 |
Spark erosion has a completely
different effect on working material than customary methods of processing. The electrical
spark hitting the work piece heats up the outer layer of the steel so much (about 10,000° C) that the material evaporates.
The metal gases formed then condense in the dielectric, usually in the form of hollow
balls, open on one side and having a sharp edge. In the work piece itself depressions,
shaped like craters, are formed. How great is the danger for the working material to be so
unfavourably affected on the surface, that the serviceability of the tool suffers? And
what about tool life, resistance to wear and buffability? Figures 1, 2 and 3 show surface
roughness, electrode wear and metal removal in relation to the firing period.
|
Apart from metal removal, surface
roughness and electrode wear, the effect on the surface quality of the working material is
of utmost importance. In most cases it was shown that there was no effect on the
functioning of the tool. In some cases, e.g. in a cutting tool, it even became more
resistant to wear, in others, however, tools broke prematurely. All changes that could be
detected were due to the high temperatures that were produced on the rim. In this rim the
structure hardness, state of stresses and carbon content of the steel are influenced. Fig.
4 shows a section of a surface that has been roughened down by spark erosion, showing the
various structural changes, which are typical of such a rim.
 |
The melted zone (Fig. 5) shows
clearly that it has solidified very quickly. Columnated crystalls have grown vertically up
out of the metal surface during solidification. A crack that has formed in this layer runs
inward along the line of crystals. The melted layer is usually about 15-30 um thick after normal rough work. In the hardened zone the
temperature rose above that needed for hardening. A hard and brittle martensite has
formed. In the annealed zone the temperature was not so high as to harden the
steel. It has only been tempered. Underneath is the unaffected core. The thickness of the
various layers appears to be unrelated to the type of steel used and the electrode
material. However, there is a very clear difference between hardened and softened
materials. In softened steel the layers are thinner and there are fewer cracks. The
brittle, hardened layer is almost non-existent. During rough work the thickness of the
layers varies much more than during finishing. The longer the firing period, the thicker
the melted and hardened layers become. Further research has shown that the strength of
current has basically the same effect as the length of the firing period. Steel with a
high carbon content gets the most cracks. Steel with a low carbon'content only develops
few cracks in the melted layer. About 20% of the cracks extend into the hardened zone and
only a few reach the core. In the core there are seldom cracks longer than 10 um. These cracks in the core are usually found in high alloy
tool steel and in high alloy high-speed steel.
 |
 |
 |
 Fig. 6a Thicknesses of layers and amount of cracks in the rim after spark erosion on hardened (52 HRC) UHB Orvar 2 microdized at different lengths of firing period |
 Fig. 6b The same after spark erosion on UHB Orvar 2 microdized which has been annealed |
The cracks are caused by stresses, which result from the repeated, rapid chilling
of the work material by the dielectric, as well as from the differences in volume between
the various structural parts in the different layers. If erosion is properly done and
includes the final finishing process, the surface errors that result from rough work can
largely be corrected. Where finishing is not possible, the following procedures may be
used:
a) stressfree annealing at about 15° C less than before.
This decreases the hardness of the surface without influencing the core.
b) softening and renewed hardening and annealing leads to an almost complete
restoration of the structure (cracks however remain)
c) grinding or scouring removes the surface structure together with the cracks. The
rate of cut is important here, and should be about 5-10,um
In summary it may be said that the structural faults caused by rough work can be corrected
during the normal process of spark erosion, which includes rough work and finishing. A
certain amount of structural changes will, of course, always remain. However, in
most cases they are of little importance. There are even instances in which the great
hardness of the hardened layer improves the tool's resistance to wear. In others, the
craters on the surface of the work piece provide a better hold for lubricants, which also
increases the service life of the tool.