Instability of arc current, voltage and light emission.

The instabilities of discharge parameters such as current and voltage are related to cathode discharge processes (M B Bochkarev A M Murzakaev 1998), hence the duration of these instabilities can give an idea of the time scale of the ectons. The possibilities of digital technology allow us to make an attempt to find patterns in the fluctuations of the discharge parameters by means of mathematical processing of signals from the arc. It is believed that for the analysis of non-stationary processes, much more than Fourier decomposition, the now fashionable wavelet method of signal decomposition is suitable. Wavelet is an elementary single wave (burst) of a certain duration and shape. In its practical sense, wavelet analysis of a signal is a procedure of searching for bursts in a signal by duration (as well as by form) close to the duration (time scale) of the wavelet. By assigning different values to the wavelet duration and analyzing the signal again and again, it is possible to obtain the wavelet spectrum of the signal. We performed a wavelet analysis of the voltage current and arc light waveforms (Fig.6.) at currents close to the threshold for two materials with very different thermal properties W and Cd.

The spectra obtained by us show that the fluctuations of arc parameters are not chaotic, they mainly consist of bursts of about 20 ns duration for W (Fig.7) and 40 ns for Cd (Fig.8).

Fig. 6. Example of waveforms for tungsten cathode

Fig. 7. The wavelet spectrum for a tungsten cathode at a current of 3 A

Fig. 8. Wavelet spectrum for cadmium cathode at a current of 0.1 A