Spark testing is a simple method to globally determine the main constituents of a sample of cast iron, carbon steel or alloyed steel. The method can also give information about the heat treatment the sample underwent (such as annealing or hardening). The test is simplicity itself. The piece of steel to be tested is simply held to a grinding stone and the pattern and colour of the sparks is observed.
The grinding stone to be used is an ordinary grey type (alumina) of medium grain and hardness. Material removal is of course not the aim of this test. On the contrary, it is essential to fully develop the sparks. For easiest observation, the sparks should not exceed a length of say 1 metre. Therefore, circumferential speed of the grinding stone should not be chosen too high. About 20 m/s is best. This would result in 1,900 rpm for a 20 cm grinding stone, which is quite low. It may be better to choose a small stone, say 12.5 cm, to run at 3,000 rpm.
It is usual, when testing an unknown sample, to alternate the sample with standard bars of known specification. This helps to quickly identify the sample. Such standard bars can of course be made from whatever piece of material comes to hand in the workshop, as long as its specification and treatment are known and documented. The only requirement is, that the standard bars - as well as the sample bar to be tested - have (roughly) equal size and shape. Most commonly, the standard bars and the test bars are made square section and rather short length, for easy handling. In the well-equipped shop, the standard bars are identified by a punch mark and arranged neatly in a small cabinet.
During testing, the standard bars and the sample are in turn held against the grinding stone. They are pressed to the stone with equal force and in the same manner. The level of lighting is kept low during the tests, to better observe the colours of the sparks. The grinding stone has to be kept clean by regularly dressing it. This prevents faulty analysis due to contamination.
Distinguishing spark patterns to look for in identification
The main rays are either continuous or dotted. Along their path, they usually fork. At the end of their path, they break up into multiple rays, two, three or a larger number. Along a single ray, "explosions" may occur. After forking, the separate rays may continue straight-lined, or they may start to curve. After breaking up, the shape of the multiple rays can take a number of basic patterns. Some of these patterns take their names from the martial arts, it seems: bludgeons, spears and lances. Others are named more ordinary, like clusters and drops. The illustration on the right shows all the various patterns mentioned.
From pattern to identification - some general traits
The menubar lists a number of typical ferrous metals. Clicking them shows the pattern they generate on the far right side of this page.
To determine whether a carbon steel is hardened or soft (annealed), look at the intensity and density of the pattern. Soft steels produce less rays, of lesser intensity, than hardened steels. The spark pattern itself does not depend on the hardness of the steel.
It should however be noted that steels alloyed with manganese show markedly increased spark activity, both in number, thickness and intensity of the rays; while steels alloyed with chromium show the opposite effect, i.e., less activity. Nevertheless, it is not difficult to distinguish them from oridinary carbon steels, as these show a multitude of explosions, which the manganese and chromium alloyed steels do not. The higher the carbon in the carbon steel, the more vivid the explosions are. Steels alloyed with tungsten, finally, show a markedly red colour.