The process we use to recondition rockers is somewhat involved but the tool is pretty simple.

We made a simple arm that allowed us to utilize the valve grinder to surface rocker faces.

The trick was coming up with a surfacing attachment that would allow us to square/surface the grinding wheel with the rocker attachment.


http://forums.pelicanparts.com/uploa...1485531855.JPG
http://forums.pelicanparts.com/uploa...1485531880.JPG

Rocker process:

We remove the old bushing and adjuster (if present)
Clean adjuster threads with 8x1.0 forming tap.
Ream the bushing hole
Black Oxide the rocker
Press in the new DP-4 bushing and burnish
Surface the contact pad
Coat the contact pad with Moly dry film (TLML 2)
Treat the rocker to a Cryogenic conversion process.

http://forums.pelicanparts.com/uploa...1485531972.jpg
http://forums.pelicanparts.com/uploa...1485532006.jpg

Some observations:

1. The rocker must be (after honing/reaming bushing to dim), I'd imagine 0.0005" to 0.001" slip fit to the pivot pin?
2. The rocker must be clamped on both sides of the bushing to also ensure squareness
3. The pivot arm must be square to the stone face using a dial indicator?
4. The stone face should be dressed before each rocker to ensure flatness?

Are these the normal procedures?

Its great to see all of these processes in repairing the rocker arms. However, something that is missing is dealing with the cause of the failure. I read these posts showing how some are repairing the rocker faces, but not one is addressing the cause of the failure and how to solve it.

What is the point of repairing rockers if the same damage re occurs?

It is an oil film issue, or lack of one. Early 911 engines had oiling from inside the shafts to lubricate the rocker faces. The cost this was too much for production possibly so the spray bar was included. There has been developments in oiling but I see nothing here?

I don't think any of these are repaired due to failures/excessive galling in this regard. Rather to true up pads for new cams.

You imagine correcto. Porsche spec book says the clearance for a new rocker shaft bore to a new rocker shaft is 0.016mm to 0.035mm = 0.00063" to 0.00138" Pretty tight.



http://forums.pelicanparts.com/uploa...1485545195.jpg

also agree with Tippy on the failure. Failure is not due to lack of oil in the rocker area. The spray bars deliver PLENTY of oil to the rocker pads. Some argue too much oil and that goes on a tangent to the camshaft housing oil feed line adapter fitting, which is an argument for a completely separate thread that already exists.

A common failure of the pads is slight misalignment due to folks changing cams and thus running two parts with different wear patterns against each other which creates localized high stress areas and shows uneven wear on the follower pad.

More common seems to be surface pitting on the rockers and the cams. Which seems to have come from using an oil that doesn't have enough anti-wear additives like ZDDP "zinc".

When we machine our new rockers we don't allow the arm to pivot and move.

We fix the rocker into a custom made fixture and machine the radius relative to the centre that we determined from the CMM measurement.

By rotating on a shaft the variation in clearance will affect the curvature and the 'squareness' of the heel of the rocker,

We machine the rocker to a precisely controlled surface finish and then isotropically superfinish.

Prior to machining we double temper the rockers at 650 degC.

The surface finish of the rocker pad is relatively important in terms of oil film retention with a cross-hatch pattern being the most effective.

This is difficult to achieve on a small volume production basis so we use a surface treatment process that not only produces a hard surface (1000 HV) but also creates a slightly porous layer that holds an oil film and improves lubricity.

This surface treatment could be applied to a cast steel rocker but is only cost effective on a batch basis. It does produce a limited amount of growth of around 0.003mm per surface. (0.11 thou)

The surface treatment uses a diffusion process carried out in a PVD Chamber at 595 degC which eliminates any thermal distortion.

A typical Chill Cast Iron Camshaft has a surface hardness of 60HRC + 9750 HV) so wear will distribute evenly between the rocker and the cam.

A DLC Treated rocker would have a surface hardness of around 2500HV and would wear the cam approximately 2.5 times faster than the rocker arm.

It the surface of the rocker is significantly lower than the surface of the cam then adhesive wear 'galling' can occur and the rocker would then wear rapidly.

The bush used in the shaft is also quite important and we use the original Glyco Semi-finished bush which is honed to size after being fitted into the rocker arm.

Rocker arm bushes are only partially lubricated and are not pressure fed so their characteristics are quite important.

The Bi-metallic Glyco bush has a PV of about 2.5 times that of a GBB Metal Polymer bush and should provide much better fatigue life at engine temperatures.

The shaft clearance needed for successful operation of a metal polymer bush is significantly greater than those specified when using a Glyco bush and this is a potential banana skin.

It is customary to reduce the diameter of a the shaft to obtain the recommended clearances but as rocker shafts have to expand into the cam housing this isn't feasible.

It is possible to increase the bore diameter of a Metal Polymer bush by burnishing but this does reduce the bearing life.

Increasing diameter by 0.001" reduces life by 30%, by 0.015" life reduces 60% and a 0.002" reduces life by 80%.

An AMPCO 18 material could make a suitable alternative to the Glyco bush and would use standard clearances.

Having personally watched the oiling of these 2V engines on a friends Spintron, I can tell you that it is about a oiling issue. Yes there is a lot of oil sent up to the cams but most of it misses where it needs to go. Add to this, the holes in the spray bar are drilled with sharp corners which makes the direction of the sprayed oil go in all different directions.

I'm sure there is something to rockers repaired poorly, Camshafts hard welded, poor lobe designs with high stress factors, but the oiling of the rockers is a very important too. So much emphasis is placed upon the lash being correct to give the Rocker its oil film, but it only works if the oil goes where its needed.

The pitting that is commonly reported is also a potential problem and is commonly blamed on the lack of ZDDP additives.

Whilst it is reasonably well established that wear levels will increase when ZDDP levels drop below 1200ppm I am not sure that this results in pitting.

Pitting can also result from excessive quantities of zinc being present in an oil.

I believe that ZDDP is monitored by measuring phosphorous levels and that ZDDP additives can contain different levels of zinc which could be an issue and cause this potential problem.

I am sure there is someone with better knowledge of additives that can shed light on this.

Pitting can also assist with the formation of fatigue cracks which would almost certainly result in premature failure.

IMHO it would be prudent to apply a controlled shot peening technique on any refurbished rocker.

A 200% coverage by a shot peening method to a Almen A level of between 6 and 10 using an S230 Grade Stainless Steel Shot would be appropriate.

This treatment will not only remove any accumulated fatigue damage from the surface of the rocker but will also introduce residual surface compressive stresses that will enhance future fatigue life if the rocker is being used with aggressive cam profiles and high rate valve springs.

The surface treatment we apply to our forged rockers introduces a similar level of residual compressive stress so provides a similar improvement.

I would shot peen prior to re-grinding the pad.

Chris, since you're educating us, I'd like to bring up a point I ponder. I grew up around American iron V8's. These engines use pushrods and heavy valvesprings.

Here's the thing, the lifter is riding on half the cam lobe so it'll spin.

When I first saw a 911 config, I thought, how could that ever wear out??

I mean, you have low pressure valvesprings with a very wide contact area (rocker pad to cam lobe) plus far less total valvetrain mass (no lifter nor pushrod).

But wears seems to happen.

Been wondering.

Wondering the same thing???
Subscribed

I am not sure that I know enough about wear to try to explain the reason for the differences that we see in terms of valve train and cam interfaces.

Sadly wear is not a very exact science and it is very difficult to produce accurate mathematical models of this type of behaviour.

Resistance to wear is not really an [I]intrinsic[I] property of a material and developing test methods that characterise materials in the same way as a simple hardness test has proven to be almost impossible.

Changes to a metals surface or near surface properties can have a significant influence and the many variables involved make comparisons difficult.

For a given set of circumstances it is normal to consider two different stages to describe the wear process.

Stage 1 would be break-in where the situation is very complex and changes to surface condition due microscopic plastic deformation of asperities and micro cracking can lead to a significant amount of instability.

The steady state condition which occurs after break-in wear rates can become very stable and it is much more likely that meaningful comparisons can be made.

The next problem is that wear does not directly correlate to friction.

Early work by Amonton seemed to imply some relationship but Coulomb's work modified this significantly, particularly when considering metals and he modified the basic model to include adhesion but took no account of energy dissipation.

In fact adhesion is extremely significant as contact areas is solely dependent on the normal load applied for both elastic and plastic behaviour and both friction and traction can have a significant influence.

There are situations where increased load can reduce wear as skidding is reduced.

I would generally conclude that trying to compare the wear rates of cam on a V8 pushrod engine and a Porsche 911 could involve a substantial amount of time using a Spintron and that the conclusions that could be drawn would show that they may differ significantly.

It is unlikely that a simple analysis of PV will provide any useful correlation.

When we developed 'our' forged rocker we made a decision to coat the wear surface based on the fact that Porsche used a hard chrome.

This has a hardness of around 850HV and should also keep adhesive wear to a minimum.

The 'safest' approach may have been to replicate this process but it is increasingly difficult to fing sub-contractors to do this work.

The process we selected is well established has comparable levels of hardness and is commonly used in 'sliding' wear applications. The surface chemistry of the coating is resistant to adhesive wear as it is incompatible with the surface of the cam and is unlikely to weld.

We have made these rockers for 3 years now and so far wear has not been an issue.

I would also agree that wear of rockers is not a major problem.

I am also interested in the position of the 'holes' in the spray bar which seem to try to spray oil directly into the gap between the rocker and the cam.

When we used to spray lubricate gears we always used to inject the oil into the trailing side of the tooth form so the coating of was carried into the mesh.

We also used to oil jet lubricate rolling element bearings and again we always injected to oil into the trailing side.

I think that there is a difference in the spray pattern and the oil hole position on early cam carriers (just after the centre lube was replaced) and the later carriers.

I will dig out some old carriers and see if this is the case.

Gentlemen, thank you for the education...fascinating. John in CT.