stock 911 turbo dilivar head stud torque?

[JFairman]

Hi, I have an '87 911 turbo that has stock smooth dilivar head studs in it.

The previous owner had the top end done around 40,000 miles ago and he replaced the bottom row dilivar studs according to the receipts I have and there is some oil leaking out between some of the heads and cylinders.

Today I adjusted the valves with the motor in the car and I would like to try retorqueing the head studs because of the oil leakage. It doesn't look easy but I could lower the back of the motor down around 7 inches on to a jack stand and remove the engine tins for better access.

I've heard so many things regarding doing this and I am undecided at the moment. I've also read and heard different torque values between 24 foot pounds and up to 36 ft lbs for the stock smooth finish dilivar head studs.

Is this a good or bad idea?

If I do it should I spray a little WD40 on the barrel nuts, wait a little while and then loosen them just a tiny bit to break the barrel nuts loose and then retorque them or just torque them without loosening them at all first and see if they turn and tighten a little?

I have Wayne's engine rebuilding book with the diagram showing the head nut torque sequence.

And most of all, how much maximum torque should 1987 stock dilivar head studs in a 930 motor running up to 1.1 bar boost on 93 octane be torqued too?

There are no sealing rings between the heads and cylinders on this 3.3 liter motor.

Thanks in advance for any helpful information on this

Jim

[Ronnie's.930]

I recently checked the torque on Raceware stud nuts and did not have to remove any engine tin or lower the engine (used a swivel head torque wrench and the appropriate length extension). I did this along with a valve lash check so I had the valve covers off and the wastgate moved aside.

Regarding retorqeing Dilivar studs - the general consensus of builders on the engine forum is to not do it as the risk for breaking them is high (you've probably read this yourself, however).

[timc]

Just my thoughts....I don't see anything wrong with setting the torque wrench to the correct spec just to verify that none are really loose, by listening/feeling for the "click" of the wrench. If the nuts are even close to the correct torque, wouldn't the torque required just to make the initial movement of the nut cause the wrench to "click" before the nut would move? I would definitely not move the nut past the "click".... used to do this on vw's and early 911's all the time.

[Speedy Squirrel]

Apply a fine film of Optimoly HT to the threads on nut and studs. Step 1)10 Nm, Step 2)32 Nm. Per the Porsche workshop manual.

Here is a brain teaser: Are you delivering an accurate torque to the nut when you use a swivel on the socket?

[Ronnie's.930]

Originally Posted by Speedy Squirrel

Apply a fine film of Optimoly HT to the threads on nut and studs. Step 1)10 Nm, Step 2)32 Nm. Per the Porsche workshop manual.

Here is a brain teaser: Are you delivering an accurate torque to the nut when you use a swivel on the socket?

The swivel is built-in to the end of my Snap-on Techwrench . . . they claim that the wrench is accurate, but I've just taken their word for it . . .

[Speedy Squirrel]

Originally Posted by Ronnie's.930

The swivel is built-in to the end of my Snap-on Techwrench . . . they claim that the wrench is accurate, but I've just taken their word for it . . .

You my man are golden. They told you right. That is a fine electronic torque wrench. Your only problem will be ... dead batteries!

[buster1015]

My 1987 turbo had a similar leakage issue. The advice above is exactly what I did, set wrench to spec and turn till it clicks. Only a few barrel nuts turned and only by a very small amount. Didn't actually solve the problem but I felt confident to continue to drive the car in the meantime.

One thing to keep in mind is that an accurate torque reading requires that the nut turns first before reaching the target torque reading. This is because the static coefficient of friction (nut not turning) is always higher than the kinetic coefficient of friction (nut in motion) and specs are based on the latter. So, your plan to loosen a bit, lubricate, and re-torque is spot on.

Since re-torquing the head studs did not solve my problem I ended up tearing down the motor and replacing them with Supertec studs which have a higher installed preload and are much stronger. Many people more experienced than myself have said that running 1 Bar or greater often exceeds the head clamping pressure achievable by the dilivar studs. The stud upgrade solved my problem.

Hopefully a re-retorque will be sufficient in your case.

[gsxrken]

Originally Posted by JFairman

If I do it should I spray a little WD40 on the barrel nuts, wait a little while and then loosen them just a tiny bit to break the barrel nuts loose and then retorque them or just torque them without loosening them at all first and see if they turn and tighten a little
Jim

Jim, the turbo shop manual says to use the Optimoly HT and tighten to 10 ftlbs, then 32 ftlbs.

Speaking strictly for myself, I would suspect that any stud that cracked just from backing off the nut was going to be a time-bomb anyway.

ON EDIT- I see that Mr. Squirrel answered a day before I did... At least I agreed with him

[JFairman]

Wouldn't you have to remove the barrel nut to put the optimoly stuff on it and then torque it back on? I don't think I want to remove one but maybe you could remove one at a time, do that and torque it back on.

Speedy Squirrel said final torque should be 32Nm and that converts to a hair less than 24 foot pounds.
You're saying it should be 32 ft lbs, and when I asked the same question over on the engine rebuilding forum someone said it should be 37 ft lbs.

A couple few years ago I asked Kevin Jeanette of Gunner Porsche Racing here in West Palm and he said 36 ft lbs. When I asked if that could break one he said it won't break.

Lastly Waynes engine rebuilding book says 24 ft lbs.

I appreciate everyone trying to answer and help but I'm just as confused as before now because of all the different answers. I wonder whats correct or best for a turbo?

It seems the only head stud torque spec everybody agrees on is 37-38 ft lbs for the ARP head studs that are shiny smooth with the rounded bullet shape tip on the end of the threads.

Not sure what I'm going to do now. Maybe it's best to just leave them where they are for now.
I have some more little projects to finish on it before I put the intercooler back on and drive it some more so there's still time to decide.

If anybody has a scan or a copy and paste from a factory manual that shows what the torque should be, please post it!

thanks,
Jim

[Ken911]

It's acceptable to remove them individually and re lube them and then retorque them. You cannot get the correct tension on the fastener or any fastener unless the threads have been lubricated per what ever spec is required for that fastener. I know it's a different application but when we torgue a manway we bring the bolts or studs and nuts or to a partucular torque sometimes several thousand foot pounds. then when all are at the final torque gap measurments are taken to ensure the cover is flat. then each one is taken back out one at a time and relubed and retorqued.

[gsxrken]

Sheesh- you're right, that's the nm spec. I'll just post a pic of the Shop Manual and shut up... ft lbs in ( )

[Ken911]

remember they are looking for a certain amount of stretch or tension on the fastener. When the spec states a particular lubricant the torque is only valid with that particular lubricant or one with the same coefficient of friction. Thats why for instance red loctite is the same friction coefficient as 90 weight oil.

[fredmeister]

The main advantage I see to using a higher torque for the studs is to create more compressive load in the cylinder heads to prevent lift off and scarring of the head/cylinder sealing surface.
The ARP are indeed 38ftlbs. I checked.
Assuming a similar Youngs modulus and stud dimensions between the dilavar and ARP then the stiffness of each stud should be similar. A 24ft lb torque for Dilavar should give less compression preload than at 38 ftlbs for the ARP.....provided the thread lube friction coefficients are similar and they are. This would make me wonder if the turbo engine should see a higher final torque as opposed to the NA engine in the factory manuals, but so far I have seen no distinction made between both engine types.
I am also assuming the BMEP on the turbo is much higher than the NA engine requiring the increased preload compression to prevent head lift off.
Just wondering now if ARP calculated all this out when they specked the torque to use. I would think so.

Fred

[JFairman]

Right on Fred!

Thats what I've been wondering in my simpler words and knowledge.

Maybe stock smooth dilivar headstud torque in a 930 motor should be higher than in a normally aspirated 911.
I have a feeling some mechanics with the experience feel that way.

[Ken911]

Probably should be higher. remember our stock engines and also the stock 3.6 turbos experiencd head lift off issues. that why porsche kept making changes ie. flame rings etc.

[gsxrken]

Fred, I doubt dilavar and ARP studs have the same elasticity. The whole point with dilavar as I understand it was that Porsche attempted to develop a stud that expanded at the same rate as the piston and head. This because the earlier engines were pulling studs out of mag cases and/or breaking.
But I agree with the rest of your point; namely that a dilavar at 24 lbs is less than an ARP at 38, but I bet the ARP is effectively clamping at something even harder than that as things heat up (whereas the dilavar would theoretically still be at 24 lbs as it expands linearly with the engine).
Thoughts?

[CaptainCalf]

Jim, don't shoot me, but have you considered the fact that many studs once stretched/torqued are done and aren't reusable? Not sure about the dilavars, but I know the ARPs can be re-used / re-torqued over and over w/o issue. I know it sounds crazy but if you're considering removing them one at a time why not go back with ARPs?

[Speedy Squirrel]

If the Dilavar that Porsche uses is the Dilavar that is commonly used in cryo fuel handling equipment, it is a 35% Nickel/Iron alloy. It is non-magnetic. I believe ARP uses a magnetic steel. I imagine they heat treat it to get high strength out of it.

Dilavar is a lower thermal expansion steel than normal stainless or most other steel alloys. The problem for the Porsche engine is that the head studs are looooong. If you use normal steel, the length increases too much when it heats up, and you loose the clamp load. The idea of Dilavar is that it will not grow so much when the engine heats up, so the clamp load is retained.

Modern steels can be heat treated. They are so strong that they can have a very high pre-load. They do grow when they heat up, but not so much, and in any event the pre-load is higher so it does not matter if some is lost.

I do not think either ARP or Dilavar studs are tightened enough to put them in the inelastic range, which would make them non-reuseable. I just think the corrosion problems with Dilavar are such that most people would not want to risk re-using them.

[fredmeister]

Originally Posted by gsxrken

Fred, I doubt dilavar and ARP studs have the same elasticity. The whole point with dilavar as I understand it was that Porsche attempted to develop a stud that expanded at the same rate as the piston and head. This because the earlier engines were pulling studs out of mag cases and/or breaking.
But I agree with the rest of your point; namely that a dilavar at 24 lbs is less than an ARP at 38, but I bet the ARP is effectively clamping at something even harder than that as things heat up (whereas the dilavar would theoretically still be at 24 lbs as it expands linearly with the engine).
Thoughts?

Disagree with you on one point. The Dilavar and ARP could have the same elasticity/stiffness provided E is the same for both....my assumption. I dont know the material properties of Dilavar, just assuming its high grade alloy steel like the ARP. Thermal coefficient has nothing to do with this value. Thats ok if they do. The thermal expansion rates are definitely different. The main reason Dilavar was used as you stated is to keep the 24ftlb preload and conversely the clampload the same as the engine heated up.
In my mind the ARP has an advantage in that clampload increases yet again as the engine heats up. Less likely to have head lift off when BMEP rises.
The stronger tensile strength of the ARP prevents early yielding and increases the fatigue life capacity as well.

Fred