Unusual leak down test and compression

[Ravenous]

I took the car down this morning to rebuild my turbo and throttle body. Before I broke it down I decided to do a leakdown test, the poor man's version with the compressor hooked to my compression tester hose. My leak down test results made little to no sense.

Cylinder 2 leaks into 3 and 4
Cylinder 3 leaks into 4 but not 2
4 and 1 don't leak at all.

Now I did the compression test just to see what the numbers were:

1- 150
2- 125
3- 125
4- 150

Then I added some oil to 2 and 3

2- 150
3- 200 (actually pegged the gauge)

My motor is using Manley 9:1 pistons on a decked block and a 2nd hand head without any machining with stock cams and springs. The compression test was done warm with the throttle open.

My thoughts are that it could be rings, valves, or a warped head/blown gasket. (Captain Obvious)

The block was bored/honed/decked and the pistons and rings were installed and broke in motoman style so I would like to rule those out but they are a possibility.

The valves were lapped before the head was installed but the decked block and unknown condition of the 2nd hand head (could of been decked previously) could be throwing off my timing. The alignment marks are just slightly off when setting timing on the rebuild. It was a pain in the ass doing the leak down test as TDC on 2 and 3 never sealed but just slightly beyond tdc 3 would seal off from 2 but leak into 4. Since 4 doesn't leak into 3 at tdc I am figuring valves are most likely at fault but the wet compression test would indicate rings. With the higher compression pistons would the valve reliefs deposit oil on the valves during the wet test... especially if timing is off?

Headgasket or warped head- This is my 2nd head gasket since the rebuild. I was burning coolant so I changed the gasket and upgraded to ARP studs about 1500 miles ago. I no longer burn coolant but I was suspecting that I may be pushing it (thermostat or leaked compression) however when I took the car down today I found that the water pump had been leaking for some time. I checked the head for flatness with a straight edge during the last head gasket install but it wasn't a high precision piece.

I am going to pull the front timing cover to verify that I haven't jumped a tooth on my timing but if anybody has any input or insight I would appreciate it as the leak down results have me baffled.

 

[Calan]

What kind of head gasket, and how much over bore are you running?

If it is an MLS gasket and you didn't surface the head to a low RA, you probably have some blowby at the fire ring between cylinders...especially if you are over-bored. It's also possible that you are getting past the rings, although it would be more likely that air getting to the bottom of the crankcase would blow out the VC vent, rather than fight it's way back up into another cylinder.

Are you sure the intake valves were completely closed? The best way to make absolutely sure is to remove the VC and pop the rockers off #4...then rotate the engine until all the other valves are completely closed.

 

[Ravenous]

I am running a composite headgasket and it was bored .020 over. I just pulled the timing cover and the exhaust gear is advanced one tooth. I don't think it jumped I bet that I botched the job when I redid the head gasket this last time. I am going to re-align it and see if it makes a difference in the leak down.

Redid the timing belt and got everything lined up again and the results of the compression test are much better this time around:

165-155-165-160

(I did run it for a while to blow out the oil from the wet test)

I am going to have to start building my spare head as the leak down test is still indicating leaking valves.