OK Dan, you asked for it.
Here is a Purdue/Stanford EE's method for a leakdown test.
A standard leakdown test measures the cylinder air leakage at 100 psi by measuring the differential pressure drop across a small orifice as the leaking air passes through. This method measures leakage directly using PV=nRT which you learned in Physics class.
Calculate the amount of air in your compressor tank using PV=nRT at the known pressure when the tank is 'full'. I was using my small compressor which has a 7.5 gallon tank at 103 psi. Adjust for atmospheric pressure if you are not at sea level: I live over 6000' AMSL so I calculated using 12.25 psi as 1 atmosphere, NOT 14.696. That gave me 115.25 psia in my tank or 222.56 liters of air if it were at STP. You can get really precise and add in the amount of air in the hoses, etc., but that is not usually very much. My 50' hose only holds another 0.15 liters of air - not worth considering.
For each cylinder, time the discharge of air into the cylinder (with the compressor unplugged so it won't cycle back on) and record time and tank pressure. For each pressure you can calculate STP liters of air in the tank so you will know how much air has discharged through the cylinder vs time. You will essentially be measuring the time derivative of the amount of air in the tank in liters/sec.
Different brands of leakage testers use different sized orifices so the percentage numbers they give are different for the same engine. Since they use 100 psi in, they all call the pressure drop in psi across their orifice as the 'percent leakage'; i.e. a 5 psi drop is 5%, more leakage producing a 10 psi drop would be 10% leakage. The best I can determine a 20 liter/sec leakage rate will be read as anywhere from a 2.9 to 11% leakage from various commercial leakdown testers which gives you an idea of what is good vs. bad. Less leakage is always better but anything in the 10-20 liters/sec is not all that bad.
My various engines that I have measured take anywhere from 39 seconds to 3+ minutes to discharge my 7.5 gallon tank from 103 to 70 psi. 39 seconds is God awful with all broken rings. 3+ minutes is nigh onto excellent. I actually calculated the exponential pressure drop in the tank vs time and leak rate so all I had to do was measure the time from 103 to 70 psi (when my compressor goes 'click' to start up again) and not have to measure any other pressures or times. Both pressures are determined by the hardware; all I have to do is measure stopwatch time between those two points and not even look at the pressure gauge. The Excel sheet then tells me what kind of a leak rate was required to discharge the tank in that amount of time. For really leaky cylinders there isn't any time to measure much else except the 103 to 70 times.
A big huge Excel sheet is useful to do all the calculations of how much air is in the tank vs pressure, etc. It sounds complicated but not really. Once you have done it a few times you realize that the time to leak from 103 to 70 is as good a metric as any other and you will end up comparing your known good engine times to the one you are testing. If it goes to 70psi in 10 seconds you should, or may already be walking. If it takes 5 minutes your engine is just fine, etc. The Excel sheet will give you the liters/sec leak rate if you need to know more accurately what you have measured.