[Reiver]

Chris_Seven's comments in the build thread on studs...

I have contributed to many of these discussions over the years and will make an attempt to summarise the issues that need to be considered in order to make a sound judgement.

The first issue is that it is never wise to consider a stud in splendid isolation - the joint design is really the core issue and the stud should be designed/selected to suit the clamping force needed to maintain the integrity of the joint for long periods.

If we consider the cylinder head joint and its interaction with the studs it is easy to realise that the stud generally behaves as a spring and in order to resist peak cylinder forces must be capable of developing a sufficiently high preload.

It is also clear that the stud is much more elastic then the cylinder head and barrel.

The quality and 'fit' of the thread of the stud and associated nut/washer will affect the correlation between torque and preload and the characteristics of these components and the manner in which threads are lubricated all have an impact on the integrity of the cylinder head joint.

The design of the 911 cylinder and head is, however, relatively forgiving compared to heads on water cooled engines.

Once the head is torqued down the ability of the stud/nut to maintain preload preferably without the need to re-torque is also quite important.

The preload requirement of the stud is relatively straightforward to determine and clearly all of the studs available for use with a 911 engine will meet this basic need. The basic tensile strength of all of the studs available is well within the design requirements of the 911 engine so buying ultra high strength studs such as H11 Tool Steels (260ksi) is of no merit.

One we run the engine and expansion occurs the situation becomes more complex.

The use on Nikasil cylinders obviously increased the expansion that occurs as the engine warms up.

The difference in the coefficient of expansion (CoE) between the materials used in the components of the joint and the studs has a direct influence on the force applied to the thread in the engine case.

In very simple terms the greater the difference the greater the increase in the applied force.

As most 'ferritic' steels have very similar CoE then the choice of stud using these types of material has little influence on reliability.

High Strength steels, Martensitic Stainless Steels and even Titanium Alloys are all very similar in this respect and if CoE is the only consideration there is little to chose.

The Nimonic family of alloys does have a slightly increased CoE so will reduce force due to expansion by around 5% but this is not very significant.

Dilavar, which is clearly, an Austenitic Stainless Steel, has a CoE which is 50% greater than ferritic alloys and this does have a significant impact on the 'pull out' forces produced by expansion of the joint.

There are, however, two other characteristics that need to be considered.

The first is Young's Modulus (E) of the stud material. The E value of the stud material will affect the force applied to the screw threads as the 'stiffer' the material the greater the force that will be required.

With the exception of Titanium all of the materials used for studs will have an E value that will not vary by much more than 2% so we can ignore this effect.

The modulus of an Alpha-Beta Titanium Alloy is about 50% lower than Steel so force increase due to expansion will be greatly reduced by the use of this material.

The next consideration is a diameter of the shank of the stud. The smaller the diameter of the shank the lower the force produced by expansion.

In high temperature design it is custom and practice to reduce the shank of a bolt to attempt to control these forces.

The standard steel head stud has a shank diameter of approximately 7.6mm.

Some of the aftermarket studs have diameters of up to 9.5mm and the increase in pull out force will be directly proportional to the difference in the square of these diameters.

I would always try to select a stud with a relatively low increase in pull out force as this will give the thread in the case the best chances of long term survival although SC cases do seem to be totally immune to this type of problem.

There is one other engineering detail that should be considered and that is the use of steel studs mixed with Dilavar studs on a single head joint.

I find the concept of using materials with a significantly different CoE on either side of a single joint to be 'bizarre'.

To suggest that there is a 50% difference in the temperature of the studs on either side of the head so it balances the forces seems a bit optimistic and is something I would consider risky at best.

To justify this on the basis that if Porsche used this approach it must be correct and then to say Dilavar is a poor choice of material is a bit contradictory. I would never advise mixing materials in this manner as the uneven load distribution will inevitably cause must result in problems.

There are also some other metallurgical considerations that should be considered.

It is quite clear that Dilavar suffers from chloride induced Stress Corrosion Cracking (SCC) and for engines used on roads that are heavily salted is not sensible.

I am sure that this is the reason that later Dilavar studs are resin coated.

I would always select studs manufactured from a material that is not sensitive to SCC.

For example ARP advise against the use of L19 in this type of environment. H11 Tool Steel is also poor in this respect.

Inconel 718 does not suffer from this type of problem and is commonly used for the manufacture of high strength fasteners in marine environments.

Some of the Grade 5 Titanium Alloys can suffer from SCC in salt water environments but there is an ELI version that eliminates this specific issue and is used where there are concerns.

Precipitation Hardened Martensitic Stainless Steels can also exhibit sensitivity to SCC in a H960 and H1150 tempers but in HH150 form would not be a cause for concern.

There was some evidence that Dilavar also suffered from Hydrogen Embrittlement (HE) but I find this unlikely due to the high Molybdenum content of this material but there is a rumour that a large number of spurious Dilavar studs did get into the supply chain and this could offer an explanation for this observation.

My overall conclusion is that for SC and older sand cast cases the standard Steel stud is quite fit for purpose and provides a good solution at a good price.

For magnesium cases I would consider using Dilavar as I would like to use studs with a reduced pull out force due to expansion.

We have recently been manufacturing Titanium Studs for this application using a 6AL4VELI alloy which has been aged to 170ksi tensile strength. We have fitted around 10 engines with these studs but so far only have around 8 months of use. They are much more cost effective than Dilavar.

There was also a suggestion made above concerning 'creep' of head studs and I would like to make a comment.

It is vey unlikely that there is a sufficiently high temperature present to activate any of the damaging creep mechanisms that could occur during the life of an engine - say 5000 hours ate running temperature.

There has recently been some work being carried out to determine the effect of room temperature creep on high strength steels but from what I have seen so far I am not very persuaded that the work is very well established.

I would, however, be more than happy to accept that Magnesium Engine cases form 2.7 litre engines used with Thermal Reactors could well have suffered from Stress Relaxation and have lost up to 50% of their strength in the threaded area surrounding the cylinder head studs and again this would seem to justify the use of studs which create lower pull out forces.