996 IMS Issue - How real is it?

[dan_189]

GT3 and Turbo engines don't have  IMS bearing or an IMS.

The purpose of the intermediate shaft is to drive the camshafts indirectly off the crankshaft. By using an intermediate shaft, the speed of the chains are reduced, which is better for chain life. This basic design was used throughout the entire lifespan of the aircooled six-cylinder Mezger engine used through 1998. The inclusion of an intermediate shaft which drives the camshafts indirectly off the crankshaft has been a mainstay of the horizontally-opposed flat 6 engine utilized by Porsche.

https://www.total911.com/technology-explained-intermediate-shaft/ 

https://www.oregonpca.org/wp-content/uploads/2014/06/ORPCA-IMS.pdf 

However you're correct that they do not have a the same IMS ball bearing the M96 non metzger engines have. The GT3 and Turbo have an internal oil fed plain bearing inside and that is where the idea for the IMS solution came from that Jake Raby invented and developed through LN Engineering.

[Peter M]

But wasnt it the Japanese (Toyota?) that 'helped' design the heads for the Boxster / 996 so they could be manufactured very much cheaper than a traditional flat 6 engine?

Actually, the new CEO realised that Porsches future relied on reducing costs by increasing quality (as defined by reduced defects in manufacture and less rework needed).  To do this he engaged a Japanese consultancy that had been formed by ex Toyota luminaries of the Toyota Production System to introduce Lean Engineering principles to Porsche.

At the same time Porsche was running out of money - losing $150M in 1992 on only $1.3B in total sales - so they had to simplify and reduce production costs and this is why they went to one cylinder head design that could be used on both sides of the block.  The decision to adopt the same front structures on both the 996 and the Boxster is another example of this cost cutting.

So I don't think anyone can blame the Japanese for the IMS issue, that surely has to lay at the feet of Porsche.  

In fact it was only through Lean that they were able to build the first perfect car on 27 July 1994.  By that I mean that every car they produced before this time had to be rectified in some way at the end of the production line before it was ready for sale.  (The company had been knowingly producing faulty cars on the production line and they had managed these faults by having an army of craft men waiting in a vast rectification area to reinspect and repair the problems hopefully before they got to the customers.)

Porsche was so successful in adopting Lean that they ended up making money selling their gained knowledge to other companies.

[ANF]

In fact it was only through Lean that they were able to build the first perfect car in 199? (Sorry date escapes me right now!).  By that I mean that every car they produced before this time had to be repaired in some way at the end of the production line before it was ready for sale.

Porsche was so successful in adopting Lean that they ended up making money selling their gained knowledge to other companies.

Is this info available freely Peter? I would like to know more about this "repaired in some way at the end of the production line before it was ready for sale"

[Peter M]

Adam,

Let me dig out a case study from one of my Lean textbooks that I will post here.  The Toyota connection gets raised enough here that I think there will be enough other anoraks around to be interested in it.

[Pork Chops]

Adam,

Let me dig out a case study from one of my Lean textbooks that I will post here.  The Toyota connection gets raised enough here that I think there will be enough other anoraks around to be interested in it.

Some Toyota Lean reading references below.  No idea if they have Porsche references.  Pre 996 cars were essentially hand build like you would do in your garage, but a little more organised.  996 onwards was a production line as you'd imagine it to be.

Toyota's rigours in regards to lean principles is staggering but seriously who CBF with doing that all day long...

The Toyota Way Fieldbook: a practical guide for implementing Toyota’s 4Ps
Liker, J. K. and Meier, D. (2006).New York: McGraw-Hill

Toyota Production System "One-by-One Confirmation"
Kitano, Keynote Address. University of Kentucky, Lean Manufacturing Conference (May, 1997).

Toyota Culture - The Heart and Soul of the Toyota Way
Liker, J., Hoseus, M., and The Centre for Quality People and Organizations (2008), New York cGraw Hill

[Peter M]

Adam,

Let me dig out a case study from one of my Lean textbooks that I will post here.  The Toyota connection gets raised enough here that I think there will be enough other anoraks around to be interested in it.

Bugger!  It's a fair bit bigger than I recalled at 32 pages.  Its a chapter from: Lean Thinking: Banish Waste And Create Wealth In Your Corporation by James P. Womack, & Daniel T. Jones that is specifically about the transformation at Porsche in the early 90's.

If you and anyone else wants a PDF, just pm me your email address and I'm happy to send you a copy.

[TINGY]

Bugger!  It's a fair bit bigger than I recalled at 32 pages.  Its a chapter from: Lean Thinking: Banish Waste And Create Wealth In Your Corporation by James P. Womack, & Daniel T. Jones that is specifically about the transformation at Porsche in the early 90's.

If you and anyone else wants a PDF, just pm me your email address and I'm happy to send you a copy.

Oh f#@k me, we are doing this stuff again at work, yes essential reading for anoraks and any budding six sigma aficionados!

[Mike737]

Mike,

The point I'm making is that a plain journal bearing copes with the torsional vibration that a bearing with rolling elements sometimes can't endure.

Imagine the wildly varying load at low engine speed that driving the valve train of 3 cylinders places on the shaft that is also compounded by its drive from the crankshaft being highly variable in its rotational accelerations as well.

Forget that plain bearings are often pressure fed, forget lubrication, it's the suitability of bearing type that is my point.

These issues are well known elsewhere in automotive engineering with the Ford DFV valve train reliability challenge probably the most famous example of destructive TV in practice.

 

 

Ah got you now Pete. I thought you we're saying that LN IMS 'solution' was the one you would go for over the standard retrofit when you said solution. Bit of a roger victor clarance moment.

 

[Peter M]

Ah got you now Pete. I thought you we're saying that LN IMS 'solution' was the one you would go for over the standard retrofit when you said solution. Bit of a roger victor clarance moment.

Mike,

Sorry I must be talking in riddles!

Yes, the IMS Solution is the one I would go for because it solves the issue once and for all. 

However I accept that this product is sold at a cost premium and that sometimes we have to compromise and use one of the many cheaper but temporary ball or roller bearing retrofits on the market that will need to be replaced every so often.

[dan_189]

6 years or 75,000kms for the LN dual row pro, less for the classic. LN also created a more economical roller bearing option but it's got a recommend lifespan of only 4 year. 

[ANF]

What we have now, or very soon will, is a culture where everyone who buys a 996/ etc will replace the IMS bearing regardless as there is so much conflicting information out there and so many varying lengths of "repair". All for something that ultimately has a very low failure rate.... quite bemusing really!

[dan_189]

What we have now, or very soon will, is a culture where everyone who buys a 996/ etc will replace the IMS bearing regardless as there is so much conflicting information out there and so many varying lengths of "repair". All for something that ultimately has a very low failure rate.... quite bemusing really!

What a selling tool fear is! I bought into it, to each their own 

[Peter M]

All for something that ultimately has a very low failure rate.... quite bemusing really!

Just another example of how irrational we all are.

6 years or 75,000kms for the LN dual row pro, less for the classic. LN also created a more economical roller bearing option but it's got a recommend lifespan of only 4 year. 

Dan,

Why would they be time limited?  

I can understand the km, but deterioration based on age?  Really?  I know of no other bearing that is replaced for that reason.  

[LS1RX7]

What we have now, or very soon will, is a culture where everyone who buys a 996/ etc will replace the IMS bearing regardless as there is so much conflicting information out there and so many varying lengths of "repair". All for something that ultimately has a very low failure rate.... quite bemusing really!

It may seem bemusing, but a $2,000 fix to insure against a possible $20,000 + repair bill is a good investment in my opinion. Not that I intend to sell mine anytime soon but anyone looking to buy a 996 will surely give greater consideration to a car that has had the bearing replaced over one that hasn't. The IMS issue is also something that has made 996's an affordable option for many, and while the 996 will never see the stratospheric rise in value of air cooled cars, I believe they will again rise in value as people realise that they don't necessarily have to remain potential grenades especially C2 manual coupes as they seem to be harder to find than tip's and convertibles. I find it more bemusing that someone wouldn't remove a known potential risk of a $20,000+ repair bill for a $2000 preventative maintenance spend which would be even cheaper if it was done as part of a clutch replacement job. Just my opinion, but there is also a lot of value in peace of mind.

[Avon]

Original IMS bearing code marked on the seal is NSK 6204DUA17, so maybe they are still available.

[LS1RX7]

I tried to purchase an original bearing from a couple of different bearing suppliers when i replaced mine but they were listed as not available to supply to other user meaning porsche have a licensing agreement for that particular part. Porsche don't sell the bearing separately either, they will only sell the complete IMS assembly for about $3500AU. 

[dan_189]

There are better solutions than the original anyway. Even the pelican IMS which is made by LN and cheaper than the IMS pro would be a better option than the original. 

[LS1RX7]

Agreed, if any component has a history of premature failure no matter how common or uncommon and there is a upgraded option available, you would be mad to use the original.

 

[Peter M]

I reckon its related to engine speed.  At lowish revs the torsional vibration in the IMS causes the balls to skid rather than roll and this causes the bearing to fail.  With the revs up, the vibration reduces and the bearings last. 

Realised I've only given a partial answer.

Even without considering the effect of torsion vibration, I think the lack of preloading on the IMS bearing is the biggest culprit.  It is Good Bearing Practice 101 to have the inner race axially loaded against the outer race to reduce ball skidding.  Usually this is done by spring loading with a wave washer for example or by shimming like in gearboxes. Often on double race bearings, each race is preloaded against each other to do this.

Im now wondering if Porsche did that with the last larger dual race bearing revision and this is why these one are regarded as trouble free.  Pity no ones had to replace one so we could see!

[Peter M]

- so they had to simplify and reduce production costs and this is why they went to one cylinder head design that could be used on both sides of the block. 

I'm wrong.

They never "when to one cylinder head design [for the 996]" at all.

Porsche had always used one cylinder head design for both sides of all 911's. Both sides of the 356's and even the 928!

I should have said "Porsche continued to use one cylinder head design for both sides of the engine block as that's what they had always done so it shouldn't be a surprise they continued this engineering tradition with the 996."