Mark R.

Wow. That is a very compelling picture Dave! Thank you.
I have never seen or experienced anything that severe before.

It's not clear to me how that can possibly happen on a manual.
What exactly does it pull against in the front..?!?

Dave928S

^^ My guess/theory is that, on WOT/max torque acceleration events, the stub shaft is under such torque from the clutch plate that it 'holds' the shaft and stops it sliding momentarily, even if it has been properly lubricated (my 82 hadn't). At that same moment in time, the shaft is experiencing 'wind up' shortening due to the max torque applied, and it moves in the direction of least resistance, which will be forward, if the rear clamp isn't resisting that movement.

A also think that my 82 was so severe, because no one had replaced the rear bolt (notice factory red paint dot on the head), after 245,000km !!!!!! A person who represents himself as a Porsche enthusiast/expert owned the car before the previous owner, who sold it to me .... expert in BS dodgy repairs was what I found when I stripped the car. e.g; past catastrophic failure of aluminium front ball joint .... 'fixed' by replacing with a secondhand aluminium ball joint.

JET951

Hi Mark R , its quite simple when you realise that with a manual trans version 928 , once you accelerate in gear (any forward gear or reverse) the twin clutch plates ( 1978-1986) or the single clutch plate version (1987-1995 ) will be locked onto the clutch input shaft ( male against female splines ) they are locked solid with the engine torque being transmitted through this point , particularly under hard acceleration in the lower gears

And its this hard acceleration in the lower gears ( from standing start particularly ) , the same principle that we were talking about with the auto trans version & the flex plate etc , only this time its a bit worse
On the manual trans version , naturally there is no" front flex plate" that will give a little as the T Tube shaft twists & " Shortens " , so here we are , manual trans , standing start hard acceleration ( say in 1st gear ) , the T Tube shaft momentarily twists under the stress of the engine Torque and vehicle weight / rolling resistance etc ,but up front the clutch plate/s are locked to the input shaft splines , BUT the rear of the T tube coupling is at this moment in time not tight enough( has not been maintained properly ), just like on the auto trans version , something has to give & it does

The rear of T Tube shaft pulls forward & pulls one end of the half moon cut -out in the spline shaft

You will also see the "manual trans version" break the cut in three places" hollow input shaft ( who the hell would design something this STUPID ), this is VERY VERY expensive , because it means a completely NEW 5th gear matching set $$$$$$$$$$$$$$$$$$$

Seen plenty of those over the last few decades & completely avoidable by periodic maintenance in the rear of T Tube rear coupling and its VERY STUPID single ( 1 only ) in-hex bolt ( quite a bad design ) and this bad design has to be maintained or it becomes VERY expensive

Again I have been warning people about this stupid rear coupling for years & years & years & years & years ( its making me feel my age )

FredR

Akram,

My concern is that once the rear bolt let's go damage occurs along the lines mentioned by Bruce and as shown in Dave's photo. If the damage permits travel of 2mm then in terms of this system that is a country mile and thus my indirect concern [that I preferred not to suggest at this stage] was whether some mechanical damage has been inflicted to any of the components.

Bruce advises there is a bit more movement there [naturally] than superficially seems to be the case and if so fair enough. I agree with Bruce that if the components are in good nick and maintained that way the system can and does work. However, once mechanical damage has been inflicted I doubt that the system will ever hold again due to the tremendous loads inflicted on them.

As for the Loctite method folks have to remember that when this problem with the front clamp first arose there were no solutions for the issue. My problems started in early 1999 just after I got the car. I felt a vibration at 3050 rpm that was anything but natural and was extremely annoying to me. I feared something was wrong and sure enough one afternoon driving through the Hajjar mountains - a beautiful location on a super section of road in the middle of nowhere I went to overtake a comedian and bang the drive shaft snapped on a top end car that had done less than 50k miles- rediculous. Porsche had no mercy on me- just coldly told me "these things happen" -yeah- right- to folks dumb enough to buy their products! I tried to raise a goodwill claim but no joy unless I purchased from them and then no guarantees of course- but the local agents were most supportive to their credit and wanted to help.

That was where the fun started. At that time we had just been opened up to public internet access so I signed up and immediately found Rennlist. Told the good folks what had happened and was aghast at the number of incidences of reported issues on the S4/GTS autos- what had I purchased?

I got a new torque tube, changed the torque converter bearings and the flex plate was inspected and deemed fine. The main agents fitted it all and everything was fine- for two weeks! Then the vibration problem re-emerged. So back to the agents but this time I asked to witness release of the front pinch bolt- we put an index paint mark on the crank side of the front clamp - released it and sure enough it sprang about 2mm to 3mm - urrgh! They immeditely reset it, got the torque wrench out to prove it was tightened correctly and sure enough the vibration dissappeared- great! This time we put another index mark [on the shaft] on the gearbox side of the front clamp. Two weeks later the vibration reappeared- back to the agents and sure enough the shaft had moved to the front index mark exactly as before.

The main agents immediately contacted Herr Ferdinand asking what to do and within 30 minutes Germany told them to fit a new pinch bolt and over torque it to 110% of rated. Again vibration gone and back again a week or so later. From my point of view I had a 928 that was likely about to suffer TBF unless something could be done. Porsche suggested to fit a new flex plate [at my cost] but I told them if they could not identify anything clearly wrong with it why should I pay? At least they offered a refund if it did not work.

In the meanwhile I was chatting with Earl Gilstrom on the list and he then introduced the idea of using the Loctite 290 formulation. Not sure if he did this on his car first but mine was the second 928 to try it as I immediately went for it. The problem disappeared never to re-occur and held for the intervening 5 years until I lost my 928 when some tw@t using a mobile phone clipped me when I was overtaking the dick brain on a fast off camber bend and I was launched into a parapet at about 75 mph - not nice. Fractured L1 vertebra, whiplash and knocked out instantly, my baby protected me from something much worse and when I came round a few seconds later the door opened like normal- a measure of how great these machines truly are!.

Later I acquired my current 928 purchased knowing the GTS motor in it was history due to TBF caused by the front clamp letting go on the poor chap who owned it at the time. I have seen in total 5 such engine failures over here all caused by the same thing - a significant proportion of the total fleet.

I concluded years ago that TBF seemed most likely to occur when the car was several years old and had covered in the range of 60k to 100k km [40k to 60k miles] but of course it can happen anytime once the clamp slips. I have never seen TBF caused by the rear clamp slipping and I have never had a problem with the rear bolt but then thanks to folks like Bruce I have checked and last year replaced the rear bolt even though it was perfect physically and dimensionally. My rear bolt has never slackened off and there has never been any signs of damage and hopefully this is what due diligence facilitates.

My front clamp was also loctited when we built this example some 12 years ago. I have retained this as long as I do not need intervention and if/when I do in will go Constantine's clamp. As I stated earlier I also trust that if I have a problem with the loctite leting go I will know it due to remergence of the 3050 rpm vibration. I occasionally check the front clamp visually and needless to say it has not moved in some 12 years of use but I do not do a high mileage using the car mostly for pleasure only.

Trust the above informative. The Loctite has served me well despite its limitations/perceived nuances. I have come across many who have used it and to date have never heard of a single instance of it having failed but I fully expect that sooner or later a failure will occur- maybe it is a good idea to change the title of your interesting thread to indicate the Loctite did not fail.

Do keep us updated with your progress.

Rgds

Fred

JET951

Hi Fred , hate to inform you , but 928 & 944 ( yep we have seen them ) thrust bearing failure can be completely avoided by a high oil film strength engine oil ( period )

This subject absolutely fascinates me , its a this century thing , never ever ever ever saw it in these last century Porsche's last century AT ALL , its a this century thing because humans are installing the wrong engine oils , even Porsche dealers could wait to install this the higher profit margins low viscosity so called synthetic engine oils in these and other deep last century Porsche's from the late 1990's onwards ( what a disaster that turned out to be ) ,isn't it interesting that Porsche is NOW pushing their 20w-50 engine oil for last century Porsche's , why weren't they doing this 20 years ago ( what a joke )

But no complaints from me , because in the last 20+ years it gives us so much work that is Completely avoidable if these same Porsche's were on 20w-50 back in the late 1990's & early 2000's

To cut to the chase , we have seen plenty of 928S4 Auto's ( example only ) that have had " MAX Flex Plate Deflection " for years , meaning the rear T Tube coupling was not tensioned enough ( rear coupling in-hex bolt stretched and none ever checked) but when we checked ( having never seen the said 928S4 before ) the Thrust bearing wear was within tolerance , meaning OK

But Why I hear you ask ?

Simple , the 928S4 in question had been ( very Lucky ) a decent Oil Film Strength engine oil the whole time , be it a 15w-50 or a 20w-50 & these engine oils had the Oil Film Strength to keep the engine thrust bearing protected

But yet we have seen 4.7L 928S2 ,928S3, 928S4, 928GT,928GTS , 944 with severe engine thrust bearing wear and ALL OF THEM had been on a crap oil film strength ( this Century ) so called synthetic with LOW levels of ZDDP , the oil had no chance of keeping the thrust bearing protected & all these oils were either ( going by previous service records ) 5w-40 , & 10w-40 engine oils

We have never ever seen engine thrust bearing wear when these engines have been on a high oil film strength engine oil , same goes for camshaft pitting & cam chain sprocket wear ( twin cam heads )

FredR

Bruce,

I am with you 100% on the use of 20w50- you may remember I also throw a tub of STP in to help the ZDDP levels. I use Shell Helix Super 20w50 mineral oil and change the oil/STP combo after 5k miles or 1 year.

Your point of view that engines using 20W50 will not suffer TBF I am less convinced about but nonetheless open minded that such just may be the case. My focus is on ensuring my motor does not get into the position where I find out the hard way. For sure a higher film strength oil will defer onset of failure but when the thrust bearing is converted into a disc brake with the flex plate mangled beyond belief one would be expecting way more from the lube than it can deliver or so I perceive and of course such can happen. As I am aware our engines were designed by Porsche for 20W50, especially so in warmer climes and it is downright dumb to run anything else in my opinion but doubtless many will disagree with us on that point.

My cam are also in excellent shape or were last I looked a couple of years ago. I run the GTS inlet cam with the S4 exhaust so I do not have the lift of the GT grind. I would not be at all surprised if GT cam pitting failures are related to witch **** lube. One of these days I might just get round to modifying the set of S3 cams I purchased a few years ago and fit them!

Rgds

Fred

Adk46

To a casual observer like myself, this is all extremely confusing. Other TBF-related threads have also confused me.

Just for some grounding in the basics, let me try out some assertions that I think are true and non-controversial:

1. If all clamping is relaxed and the system goes "limp" (no axial loads, flex plate flat, etc.), and then you tighten the clamps so they will not slip under any circumstances (may require different clamps or special procedures), you will not suffer engine thrust bearing failure.

2. On any actual car, without this hypothetical slip-free capability, circumstances may arise that cause the front clamp to slip, and lead to TBF.

3. Same as 2, but referring to the rear clamp.

4. The nominal pin-in-hole design of the rear clamp shouldn't allow movement, but it can slip anyway because of its actual design, tolerances, installation, wear or deformation.

5. The above-mentioned circumstances causing slip are numerous, or subject to opinion/debate, or of a nature that only extensive research will reveal with certainty.

Constantine

Takeaway from this thread aside from the oil discussion is:

The 928 drive shaft within the torque tube will wind up and get shorter under torque loads.

Unless the clamp bolts securing the drive shaft in both 928 automatics and five speeds are correctly torqued to WSM specs, the drive shaft can move under contraction and get hung up or damaged.

Periodic inspections of these bolts are needed to ensure they are still holding the drive shaft correctly.

What I would add is that in 928 automatics, due to the design of the front flex plate clamp, the drive shaft has a higher propensity to pullout and get hung up in a loaded state which can cause TBF.

Do not want to devolve into an argument about this and am not trying to promote this view just to sell our Super Clamps. This is fact well established by many owners who have arrested this front flex plate movement by using Loctite, the Ritech clamp, Porken's PKlamp as well as the Super Clamp.

This is very important for 928 automatic owners since once TBF becomes present, the 928 is a write off unless the owner wants to spend thousands of dollars and months/years to fix it.

Cheers,
Constantine

touay001

Thanks Fred for your reply. I have rechecked the shaft, clamp and bolt when I came back from work today. There is no damaged to any component as far as I can see. The bolt is free from any marks to the body. The clamp is intact. I used a baroscope inside the clamp into the detente and I could see no unusual wear. There is no red powder in the bolt threads but, the bolt is definitely stretched. The end play thankfully is within spec. The loctite has definitely fused the hub to the shaft and the loose rear clamp caused the flexplate to be preloaded. I have stated that the flexplate with bowed approximately 2mm. what I get from your post that if I remove the rear bolt and there is a movement of the shaft then, I shouldn't be able to reinsert the bolt. So if I'm able to reinsert the bolt then one of the following should hold true:
1- Damged shaft detente
2- Stretched bolt
3- While prying the flywheel and later realigning the rear clamp hole to the shaft grove the bolt was able to be reinserted.
4- The combination of 2 and 3.

Point 1 could be eliminated since a damged detente will cause damge to the bolt. Point 2 is proven. Point 3 and 4 are a posability. I'm I correct?

FredR

The real problem in all this is why some examples seem to hold on the front flexplate and some do not.

For sure if the bolts are not torqued correctly they will slip but some [like mine] slipped when they were torqued correctly [and with controlled over torque] and as I am aware no one has been able to conclude why this happens. Possibilities include quality control & inadequate tolerance specs to name but a couple.

Rgds

Fred

FredR

The crank end float is somewhat irrelevant to the issue at 0.18mm. I suspected possible damage to the detente but you have eliminated that. Inspection of my spare unit and experience with my current setup suggested that float on the rear coupling was not likely but Bruce advises it is possible so therein may lay the answer to this conundrum. Whereas your bolt may have streched and necked in the process there is no way I can envisage it has stetched to the point it would allow 2mm of float- it would be so grossly deformed you would be posting pics showing how crazy it looked.

Rgds

Fred

Mark R.

If Akram's rear shaft spline looks anyhting like the one in the picture Dave posted (the one I reposted in Post #46), then I can now see how the bolt could easily be reinstalled after the shaft was released, and had moved rearward by 2mm. It simply means that the shaft, when released, returned back to the proper location which lines up the original cutout for the bolt again. Prior to release, the bolt would've been sitting in the newly deformed indentation, and then, upon release, it moved back to the correct location.

The question I would have is whether a brand new bolt by itself will be sufficient to take care of this problem. It certainly seems like the damaged shaft would no longer have enough of a land to keep the new bolt from riding into the deformation, and therefore would quickly migrate into the new bolt...

.

Mark R.

All of this tells me that the exact same "ratcheting" mechanism (for lack of a better term) that occurs at the front coupler can also occur at the rear coupler.
Ratcheting as in the shaft pulls out of the coupling, but then is unable to push back in, which over time leads to the noticeable migration of the coupling on the shaft.
The difference is that the front coupling relies strictly on clamping friction, whereas the rear coupling relies on clamping friction PLUS the bolt cutout groove.
So it requires even more of a failure at the rear for the clamp to migrate there.

I have always been under the impression that the rear groove does it's job, and have been focused on dealing with the front clamp.
I think I may have just been lucky that my rear clamp bolt has remained tight, and has performed its task as designed...

.

bureau13

For us 5-speed folks, is there a way to check that things are OK without removing the rear bolt? Back when this came up last, I bought new bolts but never put them in...I'm just leery of removing something that "seems" to be holding properly. And by seems, I mean I don't sense anything out of the ordinary when driving it, but I haven't measured or observed anything directly, because I'm not sure how to do that, other than by removing the existing bolt.

Dave928S

Mark ... when the migration gets as bad as seen in my pic, you'll see, if you look carefully, that the splines have taken a hammering as a consequence of the lack of clamping .... this is the limiting factor on whether you could just use a new bolt. The shaft on mine was too far gone, and would have just moved again because the splines were damaged and sloppy. If the splines had been in good shape, I would have just inserted the shaft to give a fresh detent position, and used a new bolt.

I agree with Constantines summary. We all have our own theories as to why this movement occurs, and the methods to deal with it .... all that matters is that we are aware of the issue, and that we all include front and rear checking, and bolt replacement and re-torquing, in our routine maintenance schedules.

Edit: and Freds confusing issue ... simple ... he must be a lead foot ....