Geoffrey

This topic has been discussed before as well as the plethora of aftermarket chips. As we dissassembled another engine due to it stopping suddenly and began to diagnose the root cause, it became obvious to me as to what had occured. This was a stock 964 engine. The first thing we found was that the gear teeth on camshaft sprocket were missing due to something stopping the cam. That something was a broken rocker arm. Previously, we had another broken rocker arm which should have been a clue. We then found a piece of the outer spring in the bottom of the oil drain plug. So, we know a valve spring broke. When then heads were off, it was clear that at least 2 valves hit the pistons from the cam being stopped and the crankshaft still turning. So now the entire engine needs to be split down to the case.

Upon further disassembly, the valves would not slide through the valve guides and were getting stuck at the valve keeper grooves. Why would this occur? Well, valve springs work on the physics principle F=MA where the force of the spring must exert enough pressure to control a valve with a certain mass to keep in contact with a rocker while accelerating at a certain rate. A stock valve spring installed at the factory installed height is about 45lbs and if you take the stock camshaft profile with its acceleration ramps and the stock valve masses and plug it into a program, it will tell you that the 45lbs seat pressure is enough to keep the valve in contact with the rocker arm up to about 6800rpm. No more. So, when Porsche set the rev limit at 6800, they mean it.

So what happened in this case? Did it get overreved? I think not because I would expect to see only damage on the one cylinder where the valve spring broke. But instead, I see that all of the valves have damage indicating that it had an issue for awhile. The valves were floating and the springs were coil binding which initially broke a rocker arm which was replaced, and eventually breaking a spring and another rocker arm which lodged under the cam lobe and stripped out the cam sprocket.

Upon further investigation, the car has a chip that allows for a 7200rpm rev limit. So, the next time you think about buying a chip, I would suggest ensuring that it has the factor rev limit. I know I've said it before, but I wanted to share with you real life evidence as to why.

Ritter v4.0

Good to know.

Also, when and if you dyno your car, especially with an aftermarket chip you can see where your car is making power- there may be little advantage to reving above a certain limit. My car peaks (hp) about 6300- so why shift beyond there just because I can?

But I don't completely understand Geoff- you say you don't think it was overreved then build and argument to suggest that it was- if it wasn't the root cause what do you think it was?

Geoffrey

Let me clarify, I don't believe that the damage was caused by overrevving a downshift, but rather by running an increased rev limit. So, you are correct in that the engine was revved over the factory rev limiter or revved beyond what it was designed for.

Indycam

"coil binding" ?

"My car peaks (hp) about 6300- so why shift beyond there just because I can?"
I think the idea is to go past the peak by 500ish rpms .

Geoffrey

Coil binding occurs when the spring is compressed so much that all of the coils are compressing and you cannot compress the spring any longer. The spring is mechanically locked and the spring rate becomes infinite.

amr89c4

When you shift gears you lose significant rpm between the shifts. Speaking as an enthusiast and not an expert, shifting at the upper limit prevents you from falling way below the hp/torque curve. If you shift at the curve peak, you fall even further below the curve and it takes more time to build the rpm back up. In my stock 964, doing that in third gear or bumbling the shift just kills the accelleration momentum. Keeping the rpm up keeps the engine closer to the band you desire to be in. Just my two cents.

Geoffrey makes a strong case for addressing the springs during a top OH or as a good reason to go ahead and get that headwork done. While I don't have a chip, I will certainly demand the limit be left stock if I were to order one with a stock valve train, at least until heavier poundage performance springs are installed.

TR6

The biggest reason that I've been told that track cars are often set to go above 6800 rpms is because it can often allow you to stay in a given gear all the way to track out of a corner without worrying about upshifting. For example, there is one turn that I am on the rev limiter at 6800 before I hit track out. I would like to take the corner in 2nd for the torque going uphill, but the tradeoff is that if I want to carry more speed by trackout, I will have to go up to third gear and forget about taking it in 2nd. If you had, say, a 7K limit, that extra 200 rpms could get you to trackout before shifting. Not a big deal in a DE, but I've been told that it can give you a slight edge in a race.

Having said that, I have no desire to flog my engine above 6800, so I'll have to take that corner in 3rd.

Laurence Gibbs

So did the rocker arm Jam in the valve train? Just a bit confused here, you seem to be saying the rocker arm was the cause of the stripped teeth on the sprocket, only way that can be is if the rocker jams, any other way and the valves would close and stop the piston hitting? The springs on the other hand could allow the valve to contact the piston and sieze the sproket if they broke. The springs have a finite life and are know for this i think ?

Indycam

"Coil binding occurs when the spring is compressed so much "
I was asking how it was happening .
Excess rpms causes coil bind ?
How so ?
The cam is launching the valve to the point of coil bind ?

Geoffrey

Lawrence, the rocker arm broke and wedged between the cam tower where the oil spray bar is and the cam lobe preventing the camshaft from turning. This also allowed 2 valves to contact the pistons. The springs to have a finite life and should be replaced when the heads are rebuilt. The stock springs are good springs, it is simply the installed height that does not provide any excess seat and nose pressure.

Indy, the springs were coil binding because the force of the spring could not exceed the force require to control the valve. Therefore, the cam lobe pressed on the rocker which pushed on the valve which overcame the spring pressure and collapsed the spring to the extent that it bound or bottomed.

Laurence Gibbs

Ahh that explains it. That's a bit unlucky. I guess had the rocker just fallen away far less damage would have resulted. it was my understanding that the rockers are a little "sacrificial" so if coil bind and snapped spring do occur the rocker took the brunt not the piston , as it gives way sooner?

Laurence Gibbs

with regard to coil bind i think of it like this (don't know if it's wrong or right!) think of the cam as a light switch and the valve and spring as the light. If you switch the switch on and off the light goes on and off or the valve and spring goes down and up. If you switch the switch faster on and off the light flashes or the valves go up and down quicker. If you switch the switch on and off fast enough you stop the switch acting like a switch anymore and the light "appears" to stay on or the valve "appears" momentarily staionary. Basically you are exceeding the on off cycle of the cam and spring to a point where they are not able to perform as a unit.

Indycam

The valve has to come off the seat farther than the normal lift for the coil to bind .
If the valve is floating , its not floating above the normal lift .
Unless the valve is being launched , I don't see it getting father off of the seat .
In normal stock form , how close to coil bind is the spring at the top of the cam ?

Laurence Gibbs

In this case I think that bind implies the coil can't overcome the pressure exerted on it to close or more accurately close quickly enough ! It can't react in time to the cams rotation.

Sire

Question if you want to rev above 6800 then with a rebuilt allso buy better valve springs?