Aside from motorcycles I also work on airplanes, the kind with reciprocating engines. To test compression the propeller is rotated (with the top spark plugs out) until you get to TDC. At this point you install a spark plug adapter and connect the differential compression tester. This gizmo has two pressure gauges on either side of an .040 orifice, you slowly apply pressure to the tester (while holding the prop) and adjust the valve for 80 PSI on the hose side, then observe the gauge on the cylinder side, if you have a good one the reading on the gauge will be within 10% of the hose side.
I really don't see how useful a single gauge "differential" compression tester intended for advanced shade tree mechanics are going to be of too much help.
Cars/bikes, the idea here is to crank the engine while holding the throttle wide open. The tester is a simple gauge with a memory that reads the maximum pressure developed while cranking the engine around a few times until the highest pressure is noted and it's not going any higher. (*gasp*) Some gauges have a rubber cone you hold in the spark plug hole with your thumb, not so good, the better ones have a hose and a spark plug adapter, better. Compare with the manufacturers specification, if it's above the magic number you are good to go. If you do all four (or more) of them and the readings differ from jug to jug this shows a problem.On a healthy engine the differential from cylinder to cylinder should all be within 10% On a high mileage engine they can be as high as 30% and it will still run okay. Anything more than that not so good.
A new way is to use a special tool that is a current clamp on the starter motor lead plus a sensor on the #1 spark plug, pull all the plug wires, you don't want the engine to start. As the engine cranks various amounts of juice will be drawn by the motor depending on where the engine is in the cycle (compression takes the most) and comparing the graph. With a good 4 cyl. engine you'll see something like 210 - 205 -210 - 75 amps. The last cylinder in the firing order only drew 75 amps, showing that this cylinder was exhibiting low compression because the starter didn't need as much current to push through compression on that one . The first time I saw this method used was with my old Ford pickup that had developed a poor idle, This was the Ford approved method to testing. Sure enough I had a leaking valve that didn't show up with my conventional shade tree methods.
If you know what you are listening for you can spot a low jug as the engine is cranking. Very handy junk yard technique when selecting an engine that's on the ground.
*Now for something you'll really like!*
Low compression on your K-Bike, first pull the valve cover and check the valve lash, if you have developed a valve issue you'll likely find no valve clearance (burned/broken valve), or if the spring is broken it will be vary obvious as there will be huge clearance. On burned valve, if the valve lash is allowed to degrade to "zero" the valve will be held open more and more, until the flame front whooshes (that's a technical term) past the valve head causing it to become severely overheated and melt. On engines of yesteryear it was not uncommon to see erosion on the valve seat that allows leakage resulting in eventual Barbeqvalve syndrome.
So, that's what I know...
Da Bee - AKA John.
Did this make any sense??