Valve head seperating from stem

[HITNRUN]

Hey guys. A friend of mine rebuilt his engine and cylinder head. He drove the car for about 40kms and started hearing a rattle from the engine. He removed a spark plug and found a large piece of metal in the cylinder. He then removed the cylinder head and found the head of a valve broken clean off the stem. None of the other valves were damaged. Has anyone heard of this happening and why? I think there may have been a defect in the valve. Thanks people.

 

[TSIfreek]

Ok you have a 10.50 car and cant answer this????
The timing belt was not put on correctly and when the piston came up a valve was still open and bang the piston wins.
Even if the belt was installed correclty it may have skiped causing the valve timing to be off.
In many cases I have seen 1 valve would bend and the other break off or they just both bend and get lucky. Depending on how far the valve was open when it was hit is a good indication of how the piston just beating on the valve untill it did break. I have heard aftermarket valves breaking like this but very rare but it could happen sure. Was the timing checked before he removed the head??

 

[GVR4592]

Ok you have a 10.50 car and cant answer this????
The timing belt was not put on correctly and when the piston came up a valve was still open and bang the piston wins.
Even if the belt was installed correclty it may have skiped causing the valve timing to be off.
In many cases I have seen 1 valve would bend and the other break off or they just both bend and get lucky. Depending on how far the valve was open when it was hit is a good indication of how the piston just beating on the valve untill it did break. I have heard aftermarket valves breaking like this but very rare but it could happen sure. Was the timing checked before he removed the head??

That's one of a dozen possible scenario's.

 

[Defiant]

And very unlikely it would have affected only one valve.
I'd guess just a bum valve.

"Sometimes things fall apart, that's God's love in disguise." -TR

 

[HITNRUN]

The valve was not hit by the piston until it came apart otherwise I would not have posted this. Valves bend before they break and this one just cleanly came apart. Just wondering how many people have had this happen.

 

[toybreaker]

I've seen a valve or 3 lose their heads on these motors, without any timing/t-belt issues.

They were all aftermarket valves. You can tell because the o.e. valves have a letter/number code on the face.

There's probably a bunch of suppliers for valvetrain parts for the 4g63, as the valvetrains are somewhat problematic, and are very sensitive to correct maintainance practices. When people base their parts choices only by cost, they open the market up to cheap parts. An unfortunate side effect is that unscrupulous e-bayers can, and do, rebox poor quality knock offs in premium brand boxes and ship them as the real deal....Several machine shops here locally have been having issues with counterfeit parts recently. Some how, cheesy parts got in the pipeline to our local supplier, and there was a rash of comebacks.

One quick test for quality is to rotate the valve in your hand and eyeball the valve margin. If you see any wobble or distortion in the tulip section, that valve is probably not even going to live in a lawnmower, much less a boosted motor such as the 4g63t. You won't see any runout in an o.e. quality valve.

I'm not sure if these valves are two pieces (stem and head) and are spin/fusion welded together at the base of the stem/top of the tulip, but I strongly suspect they are. This would explain how there can be a clean separation between the stem and the head. I've never seen any information on this for these motors, but this is *generally* how valves are manufactured. The tulip/head are held stationary, and the stem is spun up and brought into contact with the head. Friction ensues, and the 2 parts literally melt together/fuse. Final machining then cleans up any irregularaties, gives the stem it's final dimension, and creates the seat. Add some heat treating and 'viola new valve.

Manufacturers due this for several reasons, chief among them the ability to spec different materials for the head and the stem. In joining two dissimiliar metals, external welding may not give full penetration and a sound weld, whereas the spin/melt/fuse technique works every time. (or so the mercedes engineers told me on a tour)

As an interesting aside, a customer bent some valves in his tsi. Being a cheap sob, he only replaced the bent valves. When his cheesmo t-belt failed a few thousand miles later, we got to see the head again. Only the aftermarket valves had broken off, the o.e. ones just bent. It was yet another testament that you get what you pay for.

As Defiant said, sometimes, parts just break. Our only defense is to use premium quality parts, and to inspect them carefully before installing them.

Please check your valve guides carefully. The force required to bend/snap a valve off can quite easily crack a valve guide, leading to oil consumption/problems further down the road...

Good luck, and I hope it turns out better this time.

 

[HITNRUN]

Thanks for the responses. The valve guide was the first thing I checked when he called me to see the damage. It was not cracked. The cylinder head was purchased rebuilt from a reputable DSM shop. We are speaking with the owner now to see if we can clear things up.

 

[GVR4592]

Was it a Ferrea valve?

 

[HITNRUN]

Was it a Ferrea valve?

No, the cylinder head was stock OEM replacement.