Wormhole

I'm surprised they only recommend 65lbs on the MLS gasket. The 90, 90 method with the stock gasket came out to about 80lbs if I recall, and I usually did a final torque of 90lbs. Can you over torque a metal gasket?

billindenver

Hmmm. I didn't know the final torque of the 90 90 method came out to 80 ish. That might be an indicator for why the ARP's are loosening up at the 65lbs that EBS is recommending. The headgasket could be left without enough squish (highly technical term)...which would leave room for it to compress further slowly. I'm just going to crank it down to 85lbs and see where it is after the track weekend. The headgasket was just sitting around my garage anyway...so isn't costing me anything to find out. If this one fails though, I think I'll give MLS a shot.

Will report in on the findings of course. Thanks for your thoughts all...has helped me semi-think my way into this.

Porschefile

Yep, you sure can. I won't name names but, a less well known retailer/builder of 944 parts once told me that with aftermarket head studs that you should apply anywhere from ~145-185lb/ft torquing the head down!!! Lol, what a crock of sh*t. Either they have somehow managed to stay in business despite being morons, or they were attempting to apply enough pressure to friction WELD the head to the block! Typically no more than 90lb/ft should ever really be needed for anything. Hell, I would think anything close to 145lb/ft would snap something.

Bill, forgive me if I understood your post wrong. Stock type "stretch" studs or bolts do not maintain clamping force as consistantly as a good aftermarket stud like ARP's or Raceware. This is the primary reason it's a good idea to upgrade studs/bolts to a non-stretch type in certain cases (high hp, high rpm, etc), as more consistant torque can be maintained so the head will be less likely to lift under high load circumstances. The only issue anyone ever has is that some MLS gaskets are of sub-par quality to others (Cometic definitely isn't one of the best MLS gasket makers out there), and on top of that there is next to no room for error with an MLS gasket. All contact surfaces have to be pretty much perfectly flat and proper torque/torque sequences need to be carried out. 9 times out of 10, when people have problems with MLS gaskets leaking it's because they didn't install them properly or they are of crap quality.

Despite the large number of misconceptions about subjects such as this, having any type of "play" or stretch in the studs or headgasket is never a good thing. You don't want any type of play whatsoever as the less structurally stiff these components are, the more likely you are to get head lift. The whole "fuse" theory holds no water and is an urban myth. That being said, sometimes you can get away with a lot on a stock headgasket, and from what I've seen the widefire is a pretty decent design as far as stock headgaskets go but it's still a fiber gasket and has all of the inherent downsides to a fiber gasket. Personally I'd go with an MLS and ARP's done right the first time and never have to worry about it again. Just make sure to deck the head AND the block as sometimes people are lazy(like me! ), don't want to pull the motor, and just deck the head which is almost certain to cause leaks.

Oh Bill, one more thing. Just FYI but, in a wide variety of other cars I've seen/used ARP's in, it seems like usually the recommended torque is always around 80-90lb/ft for a good, consistent seal. 65lb/ft definitely sounds a bit too low.

2bridges

used ARP on ten non-porsche engines I have built. Never an issue. My machinist has built literaly hundreds of race motors with arp's - no failures to date. One possible rod bolt failure - however run well beyond intended rpm with axle break and no rev liimiter

hate to say it, but sounds like operator error.

billindenver

Or instructional error. Pretty difficult for two of us with freshly milled heads and blocks torquing to spec's supplied by EBS to get it wrong at the same time. I'm thinking the torque spec is to light...we shall see though, as I just torqued to 80.

2bridges

nod nod

Tom M'Guinn

By the way, did you use oil on the threads and did you torque and untorque a few times -- to help ensure an accurate torque reading?

billindenver

Used the Moly that comes with ARP's, and torqued down in stages.

Tom M'Guinn

Did you use the same torque wrench? I had my old Craftsman calibrated a few years ago and was shocked how far off it was! It was under-torquing by a significant margin.

2bridges

excellent point - a cheap *** torque wrench is worse than nothing at all for this job

KuHL 951

I'm glad this issue has come up although I hate to benefit from the problems of others. I'm going to try the overnight hold at 80. Back off 180 deg. in the morning, re-torque to 80, sit for a day, repeat and finish at 80-85 depending what else I can glean from this thread and other sources. This will be with a new Wide Fire so I'm trying to get it to compress as best I can. Next time I have the cam tower off I'll re-torque if I see any leakage. I'm going to borrow a friends Snap-On, I don't trust my Craftsman although I do have an old beam type but it's so hard to read while pulling that hard. I hear the beam types don't get out of calibration like the clickers, is that true?

Tom M'Guinn

Are you using Raceware and/or ARP. If not, I'd stay closer to stock torque.

As for beam styles, raceware says the same thing -- that they are more accurate than click styles and don't loose calibration.

billindenver

Not a bad point, my torque wrench is a couple years old and I suppose it could be out of calibration as it is a clicker. I'll check the nuts with my backup beam type in the morning...just to be sure.

KuHL 951

I'm using ARP. EBS Racing says the final pull is 65#. From what I'm seeing here and from other comments that sounds too low for a turbo head. If the ARP yields past 100#, I think 80-85 should be fine. I doubt seriously that staged torquing can cause any deformation in the outer block causing the bores to ride lower than the center of the head. Most of the leaks discussed here have been coolant loss outside the cylinder and combustion chamber. That makes me think that the clamping force is just not adequate at the recommended 65# for the perimeter of the head, not the center. Then again I could be way off base as usual.

Porschefile

Just FYI but Mk3 Supras ('86.5-92) factory torque specs are 65#, and they blow headgaskets all day long. ~80# is a much more solid number.