Head gaskets destroyed but why? 74 2.7 911.

moneymanager · 10-13-2007, 10:46 AM

I've asked one or two of you these questions privately, but thought I'd seek the wisdom of everyone.
Motor is 74 2.7. JE Pistons at 9.5 CR; Elgin modified S cams, ported, case line bored, shuffle pinned, case savers, generally blueprinted by Comp Engineering, PMO's, assembled by my long time very competent local guys. Timing was set at 6k rpm to specs agreed to by various gurus for the CR and set up I had. Car ran perfectly for 5k miles. At which point oil cooler blew. While in the area, my mechanic noticed a combustion leak at the #1 cylinder. Heads pulled. One of the tiny little spring-like head gaskets had a hole burned through it. In addition, the JE pistons appeared to have been just tapping the head at the edge...maybe a thousandth or two of non-clearance. The other head showed perfect gaskets but a similar slight piston/head contact at the piston edge. (For the record, the piston/valve clearance was checked but no one considered that there would be any other clearance issue.)
Shop debated whether to trim the pistons a tad at the edge or the combustion chamber in the head to deal with the clearance issue. Opted to do the combustion chamber, where I think about 0.020 was removed. Blown gasket put down to bad luck/poor manufacturing quality control. Motor reassembled.
I drove the car a few miles on the street, and at Laguna two months ago. There a tire wore thru an oil line and left me with a big oil leak on track. I drove off gently with oil pressure never dropping below 30 lbs, and had car towed to shop for remedial action. Mechanic made temporary repair and said "heads are loose, have them torqued."
I conveniently forgot that piece of advice, replumbed the oil line to get it out of harms way, and did Willow Springs a couple of weeks ago. Car felt, sounded fine to my untrained ear. I took it to shop for other work a week ago. My mechanic said I needed to torque the heads.
Heads were torqued; car still made same noise. Mechanic said we need to pull the heads off and find out what's going on. Early inspection showed four cylinders leaking. All four gaskets were now blown thru, with enough damage to one cylinder caused I think by gasket fragments to require replacement. The heads are now with Ollies in So Cal for repair and diagnosis. My local shop with 30 years of experience among the key people has never seen a problem quite like this on a 2.7 with the sort of rebuild it has gone through. Detonation was the first thought, though the conservative CR and timing gave us no reason to suspect it would be an issue. We're also wondering if the very slight head milling somehow removed enough material to create excessive temperatures on the gaskets and cause their failure? Any insight gratefully received.

dtw · 10-13-2007, 12:18 PM

Is 9.5:1 workable on Cali pump gas? I thought that was a no-no but not sure. Did you ever check CHTs or put a Gunson on the car?

After the abuse the head studs have possibly taken, I wonder if checking over the case & case savers would be a good call right now during the downtime.

I'm running milled heads on a hot 2.4 with no head gasket issues over several thousand miles.

moneymanager · 10-13-2007, 12:33 PM

9.5 is the ragged edge for CA gas per most of the experts. WHen I run at the track I toss in a few gallons of race fuel just for insurance.
I don't know if what kind of tuning equipment was on the car in the earliest days,none since. The PMO's came already set up and worked about perfectly when first fired up. No head temps taken ever to my knowledge... but car runs cool. Hit 250 once at track and I immediately slowed down to cool things off. Have since finally hooked up oil cooler which keeps oil temps under 220 no matter what.
I should add to my earlier description that we see no signs of detonation..sand blasting of piston top, bent plug electrodes or any other evidence of less than perfect plug performance, piston scuffing etc. Thanks for your interest.

MBEngineering · 10-15-2007, 01:54 AM

HI JH
What head studs are you using and what make/type of torque wrench are you using??, a photo of the cylinder/s and heads could be useful.

rsgards mike

moneymanager · 10-15-2007, 08:28 PM

Thanks Mike. I should have thought of a picture; unfortunately the heads are already on their way for analysis and rehab. The studs were current factory Porsche issue which I understand have been a satisfactory product. THe torque wrench question is interesting. I believe it was a Snap On, but will inquire further. Doesn't mean it couldn't have been off though.

MBEngineering · 05:16 AM

HI JH
this 3.0lt engine I had in for repair, and I found out it was built with a brand new Snap-on torque wrench, there latest digital read out one, and this happened after App' 150 ish miles,the head nuts came loose,

you need a torque wrench like this and you can then keep a check on the bolt/stud stretch, did your heads/cylinders look like this??.This was on a ally 3.0lt SC case and the late fully threaded 993 Porsche studs, the studs where tight and not pulled in the case

just check your cases in where the head studs are and the cylinders fit , as I have had the case split, and go barrel shaped , because to much lock-tight is in the bottom of the stud hole, and the pressure of the stud in to the hole has hydraulic-ed, the lock-tight has then expanded and burst the case at the side of the cylinder in the case spigot/hole for the cylinder, then the stud has come loose?

regards mike

moneymanager · 10-16-2007, 10:52 AM

Mike,
Many thanks for making the effort to post this. Very much appreciated. I have passed it all along to my shop for them to look at. No matter what they decide about my current problem, I'm going to check head torques frequently once we get this thing back on the road to be sure THIS doesn't happen! I'll pass on any comments I get from my guy or the shop which is redoing the heads.

snowman · 11:09 PM

Ollie's will not do you wrong. They are one of the best.

My humble opinion is that the head studs couldn't do the job and or you had to little clearance form the piston to head. Torque is not likely the answer. The head studs are very long and the torque will be fairly accurate, but remember torque can be off by 50%, even with the best wrench. You need to compensate by having even better studs than needed. When doing this you need case savers, because the margin in the torque may remove the studs from the case. I think detonation is unlikely, but possible. More likely is overheating. Did you have a head temp gauge? If not you need one. Extra oil cooling on this oil cooled engine is a must.

How to find the answer. If the head studs were loose you will see evidence of hammering of the cylinders on the heads. If detonation is responsible you will see it on the rod bearings, the pistons my have chunks out of them or brunt spots, the spark plug ceramic will be shattered or have pieces missing. Overheating, look for burnt oil deposits. Also look for exactly what you have, fried cylinder seals, loose head studs, piston contact with head. In any case, do what Ollies says to do.

J P Stein · 06:33 AM

Why do head studs nuts lose their torque? If/when they do, a blown head gasket is certainly a possiblity....likely even.

At risk of starting a war, here are my views.
Steel & aluminum expand at different rates....this may be the last point we all agree on. As the engine heats up, the aluminum parts attempt to expand at their acustom rate, the steel does the same. Since the aluminum is more malable it surrenders to the steel by deforming (let's ignore the case threads for a bit). We have several areas to spread this deformation over....2 surfaces at the case/cylinder juncture, 2 at the cylinder/head, one at the head/nut washer, so this may not be noticable. There is also a certain amount of "give" in the case threads. Aluminum also looses strength as it is heated which could exaserbate the deformation. Throw in a magnesium case to the equation and things don't get any better.

It is my contention that this deformation can become permanent. At some point during heating & cooling cycles, the aluminum components could become "cold worked" ....actually increase in sprength in the cold worked areas and the whole deformation process ceases as a sort of equilibrium is reached.

As a practical matter, if you're gonna use steel studs, keep your torque wrench handy & keep your engine from getting hot.

When tearing down my 2.7L after 7 years of service, the once torqued at assembly head nuts were tight (as were all joints). I used 24 of the coated/full thread Divilar studs. I also kept it cool....maybe too much so.

I'm home this morning with a sick wife and nothing to do. The above opinion is a result.

moneymanager · 10-17-2007, 01:51 PM

Thanks to all who have contributed.
Here's the verdict from Ollies... the time certs are pulling out of the case. Apparently there are small time certs and bigger ones...maybe 14mm od vice 12mm. They are now seeing the smaller units, favored to avoid other clearance issues, failing to the tune of several a month. The answer? Tear the case apart and change out the small ones. Nothing a few thousand $ won't fix.
To me it is strange that the 2.7 community has been using time certs successfully for years without a problem. And now, all of a sudden, an epidemic apparently. Maybe magnesium just dies after 50,000 heat cycles.

J P Stein · 10-17-2007, 02:34 PM

Many of us in the 2.7L community use case savers rather than time certs. Again, heat is the mag case killer.

OK, I may have gone overboard on keeping the thing cool but the Aux thermostat does eventually open at approx 185 F and feeds the aux cooler. I've seen over 100C (212F)on the gage rarely.

moneymanager · 10-17-2007, 06:45 PM

Not knowing the difference, I used the phrase "time certs." They were case savers, installed by the primo shop in the country on these issues... Competition Engineering. If you are building a 2.7, take a close look at this issue. I did everything every expert reports is required to build a bulletproof 2.7. It didn't work.

snowman · 10-17-2007, 06:52 PM

If it gets to hot, no method will work, it will fail.

dtw · 10-18-2007, 04:59 AM

moneymanager, I'm really sorry to hear this, truly. I know how frustrating it is to have an engine running, torn-down; running, torn-down. A real pain in the ass and a drain on the wallet.

On the flip side, I am sending my 2.7 case out to EBS this week for case-savers, among other things. Your experience may have helped me, and others, dodge this bullet and for that I thank you.

moneymanager · 10-18-2007, 06:48 AM

Thanks for the thought Dave. I hope my experience benefits someone. Ask a lot of questions about this...between Bruce Anderson's opinion, the experience of many others, and Competition Engineering's expertise and reputation, I thought I was home free. Apparently not so. With a firm cost estimate in hand, I'm going to consider whether it makes sense to junk this case (line bored, shuffle-pinned, boat-tailed as it is), part out the rest, and acquire a decent aluminum three liter.

Eagledriver · 10-18-2007, 07:46 PM

This brings us back to Dilavar studs. I think the only usable configuration for a 2.7 that will be worked hard is case-savers, and Dilavar studs. Yes I know they tend to break but the epoxy coated seem to be better and even if they do break it's usually after many years of service. I have only built one 2.7 and this was the configuration I used.

-Andy

moneymanager · 10-20-2007, 06:34 AM

Well we are back to square one. The heads went to Ollies for refurbishing. George there (who wasn't looking at the case) told us our problem was most probably loose head studs...probably due to use of smaller time certs rather than bigger and more secure case savers. So we pulled a stud and, lo and behold, we have case savers! The original work by Comp Engineering was exactly as it should have been. Not only that, but the stud was really solid in the case. Having ruled out loose studs and detonation, we are now faced with three possibilities: 1) the deck is uneven (unlikely as this was redone with the original motor rebuild 5k miles ago,) 2) the cylinders are different lengths (causing unevenness and hence instability,) or 3) the heads were loose due to incorrect torque when the motor was assembled or they loosened later due to heat. I'll post further reports next week.

MBEngineering · 10-21-2007, 05:33 AM

HI moneymanager,

"1) the deck is uneven (unlikely as this was redone with the original motor rebuild 5k miles ago,) 2) the cylinders are different lengths (causing unevenness and hence instability,) or 3) the heads were loose due to incorrect torque when the motor was assembled or they loosened later due to heat."

1&2 if you have any uneven deck or cylinder lengths when the engine was built , the cam's would bind in the cambox, if that checks out OK then you are only left with No'3 torque or heat, as the headstuds/casesavers are OK ?

regards mike

moneymanager · 10-21-2007, 09:35 AM

Thanks Mike. That's about where we come out. Checking deck and cylinder lengths just to be sure, but I think we're back where you suggested...a torque wrench which was badly out of whack. I'll report further details when available. Jim

snowman · 08:09 PM

Beware of torque readings. The stretch they produce in the head studs can be very uneven and the ammount uncertain by over 50%! and thats with a GOOD torque wrench. This uncertainty must be taken into account somehow if you are to be certain of how much clamping force is actually being utilized with your studs. Its a very difficult problem. The solution is to somehow measure the actual stretch of the studs. Nothing else will work without huge safety factors, factors you probably do not have. I suggest that you set up a dial indicator on one of the studs, measure the length without torque, then with. You will have to talk to someone who knows studs much better than I do to know how much stretch they need. I do know its over 0.006". Once measured on one stud, and using the corresponding torque, you can get much closer the the right clamping force than you are now. Stretch is the ONLY way to be certain your studs are tight.