darth

Hi All

I purchased my 90C2 back in 2000 with 115,000km having a slight oil weep on #3 cylinder. Lately the engine developed an exhaust tone which I can best describe as a "ratty" sound only apparent on acceleration. I replaced each exhaust component one at a time leaving the heat exchanger on the driverside as the last component to be replaced and still the "ratty" sound persisted. I then did compression / leakdown tests before adjusting the valves and discovered cylinder #3 had 30% leakage (i.e. even after a valve adjust) but good compression compared to other cylinders. The other cylinders had leakage values below 5% and I was contemplating a top end rebuild. Since the leakage showed that neither the intake or exhaust valves were leaking I suspected a broken ring. I then decided to retorque the cylinder bolts (i.e. as a last resort) and discovered most of the bolts were loose and #3 cylinder bolts were less than 30 ft-lbs. I retorqued all the cylinder bolts up to 45 ft-lbs. Leakage on that bank of cylinders are all below 2% after the retorque and it has been a few weeks without the "ratty" exhaust tone as well as no more oil weep. I'm wondering how many rebuilds have been performed to address leaky cylinders when a simple retorque was necessary and maybe the leaky cylinder syndrome on this model year without the cylinder seals is due to this issue?

Bill
90C2

dfinnegan

That's a very interesting story! I think I've read a similar one here before, too.

One question, I thought that re-torquing the head bolts was a no-no. Anyone have any comments or input?

C4Russ

This is my understanding also. However I have to ask, if the heads were torqued properly in the first place, would they loosen? I'm asking if they loose because they weren't properly taken to their correct settings to begin with meaning they may not have stretched. This would imply they are still safe to use.

darth

Yes retorquing was my concern too. I've always read that they are torqued to a specific value then tighten another fraction of a turn to obtain the required stretch but then I read an ARP article which specified a torque range between 40 an 50 ft-lbs. I figured I had nothing to loose so with the leakage guage connected while I increased the torque in small increments - tightening each cylinder evenly so as not to warp any surface I watched as the leakage value dropped on each retorque step. I also had a concern of the cam being misaligned - time will tell. My theory for them coming loose is the 17 years of heat cycles and the pressures involved that they stretched and there was still enough stretch left in them to be able to retorque them to the specified value. I was also concerned with breaking a stud in the event that they were at their stretch limit. I think I still run the risk of breaking a stud due to the continued heat cycles and pressures. I'm wondering if it is possible to remove each stud one at a time and replace it with a new one without dismantling the top end? Although I have no records of the previous owners having had any work done on the top end, I can only assume that the factory was the only wrench on these bolts unless of course a mechanic along the way intentionally loosened bolts in order to secure future buisness but then that would be my paranoidal cynical side talking! I in no way recommend this procedure to others but I would definitely like to hear from others if they decide to do it as a last resort.
Who know this cylinder leak syndrome for this model year may be dispelled.

Hope this helps someone
Bill
90C2

greenjt

Very interesting. Specifically what process did you use to re-torque the studs?

darth

James

There are basic rules to follow when torquing any engine head or cylinder or for that matter any 2 mating surfaces. One wants to bring the 2 surfaces together as evenly as possible. In other words you don't want to avoid having one bolt tighten to say 45 ft-lbs and all the others still loose at say 30 ft-lbs. Tighten loose diangonal bolts slowly in small increments -say 2 ft-lbs one at a time until they are all the same then continue tightening diagonally in small increments. On engine heads that are all one piece like most North American cars the manual usually gives a bolt titening sequence. In the case of Porsche each cylinder has 4 bolts and I treated all 3 cylinders on one side as a single surface and at the same time each cylinder as well, titening the 4 bolts in a diagonal manner. Hope I explained this without adding too much confusion.

Bill
90C2

colo964

I'll give it a shot at my 210,000 mile service (about 2-3 weeks away!). Maybe some RTV snot, too.

darth

I'd avoid the RTV snot - I've read O2 sensors don't like the smell of silcone - kinda destroys them. Just rely on the mating surfaces to seal the leak - concentrate on bringing the 2 surfaces together evenly by increasing the torque on each bolt (i.e. in a diagonal pattern) in small increments of a couple foot-lbs at a time. It would also be good to have a leakage tester connected to the cylinder(s) your torquing at the time to watch the leakage decrease - that is if there is significant leakage to begin with. Torquing the bolts won't solve a leaky valve. You should be able to pick up a good leakage tester for under a $100.00. http://www.arp-bolts.com/Catalog/Cat...talog_0026.pdf ARP listed a range between 35 and 50 ft-lbs if I remember correctly and that was for there studs, I assumed Porsche's studs could handle the same torque but I stoped before I got to the maximum 50 ft-lbs because the leakage went down to 2% by the time I hit 45 ft-lbs and I was more than happy with that. While torquing the bolts I released the pressure in the cylinders being applied by the leakage test so that I wasn't fighting the pressure being applied to the cylinder, then I reconnected the pressure after each step. I have to admit I was shaking like a bunny on each tightening for fear of breaking a stud but I had only one other option at that point and that would have been a top end rebuild. I still run the risk of a stud breaking sometime in the future - only time will tell. I'd be interested in hearing how you make out as I'm sure others would too.

Good luck
Bill
90C2

MisterRisky

It looks like you may already have a leak down tester. If not, I highly recommend the DIY approach:

http://www.xs11.com/tips/misc/misc3.shtml

Because everyone knows you get extra Porsche Points if you build the tools yourself.

-JD

darth

The DIY link above does not mention the sympton I observed. I heard a hiss but it wasn't coming from the intake, exhaust, dipstick or adjacent spark plug hole but turned out to be at the base of the cylinder, the area oil was weeping from and causing the unusual "ratty"sounding exhaust tone on acceleration.

Bill
90C2

MisterRisky

Totally true Bill. The link was just to the instructions to build your own lead down tester as an alternative to buying one.

-JD

darth

I'm not knocking the DIY - I think it's an excellent article I just wanted to clarify that the procedure I'm suggesting has been successful addressing a leak at the base of the cylinder where the pressures are nowhere near those in the combustion chamber or at the top of the cylinder. This procedure may help with a leak at the top of the cylinder but it is a good idea to diagnose the problem before tightening cylinder bolts.

Bill
90C2

JasonAndreas

The original torque spec was:

1st Stage: 15Nm
2nd Stage: 1 x 90° ± 2°

but when they changed (08/24/93) the head gasket material from graphite to metal the spec went to:

1st Stage: 20Nm
2nd Stage 1 x 90° ± 2°