IN PRINCIPLE the cold compression in a model I/C engine can be calculated from the total cylinder volume and the stroke of the piston in the cylinder. But in practice this pressure can be lower because of leakages along the piston and/or leakages at any other place in the system such as the valves, cylinder head gasket, spark plug, etc. Also the compression can be made too high and many small engines can suffer as a result. It is my experience that for small model engines the best compression is anywhere between 3 and 5 atm.
It is important to measure the real compression pressure in the cylinder mainly to prevent problems. Pressure in the cylinder of an I/C engine is very dynamic but, in fact, it is only important to know the maximum pressure when the piston is at its TDP when the ignition occurs.

Of course, air-pressure gauges exists but most model builders don't have them, or they are not suitable for this purpose. Furthermore, these kinds of gauges are mostly mechanical dial instruments in which the needle pointer will swing violently back and forth due to the strongly alternating pressures in the cylinder when the engine is turning over, for instance using a hand drill. So these kinds of instruments are not practical for measuring the exact maximum compression.
So, I made an extremely simple but effective little device to measure the maximum compression in the cylinder in a very easy, but fairly accurate, way. It is so simple that it will not be a problem for any model builder to make this little device for themselves.

As shown in figure 1 the system consists of a little ball valve that must be screwed in the cylinder head instead of the spark plug and is connected with a rubber hose to a jam jar with some water in it. The spring pressure on the steel ball can be varied with a screwable pivot on top of the ball valve. At the moment the compression pressure (P) below the ball is equal or a fraction higher than the spring pressure the ball will be lifted from its seat, causing sudden air-bubbles in the jam jar.

The compression pressure that causes this effect can be derived from the measure "D" if the relation between "D" and the pressure that lifts the ball is determined in advance. This linear relation can be determined once, and simply, by connecting the ball valve to a compressor with adjustable and readable static air pressures. Measuring the distance "D" with a caliper at different adjusted air pressures when the ball is just lifted, one can make a table or graph from which the gas pressure can be read at a certain values of "D". Figure 2 shows such a relation, valid for my ball valve.

The work-out
Because the ball valve must be screwed in the cylinder head the thread must be the same as that of the spark plug. It is, of course, also possible to make a reducing piece in case one has to deal  with different thread types. In all cases one must take care for good sealing with rubber O-rings.
The global dimensions of the ball valve will be about the same as the spark plug; but somewhat longer. The dimensions of the ball and the spring are more or less given because they must fit in the ball valve housing.

I used the standard steel (bicycle) ball of 3/16in. dia. The spring has a 4mm inner diameter and with the wire diameter of 0.6mm the outer diameter is about 5.2mm. The extended length of the spring is about 35mm. The maximum pressure of this spring is about 1 kgf when it is totally depressed. One can use a slightly different spring as long as the spring pressure is in the same order of magnitude.
The force below the ball equals PxO, where P is the compression pressure and O is the cross section of the ball seat. I wanted a measuring range of about 2 to 6 atm and a simple calculation showed me that the seat diameter then must be 3.5mm, using my available spring.

Originally I only made a bore in the brass housing on which the ball rested. But, especially with low spring pressures, this didn't seal 100% causing little and inconvenient bubbles in the jam jar while the ball was not lifted by the gas pressure below the ball. Therefore, I changed this by using a rubber O-ring for the ball seat. This O-ring has a 9mm outer diameter and a hole diameter of 3mm. The ball rests on that, on the right diameter of about 3.5mm. The graph in figure 2 of the calibration proves that this theory was according to the practice. For this calibration I used my refrigerator compressor that can make a maximum static pressure of 6 atm and on which I connected a manometer from a bicycle pump.

I made the bore below the ball very small (1.5mm) to add an insignificantly small volume to the cylinder content.
One could fit a kind of pointer on the pivot to indicate the stroke of the pivot with some scale division on the ball valve housing, but making

such a scale is not that easy and measuring the dimension "D" with a caliper is easier and more accurate in my opinion.
The jam jar half filled with water with the central tube and a vent hole in the lid is not only extremely simple but also a perfect way to indicate the moment the ball is lifted from its seat. Of course, the dimensions of the jam jar are not important.

Measuring procedure
Remove the spark plug and screw in the ball valve instead. Connect the ball valve to the jam jar with a rubber hose. Turn the engine with a hand drill and screw in the pivot on top of the ball valve until no bubbles are released in the jam jar; then the compression pressure is lower than the spring pressure on the ball. Gradually screw out the pivot (with the engine turning) until suddenly bubbles start to appear in the jam jar.

That is the point where the compression pressure just is lifting the ball. Measure the dimension "D" and read the corresponding pressure on the calibration graph. This gives the maximum compression pressure in the cylinder.
See also the video here: http://www.youtube.com/watch?v=KGv4b-qjT4s&feature=player_embedded

Some results

I measured the compression of all my IC engines with this little device. They vary between 2 and 4 atm which was in line with my expectations. But the compression of one engine was even more than 6 atm. That is the 4-stroke engine with rotary valves with which I have had some troubles from time to time; when the valve discs are pushed away from each other making the engine unreliable. One has to realize that the combustion pressure is about 4.5 times higher than the cold compression, so then we are speaking about peak pressures of 20 to 30 atm! In fact, this problem was the reason why I developed this pressure measuring device. Maybe I now can make this engine more reliable by reducing the cold compression to a more acceptable level; we'll see.

>> ED’s Es
>> SUPER SUPPLIERShttp://web.me.com/david_carpenter/Site_4/News_and_Editorial.htmlTerrier_4.htmlKrokodil_loco.htmlhttp://web.me.com/david_carpenter/Site_3/Eds_Es/Eds_Es.htmlGreat_Models.htmlFeatures.htmlModels_to_Build.htmlClubs.htmlsuper_suppliers.htmlshapeimage_3_link_0shapeimage_3_link_1shapeimage_3_link_2shapeimage_3_link_3shapeimage_3_link_4shapeimage_3_link_5shapeimage_3_link_6shapeimage_3_link_7shapeimage_3_link_8shapeimage_3_link_9

Figure 1

Figure 2