An interesting experiment to demonstrate the various hardness values would be to take a piece of tool steel (such as D2) after it has been subjected to EDM, stop-grind the surface in three or more different locations, conduct very light-load surface-hardness checks and note the hardness variations.

Another experimental method would be to cut an EDM sample (using an abrasive cut-off machine with a strong coolant flow) and then pre-grind and polish the cut sample. This would be followed by a cross-sectional hardness traverse from the surface through to the core. Then note the hardness values.

You will find that the hardness value will be in the range of 63-66 HRC (and possibly higher) just below the surface. This would be indicative of a region of untempered martensite. Follow the hardness test with an etch using 5% (approximately) solution strength nital, and etch the surface for approximately 4-6 seconds. This should be followed by a wash and rinse (in cold, clean running water), and then the etched surface should be suitably dried. The sample should then be examined under a microscope at 400-500X.
You will observe many different microstructures (including untempered martensite). You will also (probably) see micro-crack propagation in the recast layer (immediate surface). The question is then asked, “What can be done to prevent this?” The answer is very short and sweet! Nothing can be done to prevent its occurrence. It is a process chance that you take.

The only thing that can be done to minimize the risk of crack propagation, either from the recast layer or the untempered martensite, is to temper the steel immediately after the EDM process. The temperature selected should be at least 380°F. The tempering temperature can go as high as 50°F below whatever is the final tempering temperature of the steels’ final heat treatment.

If you don’t have a tempering furnace, simply put the piece (if it is small enough) into an ordinary domestic oven at 380°F and temper it there. While the EDM process is an accepted method of machining, it is a method that will (and does) affect the EDM surface metallurgy.

It would behoove some of the metallurgical research establishments to conduct an investigatory research program into the metallurgical disadvantages and possible advantages of EDM.

The question often arises as to when the EDM process should be conducted – before final heat treatment or immediately after final heat treatment? There is no categorically correct answer to that question, other than to do what works best for you.

Heat treaters, please be aware of the potential problems you face if you heat treat any tool steel that has been machined by the EDM process. A potential surface crack may well be pre-existing, and you accelerate the pre-existing crack by the subsequent heat-treatment procedure. Then you are blamed for cracking the tool! To avoid that surprise, it may behoove the heat treater to conduct at least a dye-penetrant check before proceeding with the heat-treatment process.