Mercury arc rectifiers
Mercury arc rectifiers can be constructed to handle powers of hundreds of kilowatts. They are effectively mercury vapour rectifiers, and the mercury ions neutralise the space-charge effect resulting in a low voltage drop.
Towards end of 19th century the mercury vapour lamp was invented. It was found to act as a rectifier. When Peter Cooper-Hewitt took up the manufacture of such lamps on a commercial scale, it became possible to make rectifiers in useful quantities. Cooper-Hewitt also established the lines along which the steel tank rectifier would later be developed, and also indicated the possibility of a control grid. In 1908 a design was put forward for a steel tank rectifier by Cooper-Hewitt. Early designs followed similar construction as the glass types.
Early glass mercury arc rectifiers had problems associated with impurities and with foreign gasses, but modern bulbs use chemically pure mercury and graphite electrodes together with superior methods of baking-out during evacuation. Glass bulb rectifiers have a large bulb to condense the mercury vapour. The size of this bulb is due to the poor heat conduction of glass. Steel tank construction eliminates the need for this bulb structure and so the anodes, which had to be held out of the way in the glass bulb types, could be held within the main part of the rectifier. Thus steel tank rectifiers do not now resemble the shape of the glass types.
Seals were a problem, as it was difficult to make a seal that would cope with the large diameter wire needed to carry a high current. Early valves used platinum wire and lead glass, with several individual wires making up one connection. A problem with this approach is that it was hard to match the wire resistances and thus one wire would carry more current than the others, leading to failure. Later the wire was replaced by dumet wire - a nickel-steel core coated with copper. This was sufficient for small diameter wires, but with larger diameters the stresses associated with the expansion of the metal cracked the seals. Three small wires used in parallel make 50A valves possible. Later, glass was developed with a coefficient of expansion suitable for sealing with molybdenum. A 5mm diameter molybdenum rod was capable of being sealed, making it possible to construct 6-phase 250A valves.
As early as 1905 attempts were made to make iron-tank mercury arc rectifiers. It was seen as a device that could handle large currents but it was realized that developments in glass could only handle so much current. Early iron-tank rectifiers followed the basic form of the glass types except they also used water to cool the condensing chamber. Seals are a problem with such designs, and due to this they are usually connected to vacuum pumps. The cathode of an iron-tank rectifier must be insulated. There is no tendancy for an arc to strike up to the metal body, but if the metal body and cathode were at the same potential it is possible that a cathode spot may form on mercury droplets running back from the condensing chamber. If this were to happen, once the droplet was vapourised it is feasible the arc would remain in contact with the metal wall and either cause the release of gas from the metal or actually penetrate it.
In a mercury arc rectifier, the cathode is a pool of mercury, and the source of electrons is a bright spot (cathode spot) formed by an arc. This moves about on the surface of the mercury pool.
Mercury arc rectifiers require at least two anodes. Typically there are two, three, six or twelve. If there were just one anode, the arc would go out when the voltage on the anode swung to negative and would need to be restarted. With two anodes, one will take up the arc as the other gives it up. To get round this problem larger devices have auxilliary (or excitation) anodes which maintain the arc when the main anodes are idle. This excitation arc maintains the cathode spot ready for take-up by the main anodes.
To start a mercury arc rectifier an arc is struck between the cathode and an ignition electrode, and then this is maintained by excitation (or auxiliary) anodes, or by the main anodes, depending on the type of rectifier. In order to strike the arc, the ignition electrode has to first make contact with the cathode pool, and then as it is withdrawn an arc is formed. Small mercury arc rectifiers, e.g. those used in battery chargers are actually tilted to enable the ignition electrode to contact the mercury pool. Larger ones have an electrode that can be drawn into the pool by an external solenoid acting on a magnet attached to the carrying arm.
The mercury vapourised at the surface of the cathode rises up into the condensation chamber, where it cools to liquid form and runs back into the cathode pool. It is this which is the limiting factor for glass bulb devices. Glass is a poor thermal conductor, and thus the bulb needs to be very large in order to allow the mercury to condense. Iron (or steel) tank rectifiers do not have this limitation as they can incorporate water cooling. Thus steel tank rectifiers can handle massive currents.
In general the main anodes of a glass bulb are straight up to about 200V. At 500V+ they are longer and contain one or more right-angle bends. The 90 degree bends in anode arms protect idle anodes from bombardment from positive ions. Although originally following similar construction designs to their glass counterparts, steel tank rectifiers do not need to be of the same form and typically the anodes are held vertically straight up from the cathode.
Control over the load current can be achieved using control grids, if fitted. Each grid can control the point at which its anode takes up the arc.
A typical 3-phase 250A glass enveloped device is about 3 feet long and about a foot wide at the widest part of the bulb.
They found various uses, from small devices used in battery chargers, to installations running electric tramways.
This file was last modified
23:19:17, Friday November 02, 2012