Self-Starting Lamp Developments

Although the high voltage lamps described previously brought numerous advantages over the original low voltage developments, a notable disadvantage of both types in the early years was the complexity of the electrical control gear required to operate them. In addition to the simple ballasting task to control discharge current, means also had to be provided to ignite the discharge.

During the early 1930s a third family of low pressure sodium lamp was developed by German Osram in an attempt to overcome this limitation. The concept was based on the high voltage style with the addition of special features such that the discharge could be initiated at ordinary mains voltages. This greatly simplified the electrical arrangements to permit the use of an ordinary choke-type ballast, in the same fashion as the increasingly popular mercury vapour lamps.

This style of self-starting lamp was manufactured by both Osram and Philips, but its commercial success (or lack thereof) is illustrated by the fact that both companies abandoned its production after a few years. Partly this was owing to the fact that its efficacy was not terribly good, but also because it was an expensive lamp to produce. Meanwhile developments in control gear had allowed a reduction in the cost and complexity of the autoleak-reactance type of transformer which has since become commonplace for the operation of high voltage sodium lamps. Nevertheless the construction of the self-starting lamps is technically interesting and only sparesely documented, so they will be described in detail under this section.

The ignition voltage of the self-starting lamps is reduced below that of mains potential by combining at least the first two of the following four design features. In some lamp types the third and/or fourth mechanisms may be employed as well:

1. The discharge tube diameter is increased and its length decreased by comparison with standard high voltage lamps. Both of these features lead to a lower striking voltage, but bring the drawback of reduced efficacy owing to the fact that the current density of the discharge is higher than the optimum level.
2. One or two glass-sleeved isolated wires are disposed within the discharge tube, running along most of its length except for a short gap usually near one or both electrodes, and sometimes also at the centre of the discharge tube. Before the lamp starts the full mains potential is applied across these gaps and on account of the short distance, the rare gas filling of the discharge tube can be ionised at a relatively low voltage. A high series resistance is provided in parallel with these wires so that once ionisation has taken place, the discharge transfers to the main electrodes.
3. An auxiliary electrode may be provided in the close proximity one of the main electrodes and connected via a second high resistance to the electrode at the opposite end of the discharge tube. The principle of operation is same as the probe electrode in the mercury vapour lamp.
4. Finally, one or more of the main electrodes may be preheated prior to lamp starting, so as to liberate electrons by thermionic emission.

Self-starting LPS lamps were developed in two different formats: double ended and single ended. Each type is described in futher detail on the following pages.