SPARK GAP SWITCHES FOR RADAR 569 



type gap, two or more of which could be operated in series. The first unit 

 type gap had an aluminum cathode and a hydrogen-argon gas atmosphere. 

 Later, under the urge for higher peak powers, mercury cathode gaps were 

 developed. Details of this study and development will be discussed in this 

 section. 



(a) Triggering Gaps in Series 



An alternative to a rotary gap in which the timing of spark breakdown is 

 controlled mechanically was the use of a fixed gap, the breakdown of which 

 is controlled electrically. One method of accomplishing this was to use a 

 third electrode to which an impulse voltage was applied periodically at 

 double the frequency of the resonant charging circuit. This voltage breaks 

 down one gap with a discharge of energy furnished by the trigger circuit, 

 which in turn causes a breakdown of the main gap, either through a modifi- 

 cation of the field in this gap or through the addition of ions which reduce its 

 breakdown voltage. This type of gap, however, required a strong air 

 blast to de-ionize the gaps and, because of this, its use obviously presented 

 no great improvement over the rotary gap. It was well known that the 

 rate of de-ioni^ation is greater the smaller the gap, so an attempt was made to 

 trigger without air blast a number of smaller gaps which when connected in 

 series would withstand the full switch voltage as employed in the rotary gap. 

 The arrangement used was that shown in Fig. 4. Six tungsten pins, 3 mm 

 in diameter, were mounted with their axes parallel and spaced to give five 

 0.5 mm gaps. The switch voltage was divided by means of equal high 

 resistances connected across the gaps, and a highly damped bi-directional 

 trigger pulse was applied to the four middle pins through capacity coupling 

 as shown. Corona points were also connected in such a way that the cathode 

 of each of the gaps is irradiated in order to reduce the spark delay time. 

 By an appropriate adjustment of the circuit elements it was demonstrated 

 that this series of gaps could be broken down by the trigger pulse and de-ion- 

 ized with sufficient rapidity so that no air blast was recjuired. 



Although no attempt will be made here to elucidate the detailed steps in 

 the triggering of the five gaps just described, we can get a qualitative idea 

 of the process by considering a simple two-gap and three-gap circuit which, it 

 turned out, was all that was required for the various applications of fixed 

 gaps as they were eventually developed. 



In the two-gap circuit. Fig. 5 (a), if the first half cycle of the trigger pulse 

 and the switch voltage are both positive, gap 1 will break down when the 

 potential at the mid point, due to the sum of the switch voltage and that of the 

 trigger, is equal to the gap breakdown voltage, which for the moment we 

 shall consider as singly valued. This effectively shorts gap 1 and throws the 

 full switch voltage across gap 2 which in turn will break down provided this 



