Looking for writeup to replace head gaskets on s4
 

Hi, I am looking for a writeup on how to replace the head gaskets on a 928 s4 - preferably with the engine in the car.
Thanks
Dawid

Dawid,

I have not come across a definitive procedure pictorial like Dwayne's Magnum Opus for the timing belt /water pump job. There are videos on Youtube and various threads.

The desire to change out the heads without removing the motor is understandable and possible with the S4 motor and its head bolt configuration however the additional "buggeration" factor should be considered not to mention the words of those who have plenty of experience of such who invariably recommend pulling the motor [I AM NOT ONE OF THEM!].

When transplanting my 90S4 motor into my current 928 I was heavily involved in pulling the motor, refurbing the motor and installing it and all I can say is that you would do well to consider pulling the motor. This would be an excellent opportunity to do other jobs [as I did] like changing out the crank seals, upgrading the sump gasket and all the other maintenance work possibilities this opportunity presents.

Your choice at the end of the day

I would like to pull the engine but have a space issue. I have also already done the sump gasket, engine mounts and steering rack earlier this year - from underneath - so am very familiar with the buggeration factors.

I won't pretend to look forward to this, but hopefully taking my time and with help from everyone here I'll get this done.

The factory workshop manuals. Volume 1 is all about the engine.
http://www.ligeti.com/928/
Åke

The main thing is that you should make an informed decision aware of the pro's and cons of what you are proposing to undertake. I am sure the chaps on this list will give you all the help and encouragement you need as and when problems arise as invariably they will and do.

I have a feeling that 20 years is an appropriate time interval for head removal. The heads on my current 90S4 motor came off 17 years ago and were in perfect condition. Now I wonder what state they are currently in. Whereas I understand what is happening to the cylinder heads corrosion wise I am still trying to fathom out why the problem triggers. The popular belief that it is caused by lack of coolant change is invariably incorrect but needless to say if standard coolants are allowed to exhaust the chemical corrosion package the damage could be immense but when one sees severe damage to the heads and gaskets in the unsupported sections of the gasket head interface and yet no corrosion whatsoever in the water jacket and the head passages that for sure tells one that the coolant was not the problem, rather what happened to the coolant.

My guess is that you will find some corrosion damage to both the head[s] and the gasket[s]. The problem is the gap between the head and the gasket - this is not sealed and coolant seeps in and stagnates, the chemical corrosion package exhausts over time due to no flow and then the glycol breaks down to form organic acids mixed with water. Once this happens the passive film of the alloy breaks down and strange things start to happen the end result being a pitting corrosion type of attack. This phenomena [crevice corrosion] is not well known. The acids formed attack both the head and the gasket. Eventually they can nibble away at the fibrous gasket layer surrounding the fire rings. Once the problem reaches the fire rings they become vulnerable to galvanic corrosion and eventually the fire ring seal fails after corrosion preferentially nibbles away laterally at the cylinder head leading to fire ring seal failure.

If your head gasket has failed chances are this is what has happened - let's see what transpires.

To directly answer your question, yes, you can remove the heads with the engine in an S4 - I've done it in the past with no problems.

It is A LOT of extra work to pull the engine if you don't have a reason to do so.

Finally, I am not aware of any writeup, but overall it's not that difficult, just a lot of work.

It FEELS, like the hard part, is removing and reinstalling the cam gears/timing..everything is is simply bolts and washers...would that be correct?

And not counting R&R'ing the heads, an intake refresh is more a PITA than a head swap.

I've got a bad back...and I know exactly where it came from...pulling heads on 924 engines.
And those things were easy and quick to do.
However, I did many, many, back in the days.

Yeah, it is definitely possible to pull the heads in an S4, with the engine in the car.
I won't pull (or ask my guys to pull) a set of heads with the engine installed, in an S4.
Way to much bending over, for too many hours.
Sure, the job will be done and you will have saved a few hours, but a bad back is forever....

And at this point in time, 50% of the time, the engine will have to be removed for additional work, anyway.

I'm in the midst of putting my '88 motor back together. Here is a couple of mistakes you can learn from.

-Note the head numbers so you can note down which side is which. It is stamped by the exhaust ports. I wrote which was which on the heads but my marks were removed when the heads were serviced. So I guessed and got them backwards. I thought it would be ok, but the caps that hold the cams in place are machined to match the heads, and on my motor the races on the unused caps on the rear are degraded from the decay of the rubber plugs. So I cannot just swap all the caps with the heads.

-Do NOT skip the step of chasing the threads in the block. I did because I could not find a thread chaser for a reasonable price and did know I could make one. As I was doing the angle torque thing I was watching the actual torque value, and one of these things was not like the other. The torque value shot up much higher than the others as I tried to get the last 90 degrees. I think the threads jammed with gunk.

For these two reasons I am pulling the heads again and tossing a brand new set of head gaskets!
Other than that:
-Sunset has the gaskets for $71, which is a pretty good price. But make sure they have them in stock, because it is slow getting them from Germany.

-Get a good triple square tool.

-ArnnWorx has the cam alignment templates. Used to be 928MS, but not sure they exist any more. AW is cheaper anyhow. And they have a triple square tool, though maybe not the best.

Thanks for the tips, Karl.
What size triple square tool does the cam hats use? I saw in another thread that Greg suggest Hazet drives for quality, so I'll get that as well.
Also, can you remove the heads with the headers attached, or do they need to come of first?
Apparently 928 Motorsport has new owners, and they still advertise and list the cam alignment tool. I'll check out AnnWorx as well.

I did my valve covers and intake less than 500 km ago. Do not look forward to removing and replacing them again, but at least now I have experience.
If that is the hardest part of the job, it actually make me feel a bit better:)

My 87 S4 had hex bolts for the cam hats, so you probably won't need the triple square tool.
I've sent you a pm.

Cheers,
Gary.

guys - what is a triple square tool? About to do this job. Need tips and info. TIA

Not so much.
More special tricks than a New Orleans Whore.

Although Porsche says that the head bolts can be re-used, I'm pretty sure that they were talking about engine repairs during the warranty period....
(Otherwise, it would be dumb for Porsche to have massive amounts of these bolts in inventory, which they do.)
It's common to see these head bolts suffering from Hydrogen Embrittlement, which translates into: "Don't stretch as much as they originally did".
If all the bolts are still gold in color, you are generally in good shape. (Extremely rare at this point in time.)
If the bolts have become dark/some of the bolts have become dark, you need a new set.

It's called a "XZN" tool.
Do yourself a favor and buy a Hazet 8808-8. This is a tool that your grandkids kids kids will be able to still use on their electric car.
I've been using the same one for....30 years?
I can't wear it out....and because it doesn't deteriorate with use, it has a very low rate of "rounding out the fasteners."
And the "new version" has Titanium Nitrate on it, making it more durable!


I've also got 3 or 4 Snap-off versions, which must be replaced after every set of heads...
....along with a handful of fasteners.
(These are left over from 914 C/V joint bolt days...and you had to have 3-4 of them to finish some jobs. Sometimes, a non-thinking mechanic will grab one of these out of my box to pull the cam cap bolts on a 928 engine. It's almost comical to listen to the swearing.)

I reused head bolts before on my S4. End up with a steady oil leak from the passenger side head on the back where the oil feed passage from the block feeds the head. Took the engine out again, replaced both headgaskets again and got a new head bolt set. Leak went away. There was nothing wrong with the headgasket so I figured the bolt must have yielded and wasn't taking the proper amount of torque anymore, so took it apart and replaced all of them. Back then they were about 15-17 bucks I think (10 years ago). They are not going to get any cheaper so replace them.
I would personally take the engine out. The head is heavy and you don't want to gouge the surface going into the engine bay. Plus if I remember right when I took the head off the drivers side the brake booster was in the way of one of the head bolts. Also the shock towers will be in the way of getting a wrench down there to take the cam caps off and you don't want to damage those bolts, they were about 10 bucks each back then not sure what they cost now. Take the engine out get the Hazet socket Greg mentioned above to take the cam cap bolts off without damaging them (that's the one I used per his recommendation back then and still have it, don't buy the cheap ones at your auto parts store it cost me 3 to 4 cam cap bolts...), have to make sure the tool is fully seated in the bolt heads before you break them loose with a breaker bar, may want to clean the gunk out with Q-tips before you insert the tool.

High strength steels are indeed susceptible to Hydrogen Embrittlement - perhaps you might be kind enough to advise what you saw that led you to understand that such degradation had happened.

Is this it Greg?
https://www.toolsid.com/hazet/3-8-dr...pn-8808-8.html

1. I had a few used bolts snap off when tightened. This concerned me greatly, as in the late '70's Porsche changed the head studs on the 911's to unpained/coated Dilivar. These things broke very frequently. If you tried to re-use them in an overhaul, you could sometimes walk away from the engine after torquing the heads and hear the studs shear.
2. When tightening 928 heads with old bolts (and sometimes with studs), I noticed quite a bit of difference between the effort to angle torque different used bolts/studs.
3. So, I bought a torque wrench that measured both angles and foot pounds.
When set to an angle, once it reaches that angle, this tool automatically tells the user (as soon as you remove the pressure) how many ft.lbs. it took to achieve that angle.
Once I could see that data, it was pretty simple to logic out that there was something radically wrong. (50% variance was common.)
Buying new head bolts solved this issue....on most engines. Angles and the resulting torque is always with-in 10%....or there is something else wrong.
(It also became obvious why Porsche says to never allow the washers to rotate.)

I've never understood the need to prevent the washers rotating when the bolts are being angle torqued.

Torque to a specific ft lbs yes. Some of the torque applied to the bolt head gets "lost" due to friction in the threads and between underside of the bolt head and the washer or between the washer and cylinder head which presumably is different. And is why you shouldn't lubricate bolts that are specified as torque dry.

Angle torque by 90 degrees say and the bolt shank immediately under the bolt head rotates 90 degrees. This applies whether the washer rotates or not. I've never understood where the washer comes into it.

Interesting response- point 1 is precisely what I might have expected to hear.

If a used bolt experiences embrittlement I would expect the stress to rise with a steeper gradient on the stress strain curve- thus if one then rotates the bolt to the Porsche specified position it is in effect over rotated in a weakened condition so tensile failure would be no surprise at all.

Point 2 is the give away - I note that you did not comment on whether a "stock bolt" or the "degraded bolt" put up the most resistance.

Regarding point 3 it would be of interest to me if you can advise what torque numbers you typically saw with "degraded bolts" and those bolts that performed "as expected".

In reality there are two ways the bolts could be degraded- one is in the forming process and the other due to process exposure during service. Not long after I graduated there was a step change of thinking in the oil patch regarding the impact of hydrogen in high tensile steels. This came about after a major explosion and fire in a Texas oil refinery where the dished end of a large process vessel literally blew apart during normal service - as I recall this happened around 1982. After that event there was suddenly a lot more focus on material specification and quality control measures. The root cause of the problem was Hydrogen Induced Cracking [HIC]. Suddenly there was a lot more focus on known similar issues such as SSC [sulphide stress cracking], amine stress cracking & caustic embrittlement. If we had to modify piping that had been in service with exposure to hydrogen the metal had to be heated to over 200C for 30 minutes or so to expunge the hydrogen prior to working the metal- failure to do so would cause cracking in and around any weld areas and particularly so in the HAZ. Ultimately these problems led to a new ASTM specification to cover high tensile fasteners that was introduced [as I recall] around 1994- i.e. as the last 928's were being built. This led me to wonder whether the 928 head bolts are up to scratch - maybe bolt specs have been updated since then?

If there was a fundamental problem with what Porsche originally supplied then bolts would have been going twang left right and centre and as I am aware that is just not the case. As you may recall my motor is from my late 90S4. Back in 2005 after I lost my vehicle in a big smash, We refreshed the motor prior to installing it into the GTS chassis. There were no signs of corrosion on the heads or gaskets and everything in the crank/rods area looked fine so was not disturbed at all. After fitting two new inlet valves to replace the bent ones on No7 cylinder, it was time to mount the heads. At the time I asked the agents whether we should replace the head bolts nd they opined it was perfectly OK to re-use them. 17 years later they are still holding up so presumably a sound decision.

The above begs the question if you see such variance in torque values when trying to re-use bolts why is this? It cannot be because the original supply was defective so the degradation has to be because of some exposure but what? The only thing that I can think of might be connected to the problem I have opined about with respect to cylinder head corrosion. The evil brew that forms between the gasket and the heads that attacks both the heads and the gasket material could conceivably be liberating free hydrogen. The trouble with hydrogen is that the mol weight is 2 and the stuff can migrate through steel yet alone gasketed joints that are under attack. I thus suspect that to some random extent, if you see the heads being attacked there may well be a causal link between this phenomena and the bolts you see degraded. In your post I responded to what caught my eye was your comment about how bolts with a "changed appearance" are suspect. I have no idea what the bolts looked like when originally installed but your observation suggests that some form of corrosion, however superficially mild, may have taken place and if so that may well be the reason that such bolts will not perform as expected.

So- what do we conclude from this? Bolts can be reused if they retain their original properties. However if there is an intent to do so, then some form of tensile testing is required to verify such is feasible. If you have some performance data, maybe we can agree a suitable test value that should be attained if for example we undo the bolts and then before removing the heads we torque to spec and measure the resulting values. Given each head has 10 bolts and new bolts are now around $35 a pop or $700 a set there is quite some financial incentive. On the other hand one does not want to risk paying someone like yourself to pull the motor, rebuild it, reinstall it and then find out the thing is leaking and then the entire process has to be repeated as per Mr NamasGT's experience.

In my spare parts bin I have a set of GTS head bolts that covered about 100k km before the motor TBF'd. The original heads had suffered corrosion damage prior to the TBF event- they look black in outward appearance terms. I am now thinking they may be a waste of space!

Regarding the washers and rotation, the only logic I could figure was that the initial torque setting is based on the frictional resistance coefficient between the washer and the bolt. If the washer is sliding over the surface of the head, maybe the starting point for rotation in effect becomes corrupted and the end point is false. Trouble with torque values is they are inferential and can be way off depending on frictional resistance. The only fool proof way to torque a bolt is to mechanically stretch the bolt, and then lock the nut in that position. This is what happens on many critical joints these days but not possible with the fixed head bolts used in the majority of 928's.

A much more basic question: How do you stop the washers rotating when tightening?

To stop the washers from turning surface them with 400 grit paper placed on a flat surface


Note to chase the threads in the block use an old head bolt that has 3 flutes cut into it the first 3 threads are sufficient

using a thread chaser or tap could cut metal from the holes this will make the studs have a loose fit and could cause the threads to pull out

It is confusing, to me, also.
You'd logic that when angle torquing, 90 degrees is 90 degrees, regardless of what the washer does.
However, experience tells me that when the washers do slip, the torque value drops...significantly.
And a reduction in torque value has to mean (to me) a reduction in stretch.

But beyond what I personally understand:
When Porsche specifically makes a notation about a certain procedure and writes a caution to not allow something to occur....
I figure they must know more about the subject than I do.

Fred:

Here's a picture of the head bolts out of one of the GT engines which Kyle took apart, recently.
I think this picture might tell you more than I could ever tell....all of those bolts started out that nice golden color.
Besides the obvious color change, note the rust stain on several of the bolts, where coolant was obviously getting past the silicone bead, on the head gasket.

I've got a box of bolts like this, which must weigh a couple of hundred pounds.
I saved them for when the inevitable occurs and Porsche no longer has new bolts and I am forced to "select/find" decent used bolts.
If you figure out a way to recondition them, I'm all ears.
https://cimg1.ibsrv.net/gimg/rennlis...20fe6576ce.jpg
Head bolts from '91 GT engine with terrible head gaskets and large amounts of aluminum deterioration.
https://cimg2.ibsrv.net/gimg/rennlis...a73e2ad076.jpg
Head bolts from '91 GT engine with terribly deteriorated head gaskets...note the rust stain on the bolt just right of center.

Greg,

Thanks for posting- very illuminating- to me at least!

The message to the OP is very clear- use new bolts unless those removed look like new- most unlikely!

Rather than clutter this thread I will start a new thread covering cylinder heads and corrosion as I perceive "interesting scope" for further discussion/learning.

Update: Heads are off! :cheers: Took a bit longer than I thought, but they are off and we can see what I am dealing with. Thanks for all the hints and tips so far, I wouldn't have managed without it.

Some photos:
Left hand side head.
https://cimg6.ibsrv.net/gimg/rennlis...f16be4ae34.jpg
I note that on three combustion chambers the intake valves are clean, and on the forth they are black, with carbon on them. Any ideas why?

Right hand side head
https://cimg7.ibsrv.net/gimg/rennlis...3c48befb78.jpg
Here all the inlet valves are clean.

Individual combustion chambers - with the remains of the head gasket in place.
#1

https://cimg9.ibsrv.net/gimg/rennlis...33aa6147ae.jpg
#2
https://cimg1.ibsrv.net/gimg/rennlis...a760a7b06c.jpg
#3
https://cimg4.ibsrv.net/gimg/rennlis...a4deb66b46.jpg
#4
https://cimg6.ibsrv.net/gimg/rennlis...ac0f915ca7.jpg
#5
https://cimg8.ibsrv.net/gimg/rennlis...3d9fddd98d.jpg
#6
https://cimg0.ibsrv.net/gimg/rennlis...32493b62c9.jpg
#7
https://cimg2.ibsrv.net/gimg/rennlis...957e710108.jpg
#8
https://cimg4.ibsrv.net/gimg/rennlis...42ba3db926.jpg

Block Left hand bank

https://cimg6.ibsrv.net/gimg/rennlis...1253fe1a7b.jpg

Individual bores / pistons
#1
https://cimg8.ibsrv.net/gimg/rennlis...7f49478d4f.jpg
#2

#3
https://cimg2.ibsrv.net/gimg/rennlis...c745fc239b.jpg
#4
https://cimg4.ibsrv.net/gimg/rennlis...1165c7ded3.jpg

Right hand bank

https://cimg6.ibsrv.net/gimg/rennlis...ae9cc27c01.jpg

Individual bore / pistons

#5 - With pieces of the gasket that stayed behind.

https://cimg8.ibsrv.net/gimg/rennlis...82cc6da177.jpg
#6
https://cimg0.ibsrv.net/gimg/rennlis...3ff699f478.jpg
#7
https://cimg0.ibsrv.net/gimg/rennlis...b48ae831f0.jpg
#8
https://cimg4.ibsrv.net/gimg/rennlis...29a8f736c4.jpg

Gaskets- or what is left of them
https://cimg1.ibsrv.net/gimg/rennlis...2b4e4228ba.jpg
#1
https://cimg3.ibsrv.net/gimg/rennlis...e5d60225d6.jpg
#2
https://cimg4.ibsrv.net/gimg/rennlis...f4507fcca1.jpg
#3
https://cimg6.ibsrv.net/gimg/rennlis...2e89c18208.jpg
#4
https://cimg7.ibsrv.net/gimg/rennlis...393fa95182.jpg
#5
https://cimg9.ibsrv.net/gimg/rennlis...ca72e5a811.jpg
#6
https://cimg1.ibsrv.net/gimg/rennlis...0a4027f493.jpg
#7
https://cimg2.ibsrv.net/gimg/rennlis...f95e9c4965.jpg
#8
https://cimg3.ibsrv.net/gimg/rennlis...5fe9a750ad.jpg

As I originally started this job because my cooling system got pressurized and the coolant escaped via the overflow in the reservoir - does these look like combustion chambers / pistons that was exposed to coolant while under pressure?

It is hard to tell with the gaskets in the state they are in, but it doesn't look as if there was blow by across the round metal rings.

On another note, I caught worn cam gears and a leaking tensioner. So that is good.

Next step is to clean everything.

Any hints on cleaning the faces on the heads?

Thanks

^^ yeah, have a machine shop deck them lightly and while they're at it replace the valve guides and seals.

The head bolts have definitely changed color! And they made a hell of a crack coming loose. They will not be going back in.

https://cimg8.ibsrv.net/gimg/rennlis...9bbd69c3f8.jpg

Dawid,

First of all many congratulations on successful removal of the heads- well done! Engine out or Ruptured Russian technique?

Now we need to find a photography course/tutor near you!

From what I can see of your photos you have dodged a bullet in the nick of time pretty much as expected. Please post some photos of the heads without gaskets and preferably in focus and with correct exposure. It is clear to me that your gaskets have been eaten away in typical fashion and now we need to see whether the heads have taken much in the way of attrition. The block side looks to be in excellent condition to me but I will defer that one to our more knowledgeable friends as they wake up. As of your post I cannot see any evidence of a head gasket leak as yet but the photos really need more clarity to make a good interpretation.

Your comment about the bolts was also interesting. The crack you heard was most probably and quite literally a crack taking place. In his earlier post Greg commented about hydrogen embrittlement and asked if there was a remedy. The simple answer is both yes and no. Assuming the bolts have been embrittled by hydrogen from the corrosion process that was taking place, were it possible to get the bolts out without torquing them then cooking them at 200C for half an hour or so would get the hydrogen out and reverse the process. However, in the process of releasing them the embrittled condition would leave them in such a state that in the process of applying breakaway torque the bolts will [I suspect] have quite literally "cracked" and probably caused the "cracking noise" you heard.

It's an audible crack, not a literal one. You load up on a 40" breaker bar to loosen a head bolt, and the loaded up force suddenly overcomes the resistance to turning and the bolt rotates quickly, making a cracking noise that you can feel through the bar. Not sure whether it's

A. the dried crud sitting around the top of the threaded portion that's fracturing apart,
B. the bolt is twisted from the force of undoing it and as the threaded portion turns suddenly it makes a grating noise, or
C. you're hearing whatever corrosion there is between the bolt and the block suddenly let go and allow the bolt to rotate.

Or D, none of the above. Dunno.

The top of the bores on some of the cylinders appear to have corrosion damage (like #4 & #8). Have you cleaned it up yet to see?

Since '87 and '88 heads are prone to cracking (this seems to occur more frequently on the '88 models), make sure that your machine shop pressure tests the heads before you have much work done.
(Have them strip the heads, clean the heads, and then pressure test them.)

Worth mentioning, since the cracking occurs right at the edge of the 90 degree counter cut where the head washers sit, these heads can also crack when they are re-installed and tightened.
(Cracking during re-installation is more common when people not familiar with these engines fail to replace the head bolts and the resulting pressure on the head is higher due to the old bolts not stretching properly.)

One thing at a time, Fred. Maybe after I learned to strip and replace heads:evilgrin:

A short video of the head bolts "cracking" as they come loose.

Thanks 545svk for this thread.
As my car (like may others) will probably need this someday, it gives me some confidence to do it myself (with some help from friends too of course!).
Keep posting.
Keep up the good work! :thumbup:

I've never met a New Orleans Whore, but I've met doing heads on an S4. That would be a number of tricks right there Greg.

Sounds the same as when I loosened my head bolts.

Cheers,
Gary