they are right - although there is always something new I can learn - but if so - why are failing?
Baz
There is no obvious answer to me, logically an open bearing with fresh oil would be the way to go.
The thought has crossed my mind that it might be a production issue - for example, if the gearbox is spun up after assembly with no oil in it to test for balance/noise/whateva.
Is the design of the GT3/Turbo powertrain different in this area ?
The thought has crossed my mind that it might be a production issue - for example, if the gearbox is spun up after assembly with no oil in it to test for balance/noise/whateva.
Is the design of the GT3/Turbo powertrain different in this area ?
I've had a lot of experience in this field in the past. It would most likely be a contamination issue, obviously there is something "special" to Porker boxes that wears out & creates bearing problems. The seals will be high temp affairs (or should be) & in theory if designed & specced correctly should in theory not fail early.
It may well be a secondary issue thats causing bearing failure, has anybody got any info on the diagnosis of the failed units? whats the spec of the bearing? some numbers would be a good start & a manufacturer.
It may well be a secondary issue thats causing bearing failure, has anybody got any info on the diagnosis of the failed units? whats the spec of the bearing? some numbers would be a good start & a manufacturer.
It is a very difficult situation to criticise Porsche designs - for many varied reasons - and we generally love the cars and enjoy working on them as excellent examples of performance automotive engineering. So I don't want to get into too much specific detail about the bearings concerned. Rest assured in both cases - if it was a contamination issue - other parts would suffer even more - so I cannot imagine that is the reason.
It could well be that it protects the bearing from ingress of foreign bodies during assembly - as that makes some sense to me - more so than that it was a good idea technically. Or it may result from problems with other models (like the 968 pinion bearing) that unexpectedly also sometimes failed prematurely.
I also openly and humbly admit my own short comings in such very specialised areas and that is why - although I just think the design is wrong - I would love someone to put me right - explain why I am wrong and therefore enable the whole issue to be resolved.
There have been many small detailed changes to the original design in these engines and deviations from previous practices in the transmission - that make it perfectly clear that it was not 100% designed or neccessarily specified right first time and many customers have paid for failings that I often feel should not have occured. On the other hand it may be that mileages of 50K before failure - are felt to be acceptable in this new highly profitable regime.
In exploring this issue I am trying to learn, making suggestions that - who knows - may result in improvements - while perhaps supporting some owners suffering from early failures.
Overall it would be great to find a general argument supporting the shielding of heavily loaded bearings from a fresh clean oil supply and an explanation as to why that is better than past established practice.
Baz
It could well be that it protects the bearing from ingress of foreign bodies during assembly - as that makes some sense to me - more so than that it was a good idea technically. Or it may result from problems with other models (like the 968 pinion bearing) that unexpectedly also sometimes failed prematurely.
I also openly and humbly admit my own short comings in such very specialised areas and that is why - although I just think the design is wrong - I would love someone to put me right - explain why I am wrong and therefore enable the whole issue to be resolved.
There have been many small detailed changes to the original design in these engines and deviations from previous practices in the transmission - that make it perfectly clear that it was not 100% designed or neccessarily specified right first time and many customers have paid for failings that I often feel should not have occured. On the other hand it may be that mileages of 50K before failure - are felt to be acceptable in this new highly profitable regime.
In exploring this issue I am trying to learn, making suggestions that - who knows - may result in improvements - while perhaps supporting some owners suffering from early failures.
Overall it would be great to find a general argument supporting the shielding of heavily loaded bearings from a fresh clean oil supply and an explanation as to why that is better than past established practice.
Baz
Baz
I agree with you , In fact some engines go on to high mileages , Probally the seals have long since failed and the bearing is getting better lube from the engine oil. Honda used to positive feed the bearings on the early bike and car race engines and on the British bikes they used a roller bearing on the driveside casing. But now the oil tech as moved on so much that nearly all the bike engines rev to 15k plus on plain bearings on the crank leaving the ball bearings to the transmission.
On the engines that you have pulled apart was there any signs of the grease still being present in the bearing. I presume the bearing that fails is the drive end of the shaft which should idealy be of a larger capacity.
I also think these engines respond well to being used everday as when i used to leave my Boxster for a week the noise from the chains as the tensioners primed themselves used to make me cringe but was not present at all when used daily.
Maybe you can try running a test engine with plain ball bearings or may be replace it with a roller bearing,no doubt you will find a cure for this problem to go with your other mods.
Thanks to the lads Bob.
I agree with you , In fact some engines go on to high mileages , Probally the seals have long since failed and the bearing is getting better lube from the engine oil. Honda used to positive feed the bearings on the early bike and car race engines and on the British bikes they used a roller bearing on the driveside casing. But now the oil tech as moved on so much that nearly all the bike engines rev to 15k plus on plain bearings on the crank leaving the ball bearings to the transmission.
On the engines that you have pulled apart was there any signs of the grease still being present in the bearing. I presume the bearing that fails is the drive end of the shaft which should idealy be of a larger capacity.
I also think these engines respond well to being used everday as when i used to leave my Boxster for a week the noise from the chains as the tensioners primed themselves used to make me cringe but was not present at all when used daily.
Maybe you can try running a test engine with plain ball bearings or may be replace it with a roller bearing,no doubt you will find a cure for this problem to go with your other mods.
Thanks to the lads Bob.
Hi Dunit,
Never any grease inside that I have seen.
The odd thing is that inside bearings that fail the debris is of course trapped inside the bearing and forced to run around with the old oil also trapped inside - like a kind of grinding paste.
In the case of the gearbox pinion bearing it is usually the surface of the outer track that has lost small microscopic pieces of metal over a period of time (looking like a mouse has chewed it away). This is usually because loadings exceed the fatigue rating of the surface material which fatigues away from the substrate. But those little bits will stay inside for a while until they get out and it would surely be better if they imediately floated away to be caught by the magnet. Of course ball bearings transmit a higher point loading at the ball face than either a tapered or plain roller bearing so under the same high dynamic load would be expected to fatigue the surfaces sooner.
In the case of the intermediate shaft, usually the cage has broken - suggesting that the relatively high speed and poor quality of the trapped oil inside has provided a drag to the bearings that has over loaded the cage - or it may be that the oil quality reduced due to the heat generated as the oil is not circulating and that caused the ball bearings to skid in the track more. The actual dynamic loads on this bearing are relativly minor. I have seen one in which the cage was broken and the balls running around crowding on to each other to offset the load and wear against each other (by rotation).
I have tried removing the seals on some cars but unfortunately - since it took many thousands of miles (and several years) before the problems emerge - it will be a long job waiting to see if it was better or worse - which is why I am trying to seek some guidance to find out if there is a legitimate reason for leaving the bearings fully sealed in the first place.
Baz
Never any grease inside that I have seen.
The odd thing is that inside bearings that fail the debris is of course trapped inside the bearing and forced to run around with the old oil also trapped inside - like a kind of grinding paste.
In the case of the gearbox pinion bearing it is usually the surface of the outer track that has lost small microscopic pieces of metal over a period of time (looking like a mouse has chewed it away). This is usually because loadings exceed the fatigue rating of the surface material which fatigues away from the substrate. But those little bits will stay inside for a while until they get out and it would surely be better if they imediately floated away to be caught by the magnet. Of course ball bearings transmit a higher point loading at the ball face than either a tapered or plain roller bearing so under the same high dynamic load would be expected to fatigue the surfaces sooner.
In the case of the intermediate shaft, usually the cage has broken - suggesting that the relatively high speed and poor quality of the trapped oil inside has provided a drag to the bearings that has over loaded the cage - or it may be that the oil quality reduced due to the heat generated as the oil is not circulating and that caused the ball bearings to skid in the track more. The actual dynamic loads on this bearing are relativly minor. I have seen one in which the cage was broken and the balls running around crowding on to each other to offset the load and wear against each other (by rotation).
I have tried removing the seals on some cars but unfortunately - since it took many thousands of miles (and several years) before the problems emerge - it will be a long job waiting to see if it was better or worse - which is why I am trying to seek some guidance to find out if there is a legitimate reason for leaving the bearings fully sealed in the first place.
Baz
Baz
Just had a look at the w/shop manual as I could not remember the reduction ratio that the int shaft runs at , Which is roughly 2/3 rds. So why you would need a grease filled sealed bearing that is running totally immersed in engine oil at a max speed of 5k beats me.
Of course the other point to consider is the quality of the bearings as this can vary greatly from different manufactures. You used to be able to get h/duty bearings that had the same dimensions but the track was ground deeper to increase the load area.
Bob
Just had a look at the w/shop manual as I could not remember the reduction ratio that the int shaft runs at , Which is roughly 2/3 rds. So why you would need a grease filled sealed bearing that is running totally immersed in engine oil at a max speed of 5k beats me.
Of course the other point to consider is the quality of the bearings as this can vary greatly from different manufactures. You used to be able to get h/duty bearings that had the same dimensions but the track was ground deeper to increase the load area.
Bob
Baz , can you confirm whether Porsche have continued this design of bearing with their 997 engines?
I've heard that the 997 3.6 engine is virtually the same as the 996 one , and have asked you in the past about the 997 variants but at the time it was early days and you'd yet to work on one.
With the 997 in circulation for over 3 years now , I'd assume that if they've the same internals (and the same problems) some of them would have departed the OPC mainstream and found themselves in indy hands when problems occur.
Or , surprise surprise , have Porsche had a change of heart and used oil lube bearings on their newer cars??
I've heard that the 997 3.6 engine is virtually the same as the 996 one , and have asked you in the past about the 997 variants but at the time it was early days and you'd yet to work on one.
With the 997 in circulation for over 3 years now , I'd assume that if they've the same internals (and the same problems) some of them would have departed the OPC mainstream and found themselves in indy hands when problems occur.
Or , surprise surprise , have Porsche had a change of heart and used oil lube bearings on their newer cars??
Hi Ballcock,
still not seen inside a 997 yet so cannot help.
The speed rating of the int shaft bearing in oil is OK for the application but the gearbox pinion bearings is marginal at top revs.
I think however the evidence of unsuitability comes from the number of failures and the only explanation I can come up with is that the oil overheats, deteriorates or becomes contaminated from being trapped inside the seals. Certainly the old oil that seeps inside the hoolow shaft is very poor quality and very "sludgy" (that's the technical word for it anyway).
The assembly method in the gearbox input shaft bearing may indeed cause a problem if anyone knocks the parts on the shafts using the bearing to restrain the shock. We have come up with a different method but I cannot imagine Porsche doing any banging in their assembly proceedures.
Still waiting for a good reason why the bearings are shrouded from fresh clean cooler oil!
Baz
still not seen inside a 997 yet so cannot help.
The speed rating of the int shaft bearing in oil is OK for the application but the gearbox pinion bearings is marginal at top revs.
I think however the evidence of unsuitability comes from the number of failures and the only explanation I can come up with is that the oil overheats, deteriorates or becomes contaminated from being trapped inside the seals. Certainly the old oil that seeps inside the hoolow shaft is very poor quality and very "sludgy" (that's the technical word for it anyway).
The assembly method in the gearbox input shaft bearing may indeed cause a problem if anyone knocks the parts on the shafts using the bearing to restrain the shock. We have come up with a different method but I cannot imagine Porsche doing any banging in their assembly proceedures.
Still waiting for a good reason why the bearings are shrouded from fresh clean cooler oil!
Baz
Gassing Station | Porsche General | Top of Page | What's New | My Stuff