The Results

The important spark characteristics (Fig. 2) are color, volume, nature of the spark and length. Note that the length is dependent on the amount of pressure applied to the grinding wheel, so this can be a poor comparison tool if the pressure is not exactly the same for the samples. The grinding wheel must also be dressed frequently to remove any metallic buildup. In general, the following observations apply:2

  • Wrought iron – sparks flow out in straight lines. The tails of the sparks widen out near the end, similar to a leaf.
  • Mild steel – sparks are similar to wrought iron's except they will have tiny forks and their lengths will vary more. The sparks will be white in color.
  • Medium-carbon steel – sparks have more forking than mild steel and a wide variety of spark lengths, with more near the grinding wheel.
  • High-carbon steel – sparks have a bushy spark pattern (lots of forking), which starts at the grinding wheel. The sparks are not as bright as those for medium-carbon steel.
  • Manganese steel – sparks have medium length and fork twice before ending.
  • High-speed steel – sparks are a faint red that “flair” at the tip.
  • 300-series stainless steel – sparks are not as dense as the carbon-steel sparks, do not fork and are orange to straw in color.
  • 310-series stainless steel – sparks are much shorter and thinner than the 300-series sparks. They are red to orange in color and do not fork.
  • 400-series stainless steel – sparks similar to 300-series but are slightly longer and have forks at the ends of the sparks.
  • Cast iron – very short sparks that begin at the grinding wheel.
  • Nickel and cobalt high-temperature alloys – sparks are thin and very short. They are dark-red in color and do not fork.
  • Cemented carbide – sparks under 3 inches, which are dark-red in color and do not fork.
  • Titanium – sparks are a very brilliant, blinding, white color and easily distinguishable from ferrous metals.

The main limitation of spark testing is the inability to positively identify the material, although in the hands of the right person, the writer has seen an individual separate 1038 form 1045 carbon steel! If an exact chemistry is required, chemical analysis in some form or another is needed.




1.      The Ohio University, Spark Testing Laboratory Experiment

2.      Wikipedia (

3.      Pye, David, “Spark Testing: Seeing the Spark,” The Experts Speaks Blog, 2 July 2010