What IMS bearing have you fitted? 3.4 C2

[Niall996]

I've had a look at mine today. Dual row, 1999 MY C4 with just shy of 300K miles on it. I was planning to change it but there is no point. The oil seal wasn't much of a seal anymore so it's been fed engine oil for who knows how long anyway.

The dual row bearings have nearly twice the load capacity of the single row bearing and in reality don't fail unless you are really, really unlucky. I think the failure rate is something like 0.6% for dual row at 100K miles vs 8% for single row.

Steel ball bearings operating within their load rating either fail in very short order or run forever.... unless they are ceramic bearings, I've used ceramic silicon nitride and zirconia based ceramic bearings in motorsport applications (e.g. clutch basket bearings on gearbox front covers) and whilst they offer advantages in weight and friction they are not stronger than steel bearings despite what people think (hardness is not strength) and are prone to developing play from the ultra hard ceramics running against the steel race if the lubrication supply is not controlled accurately. In an application like the IMS bearing play becomes failure quite quickly. To give an example we used to life the ceramic bearings carrying the clutch basket in one of our F1s at 500KM, these days we use steel in the same application and the oldest has 2200KM on it and is still going strong.

My vote if it's a dual row just leave it alone.

Why would Porsche have moved from the dual row bearing with twice the load capacity to a single row with, by extension, half the load capacity? What precipitated that change or necessitated it?

 

[Alex]

Could be various reasons - cost, supply & availabilty, size, etc.

In the grand scheme of bearing manufacture, 996 production figures are small fry.

 

[Chris_in_the_UK]

The swap from double to single row was to a larger bearing.

 

[Alex]

Good info Alex. Much appreciated.

Does flipping the seal open on the outer side (the side you can access from flywheel side) actually get much oil in that area? I've only seen photos so far so am struggling to picture it.

So you don't believe the spinning bearing will force oil away rather than take it in? As per "that" video. :grin:

The threaded spigot is definitely something I've read very little on. So is it that the spigot is screwed into the end plate/cover? I'd presumed it would be machined as one piece?
If so are you saying the old standard spigot should be removed then the thread opened up and a larger thread machined in for an updated spigot to be fitted? If so where does the larger threaded spigot come from?

Excuse all the questions but it sounds like you've done all the reading up I'll have to so if you don't mind me picking your brains. Sounds like inspection is the way to go first then.
I've mentioned before on another thread though my luck with cars is terrible. If something can happen it will happen on me. Snapped head studs on my old 70s 911, and again on my 964T, etc etc. So I understand the risks with any 911 but it would be nice to not have to rebuild the engine for a couple of years.

With roller or ball bearings, there is very little friction created as all faces roll against each other, instead of rubbing against each other like with a solid phosphor bronze bearing, so the levels of lubrication required between roller and solid is massive. Solid bearings need a constant flow of oil to ensure there's always a film of oil between mating surfaces, where frictionless bearings don't need anywhere near that. I used to run many roller bearings dry as they were used in a cleanroom environment so no oils or greases were permitted and the life span was months/years, whereby running a solid bearing dry would probably last around 5 minutes.

Within the 996 engine, the ims bearing is actually sat below the oil level when the engine isn't running so whilst not in motion it's fully submerged. Once the engine is running, there is so much oil being splashed/sprayed around the crankcase that an oil mist is created. This mist is perfectly adequate to ensure there is enough lubrication to the ims bearing, so long as it doesn't have the shield still fitted. If the engine has done many miles and the seal quite worn - like poppopbangbang's then the ball bearings would still get lubricated by the engine oil. It's your call if you inspect it and find the seal worn, whether to remove it or not, but it seems like a no-brainer to me whilst you're in there. The seal on the otherside of the bearing does not get oil flowing to it as it's sealed within the inter-mediate shaft/tube.


The spigot I was talking about is the centre piece that presses in the bearing and screws into the cover plate seen below:

It's recommended that if you change the bearing, you fit a spigot with a lager diameter thread to avoid it snapping. See this photo below showing at top the original one and below one installed in the cover plate that has a larger thread (I think it's something like M12 instead of M8)

All photos courtesy of infrasilver

 

[Alex]

You can probably see the difference better here:

 

[cobrars]

HI there,

I have a 1999 996 Carrera 2 Manual that has had a full Hartech rebuild in 2008 (not because of a failed bearing) That was done at 65k. The car now has just short of 90.5k and is due it's service in April. The rms is slighty weeping so while this is just a minor service and relatively cheap, i thought i'd have it repaired.

Now, i would assume Hartech replaced the bearing with a new one supplied from Porsche but with the dust covers removed...... I did ring Hartech and asked but they said they could not remember what had been done as it was so long ago....

So, this bearing has about 25k miles on it with no signs of damage (filings in the oil or filter) and i assume they did replace it with a dual row, which again Hartech couldnt tell me if they did or not....

I was looking at the "DOF" kit that "Pedro" goes on about, Design 911 sell them for £599 plus the vat....

Does the community suggest i just get the bearing inspected, if it seems fine, leave it and replace the RMS, new clutch and just leave it there ???

Thanks
Craig

 

[Alex]

Hi Craig,

To start with - how do you know the RMS is leaking? It could just as well be the ims cover plate leaking (as mine was).

Either way, when you get the leak fixed, just have the bearing checked for any play and to see what type is in there. Remove seal if it hasn't already been done.


Just another note. Fitting a direct oil feed system robs vital oil from other working components within the engine that require it (and it's pressure) to function, like hydraulic lifters, etc.

 

[cobrars]

Hi Alex

I was informed by my specialist that the RMS was at fault... but like you say it could quite as easily be the IMS cover, either or both will be fixed as dripping oil is a pet peeve of mine... hate it... :hand:

Well as long as it seems ok i'll just leave it then, you've most likely just saved me £720... plus fitting... ha, thanks :thumb:

You know what, i never looked at it that way... I just thought there'd be enough to go round... DOF is now off the cards...

Thanks for your info on this topic.... much appreciated... :worship:

Craig

 

[kurlykris]

Both of my Porsches have factory fitted IMS bearings, one 2006 with 55K on the clock, the other 1998 with 103K on the clock, if they were going to fail, they would have done so by now I think.

Monitor the old engine oil and filters, fit magnetic sump plugs and examine on service day :grin:

You risk doing more harm than good fitting " snake oil " expensive aftermarket bearings for no reason whatsoever.

Just my 2 penneth :dont know:

 

[poppopbangbang]

The swap from double to single row was to a larger bearing.

No it wasn't. The orignal single row was about 55% the load rating of the dual row bearing. The later 2006 onwards, none replaceable in situ single row bearing took the bearing load rating back to essentially the same as the original dual row.

 

[poppopbangbang]

Here's a side by side of the dual row early 3.4L bearing and the later 3.6 single row bearing

Both the above bearings are 6204 types with an OD of 47mm. The later post 2006 bearing is a 6305 type with an OD of 62mm.

When you see the single and dual row bearings side by side it becomes clear why early 3.4L cars are essentially bomb proof and later 3.6L cars aren't.

There are quite a few dual row bearing cars now well north of 250K on the original bearing.

IMHO the biggest issue is that in most bearings, certainly the 6204 types, in most applications only a percentage of the balls in the bearing are actually carrying load (between 25 and 50% depending on the load path) which means in the case of the single row bearing you have exactly 50% of the load capacity in the bearing that the dual row has. In addition to this manufacturing and balance tolerances are always going to have an effect on the static side load the bearing experiences i.e. is it running with a greater/lesser side loading than other installations of the same bearing in the same application. I suspect most dual row bearing failues are due to this, when the tolerances stack up well the bearings last forever...... but of course you don't know this on a fresh build (that'd be the 0.6% then) but you do if it's done 100K miles!

In short if it's not ***** itself spectacularily yet, it has plenty of miles on it and it's a dual row leave it well alone. The liners will fail or it will run out of valve seats way before the dual row IMS does.

The engine was designed to run with a dual row bearing in this location, the designers were well aware of the load requirement. I seem to remember lack of reliable availability of the dual row bearing lead to the use of the single row bearing as all the castings stayed the same bar the IMS cover. In theory it should have still been strong enough and in a lot of cases it is but it's not a patch on the dual row in terms of strength.

 

[Chris_in_the_UK]

The swap from double to single row was to a larger bearing.

No it wasn't. The orignal single row was about 55% the load rating of the dual row bearing. The later 2006 onwards, none replaceable in situ single row bearing took the bearing load rating back to essentially the same as the original dual row.

Thanks for the clarification - I was referring to the 2006 version onwards.

 

[bazhart]

Wow - what a subject for all to contribute their own particular info on and done to death over the years.

The early double row bearing was located with a snap spring ring that on removal with a puller can sometimes catch and damage the housing requiring a strip to replace the shaft. We have not had this problem but have had to rebuild a couple of engines that did - so it can happen - if rarely.

You can support the shaft with a machined out standard housing to maintain its position and use a sliding hammer to remove the bearing. This enables the spring ring to "jump" and clear the groove.

The problem is mainly in replacing it when you have to tap it beck in and this allows the shaft to move further in and can damage the plastic chain runner guide tracks.

The inner seal - as it wear - allows a small amount of oil to enter the inside of the IMs tube where it sits for years - only getting out when you remove the original bearing.

The original double row bearing was a special and I have not managed to obtain load ratings for it but it is not as high as two single row bearings (i.e. not twice the single load rating) because with the groove dimensions reduced it is more like somewhere between a double row self aligning bearing and two single rows side by side - at a guess probably more like 20 to 40% better load rating. A narrower double row bearing of same overall dimensions had a load rating some 25% less than a single row due to this reduction in groove dimensions.

The problem is that the original bearings trapped the fine small "running in" particles of metal that mixed with the grease for form a very fine grinding paste. We were lucky to find two over the years that still had grease inside and the grey colour with fine metal particle inside. The double row would reasonably be expected to create more particles than a single row and run hotter as a result and so we concluded that the early double row failures were more due to trapped debris than load rating.

The larger 6305 bearing that replaced the 6204 single row rates at 77% better load rating - so compared to the non standard double row bearing is possibly 50% better.

The video referred to is IMHO misleading because a ball bearing requires minute amounts of oil to run perfectly (as a previous contributor correctly described) and although the outside of the bearing is rotating - not only is it under the oil level of the sump but when running all the oil inside the engine is being churned around and creates a mist (which is ideal to lubricate a ball bearing) and right next door to the bearing is a sprocket or hivo gear of much larger diameter running at the same speed but churning up oil and throwing it around with much greater force than the bearing is forcing it out with. Add to that the chain running at up to 40 mph in the oil bath and you get some idea of how perfectly adequate the spray oil supply is without the seal in place.

When the dual row bearings first failed (at relatively low mileages) we managed to obtain the remaining stock of the special bearings from the manufacturer and fitted those to customer engines until the supply ran out.

when they ran out - we concluded that the next best then was a single row bearing (which we fitted with a spacer and a stronger central shaft) which was later supplied as standard by Porsche in later engines. 8 years ago we may not have recorded the specification of the bearing we fitted then but it will have had the seal removed and therefore either way should be acceptably reliable.

We have since had to rebuild engines for IMS failure of various types and concluded that basically the space for the bearing was too small and the bearing under-spec'd and therefore even with spray lubrication - may not outlast the engine. However IMS bearing failure is a very low number compared to cracked and scored cylinders meaning you are much more likely to have your engine stripped to repair those failures than an IMS bearing failure and therefore can change the shaft for one with the larger ball bearing at the same time - which is the perfect solution - especially with the outer seal removed.

So Hivo cam-chain drive cars could fit the Porsche larger bearing shaft during a rebuild and we have manufactured our own re-manufactured shafts for both Hivo and roller chain crankshafts with that larger bearing. In doing so we have plugged the inner IMS tube so old oil cannot enter it or interfere with the lubrication system - but this can only be fitted during a rebuild.

The small number of IMS failures now reflects that most small bearing engines have done enough miles to allow fresh oil to enter and exit the bearing and keep it going for much longer.

A plain bearing is ultimately going to provide better life expectancy but requires a continual supply of fresh oil - potentially robbing oil supply from other vital engine parts on hot tickover where the rather crude standard oil pump supply is poor.

If an engine has throughout its life been driven modestly - and the original bearing has survived - then removing the seal will probably provide the best overall outcome at minimal cost.

If the engine has been used spiritedly then cracked or scored cylinders, worn crankshaft shells etc may well be the first reason it may need a rebuild during which time the best solution is the larger ball bearing.

Weighing all this up - unless the existing smaller bearing is showing definite signs of wear - I would recommend removing the seal and awaiting a long life with it or a rebuild for other reasons rather than possibly waste money on a solution that may well not be any better and also find the engine needed stripping and rebuilding later on anyway for another failure reason.

With care - replacing it with a newer bearing of the same specification is going to work most often but if one failure occurred just after a business fitted such a replacement bearing and they felt obliged to repair the damage FOC - they would need to fit hundreds if not thousands of bearings successfully to pay for that one failure - so it is a risky business (unless the business involved would simply not cover such an eventuality anyway and expect the customer to pay for the rebuild) and "YES" we have been involved later on in fixing a few engines where just such a failure has occurred (often with litigation between the owner and the original bearing replacer resulting in some compensation) hence our own caution at offering such a solution when we would honour our obligations to our customers.

So a complicated situation not brought about by the independent businesses trying to help. Respect to those coming up with what they believe to be viable solutions - caution to others over the whole situation.

If you are considering an "upgrade" (which in reality is unlikely to be any better than a simple replacement) it is usually because you intend to keep the car for many years and as such the chances of also needing a rebuild along the way are quite high - when you can solve the problem with the larger bearing - so if you are considering saving for a rebuild one day anyway - why waste money on a short term solution that might not work anyway our outlast the rest of the engine?

Any replacement carefully fitted will probably be a bit better than the original - but is it worth the risk and expense - taking everything into account?

Baz

 

[Marky911]

Sorry for the deathly silence lads. Been busy. I'm on my phone so will keep it brief. Thanks for all the replies and pictures. Good stuff.

I'll basically get the bearing checked at this years service next month or chance it for a year until next spring.

I must admit due to the expense of checking it being almost the same as swapping it, if I could find an identical bearing I would probably still swap it if we were in there that far. As it is I believe you can't actually get a replacement bearing of exactly the same dimensions as the original dual row. The ones available are a touch narrower and require a little spacer ring which isn't really ideal.

So I'll get it looked at and flip the seal open. The original one went at 55k. The car was fitted with a replacement engine from an identical year car but with 10k less miles on. A new clutch and dual mass flywheel were fitted at the same time. The car has now covered 75k (engine 65k). I'm hoping then that I'm in the safe zone. It never bothered the guy I bought it off who owned for 5 years so I should just forget it, but I know my luck with these things.

I say the price is the same whether I fit a new bearing or just check it due to the fact that once the clutch is removed, if you reuse it it can sometimes go bang a short while later, due to stresses involved in untensioning it then refitting and re-torquing. It's rare but does happen, hence I'll be factoring in a couple of hundred for labour plus the price of a new clutch etc, just to check it. I'm happy with that it just seems daft not to just fit a new bearing if we are that far in.
I won't though if the existing one looks fine.

One last thing id have thought it prudent to stick a new RMS in while we are in there so it will have had IMS inspection and seal removal, new clutch and new RMS, hence all done in there for a while, hopefully!
Or should I be saying if the RMS ain't broke, don't fix it. Can't see any issues in sticking one in though?

Once again cheers for all the info. I knew the basics but have learnt a bit more. My last 996 had the Mezger, so none of the worries. Oh, and the oil filter was on the top of the engine so you could check it to your hearts content. :grin:

Anyway thanks. Some great 996s on here. Kurly Kris I seen yours for sale when you bought it. What a difference getting it on the split rims has made. Looks stunning now. :thumb:

PS Cheers for the expert input Baz. Always welcomed. I agree the topic has been done to death although in fairness I was after what bearing to buy and where to get it, it just ended up going down the usual route which to be honest has been no bad thing for me. Cheers.
:thumb:

 

[Alex]

RMS's are funny little creatures. When I thought mine was leaking, turned out it was the ims plate. I told my mechanic to leave the rms as they can easily not be fitted correctly and start leaking.

I guarantee the bearing that's in is perfect. I'll also put a wager on it (lets say a new engine) that it won't go pop in your ownership. If it doesn't - you owe me a new engine. If it does - I'll buy you one. How about that? :lol: