In the operation of a TEA CO₂ laser, the main energy storage condenser normally discharges its energy into the load with the help of a fast, high current, high voltage switch, conventionally a spark gap or a thyratron. In the repetitive operation, thyratron is known to be a better choice as spark gaps operate in the arc mode and suffer from recovery problem. Thyratrons, however, are expensive and have limited life. As a result there is a growing interest in replacing these switches by all-solid-state-exciters (ASSE) in conjunction with magnetic pulse compression (MPC) [1]. Although such systems have long life and high degree of reliability, they suffer from low wall plug efficiency and bulkiness [1]. Efforts have, therefore, been expended to eliminate the main discharge switch altogether in the operation of CO₂ lasers [2–4]. In these methods the main discharge condenser is directly connected across the laser electrodes and is charged to a voltage below the self-breakdown level of the laser gas mixture in the inter electrode volume. Automatic switching of this condenser occurs when the inter electrode volume is preconditioned by UV [2] or X-ray [3] photons or electrons from an external source [4]. In all these methods although the main discharge functions without a switch, the preconditioning of the inter-electrode volume is initiated by a switch. We report here the operation of a UV preionised TEA CO₂ laser, which does not require the service of any external switch. The principle of operation of this scheme is described below.

In a typical UV preionised TEA CO₂ laser, the UV photons emanating from the spark channels placed along the length of the discharge precondition the inter-electrode region. In actual operation with a sequential spark type preioniser, all these spark channels are overvolted sequentially leading to their closure with the help of an external spark gap. The spark gap itself operates in arc mode and can be a source of UV radiation when in conduction. Our intention was to make use of the UV radiation emanating from the spark gap itself for preconditioning the inter-electrode volume. We achieved this by segmenting the main spark gap into smaller adjustable gaps (see Fig. 1) and placing this integrated spark array along the length and to one side of the discharge as shown in Fig. 2. A resistance (R) was connected across each of the gaps. We have successfully operated a mini TEA CO₂ laser where this spark array served the dual purpose of a switch as well as a source of UV photons for preconditioning of the inter-electrode volume.

 

Fig. 1. The configuration of the spark array cum switch

Download Full Size | PDF

The laser head comprised of a pair of cylindrical electrodes defining a discharge of cross-section 0.3 cm × 1.1 cm and length 8.0 cm. The electrodes and the spark array were housed in a leak tight Perspex chamber, the two ends of which were O’ring sealed with a gold coated copper mirror (1m ROC) and a 90% reflective ZnSe output mirror that defined the resonator cavity of length 16.5 cm. The output energy and the power of the laser pulse were respectively monitored by a pyroelectric Joule meter (Gentec, Model ED 200) and a fast room temperature detector (Photonic Solutions, Model PEM L3).

The pulser circuit used for the excitation of the laser is shown in Fig. 2. A condenser C was resistively charged to a suitable voltage ‘V’. The resistance ‘R’ ensured that the entire voltage ‘V’ appeared across the first gap of the spark array leading to its closure. The voltage then appears across the second gap resulting in closure of this gap and subsequently of all the remaining gaps in a cascade. The UV photons emanating from these sparks preconditioned the gaseous medium in the inter-electrode volume. Alongside preconditioning, the preionisation current also charged up the spiker capacitor ‘C_sp’ to an elevated voltage, which was impressed across the discharge electrodes. This high voltage impulse caused the breakdown of the preconditioned inter-electrode gap. The inductance ‘l’ provided the required delay between the pre and the spiker discharges. The remaining energy in the condenser ‘C’, the voltage across which was reduced as a result of powering the pre and the spiker discharges, now appeared in the discharge at a rate decided by the value of ‘l’ causing its sustenance and excitation of the active medium.

 

Fig. 2. The laser head, the spark array cum switch and the pulser circuit.

Download Full Size | PDF

The laser was initially operated with a gas mixture of CO₂:N₂:He::1:1:3 and the values of C, C_sp, and l were optimized respectively to 2.7 nF, 300 pF, and 3.3 μH by monitoring the output energy of the laser. While optimizing the gas mixture we found that the operating voltage of the laser too changed. This is because the operating voltage of the laser is determined by the break down voltage of the first gap in the spark array and as the array is exposed to the gas mixture any change in its composition also changed the breakdown voltage and, in turn, the operating voltage of the laser. The most optimized and reliable performance in terms of energy (viz., 38mj @ 7 % electro-optic efficiency) occurred for a gas mixture of CO₂:N₂:He::1:1:2.5 for which the operating voltage was ∼20 kV. The efficiency would be higher if the energy expended in preionisation is accounted for. By virtue of its short cavity length, the laser emission was expected to occur on single longitudinal mode, which was corroborated by the absence of any beating at the cavity round trip time (t_R) in the temporal profile. In order to remove the dependence of the operating voltage on the gas composition, the preioniser cum switch needs to be isolated from the laser head by making use of appropriate windows for coupling the uv radiation into the inter-electrode volume.

One obvious disadvantage of this sequential spark preioniser array is that the entire discharge current flows though the preioniser. This would limit the life of the array in repetitive operation. A possible remedy would be to use a parallel spark array with mutually coupled ballast inductances to achieve simultaneous closure [5] of a number of parallel gaps without the need of an external switch.