I built two secondary coils: Small coil D25.4/H119 (Diameter=25.4mm and Height=119mm) and Medium coil D73/H228. I experimented with the spark gap, and observed that a series spark gap is better in terms of power handling, noise, and spark discharge length. The capacitance is very important. It should be able to pump a high current, so a low ESR and a low inductance pulse type capacitor is required.
I tested this coil with two different primary caps. First I constructed a water filled jar cap using aluminum foil as shown in Fig. 3, and measured its capacitance as 352pF. The estimated voltage rating was about 20KV. I run a
bunch of tests by varying the supply voltage, changing the spark gap configuration, and putting different discharge terminals. A series spark-gap configuration involving four or more gaps gives the best performance.
However, the total gap spacing must not be larger than 5-6mm to protect the primary capacitor. A series arrangement is quieter. The streamers emanating from the discharge terminal are quite visible in the dark. And I measured a spark discharge of about 3-3.5cm using a grounded rod. The estimated voltage output was about 70KV. The coil and the streamers are shown in Fig. 2 and 3, respectively.
Then I tested the same secondary using a 2.53nF/16KV MMKP array as shown. Primary inductor was replaced too. Since the cap has been increased, a much lower primary inductance was needed. The spark length was about 4.5-5 cm. However, the spark gap firing rate has been reduced. A higher current is needed from the HV transformer.. By 4beowulf7 - firstname.lastname@example.org