[Smallblock454]

Hello,

Originally Posted by Gelbster

Markus,
…Your list of 7 IMSB problems/causes is interesting. I have never seen such a list before. The pragmatic question is - what can we do about it?
Any of the 7 issues, if severe enough will eventually cause IMSB failure. Let's suppose I have a still-running engine with one or more of these issues(but have no way of measuring it). I remove my original OEM IMSB at 90K miles and find the seals 'blown, the tube with oil in it, the bearing a little wobbly, no ferrous debris in the filter or pan. I have 'caught; it just in time.
So what do I replace the IMSB with - suspecting I have one or more of the 7 deadly problems with the tube?
The answer is - the most fault-tolerant bearing available - The Solution. But will it pass "Pre-Qualification" for Pro installation+warranty?
If not, what to do? The second best choice is debatable - as are the tolerances for all the 7 deadly IMSB sins in your list and the way they do/do not interact to cause premature failure of the IMSB.
I guess the second best choice is DIY install of "The Solution" in this scenario or a cylindrical roller bearing if on a budget?
Thank you for your interesting perspective that may clarify this mystery a bit more.

Hi Gelbster,

well first of all there is a myth that the IMSB never was designed to be replaced while the engine is in the car. Because there is no official Porsche special tool offered? Simple question: would some shurely not too dumb engineers would have designed the mount and bearing this way?

Best solution:
There are different aspects. I think everybody has to decide himself what he or she wants to do and what amount of investment and risk to take.

Pre qualification: i completely understand why there is that process. If you have hard metal debree in a soft full aluminium engine with Lokasil coated cylinders you have a major problem. Some of these engines will still run for some time (maybe if you're lucky and the metal debree amount wasn't high forever), but sooner or later you can run into problems. A new IMSB will not solve the problem. And as said there are a lot more causes why the engines can fail.

Cost:
Does it make sense to put a 2K-3K solution in a 10K car? Would you think one minute if it would be a 100K car?

On the other side:
A 986 engine rebuilt in Germany starts at about 12K Euros and grows up to about 15-20K Euros for a more optimized engine.

An used 986 engine is available at about. 3-4K Euros. But you'll never know what you get. And to make it a good one you also have to put some additional money into it precautionary.

Installation:
From my personal POV an installation can definitely go wrong if not done properly.

Perfection / perfect world:
If you want to make 100% shure that there is no problem you have to rebuilt the engine every 50K miles. But as always - also a rebuilt engine can fail sometimes.

Conclusion:
I don't have a solution for that.

My personal opinion:
A proper installed high quality roller ISMB with every clutch replacement. And if your DMF is worn even if the clutch isn't, replace the DMF, clutch and IMSB all together. And: if the engine fails, it fails - there are a lot of causes why the engine can fail. IMSB is only one.

Originally Posted by Gelbster

Here is a hint of how to qualify a would-be M96 engine rebuilder: "Would you let me see the IMS runout test on my shaft?
And what about the other 6 items on the list of 7 supplied by Smallblock.
My guess is that only well equipped and trained machinists would be able to do this test. The average engine rebuilder would not have the correct equipment, nor even know what to look for.
The pragmatic suggestion is to use a new M97 shaft when rebuilding an old M96? How else do you buy an M96 shaft that is straight? Then you do not need such a fault-tolerant/expensive bearing? Dang - I did not know this when I rebuilt mine - next time... :-)

As for installation and quality tests.

@ JFP in PA: If you have got new parts from Porsche that are out of tolerance, sent them back, tell them what you found and ask them to sent you a new part.

I see a lot of causes why a new part can be out of tolerance. Example: i lately ordered some Brembo brake pads in Italy (around 600 miles to my home). I asked kindly to package everything very safely before shipping. What did the idiots do. They throwed the OEM packaged brake pads (just put in a card box) in another card box and shipped them this way. The result is expensive garbage you can throw into a trashcan.

Also there are tolerances. These engine parts are mass production. Pistons and rods are not equally weighted before installation. Crankshafts are not finely balanced… If you want a 100% perfect spec engine you'll need to bring it to the 100% perfection yourself or by somebody else. Not to mention you'll need the machinery and know how to do it – in your case i have no doubts that you can do it.

@****Gelbster: For the DIY man who is doing a repair or engine rebuild in you own garage (like you did) i don't think there is a simple way to get everything perfect. Shure with time, know how and patience you can build a really good engine. But you need very expensive machinery and also know how to do and test some things. One solution could be to bring the parts to a local machinery shop and let them do the tests and optimization for you.

Originally Posted by flaps10

Sorry Markus, I don't mean to pick on you. This is a great, healthy discussion and I just want to make sure that some assumptions die where they should, so thanks for contributing to the discussion.

CNC machinery (which I've spent my entire working career around) only make it possible to make the same exact mistake a shocking number of times. People like to assume that CNC means perfect when compared to an old man on a 100 year old conventional mill or lathe. Accuracy and repeatabilty are functions of machine rigidity (stiffness) and not the black box hanging on the side of the machine giving orders.

The case halves are cast (and rather poorly if you examine them), which are then machined in several operations before they are assembled into an engine. Each time you remove the casting from one machining operation a human puts the casting into the next machine. There is room for error and variation every time you touch the part, and depending on how the part is "registered" (that is, how they pick up previously machined features to index the next operation from) it can have a huge effect on the consistency of the finished product.

A loose chip in the fixture will cause variation, a cutter that is getting dull or has runout, how tight one operator clamps the part in the fixture compared to another guy, etc. I've seen marks in a part after a train went by the shop I worked in at the time and older machines could just take a left through the middle of a part on a hot day for no reason.

IMS runout doesn't surprise me since the thing is pressed together, and having a sprocket slip isn't uncommon.

@ flaps10. That is what i've wrote / meant. The aluminium engine case is made by an aluminium cast. Then it was CNC machined. Then it was Lokasil coated. Think i've wrote that, no problem at all. If there is something misleading or not understandable just ask. My english skills are just the way they are. Always a problem to be exact and precise for me.

I agree that CNC machining doesn't mean there can be faults and errors. But if you take a look at modern engine production plants you'll find the processes everywhere. Also every 986/996 engine at Porsche were assembled by skilled and educated mechanics. So there was a quality control by eye and also by machines that controlled the results - shurely not at todays standards.

I could go more further in that production process, but i think that would be much to much for that post and honestly it would take me a long time to translate the terms and processes precisely.

Originally Posted by JFP in PA

The cost benefit calculation for the IMS Solution is multi faceted. First, and perhaps most importantly, it is a permanent fix that never needs replacement. So if you plan on keeping the car for a longer period, you need to consider the cost of multiple IMS retrofits vs. the one time cost of around $1700 for the Solution. Secondly, the IMS Solution is tolerant of conditions that no other retrofit would tolerate, such as high run out shafts. As I keep my cars for a long time (my daily driver, purchased new, will turn 20 next year), when I retrofitted both my and my better half's M96 cars (both purchased new), I did not even consider another choice, they both got IMS Solutions (one dual row, one single row), and I could not be happier with the outcome. And as we do PPI's for customers, I can also tell you that buyer's are aware of the value of the IMS Solution as well. People tend to forget that even a quality IMS retrofit that has 60K miles on it looses some of its value at resale, as the buyer has to factor in doing it again in the foreseeable future. The IMS Solution does not.

When I refer to run out, I am looking at how true the surface that hold the IMS bearing stays as the shaft rotates. Exactly where the movement comes from is a more complex issue than just bending. Because the bearing mount surface is machined into the shaft, its center line may not be on the shaft's center line, which could give you a straight shaft that turns true, but that is not what the bearing sees. Some shafts may also not be true themselves (perhaps from the process of pressing the chain gears on), causing the same issue even if both centerlines coincide. And while I have obviously not personally seen the numbers of these units that Jakes has, he has shown photos of shafts where the run out was well beyond 0.010 " and mentioned how he has had to develop screening procedure's to test shafts going into the engines he builds. Just another one of the joys of the M96 engine.

@ JFP in PA: These are interesting aspects.

Let's calculate some things:
LN premium solution: $1700 plus professional installation or premium tool $600 and plus cam locking… tools plus DIY. I would estimate 2.5K plus.
Conventional high quality ball bearing plus new updated mount: $200 plus materials for tools made yourself $50. I would estimate $ 250 plus.

Clutch needed around every 60-70K miles. AT-car - other calculation.

IMS run out:

I'm not an engine engineer, but i think it's an interesting aspect. Is a ball bearing capable to handle more IMS tolerances than a plain bearing? I did a research and from what i understood i would say yes.

@ all / for those who are interest in bearing technologies and want to get more into it:

Some infos about bearings in general (english language)
Schaeffler Germany | Products & Services | Rolling and Plain Bearings

Some infos about roller bearings (sorry, only german language)
Technische Grundlagen

Regards, Markus

PS: If you have access to a spare engine and a 3D scanner and a 3D construction program you can do funny things.