Exhaust valve contact all four pistons! PICS ADDED

[wil892]

Before I post my main thread showing the rebuilding of my GTD

I have taken the engine out and stripped the head off. Since replacing the cylinder head with new 17,000 miles ago there are now exhaust valve indents on all four pistons.
I am not sure how this is possible as the engine was timed perfectly, confirmed about twice during this 17,000 period, using the correct locking tools.

The indents themselves are not just "kissing", they have made a noticiable circular ridge on all pistons. The head gasket was replaced with a 3 notch, the same as the old one. There are no dents in the lifters or abnormal wear. Also there was an even layer of soot all over each piston so I assume the valves aren't hitting the piston on every stroke.

I have noticed that cylinder 3, which had low compression has more soot in the exhaust port and oil coming from the guide in that port, which would explain my oil consumption. Whether the low compression and oil are due to a bent valve and guide I don't know.

The only thing I can think of is that the new hydraulic lifters didnt bleed down properly when first started up, but that wouldnt explain why its all the exhaust valves.

Anyone seen this before, and why it has happened?

Thanks, will get some pictures up tomorrow

[nathan_b]

wrong head gasket?

funky rods? did you spec the piston protrusion?

valve seats sinking?

are you sure it's not just the cast valve indentions in the pistons??
one can hope right??

[Henchman]

Was there carbon buildup in the indents or were they shiny?  If they have darkened, it may have been a condition that existed breifly at first startup.....

Did you shave the head?

Ian

[Mark(The Miser)UK]

All 4 exhaust valves hitting is a timing issue. Cam retarded.... Or not.
My head does this.

If your cam is set correctly, then the problem likely to be because the valves are not sunk enough in the head, which is the reason VW had the policy of early times when heads were still being manufactured  " Heads not skimmable" Is this head skimmed? Because of valve overlap, clash  may be unavoidable [my case]

The burrows in the pistons may be less deep than they appear. What makes them look worse is the raised edge around the circumference. Try scraping it off. True clash halts casm and snaps belt...
Why have you taken the head off? Was it just the oil burning?

How low was the compression?

[wil892]

The head gasket was a elring 3 notch gasket. The piston protrusion was not measured as the original gasket was three notch as well. Its definitely not just cast in. Clear marks and indents on all pistons with little hole bit in the middle of the valve.
There was an even carbon layer on all of the pistons, I didnt see the indents immediatly, but looking closer I did. Then i brushed off the carbon and the indents can be seen clearly.

The head was a new casting (no-name brand probably chinese) that was built with new OEM valvetrain at a respected VW diesel engine shop. The head was supplied to me as brand new, as was timed properly from the beggining with a new Dayco timing belt. I used feeler gauges on the cam lock to ensure it was fully accurate, then rechecked at least two times since fitting the head when trying to sort my injection timing.

I measured the compression on all four on a cold engine, which showed 250psi on all apart from number 3 cylinder which showed 100psi. I believe the tester wasnt showing accurate results because the engine normally wouldnt properly run with these results. However the engine fires easily even with low cranking speeds, it has just been missing on one cylinder when first fired and consuming oil, and also trickles smoke when cold idling.

The engine had a number of oil leaks and I have removed it and taken it apart for a full going over, it will have a new injection pump and the turbo refurbished as well.

I initially thought the low compression on one cylinder was sticking rings but seeing the valve indents it could be a leaking exhaust valve. The bores show clear honing marks and there is little carbon buildup. I will be measuring the bores when I pull the pistons.

[rabbid79]

This is very strange.  I kind of like your theory about the hydraulic lifters not being drained down before the initial start.  The other possibilities could be that the valves are too long (e.g. is it possible to put the keepers on wrong?), the distance between the deck of the head and the cam is too short (e.g. cheap aftermarket head),  or the cam is not stock dimensionally (e.g. base circle reduced for higher lift of valves).

However, it seems like the problem has gone away on its own, which goes back to the lifter issue, since none of these other issues can fix themselves.  If you had problems initially with the timing, there could have been a lot of banging of pistons and valves going on inside the engine that would have been masked by the sound of an improperly timed pump.  That could have been why you never noticed it.

[rabbitman]

Well I don't think the keepers can get put on wrong enough to hold the valve different, but either way, the lifter pushes on the tip of the valve and that dimension has to be in a certain range for the lifter to work properly.

Also I believe the book says to squeeze the lifters with a c clamp to make it shorten as much as possible. If that step was skipped then I can see the valves hitting the pistons at first.

[wil892]

rabbid, the engine was cam to crank timed perfectly before I started it up for the first time with the new head. I did not have a dial indicator at the time so I couldnt set the injection timing.
The engine was started until it fired regularly, which took a few seconds. As I remember there were no banging sounds just lifter ticking as the whole head was new. It was then switched off and towed to the local diesel specialist who set the injection timing to stock. I towed the car home and then started it and it warmed up, then I did the last torque sequence.

I cannot think of a single event when valves hit pistons. And I don't know if at high rpm they have been hitting them all the time. I cant believe the valves haven't broken up in 17,000 with the dents they have made in the pistons. Also the dents have definitly come from this head as they had no marks when I installed the new head.

My bentley manual says leave lifters for 30 minutes to settle after installing to prevent valve contact. It took me at least a few hours to attach everything again and then start it. In fact I think I left it overnight after installing the head until starting.

[rabbid79]

How about weak exhaust valve springs?  Maybe they're floating a little at higher RPMs?  And these engines are so loud at high RPMs anyway, it might be kind of hard to notice.  It might also explain why the engine started ok the first time you fired it up.  If there really had been bad piston/valve clearance issues, it could have got bound up when you first cranked it.

Don't these engines use 2 springs (inner and outer) for the valves?  Maybe you're missing some springs?  Intake valve springs could be weak too, but they wouldn't be a problem in this case since the piston's low in the cylinder when the intake valve closes.

[rabbitman]

How about weak exhaust valve springs?  Maybe they're floating a little at higher RPMs?  And these engines are so loud at high RPMs anyway, it might be kind of hard to notice.  It might also explain why the engine started ok the first time you fired it up.  If there really had been bad piston/valve clearance issues, it could have got bound up when you first cranked it.

Don't these engines use 2 springs (inner and outer) for the valves?  Maybe you're missing some springs?  Intake valve springs could be weak too, but they wouldn't be a problem in this case since the piston's low in the cylinder when the intake valve closes.

Yup

I don't think it would run very good if the valves were floating.

[wil892]

Ok, here are some pictures. Just to say I don't think the valves were floating as the engine used to pull really well above 4000 rpm but was quite smokey.

The state of the conrod shells is bad, compared to the mains which have only slight wear.

[rabbid79]

Just to say I don't think the valves were floating as the engine used to pull really well above 4000 rpm but was quite smokey.

So because the engine pulls well above 4000 RPM means the valves weren't floating at 5000 RPM?  I'm just suggesting ideas here, but you've pretty much shot everything down without even checking.  It sounds like you're looking for a silver bullet, but I think if there was one, someone would have suggested it by now.  Figuring out why your valves made contact with your pistons is going to take some time, and require that you check a few things.

I would suggest removing the cam and lifters to make sure both sets of springs are there.  How hard are they to press down?  It should be pretty hard.  Make sure the valves are actually opening when you do this, and that it's not just the lifters giving.  Have you checked the length of the valves?

[Mark(The Miser)UK]

So you've taken the engine apart now, probably needlessly. Those impressions are historic, and likely trivial, although naturally alarming to many. The reason I cn say they are historic, is because of the carbon granules and the lack of concentric circles.
Here is a picture of on going clash:

http://imageshack.us/photo/my-images/844/imgp9926.jpg/
& one of its sisters
http://imageshack.us/photo/my-images/26/imgp9923.jpg/

Note those circles I mentioned. They are there because 1000's of hits by the corroded exhaust valve as it rotates.
Also note the shadow of the inlet valve on the second piston. Meaning I have nowhere to go.
This is my engine, and it is still running and getting 50 to 60 mpg [uk]    

[wil892]

Mark thanks for the info about the concentric circles, that makes sense to me.
I really don't think I took the engine apart needlessly though. There is some conserable wear on the connecting rod bearings and compression is low on number three cylinder (hence me taking the thing apart). There is oil everywhere from failing seals (crankshaft). I was getting 40-45 mpg.

I can only think number three has low compression due to a slight valve leak caused by piston contact. I can see an oil trail coming from the guide in the exhaust port on number three as well.

Rabbid, sorry if you think i'm dismissing everything, I have simply answered all the questions accurately with the procedure I took. I didnt mean to sound like I dismissed your suggestion because I haven't. This new head cost me a lot of money a couple of years ago and to find it needs looking at and probably the exhaust valves replacing already, is quite annoying, as I did everything exactly to the book and bought all the correct tooling.
I will check the valves when I can get the head dismantled (I have no valve compressor), but I did try and push the lifters down but couldn't with any of them, even with a lot of force.

[Mark(The Miser)UK]

"Needlessly" was a little excessive

Looking at the carbon swirls across the pistons, I'd say that the injectors were very well matched.  I'd also say that compression was good too, or at least doing an 'equal job'.

The shell bearings are not bad either I know they have, a few burnish markings, and one has a minor flake  of something. However, I don't suppose their depth/thickness has changed, and they are merely reflecting slight imperfections in con rods etc; and  are simply snuggling themselves into their location.

Possibly the best pictures of shell bearing wear and it's causes can be found on the 'Clevite' site.

As long as you give the oil pump an overhaul, I'd leave those bearings alone. A new bearing still has to bed in, if there are any slight anomalies and engines do have plenty of those.

Much variation in compression ratio can be down to differing levels of ring gap alignment.
Best initial setting for ring gaps is not the 120 deg apart put in books but 180 degrees for the top two rings and about 175 degrees for the oil ring.

As you have ther head off, to check the seating of the valves, invert the head and pour some kerosine onto the closed valve heads...Oops paraffin