The most accurate, and technically correct, is the "mechanical compression". The mechanical compression is calculated from the swept cylinder area above the exhaust port, and the combustion chamber volume at top dead center. I'll forego the formulas and intricacies of the various volume and dimensional measurements. For the purpose of this article, it's not necessarily important to know how to calculate mechanical ratios, it's only important to understand their level of accuracy. The functional factors involved with maintaining the accuracy of the mechanical compression ratio are, the overall cylinder height (with gasket thickness included), the exhaust port height, and the volume of the dome in the cylinder head (measured in cubic centimeters or "cc"). Professional engine builders go to great lengths to maintain the accuracy of these dimensions in an effort to assure that each engine they construct will have exactly the same compression ratio. Engine builders use this method because it offers unmatched consistency and accuracy.
Indicated compression, using a normal compression gauge, is much easier to measure. Unfortunately the variables in an indicated compression measurement are numerous...very numerous. The potential margin of error caused by these variables is why engine builders don't give much credibility to indicated readings. In fact the room for error is so great that most engine builders just plain don't believe "anybody" who calls on the phone with an indicated reading that they have just taken. The following, offers is a partial explanation of why they feel that way.
GETTING AN ACCURATE (as possible) INDICATED COMPRESSION READING
THE GAUGE - Most engine builders have a drawer in their rollaway dedicated to storing all the "cheap compression gauges" that they have used the poor judgment to buy during their career. However in the very front of that drawer is the Snap-On gauge that gets used regularly. This gauge is preferred not just for it's good accuracy and durability, but rather for one simple design feature. The adaptor hoses, of the Snap-On gauge, that screw into the spark plug threads has a Schrader air fitting at the spark plug tip location. That is, the pressure is sealed off at the face of the dome in the cylinder head, which gives the truest representation of the exact combustion chamber volume. Most other automotive gauges have this air seal fitting mounted in the gauge body, at the end of a 16" hose. This means that the air volume inside that hose (usually about 3-4 cc) is added to the combustion chamber volume during a measurement. The end result of his added volume is a reading that is 20 - 35 psi lower than the true reading. For similar reasons, the tapered rubber "hold - on" type gauges are virtually useless. Besides indicating the added 2 cc of the threaded spark plug hole itself, these gauges are notorious for leaking as well.
THE ENGINE - 1) All accurate readings must be taken from a dead cold engine. A warm engine will yield slightly lower numbers. 2) The exhaust pipe and carburetor must be installed. The restrictions in the carb throat and the back pressures of the exhaust system can affect the readings. 3) Have a full charge on the battery. The speed that the engine is spun, has a significant affect on the indicated reading. Having the charger connected during the test insures maximum starter motor RPM's. 4) Leave a spark plug in the cylinder not being measured. Contrary to what you may think, the engine will turn over slightly faster with the opposing spark plug installed.
THE TEST - 1) Make sure both spark plug caps have spark plugs mounted in them, and those plugs are grounded to the cylinder or head. 2) Hold the throttle wide open to admit the maximum amount of air. 3) Hold the start button down until the needle on the compression gauge is no longer rising. 4) Test both cylinders.
DO NOT squirt any oil into the cylinders to improve ring sealing for the test. The presence of added oil can cause readings 20 - 30 psi above the accurate "normal oil presence" reading.
Ideally, there should be no more than 10 psi difference between cylinders. If the difference is greater than 20 psi, you should consider removing the cylinder head for inspection. If the difference is greater than 30 psi, you should consider removing the cylinder and pistons for inspection.
HOW MUCH INDICATED COMPRESSION IS IDEAL?
The more accurate question is, "What operating temperature is ideal". Each particular engine arrangement has a different ideal indicated compression reading. This happens because compression is only one of the factors that determines the engine's operating temperature. The factors are : peak rpm ability, ignition advance, compression, and gasoline octane level. Because of all these variables, few engine builders will agree on any one specification, however almost all engine builders will agree on certain perimeters. The following scenario will help you to understand why many engine builders are so coy to make a specific recommendation.
JOE'S STORY - Joe is the proud owner of a stock 1992 Kawasaki 650 stand up that he bought brand new. After two years of reliable performance, Joe decides the boat needs more power. Joe's buddy down the street says that increasing the compression on his 650 X2 made a world of difference. Joe is the kind of guy (like many of us) that likes to get the maximum he can with the minimum cost and...without compromising reliability.
Joe goes to a local shop that builds race boats and asks the race mechanic, "What is the most compression I can run in my stock 650...while running on pump gas and not losing any reliability ?" The mechanic says "Oh, about 175 psi". Joe says "You don't understand...I want to get the max out of this thing". The mechanic thinks for a moment, and then tells him " On a totally stock 650, with a good fresh top end, and 92 octane, you could run 190 -195...max!"
With this piece of free, and very reliable advice, Joe takes his boat over to his machinist friend to do the milling work. With a Snap-on gauge in hand, they figure they'll just keep cutting the head surface and squish bands until they reach 195 psi. A few hours later, it's done.
Joe goes riding for the weekend with his pals. Right away, he notices a giant increase in overall acceleration and speed. In the early going, he drives away from his buddy that's riding a brand new 650...cool. But later on in the day, they repeat the drag race. Joe and his pal are dead even. In fact, it seemed like the longer they held it wide open, the slower Joe's 650 became. The next morning, when all the guys first got on the water again, Joe had the fast boat again. But like the day before, as the engine got hotter, it got noticeably slower. Joe and his pals figured that the milled head was pushing the two year old pistons and rings past their limit.
Back home, Joe disassembled the engine and took his cylinder and pistons to his machinist buddy. After some quick measurements they discovered that the bores had .007" of clearance and the pistons were collapsed .003"...obviously his riding buddies were right. With a fresh top end and a .003" clearance bore job, Joe went out for an afternoon of break in riding. Right away the boat had even more bottom end than before, not to mention better throttle response. After two tanks of careful break in, he met up with the guys. It was time to hang this baby wide open and blow his pal's doors off. At the start, he ran off. But about 80yards out, his pal's stock 650 caught up and rode away. As Joe was riding along wide open, noticing how much slower the top end speed seemed, the rear piston seized.
This time, Joe took his entire boat to the race shop to be fixed once and for all. The mechanic called Joe the next day and said, "We found the problem...Your front cylinder had 225 psi compression. You milled the head way too much". Joe responded "I cut it to 195 psi just like you said." The mechanic then reminded Joe that he said 195 psi on a stock boat with a fresh top end. He said "Joe, nobody sets up the compression on a worn top end. Boring a worn top end can easily increase the indicated compression 20 - 30 psi. That's why engine builders always cut heads to a particular cc volume They know which volume will give ideal indicated readings on a fresh cylinder. That same volume will give slightly lower readings on a worn top end, but it won't cause a seizure when the cylinder is bored. When you originally cut your head, the mechanical compression ratio was way too high, but the top end was so worn out it couldn't create enough compression to kill itself. The excessive mechanical compression ratio only caused it to over heat and slow down a little when it got hot. But after the cylinder was bored to the correct clearance, the indicated compression became high enough to cause an instant meltdown instead of simple over heating."
He added, "Joe, the same thing might have happened if you had put on a pipe. Remember you originally asked me what was the max compression you could run on your otherwise stock 650...and I said 195 psi, which is right. But the added rpm's of a bolt on pipe will also cause a lot of additional heat. To keep that heat from killing that engine set up, you'll need to back off the compression to about 180 psi." Joe responded, "Well, that's kind of ridiculous. How is anybody supposed to know what the ideal compression is for their particular engine and bolt on parts etc."
The mechanic replied, "Well, you either do the meltdown point testing on your own, or buy the head modification from someone who has already done it. Our shop only sells a few different engine set-ups or kits for your engine. We know the ideal head volume for those few kits that we sell...and we really don't worry about the rest."
Joe's story is not an uncommon one. Stories like this one is why engine builders often shy away from offering compression information. It has nothing to do with compression ratios being some kind of top secret. It has allot more to do with the engine builder being unsure about your measuring procedure, your bolt on parts, and the amount of wear your engine has. The engine builder understands that even with the best of intentions, his free advice can result in an expensive engine failure. For him it makes better business sense to give no recommendations rather than potentially costly recommendations.