[QUOTE]Originally posted by xviper
[B]
You put each cylinder such that all valves are closed.

 

[QUOTE]Originally posted by gernby
[B]

Perfect.

 

[QUOTE]Originally posted by xviper
[B]
Not quite.

 

it goes by a certain percentage. You want to be around 5%, and consistancy is the major thing during compression/leakdown tests. You want all 4 cylinders to be similar to one another.

 

I know this is an old thread, but it seemed like as good of a place as any for my follow-up question:

If you do a compression test and the numbers look good, is there any value in doing a leak-down test, also? Is there a possibility that it will uncover something that wasn't seen during the compression test? If so, what kinds of things might it reveal?

 

I just checked mine. When I do it I write the cyl# and comp on the back of the plug cover. At 32k with 20k under boost, I had 255 on all four cylinders.



fltsfshr

 

One thing that I've heard is that a compression test can be skewed higher by carbon buildup causing the instantaneous cylinder pressure to be higher (due to the increased CR) even though it may not be holding that pressure very well.

 

Gernby, I'm not expert, but if a compression test shows everything is within spec, I don't think a leak down test is needed. Leak downs help you pinpoint exactly where the problem is because you can follow (hear) the air coming out of the intake/exhaust side (in the case of valves), or you could have air blowing out your crankcase (rings), and maybe in very rare cases, between cylinders (crack/headgasket).

 

Shouldn't the compression test is also able to pin point the crack/headgasket by comparing the pressure between cylinders? Like within +-14psi I think. I can be wrong

 

But you can't be sure if you've got damage in two cylinders, or something inbetween them. Compression tests help you eliminate possible problems more than pinpoint them.