Updated 21-XI-2011
Sodium Vapour
Introduction
Spectral Properties
Lamp Technology
Vapour Pressure
Current Density
Gas Filling
Glass
Electrodes
Sodium Migration
Failure Mechanisms
Lamp Designs
Low Voltage Style
     Compton's Lamp
     Philora DC
     GE NA-9
High Voltage Style
     Philora AC
     SO/H U-Tube
     SOI/H Integral
     SOX/H Coated
     SLI/H Linear
Self-Starting Style
     Double Ended
     Single Ended
Control Gear
Series Operation
Autoleak Reactance
Ballast-Ignitor System
High Frequency Electronic
References
Literature

Electrodes

The electrode construction of the low pressure sodium lamp is similar to that of the low pressure mercury fluorescent lamp, although somewhat heavier to handle the higher discharge currents. The electrode is a beehive shaped triple coil of black tungsten wire, which acts as a hollow cathode so as to reduce the lamp striking voltage. It is coated with a layer of emissive materials which improves its thermionic efficiency. The method of sealing into the discharge tube is quite unique - a special design is required on account of the aggressive behaviour of hot sodium vapour towards the glass-to-metal seals which present a weak spot in the lamp construction. The arrangement of a typical low pressure sodium electrode assembly is illustrated in Figure S14.
Figure S14 - The LPS Electrode Assembly


The Emissive Coating
To reduce the energy losses associated with injecting a flow of electrons into the discharge space, and thereby reduce the temperatures and thermal losses in the vicinity of the electrodes, they are coated with a selection of electron-emissive substances. The emitter is most commonly of the triple carbonate type, and consists of a mixture of barium, strontium and calcium carbonates suspended in an organic solvent. The tungsten coils are dipped into the suspension and an electrophoretic process draws a precise amount of emitter into the spaces between the coils.

The emitter must be thermally activated before the lamp is sealed, and this is effected by electrically heating the cathodes under vacuum. The carbonates decompose and are reduced to the metal oxides, evolving carbon dioxide gas in the process:

            BaCO3 / SrCO3 / CaCO3  ----->  BaO / SrO / CaO  + CO2

It is important not to pump the evolved gases away immediately and to continue activation a little while longer. Some of the carbon dioxide then reacts with tungsten metal in the cathode to form tungstic oxide, and this in turn reacts with the metal oxides to produce the metal tungstates which forms a more durable emitter:

            3CO2  +  W  ----->  WO3  + CO

Once activated, the carbonaceous gases must be pumped away as their presence in a finished lamp would lead to short lifetime, low efficacy, and high striking voltages. The discharge tubes are then filled with pure neon and a discharge is struck while still on the exhaust machine, using special ballasts that provide a higher than normal open circuit voltage. This cleans the glass and electrode assemblies by ion bombardment, causing further impurities to be released. These are then pumped away, and before its removal from the exhaust machine each discharge tube receives its dose of liquid sodium, plus the final fill gas. Once activated the cathodes must not be exposed to air or moisture because this will poison them, and could result in premature lamp failure.

During lamp operation, particularly when starting, some emitter is thrown off the cathodes. When there is no emitter left, the lamp will not strike so easily, and this marks the end of lamp life. Alternatively, if one electrode loses its emitter before the other, the lamp could rectify (it converts the alternating current supply to direct current). During rectification large DC currents flow which can damage the ballast's windings. It is for this reason that one of the outer lead wires in each lamp is thinner, designed to fuse if the lamp rectifies, and thus saving the ballast.

It is important to ensure that the correct amount of emitter is deposited on the cathodes. If there is not enough then it will be consumed too rapidly and the lamp will have a short life. If there is too much then it can flake off in large pieces and there is some evidence to suggest that this also reduces lamp life and leads to premature blackening of the tube ends. The emitter coating should be over the tungsten coil only, and must not extend down over the clamp to the nickel lead-wires, if present.


Sodium-Resistant Seals
The point where the metal wires of the electrode assembly are pinch-sealed into the discharge tube glass represents a weak spot in the lamp construction, and if measures are not taken to keep the sodium away from the seals, they will be attacked resulting in premature lamp failure.

The first level of protection is to seal a short length of sodium-resistant borate glass over each of the metallic lead wires. 2-ply glass is again used here, but unlike the glass used for the discharge tube, this time the borate layer is present on the outer surface of the tube.

Naturally this borate coating offers good protection along the length of the glass-sleeved region, but it does not extend right over the ends of the sleeved area, where the metal wire penetrates the glass. Sodium must be kept away from this exposed area, and it has been found particularly effective to seal a small tubular bead of magnesium oxide into the glass at this location. This have been found to be entirely satisfactory in keeping liquid sodium from coming into contact with the glass-metal seal. The temperature inside the ceramic bead is a little higher than the rest of the lamp during operation, and it also takes longer to cool down after switching off. The slightly higher temperatures it maintains at this spot are effective in preventing sodium condensation here.

Since about 1985, a slightly improved seal has been employed in Philips lamps. It was first employed in the Belgian made SOX-E types, and about a decade later was extended to the rest of the range after all SOX production had been relocated to Britain. In this construction, pure borate glass sleeving is employed - it is not a 2-ply glass tube. Being composed of a single glass type, its protection is extended right over the ends of the sleeved region up to and in contact with the metal lead-wire. The use of a magnesia bead is therefore no longer necessary.

An interesting further improvement of the Philips seals is the use of an iron-nickel-cobalt alloy for the lead-in wires, instead of the classic copper-sheathed nickel-iron (dumet) wires that are employed in competitor's SOX and indeed all other lamps fabricated in soft glass. This offers a small decrease in the occurrence of seal leaks, which are a rare phenomenon that sometimes affects lamps having dumet seals. Occasionally if the copper sheath has not been sufficiently well brazed to the nickel-iron core, tiny leakage pathways may exist. Since the Fe-Ni-Co wire is made from a single alloy, such kind of leakage problems are entirely eliminated.

The two kinds of seals can be identified by the colour of the metallic wire passing through the glass. Dumet lamps always show a copper colour appearance in low pressure sodium lamps, whereas the Fernico type seals show a green-grey colouration.