I can't argue with the value of proper weight oil with a good level of zinc and phosphorus. I've always used M1 15W/50. However, as the face of the thrust bearing is not fed oil at pressure [correction from later post by Tails - it is of course fed oil from the adjacent crank bearing flow out the side], I still like to keep the preload to a minimum, and the aftermarket front TT shaft clamps achieve that.

I still don't understand the claim that the rear clamp is the culprit in developing excessive preload.

 

As I understand it, the job of a flex plate is to allow for expansion of the torque converter as engine RPM increases. This result in a constant expansion and contraction of the flex plate and a cyclic thrust force. Does anybody know how much flex is occurring during normal driving? Interesting to compare this to the preload typically found on some cars. The preload will of course come in addition to the cyclic force and give a higher peak thrust.

 

Hi ihoe,

The torque converter in normal cars is directly attached to the flywheel via a flex plate. In that scenario a expanding torque converter will ultimately push against the engine's thrust bearing.

The torque converter in the 928 is at front of the transmission housing which is located at the rear of the 928. The torque converter flex plate is attached to it and the splined coupler passes through the housing and is located by one or two bearings, depending on the model year.

If a 928's torque converter balloons too much it would start machining the flex plate into the housing. Small deviances in its size will not be felt at the front flex plate/flywheel/thrust bearing due to its isolation from the front flywheel by the rest of driveline.

Cheers,

 

Bill,

Agree with you on all your points.

 

I recall that the TC floats on the input shaft with some clearance to the rear, suspended on a rear flexplate, and is free to move rearward, so even if it expanded, it would move back over the shaft on which it floats, not forward. And even if it moved forward, it would have to press the snout out of the cover to move the TT shaft. Never heard of that happening. Also, I think the rear flexplate is smaller and stiffer than the front and is more of a locator for the TC.

 

I don't understand completely how a expansion of the converter will not be felt by the flex plate because its isolated by the rest of the driveline. I thought it was a direct axial connection between those parts. Can anybody explain what actually causing the shaft to be pulled backward out of the splines? I see people talk about "wind up" causing it, but to me the loading condition to the shaft is pure torsion. Why the pull? Is it pulling backwards because it is elastically twisted? As a new owner of a 928 automatic I think I have read almost everything here on Rennlist about this topic, but still not understand everything:-) By the way, I understand very well that the preload seen on many cars can cause problem with the thrust bearing.

 

Further to my post #60, it must be remembered that the 928 engine was developed in the early 1970 and engine technology has progressed over the last 40 years with bearing clearances reduced.

The thrust bearing is fed lub oil from the flow out from the crankshaft bearing and has to maintain full film lubrication as the pressure from the migration of the front flex plate causes an increase in load that break down this film to boundary lubrication, increase in temperature, reduction in viscosity in this area in a degenerative spiral until thrust bearing failure.

The follow is an explanation via tribology of how this failure occurs. I have experienced this type of failure on internal combustions to thrust bearings, crankshaft journal bearing and connecting rod bottom end bearing which have caused the crankshaft to have thermal cracking in way of these bearings.

Explanation:

Tribology is the science and engineering of interacting surfaces in relative motion. It includes the study and application of the principles of friction, lubrication and wear. Tribology is a branch of mechanical engineering and mineral science.

Viscosity, by definition, is an oil’s resistance to flow and shear. It is the single most critical physical property of the oil as it affects both the wear rate and the fuel efficiency.

The most common unit of measure for viscosity is the Kinematic viscosity and this is usually quoted in data sheets at 40°C and 100°C. The commonly used unit of measure is centistokes but the correct SI unit of measure is mm2/s.

Absolute Viscosity is a measure of a fluid's internal resistance to flow and may be thought of as a measure of fluid friction and of the oil's film strength to support a load.


High-temperature high-shear-rate (HTHS) viscosity is an indicator of an engine oil's resistance to flow in the narrow confines between fast moving parts in fully warmed up engines. The most common test for this is ASTM D 4683 as it closely mimics the conditions found in an engine's crankshaft and connecting rod journal bearings, as well as other narrow regions such as between the cam and follower on flat bucket tappets. This measurement influences such factors as fuel consumption, valve-train wear and bearing protection.

The viscosity will determine how easily the oil is pumped to the working components, how easily it will pass through the filter, and how quickly it will drain back to the engine. The lower the viscosity the easier all this will happen. That is why cold starts are so critical to an engine because the oil is cold, and so relatively thick.
But, the lower the viscosity, the less the load the oil can support at the bearing on the crankshaft. The higher the viscosity, the better the load it can support. Even this, however, has a trade-off, since the higher the viscosity, the more the drag at the bearing, and hence, potential power loss, or increased fuel consumption. So a compromise is chosen to minimise power loss, but maximise load support.
For domestic use, engine life is important, and in the main you should adhere to the recommended viscosity for your engine. For motorsport, engine life is not critical, winning is, so these high performance engines can use lower viscosity oils to maximise power output to the wheels, but then again they generate a lot more heat so may use a higher viscosity anyway.

There are three different types of lubrication: boundary, mixed and full film. Each type is different, but they all rely on a lubricant and the additives within the oils to protect against wear.

Full-film lubrication can be broken down into two forms: hydrodynamic and elastohydrodynamic. Hydrodynamic lubrication occurs when two surfaces in sliding motion (relative to each other) are fully separated by a film of fluid. Elastohydrodynamic lubrication is similar but occurs when the surfaces are in a rolling motion (relative to each other). The film layer in elastohydrodynamic conditions is much thinner than that of hydrodynamic lubrication, and the pressure on the film is greater. It is called elastohydrodynamic because the film elastically deforms the rolling surface to lubricate it.

Even on the most polished and smooth surfaces, irregularities are present. They stick out of the surface forming peaks and valleys at a microscopic level. These peaks are called asperities. In order for full-film conditions to be met, the lubricating film must be thicker than the length of the asperities. This type of lubrication protects surfaces the most effectively and is the most desired.
Boundary lubrication is found where there are frequent starts and stops, and where shock-loading conditions are present. Some oils have extreme-pressure (EP) or anti-wear (AW) additives to help protect surfaces in the event that full films cannot be achieved due to speed, load or other factors. These additives cling to metal surfaces and form a sacrificial layer that protects the metal from wear. Boundary lubrication occurs when the two surfaces are contacting in such a way that only the EP or AW layer is all that is protecting them. This is not ideal, as it causes high friction, heat and other undesirable effects.
Mixed lubrication is a cross between boundary and hydrodynamic lubrication. While the bulk of the surfaces are separated by a lubricating layer, the asperities still make contact with each other. This is where the additives again come into play.
With a better understanding of this process, it should be easier to define what lubrication actually is. It is a process of either separating surfaces or protecting them in a manner to reduce friction, heat, wear and energy consumption. This can be accomplished by using oils, greases, gases or other fluids. So the next time you change the oil in your car or grease a bearing, realize there is more going on than meets the eye.

 

The automatic 928's drive shaft, a 25mm or 28mm steel rod, is pinned at both ends. Under torque load it twists and the drive shaft contracts some due to this twist. The OE front flex plate clamp by itself cannot hold the drive shaft under this contraction and the drive shaft pulls out a bit. This walking out of the drive shaft usually stops at about 2mm-4mm of forward bow on the front flex plate.

This forward bow then puts pressure on the flywheel, which is connected to the crank, which drives it into the rear of the engine's thrust bearing.

Some people find it hard to conceptualize this happening until one studies how torsion bars work in suspensions. Their twisting is why they are used instead of standard springs and they have been in use by many manufacturers for years.

This property was called on by the drive line engineers of the 928 to soften some of the harshness of the 928 drive train. Other manufacturers like Chevy, who use a hollow aluminum drive shaft in the Corvette's transaxle drive line, place rubber "guibos" in the drive line to soften this harshness. Audi, BMW, Bentley and others use rubber guibos too. Guibos wear out and need to be replaced with some regularity. A search on Youtube will yield some videos on replacing rubber guibos. "Guibo" is actually a misnomer for the actual part, but "guibo" is used widely as its name.

You really need to study the manuals and talk with real drive line engineers to fully appreciate the beauty of this design for the 928 drive line. However Porsche screwed up a bit when they took away some parts of the front flex plate clamp design in 1984 and never changed the front clamp afterwards.

HTH,

 

Hi Constantine , I am glad that is you're first name I only like referring people by there first name , actually I took the comment about me not advising my clients about the T Tube pull out personally , so I responded in that vain , but we ( both of us will stop that here & now )

Now the words " shaft Pull Out " , that sounds like the shaft is pulling back away from the engine flex plate (at the front of the T Tube ) , now do you remember the photo I got Sean to put up on a previous thread about this subject( Rennlist ) where it shows the teeth cuts on the side / flank of the rear coupling inhex bolt ( remember ) , I think Dave posted a similar photo .
This was the rear inhex bolt stretched to the point where the grip / tension it gives was ( for want of a better expression ) no longer , the teeth marks clearly show the simple fact that the bolt has not turned ( it stretched )

The Teeth marks( bolt flank) are the result of the T Tube shaft moving forward from the rear coupling , now the bolt can not move the same direction or any direction because its in the coupling clamp hole.
Now as the T Tube shaft ( quill shaft ) moves forward & the flank of the bolt starts to ride up onto the male splines of the quill shaft , this partly destroys the splines as the splines dig into the side flank of the inhex bolt

Think about it , you need something to hold a bit of ( say ) grip to the quill shaft to hold it in the forward position , its the black nylon surfaced sleeves ( centre of the TT bearings ) that grips the quill shaft just enough to "Hold" it forward , hence the held deflection of the flexplate & if one just releases the front coupling only , then ( yes) you will release the flex /deflection of the flex plate you get a result , but its false , you need to push / lever the quill shaft back into place & amazingly the cut-out in the rear T T coupling will align directly in the centre of the half moon cut-out & one does not have any issues getting the new bolt into the rear coupling because everything aligns perfectly

Think about it just for a moment , lets say the front coupling was the culprit , OK , so in that case it loosens and the slip occurs on the splines , well why on earth does it not just slip straight back ?.
Why do you have to undo the front inhex bolt to get it back ( false resetting of flex plate ) , the reason you have to undo the front coupling is that it needs to be loosened to move it ( common sense )

Now here is the crux of the issue , the only person to see this damage ( when it gets to this point in the car) is the mechanic working on the car ( 928 ) be it manual or Auto , now if &( I am only assuming here ) if you guy's only see the Torque Tube on its own , then I can fully understand why you think that the "bit of wear on the rear of quill shaft splines " this will naturally be on the rear part of the half moon cut out of the male "rear" splines , because this is the part that "rides under " the side/flank of the inhex bolt is from someone belting( using a hammer ) in the inhex bolt through the half moon cut-out , naturally this will damage the first few threads on the top of the inhex bolt , but ( the Photo ) shows no damage to the threads , just the flank , because the shaft has moved ( with some force ) forward

So naturally you do not get the chance ( if this is the case ) of seeing all the bits of damage & the shaft being /moved forward , it looks like just like ( shaft damage ) some fool has just hammered the bolt in

We have been seeing this for about two decades now , that's a lot of 928S /928S4/ GTS

The clincher for us is that we stopped this issue completely years & years ago , by }

A) maintaining the front & rear couplings ( new bolts & reposition the quill shaft back into place , which always makes the alignment of the half moon cut out in the rear splines align in the centre position , meaning the quill shaft had moved out of the rear coupling ( that's all we ever see )

B) Checking this rear coupling " mainly " every major service ( takes little time to do as part of the service & they are always a bit loose if the owner is a bit of a lead foot ( bolt stretch )

Now , engine thrust bearings , and yes the thrust face of the thrust bearing is subject to little or NO oil pressure , that means the Thrust Bearing face Relies on Oil Film Strength ONLY, I was hoping someone would mention that , its life expectancy is only linked to oil film strength

Just before Xmas and after I have had some phone calls from a few owners of Track modified 944 guy's here in Aus ( Melb actually ) , picking my brain about their failed engine thrust bearings in their track 944's , the reason they contact us is because we build very fast 951's & 944S2's & 968's for track work here in Aus for club events etc , but our 944's 951's ( all versions ) Never Ever suffer from either
A) Thrust bearing wear / failure
B) Conrod no 2 bearings ( we call them big ends here in Aus )

Why = massive oil film strength for these last century Porsche's
And yes the oils they chose was not high enough in oil film strength & yes they were all manual trans

Steve Weiner , Rennsport Systems USA , had found the same thing with thrust bearings in the air cooled Porsche's & I think even the 996 Turbo as well ( going from memory )

The use of low oil film strength oils ( read road emission friendly low ZDDP low viscosity engine oils ) create thrust bearing wearin these last century Porsche's & yes in manual trans Porsche's too from last century

So here we have all these small things ( individually ) coming together to cause issues( thrust bearing wear) & naturally the amount of use & the type of usage will effect the time line of thrust bearing wear , meaning I could put a low viscosity , low oil film strength oil into a 928S4 & if I never start the engine or just do a few kms a year , then naturally its going to take decades to show up anything

Now we go onto the multitude of 928S's 5.0L/ 928S4's / GTS's that we have never seen / worked on before, and we measure thrust bearing wear ,out of the no or new spec wear spec ( 0.06mm to 0.192mm ) , but now into the wear limit spec , meaning worn thrust bearings ( up to 0.40mm) , but just under the wear limit , we measure it , change the engine oil & filter , but this time we install a 20w- 50 engine oil ( as mentioned in the owners manual )
Note = in every case of thrust bearing wear , the said Porsche had been on a low oil film strength oil , meaning something like a 5w-40 and even some 10w-40 oils with low oil film strength

Then a year or so later we remeasure the Thrust Bearing wear & in ever case the Thrust Bearing wear had been arrested , naturally there has to be a time when we may find the thrust bearing is just too galled , but in the 8 to 10 years , this has not happened yet , every one of the ( just with max spec ) we have been able to stabilise

We had a 1984 year type ( built in late 83) 928S Auto trans 4.7L ( euro spec ) here at work about ten+ years ago that the engine thrust bearings had completely worn through & this was one of the first of many thrust bearing failures we had come across , this engine was toast
This was the one I took particular interest in , because the only thing we could find strange was that the owner said that he had switched to the latest & greatest " new Synthetic engine oil " about two years earlier , and this was a 5w-40 , which we found out a few years later that the (famous oil company ) who made it was very proud of the fact that the AW packages in the oil was the "New " Boron ZDDP replacement , it had almost no ZDDP , hence the massive cam lobe issues , thrust bearing issues , piston to cylinder issues etc etc , anywhere there was no oil pressure

Note } worth noting that the thrust bearing specs are different on the 4.7L as compared to the 5.0L /5.4L engines

WE have just assembled & installed a 4.7L engine back into a customers car ( 928S Auto ) we did a complete engine reseal with the engine / car has done 361,000 KMs ( city car all its life so the engine hours of operation would be something like 500,000Kms ), it basically really only needed seals & gaskets( oil leaks from head gaskets , common on the 4.7L engines at old age ) , the only thing that was mildly worn were the 30 year old valve guides , the cams/ lifter faces were like new , the engine main bearings including the thrust bearings were like new , the pistons & cylinders were amazing , no measurable wear in the cylinders , not even a tiny lip
The owner requested new conrod shells (big end bearings ) & new rings ( good idea) , but the old ones could have been reinstalled , but they are inexpensive new in std size, so why not

This car has only ever been on a 20w-50 all its life & there was what I expected = no wear

That's why I mentioned the 1984 928S we had with the chopped out thrust bearings all those years ago , it was destroyed by the engine oil that just did not work & the thrust bearing destruction wrecked the crankcase ( what a wast that was avoidable ) but is only avoidable if people / owners get to know that , hence the trouble I go to to minimise the destruction if I can because I do not want these cars going off to the scrap heap on an avoidable thing like crankcase destruction on 30 year old cars

If I see Sean over the weekend I will see if I can get him to post some pictures ( I am not that computer savvy when it comes to putting up photos )

 

This just amazes me that the front flex plate and the movement on the drive shaft is even an issue or debated topic. We know what causes crank shaft thrust bearing failure on one side of the bearing. (oil issues aside) The idea that you guys down under are finding some odd rear torque tube pinch bolt to be the odd man out just is amazing.

When shown photo proof of the movement (which happens on every auto TT 928 made) ya got one guy who is asking for the shininess on the shaft....really? Someone does not know how metal on metal friction works. You get shiny when the **** is moving back and forth over a time period. Not from one end of a shaft moving forwards, and forwards, and forwards a few micro milliliters over a period of time. Oh, then when someone releases tension at one point where the front flex plate actually has to forward tension put on it by the drive shaft that it moves back towards the rear of the car by 2-7mm.

Where do you backwards Ausi guys get that all of this movement is due to the rear flex plate bolt? Not happening.

It was mentioned that the torque converter swells front to back. Sure it might in the .001 measurements but anyone who has pressed in new converter bearings will know, that torque converter is not moving forward much at all. Definitely not moving enough to cause a front flex plate to push towards the front of the car. Before anything in the front of that rear flex plate is pushed you are going to destroy the brass ring in the transmission.

I hear Bruce is the be all end all (Greg Brown of the States) but fvck me man, the rear pinch bolt is not causing front flex plate migration on the drive shaft. You can't tell me you've never seen a hundred of them move back on the splines when you pull that front flex plate pinch bolt out. You know what causes that as well as anyone else that has touched one of these.

 

All four of my automatic cars had deeply dished flexplates.

Loosening front clamp returned them to near-flat.

Auxillary clamps hold them that way.

High film forming oil is just part of the equation.

 

Hi Guys here are some Pics that Bruce asked me to post up.
As in nearly every case we have of movement this is the outcome. The touque tube shaft moves within the rear coupling enough for the bolt to ride up the splined shaft. When we remove the front belhousing cover to inspect the front coupling we find that the flex plate has been pushed forward as the front coupling is still gripping in its original location.
Regards
Sean

 

Coriolus effect?

 

That is one loose bolt. I've had my rear TT shaft clamp bolt out 5 or 6 times doing TC cover bearings, checking the TT bearings, replacing the TT damper, tranny repair, etc. It's unmarked. And I've had the same bolt out on probably 25 other automatic 928 S4s in the last 15 years. None marked up like that. I agree - that one moved. It shouldn't have.

 

On automatic later models the thrust bearing goes on the rear face when this problem of clamp slippage manifests itself. I have a shell that demonstrates this and a GTS crank that had to be refaced such that were it to be mounted it would have an end float of .5mm and most on the rear face so either a custom thrust bearing would be needed or it is toast and that before the nitriding issue is considered. That some oils may fare better than others doubtless but no regular oil can stop this kind of thrust load from grinding out the bearing face ZDDP or not- it becomes an internal disc brake . Cause and effect- plain and simple.

On lubrication theory again a simple matter. Either you have enough film strength or you do not. If you do not metal to metal contact will occur and the bearing wil fail in no time. Oil pressure is required to get the oil to the lubrication site and it also controls the amount of oil flow through a given bearing. The more flow the more heat is removed assuming it is pressure fed and the oil is cooled to start with. If the oil temperature is too high the film strength will reduce and premature failure will occur. Yet again cause and effect.

The notion that synthetic oils have lower film strength often gets kicked around. This is just not correct. Synthetic lubes have superior properties like for like the only negative aspect is the price. Viscosity is the enemy of engines- viscosity causes drag and pulls down the efficiency of the engine so manufacturers of oil tout oils with lower viscosities but credible film strengths which improve gas mileage [slightly]. Our engines were not designed with such oil in mind and as we see regularly mentioned, performance of such oils is questionable on the tappets and quite possibly similar duties such as the thrust bearing. Thus I can relate what Bruce says about potential thrust bearing failure due to poor lubrication, just not in the case of the slippage of the clamp. Is there a link between types of oil and onset of this form of TBF failure- maybe but not much I would suggest.

Constantine carried out exhaustive mechanical tests on the stock clamp and his design which is way superior. Just look at the stock thing- it looks like it came out of a 1937 Triumph Tiger 100 the design is so crude. His design provides much more clamping force. A spot of Loctite has stopped my clamp from migrating for the last 13 years- no measurable increase of the thrust bearing movement in this time. The thrust bearing is a wear item but designed to last a very long time.

With any given oil higher viscosity= higher film strength, the trick is to get what you need at the minimum viscosity possible.

Regards

Fred