The shims & cir-clip on the early T Tubes ( up to 1984 aprox ) were on the side of the engine flex plate closest to the flywheel , so with the shims & cir-clip on that side it can not have any function in stopping the T Tube quill shaft from moving forward , it can only ( if forced ) move the flex plate back , however in that mode the engine flex plate would flex" back" towards the trans direction , in the direction they never go

However if the shims & cir-clip were on the trans side of the engine flexplate ( which they never were ) , then & only then would it hold the T Tube quill shaft in the same position within the flexplate female splines if the front coupling became loose ( inhex bolt stretch ) , but of course it would allow pushing the engine flex plate forward , so the shims / circlip are of no help at all with stopping the T T quill shaft from moving forward

 

And then engine oil entered the fray...

FWIW: The PO of my car was the type who changed all fluids on the car very frequently....and got everything at Wal-Mart. When I asked what brand and weight oil was in the engine, he replied "Whatever it is I got it at Wal-Mart in 20W-50 weight". Perhaps his use of high viscosity oils has contributed to the engine retaining a 0.178 mm end-play. It's also important to note the PO bought the car in 2004 with 25K miles and I purchased it in 2013 with 85K. He used it as his daily driver for 9 years. So the car's history is well known and maintenance very consistent.

All the knowledge from the folks with decades of experience seems to indicate that BOTH front and rear clamps/bolts are contributing to the problem, with wide variability from car to car. In fact, I can see Super Clamp or PKlamp masking a loose rear bolt, by not allowing the shaft to migrate forward relative to the front clamp. I think lots of great knowledge has been gathered by many dedicated folks working on this problem - yet the problem continues to have subtle complexities to keep us scratching our heads.

 

Hi Jon , yes you are correct , its a combination of the front & rear T T couplings & in regards to the Constantine Super Clamp for the front coupling , its very high quality , I just wished there was the room in the rear of the 928 ( 84>) T Tube rear coupling area so Constantine could make something similar , but the rear coupling area is just too small an area to make something work , but then again Constantine may come up with something for the rear coupling

And on that subject ( T Tubes ), the Constantine Super bearings , be it for 928 or 944 are the best engineered I have come across as a complete package

 

Exaxctly. But shims are very helpful in preventing front clamp moving forward on center shaft when center shaft remains in its starting position. Whole reason for puting in clip and shims was to make sure flexplate has no possibility to move forward. Porsche didn't put them there just for fun of it. They saw flexplate clamp movement as a problem. Problem existed in only forward direction. By leaving them out starting in '85 MY they made TBF possible. They saw wrong installation as larger problem and deleted them without realising how big problem it would cause.

I have seen flexplate movement without rear clamp being loose or shaft position changed in any way. Its true that real clamp can work itself loose but its not prerequisite to flexplate being out of shape. And its been out of shape towards crank in every occasion I know. Never towards gearbox. Thats also reason Porsche put shims only on front side of front clamp. Those shims had no function at all in '78-84 cars if problem were at rear clamp design.

I haven't looked into this but its possible Theo's front clamp could be used also at rear or at least very similar design. It would also work in manual gearbox cars which have same rear clamp looseness problem.

 

Exactly right Captain Slow.

Then the clamp solutions are fitted and there is no more movement. That's because there is no more movement to be had on the rear splines.

Cheers.


.

 

I'm not sure if we all agree or not. Unless damaged, there is very little play in the bolt/slot rear clamp mechanism. Tighten the rear clamp in whatever position you can and the TT shaft will still migrate back through the front clamp up to 2-3 mm with ease, taking only a few months to get to that point, bowing the front flexplate, preloading the thrust bearing. Release and reposition the front clamp, relieving the bowing and preload, and it reappears in a few months, ad infinitum. Put a PKlamp or Superclamp on the front clamp or install an 85 or older TT shaft with the shims (I've done it in several S4s) and the 2-3mm (or more) of preload on the front flexplate never accumulates again. The plate stays perfectly flat and wear to the thrust bearing is negligible. If the shaft were moving significantly at the rear, the front flexplate would bow one direction or the other and there would be galling or wear of the bolt or slot in the shaft. If there is a fundamental difference between your conjecture and my observation, I don't know what it is.

 

Hi Bill -
I used your write up/pictorial and Dwayne's to perform my flex plate and end play check over the weekend. You've helped a lot of newbies like me do this important check. Thanks.

Is it possible that we are having different experiences? Or just using slightly different wording about the same experiences? Let's see...

You wrote:
What you say above sounds ALMOST exactly what I observed over the weekend. EXCEPT for one small detail. I found my flex plate to be deflected between 2-3 mm, as you state. But my shaft does not appear to have migrated through the clamp. It does appear that the shaft has migrated forward, deflecting the plate forward 2-3 mm. When I released the bolt on the clamp the flex plate returned rearward to it's neutral flat position (moving the camp rearward 2-3 mm). I didn't see any signs that the clamp had migrated during the years of driving. It seems that in my case (especially given how little of my splines are still visible) the shaft is migrating forward, and releasing the clamp bolt is simply allowing the plate/clamp to travel rearward to the neutral position.

The fact that Dwayne measured about 3/4 inches greater distance to the forward TT bearing and he has about as much more splines showing than I do, leads me to think the TT bearing on my car is migrating forward with the shaft. I fear that the next time I make this check I may find 2-3 mm of deflection and the same small amount of splines showing as shown in my after-release-photo. Then if I release the bolt again I might see the flex plate/clamp move rearward to the neutral position - causing the remaining splines to disappear. Thus, I'll be checking the rear clamp position soon.

I'm starting to wonder if there are a variety of circumstances involving the front clamp, rear clamp, both clamps, driving habits, proper weight engine oil, wear and tear on TT and TC bearings, engine and trans mounts, etc.; all varying from car to car, that all work in concert to varying degrees to manifest the range of observations.

I'll inspect everything again, including rear clamp, during first week of April. This phenomenon can be a worrisome, frustrating, and produce some tragic ends to 928 engines...but it sure is fascinating.

 

Hi Erkka , "if "we saw / experienced mainly the front T T coupling loosening its grip ( bolt stretch ) then and only then you would be correct in regards the the shims & circlip ( meaning the shims & circlip would hold the flex plate) , but its not , in fact its the exact opposite , we mainly / only see the rear T T coupling loose ( bolt stretch ) and the front T T coupling not loose or sometimes 3/4 of the torque that its meant to be, meaning the front coupling is tight enough , certainly enough to still grip the splines

So with the rear T T coupling becoming loose ( T T bolt stretch )or a lot looser than the front , this has become the normal thing we see over the last 20 years or so , meaning if one replaces both the front & rear coupling inhex bolts with new ones & then say after 20,000Kms ( example ) of a mixture of normal driving & a few hard acceleration events from the standing start etc , one will see ( via a torque meter ) that the rear inhex bolt has loosened its grip ( bolt stretch )

The interesting thing about the shims/ circlip , is that Porsche simply stopped fitting them( early on ) , so thinking logically , if they thought this would help to stop a problem ( flex plate deflection ) , they would of just kept fitting these very inexpensive little shims & circlip ( probably no more than $ 4 worth ) or if they saw an issue say by late 1980's or early 1990's , Porsche only had to send out a service bulletin saying they recommend fitting of these cheap little things as a ( in field service program ) & start fitting them again in the new cars being built , that didn't happen , because back then there was no issue

 

Apart from Me, JET951 and Captain_Slow ...

... You are all still WRONG ! ! !

Why?

Because ... if two metal surfaces under heavy compressive load slide one past the other, scrape or slide marks will be visible on each surface.

In this case, rearward of the shaft splines from where the front collar has migrated forward.

You can't drag one lump of metal over another, under pressure, and not leave marks on both pieces.

It won't be just simple dragging either. Highly likely there is also vibration involved, which will further emphasise the markings.


I'm looking for something like this:

Before:




After (my impression):





So far I have not seen any pics showing this.

Until the 'rear clamp' theory ... DENIERS ... hoo arrh harr harrr ... can produce pics showing evidence of sliding contact then AFAIC my analysis stands!



.

 

DA-dannnngggg! X Wrong. Do not pass go. Do not collect $200.

All these pics show are images of a front collar in two different positions on the shaft. The same marks on the inside of the Allen bolt head in each pic show this is not a genuine before/after situation.



.

 

Are you saying you don't believe the front clamp moves relative to the tube? Many people have reported this issue.

 

Front bolt is not enough to keep coupling in place. I have seen flexplate movement myself. Thats half of the reason why I went through trouble of sourcing and installing '84-87 MY torque tube and shims at its front end. Real bolt to not have to be loose and front still moves. It doesn't take much driving to make it happen. Only way to stop it moving is to do some changed. Easiest way which I trust to work is Theo's clamp. Movement will keep on happening no matter what is done to rear end. Its true there is hardly any markings on bores at front end when it moves. How it can happen I have no idea. In any case there definitelly is more splines visible than when front bolt was put in.


They stopped fitting them because some mechanics installed stuff back in wrong order and caused thrust bearing failure that way. AFAIK they have never acknowledged leaving them out made things worse and caused more TBF failures than mechanics ever could. They didn't see issue they created and though it was all mechanics fault. At least thats official line from factory.

Infact Porsche's official line contradicts what you're saying. Their version is that by tightening rear clamp at wrong time that clamp was good enough in holding shaft and clamp together that it pushed crank forward without any piece in the chain giving in at all.

 

If some folks want to believe the front clamp does not slip well that is their choice.

If you know that your front clamp is slipping and you are confident it will mysteriously stop at 2mm or so extension by all means believe that.

No one has ever said the rear clamp cannot slip but I have seen 5 engines fail due to TBF, none of the owners knew anything about it until after the fact and that in a sample population of about 30 or so examples. All failed because the front clamp slipped including one of them the original GTS motor in my current 928.

As I have written many times, my clamp would last about 2 weeks before it slipped- then I could immediately feel a vibration at exactly 3050 rpm. Reset the clamp it dissappeared. Then I tried the Loctite solution and the problem has never re-appeared in some 15 years - now go figure what that means.

I cannot guarantee something will not let go on my 928 but I know what I have seen and probably why Constantine, Ken and Theo have come up with viable solutions to a real world problem. Has anyone who fitted one of these solutions gone on to suffer TBF? Never read of it if they did.

I feel sorry for the hapless souls who knew nothing about the looming problem and lost their motors. However, I would have little sympathy for any "enthusiast" on this list who lost a perfectly good motor despite being advised about this problem and did not take any mitigating actions.

Whether anyone actually knows for a fact the actual mechanics that are going on remains to be seen but my take is that the clamp slips in very small increments probably when there is a sudden shock caused by a kickdown type of change, it is unidirectional so polishing does not take place. I stress this is my theory, maybe it can happen in one sudden big dollop but somehow I doubt that.

Regards

Fred

 

Hi Fred – I’m guessing after using the red Loctite you put a dab of paint on the splines and that’s how you see if your clamp moved(?) I did red locktite/paint on splines thing to my car ~4 years ago and twice a year I check the paint mark to see if there’s been any movement. Knock on wood—none yet. Just curios if you do anything else like put a straight edge on the flywheel to see if it’s bowed or anything else. I feel like I should be doing more than just checking the paint mark on the splines.

 

We just re-posted pictures initially posted by Jon aka "Captain_Slow" since they are representative of what most 928 automatic owners find when they inspect their front flex plate and have not taken steps to stop drive shaft pullout.

So you're saying that Jon staged the photos?