Okay, I'll take the dare from Swain and post what might be the more "normal" reply :-).  I think he is being a little of a perfectionist.  I don't have his credentials, but from what I have seen, a compression leakage of 74 to 76 is the most common.  A compression number up to 78 suggests that the engine is pumping excess oil past the rings and the oil is sealing that leak path.  So I would worry more about the ones that measure 78 than I would about the one that is at 75.  Assuming you check the compression every 100 hours or so, I would wait until you have more than 2 compression tests with ever-dropping values before I would even start to worry.  That will take another 200 hours of operation.  I wouldn't worry until the compression dropped below 70.  And like Dan said, it is always a good idea to warm the engine and then do the compression test as soon as possible afterward.  I've gone as far as to warm the engine with the cowl off so I could do it faster, but to just idle on the ground is not a good thing just before a compression test.  If I were really into the diagnostics I would warm the engine (cowl off) at the normal runup rpm and then run it at full power for maybe 15 seconds, pulling the mixture while at full power to get a "clean" shut-off.  Then you'll get as good a test as possible.

And why do you change oil every 25-35 hours?  With a modern engine and modern oil, I don't see any reason to change oil that often.  I would run it at least 50 hours and try to change it before 75.  If the engine is run often, even 100 hours isn't out of line.  What does the oil look like at the oil change?  Can you see the dipstick through the oil?  If so, it probably doesn't need to be changed yet.

Gary

Previously, from Swain: 

Hi Dan,

 My opinion will differ from most here. It is based on 28 years as a professional aircraft mechanic and a cylinder overhaul shop owner for 17 years. Lycoming says that NO static leakage is allowed during a compression test and I agree. What will happen is that if your ex. valve continues to leak past it's seat it will errode a gas path in the seat and valve face. This will cause a "hot spot" in that area that can eventually lead to  valve face breakage which might do major damage to one or more cyl. I say more than one because I have personally seen a piece of ex. valve leave it's cyl. through it's intake port and enter the next cyl. beside it and lodge itself into the face of that piston while bending the other cyl. intake valve. (BTW I keep that piece of valve in my desk drawer as a reminder of what can happen). The only way a valve can cool itself is through the time it sits on it's valve seat to transfer heat from the valve to the head and also through the valve stem. That's why when guides get worn and seats leak you wear out valve stems and guides much quicker, there is not as much metal contact for cooling and the psi of force on the wear surfaces increases (same force less wear surface).

 Remember one compression test is only a snapshot of what is happening in your engine. You need more than one and other information to develope a trend on what your engine is really doing.

 I would go fly your airplane at high power settings for one hour and come back and take another compression test. Then you might try some of the tricks mentioned here ( I have never had much luck with short cuts). If it was my airplane I would pull the cyl. and fix the problem and not screw around with wobble test, staking the valve, or trying to lap a valve without pulling the cyl. A gasket set is less than $20 bucks and you don't have to replace the rings if there are serviceable and you don't hone the cyl..

 I have repaired/overhauled over a 1000 Continental and Lycoming cylinders and it does not take very long to R&R a cyl.

Swaid Rahn

A&P,I.A.,ATP

LIVP-T(Walter)

LIVP-T(Garrett under construction)