I would like to know at what percentage of theoretical spark energy does sparks occur in actual conditions and how much it does vary. Does the spark energy variation follow a normal distribution?
The following manuscript may be helpful. Those manuscripts are not dealing with measuring the spark energy; rather dealing with influence of spark energy. If you will need it, you can consider
A review on influence of electrical process parameters in EDM process
T Muthuramalingam, B Mohan
Archives of Civil and Mechanical Engineering 15 (1), 87-94
Influence of discharge current pulse on machinability in electrical discharge machining
T Muthuramalingam, B Mohan
Materials and Manufacturing Processes 28 (4), 375-380
Performance analysis of iso current pulse generator on machining characteristics in EDM process
T Muthuramalingam, B Mohan
Archives of Civil and Mechanical Engineering 14 (3), 383-390
Monitoring and fuzzy control approach for efficient electrical discharge machining process
T Muthuramalingam, B Mohan, A Rajadurai, D Saravanakumar
Materials and Manufacturing Processes 29 (3), 281-286
Experimental investigation of iso energy pulse generator on performance measures in EDM
T Muthuramalingam, B Mohan, A Rajadurai, MDA Arul Prakash
Materials and Manufacturing Processes 28 (10), 1137-1142
SCHWADE, Max. Fundamental Analysis of High Frequent Electrical Process Signals for Advanced Technology Developments in W-EDM. Procedia CIRP, 2014, 14. Jg., S. 436-441.
The spark energy can be calculated relatively easily from the i (t) and u (t). To analyze which discharges are sparks and which are not called them some papers. If, for a high number of discharges, they apply this via the spark energy, then it seems to me very unlikely that this will be a normal or Gaussian distribution. There is not even a correlation about the fact that the same discharge energy must always be a spark. Since this distinction may include e.g. (Spark, arc, short circuit), the discharge energy is only a conditional analysis factor.
It is much more interesting, but it is very expensive to determine the correct distribution of the spark energy for a given electrode arrangement on the cathode case, the anode case and the discharge channel.