Valve Stem Tip Spalling

[Kevin Johnson]

I'm looking at the tip of that rocker arm and the factory finish looks horrible to me ... as cast,if my guess is correct.

That was my opinion as well. However, I purchased a new LS7 exhaust rocker arm to examine it more carefully. There are witness marks on the reverse of the casting showing that the tip is pressed into a die of sorts thereby imparting a very even surface. Also, the surface is slightly convex from side to side which would generate a point contact.

There have been at least three variations of the LS7 exhaust rocker arm casting with the mass distribution slightly altered, presumably to help with the dissipation of dissonant energy into the parts of the valve train assembly.

With respect to the change in installed height my suggestion is to correct it to factory specs, at least with respect to the LS7 which I believe to be an example of a general class of problems (with respect to exhaust valve guide wear and subsequent valve head dropping) that are too complex to be adequately modeled.

[zums]

ive used nothing but mobil 1 in everything from daily to 8000+ rpm sbc , never had an oil related problem, although when the writing was on the wall i stocked up, my last batch has all been gl-4, dont know if i would try it with gl-5, but to say mobil1 is shit is foolish

[nickmckinney]

So when it exists, what does that leave? Poor design, or poor materials?

IMHO - its poor materials/processing for that batch. The motors are tested WOT for days so the design was good at sometime. Since about the mid to late 90's the materials the OEM are using for valves has become so hard and durable some won't even bend with PTV but rather the guide fractures instead. I would bet something got missed at the valve manufacturing point somewhere, 3rd shift kinda thing. Also the OEM rockers of late you typically have to torch them cherry red first before you can get a drill bit to even scratch them.

[Strange Magic]

Its not all crap. You just have to know which are, and which aren't. They along with all the other oil manufactures/marketers were required to make less than favorable changes to their oils to meet certain specs. Some Mobil 1 oils like their 5/30 is less than it could be. Don't blame them, blame the EPA and ILSAC(auto manuf). I wouldn't run Mobil 1n or ANY GF5 certified oil in a lawn mower, let alone a performance engine specially once its out of warranty.
I run Mobil 1 in a blown BBC in an edurance boat, and never had a single issue. But then I am not trying to run some oil dictated by the EPA and auto manufs.

Yes, understood, but I was not going to get into some of the other Mobil products that have the better components because they are not the easiest to find and your average joe is looking to pull it off the Walmart or auto part store shelf. Even if easy access, I still am not fond of any of there products. There are plenty of oils that are much better on the market and easier to obtain. I see it this way, if your going to build yourself a purpose built engine (non oem, non oem rebuild) then do the research necessary on oils, rather than just assume upon the oil you pulled off the shelf, is good.

The emphasis is heavy over the past recent years for companies to operate at equal margins of the past, and in many cases they do so by degrading the product, and that goes for most anything we purchase. The buyer now, has to be more concerned than ever, and rightfully so.

[nickmckinney]

How in the hell did this get into an oil discussion, this has nothing to do with the oil. Rockers from other motors go 200,000+ miles and look polished at the same point. Something is wrong with the materials used here be it a bad material batch, bad heat-treat, etc. The OEM are more than capable of making a rocker and valve stem tip that would not wear on these low pressure valve trains even if goat pee was used for the lubricant.

[Strange Magic]

Not sure if I know to many oem applications that have .600 of lift. Please enlighten us if any know of such.

[nickmckinney]

Those exact motors were tested WOT with those same designed parts at least a year before being released. The OEM testing is way more brutal that anything we will do, and it was probably done with a few *hundred* samples of this motor before being released. So the design for the wear tolerance worked at one point with the prototype parts, why is it failing with the production parts is the real question. Here is a post from a GM powertrain engineer years back describing the testing:


"Any engine program like the LS1 for the Corvette, the LS6 currently and the upcoming LS7 are testing on dyno and in cars for YEARS before release to the consumer with the specified lubricants and coolants and such.

In the case of the Mobil 1 in the Corvette I would hazard a guess that the number of dyno engines run with Mobil 1 is in the hundreds....like somewhere between 200 and 300 all total. These are engines that run for 200, 300 , 400 or more hours at full throttle, max RPM, max power and are then torn down and analyzed in a variety of ways for wear and other lubrication performance. This is on top of the countless specific bench and dyno tests run on the specific lubrication system, cold start lube system performance, cold over pressurizaration (start at -20 and go immediately to 5000 RPM), cold start field testing and driveability, engine cooling testing, etc. Fleets or test cars are running with the production intent designs and lubes on accelerated durability, endurance, emissions, track testing, etc... The products are testing far in excess of what any customer can do."


IMHO - GM either went to another supplier expecting the same spec or the original supplier screwed the pooch somewhere. It wouldn't be the first time another supplier was clueless as how to get it done properly, I have been burned quite a few times myself that way as just a small shop.

[Kevin Johnson]

How in the hell did this get into an oil discussion, this has nothing to do with the oil. Rockers from other motors go 200,000+ miles and look polished at the same point. Something is wrong with the materials used here be it a bad material batch, bad heat-treat, etc. The OEM are more than capable of making a rocker and valve stem tip that would not wear on these low pressure valve trains even if goat pee was used for the lubricant.

I think the difficulty is that the problem is a convergence of multiple issues. Engineers are trained to look for well defined root causes and that is what makes this problem quite vexing but representative of most real world problems (which are currently treated probabilistically). Close study needs to be made of the systems that allow analog versus discretized digital input (driver input at the throttle, for example, and powertrain programs that feed back into ignition control).

You can be sure that if the problem were so trivial as a bad batch of components or a lubricant it would have been addressed decisively with a metaphorical sledge hammer. The engineering prowess at GM is at a very high level. The steady stream of engineering tweaks indicates a far more complex issue.

[MadBill]

it strains credulity that the horrible "alligator skin" finish on the rocker pad shown earlier can be deliberate. Surely a finishing step must have been omitted? I would think such a surface would not only cause fretting of the valvestem tip but would 'bite in' and apply far more side as it moved through its arc than would a polished pad, thus exacerbating guide wear.

[Kevin Johnson]

it strains credulity that the horrible "alligator skin" finish on the rocker pad shown earlier can be deliberate. Surely a finishing step must have been omitted? I would think such a surface would not only cause fretting of the valvestem tip but would 'bite in' and apply far more side as it moved through its arc than would a polished pad, thus exacerbating guide wear.

There was an extensive discussion on this on a Corvette forum. I thought as you do until I purchased and examined a new OEM specimen. This being, mind you, a specimen produced many years after GM was well-aware of exhaust valve guide wear issues.

[Strange Magic]

We are trusting GM? Take your glasses off people. The records speak for themselfs. The decisions from General Motors are consistant with thier products produced, and those products are based upon the competition. Price, Quality, Consistancy, Service. I worked for GM and have cetifications to prove just that. I also worked for Toyota as well. There quality and consistancy is poor at best when compared to the competition. You can argue this all you want, but facts are facts and the establishments that sell these products have become few and far between.

Do you really think they grabbed a shelf oil and used it for testing? You got to be kidding.

I will put zero trust and zero faith in any gm engineer.

Those exact motors were tested WOT with those same designed parts at least a year before being released. The OEM testing is way more brutal that anything we will do,

Well that's contrary to reality, because reality has GM vehicles back in more frequently than many others for warranty work from failed or poor design. Lets not kid ourselves, true testing means that it cuts into their bottom line and they would rather spin the situation, leading you to believe that they true tested their products thoroughly, than actually doing so.

[Kevin Johnson]

We are not trusting GM but simply recognizing that with some problems there is no amount of testing that will exhaust the variables present.

I would not be at all surprised if recent issues with unintended acceleration at Toyota are of the same class and equally vexing to analyze.

Now 40 years on, the problems with the Porsche 928 rod bearing failure can be seen to principally hinge on driving technique in a vehicle with near perfect 50:50 front to rear balance. It is impossible to model all driving behavior. What you can now see are attempts to over-ride driver input in order to truncate variables.

For the philosophically minded reader, Descartes' evil genius is reaching reification by virtue of our techniques of formal analysis being blunted by the class of problem. This problem will grow worse.

http://en.wikipedia.org/wiki/Evil_demon

[Mark200]

[...] IMHO - GM either went to another supplier expecting the same spec or the original supplier screwed the pooch somewhere. It wouldn't be the first time another supplier was clueless as how to get it done properly, I have been burned quite a few times myself that way as just a small shop.

I have seen photographic evidence that the original LS7 exhaust valve, as stocked by dealers, carried the MAHLE logo. At some point around this time, MAHLE aquired an Argentinian manufacturer named EDIVAL. Subsequent LS7 exhaust valves stocked by GM dealers carried the EDIVAL logo. The closeup photo of the valve stem tip I posted earlier was an OEM valve carrying the EDIVAL logo and an early 2008 date code.

As to the finish on the stem tip, I should have taken some magnified photos of the finish on a new a lash cap for comparison; didn't think of that at the time, will revisit it when I have the opportunity (probably be a few weeks). In the meantime, is there some shadetree method of dressing the finish on those valve stem tips or should that be left to a machine shop?

I'd like to run Yella Terra roller tipped rockers on this engine to hopefully minimize this and other issues (stem side loading), but that has opened up a whole 'nother can of worms that I have yet to sort out (mainly the amount of extra spring pressure needed to compensate for the extra mass) which is probably best for a separate thread.

[Kevin Johnson]

[...] IMHO - GM either went to another supplier expecting the same spec or the original supplier screwed the pooch somewhere. It wouldn't be the first time another supplier was clueless as how to get it done properly, I have been burned quite a few times myself that way as just a small shop.

I have seen photographic evidence that the original LS7 exhaust valve, as stocked by dealers, carried the MAHLE logo. At some point around this time, MAHLE aquired an Argentinian manufacturer named EDIVAL. Subsequent LS7 exhaust valves stocked by GM dealers carried the EDIVAL logo. The closeup photo of the valve stem tip I posted earlier was an OEM valve carrying the EDIVAL logo and an early 2008 date code.

As to the finish on the stem tip, I should have taken some magnified photos of the finish on a new a lash cap for comparison; didn't think of that at the time, will revisit it when I have the opportunity (probably be a few weeks). In the meantime, is there some shadetree method of dressing the finish on those valve stem tips or should that be left to a machine shop?

I'd like to run Yella Terra roller tipped rockers on this engine to hopefully minimize this and other issues (stem side loading), but that has opened up a whole 'nother can of worms that I have yet to sort out (mainly the amount of extra spring pressure needed to compensate for the extra mass) which is probably best for a separate thread.

Katech was going to perform valvetrain dynamics studies on a number of LS7 combinations. I suggest you get in contact with Jason and see where that is at.

[Mark200]

Katech was going to perform valvetrain dynamics studies on a number of LS7 combinations. I suggest you get in contact with Jason and see where that is at.

That's all wrapped up, waiting on results to be tabulated and published. A few of us lobbied -- in vain -- for aftermarket rockers to be included. C'est la vie