valve removal

[T41pilot]

Compression check on 1 and 2 cylinders were poor and I found that the exhaust valves are leaking badly. Only ten hours on the engine so I suspect they may have been leaking from day one. I want to lap the valves but upon removing the springs the valves did not want to come all the way out. They go as far as the area where the keeper grooves are and no further. I tapped lightly to see if they wanted to move further but no joy. Thought I better ask here first before I tapped any harder to get them out. I currently don't see any other way to get them out other than lightly forcing the ends past the guides. They move nicely in the guides up to that point. What am I missing here?

[Bryan Cotton]

I'd check with Sonex. I think there was a bad batch of heads and they may be able to help. Mine looked bad before I built the engine and I got them replaced.

[GraemeSmith]

In my experience that is very common. The retainers very slightly spread the groove of the valve stem. I dress all around the grooves with a very fine file. You won't even see much come off. Apply oil to the grooves and then the valves slide right out.

Make sure everything is spotless clean before reassembly.

[T41pilot]

Thanks Graeme

I have 3 of them out now. I used some fine sandpaper just in the groove area and cleaned them well before testing to see if they would start to come out. Would take several tries. Then a final sanding, cleaning and oiling and they come out fine. Actually started that process before your reply. Figured it would be OK since the rest of the stem was moving just fine and the groove area was not in the guide during normal operation of the engine. Your explanation makes sense of why that area would expand over time a little after the heads were assembled.

[Onex107]

If you are doing the 80 psi leak down test, run the engine before, and do it as hot as you can. It does make a difference. All it takes is a small piece of carbon to make them leak. You didn't say what the pressures were. I didn't do it until the pressures were in the low 60's. When I lap the valves, I don't take them out of the head, I chuck a drill on the end of the valve stem and put some lapping compound on the seat and run the drill both ways while pulling on the drill. It may be unconventional, but it works.

[T41pilot]

The original compression test was with a warm engine. One cylinder was 55. The other was 60. With the head on the bench, one cylinder still 55. The other improved to 65. These values are unacceptable to me. I'm waiting for a new tool to get the valve springs reinstalled. My universal compressor tool isn't working for installing the stem keepers. Bryan was right when he said the purpose of the hobby was for buying tools. After lapping the valves, they visually look better now. The new air test will have to wait. I built a similar Jig that Graeme did. It seems to be doing the trick.

[mike.smith]

The VW is not a tight tolerance engine. I almost never get good values on a leak down test when the engine is cold, though I do still test it that way to start. Then I go fly in the pattern for a few circuits and come back to do the test. The numbers have always been well within the 70's, and sometimes close to 80.

[GraemeSmith]

A way to perform a compression test on a cold engine is to inject (oil can with a long spout) some engine oil into the cylinder through the top plug hole, rotating the engine a few times to coat the bore thoroughly and give the oil rings something to seal on. Then test.

This is useful when you are trying to figure if staking a valve has dislodged crud from the seat - before taking it back outside to run up, get hot and do a hot test. Which can be a pest if it takes a couple of goes to clear a valve.

Done this way - I've got results that are within a couple of pounds of the subsequent hot test.

It is worth noting that the FAA AC43-13.1 "Acceptable Methods of Work" only recognizes the hot method.