US3227907A

US3227907A - Electric discharge lamp with integral pressure regulator

Info

Publication number: US3227907A
Application number: US248373A
Authority: US
Inventors: Carl J Bernier; Richard A Menelly
Original assignee: Sylvania Electric Products Inc
Current assignee: GTE Sylvania Inc
Priority date: 1962-12-31
Filing date: 1962-12-31
Publication date: 1966-01-04
Anticipated expiration: 1983-01-04

Description

Jan. 4, 1966 c. J. BERNIER ETAL 3,227,907

ELECTRIC DISCHARGE LAMP WITH INTEGRAL PRESSURE REGULATOR Filed Dec. 51, 1962 ERNIER A. MENELLY INVENTOR.

W M, ATTORNEY J. RICHARD United States Patent 3,227,907 ELECTRIC DISCHARGE LAMP WITH INTEGRAL PRESSURE REGULATOR Carl J. Bernier, Beverly, and Richard A. Menelly, Danvers, Mass, assignors to Sylvania Electric Products Inc.,

a corporation of Delaware Filed Dec. 31, 1962, Ser. No. 248,373 4 Claims. (Cl. 313178) This invention relates to electric gaseous discharge lamps and especially to fluorescent lamps of that type. In particular the invention relates to such lamps in which low pressure mercury vapor is a constituent of the gaseous atmosphere in the lamp.

In such devices, the light output and efiiciency depend upon the pressure of the mercury vapor, and generally have their maximum for a vapor pressure corresponding to that at a temperature of 30 C. to 40 C., that is about 3 to microns. The mercury pressure is generally determined by the temperature of the coolest part of the lamp, since the mercury is present in excess of that needed to exist entirely as a vapor and some of the mercury will remain in the liquid state. At very high inputs to the lamp, the coolest spot in the tube may be above 60 0., even 80 to 100 C., and the mercury pressure is so high that the efliciency is poor.

Various ways of cooling portions of the tube to keep the vapor pressure low enough have been proposed. Shields have been used between the electrodes and the ends of the tube, heat sinks have been attached to the tube, and the lamp tube has been increased in size and made with grooves, depressions and the like. It has also been proposed to use an amalgam of mercury with some metal such as cadmium, the vapor pressure over an amalgam being somewhat less than that over the pure mercury.

A more effective method of keeping the mercury pressure at a proper value is to use an alloy of indium and mercury in the tube, or simply to place a coating of indium, in the form of a ring, for example, at the center of the lamp, if the lamp be tubular.

The use of indium is very effective in fixing the mercury vapor pressure after the lamp reaches thermal equilibrium, and metals such as gallium can also be used. But the mercury pressure in the lamp will be extremely low when the lamp is first started, because a considerable time is required for the middle part of the tube, the part where the indium is, to warm up. As a result, the lamp may only emit a third of its normal light output even two or three minutes after being started, and may not emit its full light output until as long a period as 12 minutes has elapsed.

Such a long dim period, is disadvantageous for many purposes. When the lamps are used in factories, for example, it would be inconvenient to wait so long for suflicient light in which to work, and accidents or loss of useful time would occur during the dim period. The lamps could, of course, be turned on 12 minutes before the working staff was expected, but that would add to the lighting bill. The same is true of the use of the lamp in gasoline station lighting and highway lighting.

We have discovered that this long dim period can be substantially eliminated by having an additional indium spot in the lamp at a point of high temperature, such as on one of the auxiliary electrodes, sometimes in the form of wires or flat plates, the latter being sometimes called flags or flag anodes.

Such auxiliary electrodes generally rise quickly in temperature, and may reach temperature as high as 200 C. Whatever mercury is picked up by the indium in this location will be quickly vaporized into the atmosphere of the lamp and quickly diffused through it. The amount of 3,227,907 Patented Jan. 4, 1966 indium present on the electrode can be very small comparedvto that present on the indium ring, and in one example I have used only 1.8 milligram of indium, as compared to 180 milligrams at the middle of the lamp tube.

Lamps of this type have been very eflective, and have reached 90% of their equilibrium light output in a few seconds.

Other suitable metals such as gallium, cadmium and tin can be used in place of indium if the part to which they are applied reache a high enough temperature to give the propermercury vapor pressure.

When the lamp is turned 011?, the mercury returns to both the indium on the electrode and the indium on the middle band, and at equilibrium will have the same concentration on each. It will then be ready for operating in the same manner when again turned on.

The use of two deposits of indium, a small one at a hot spot and the main one at a cool spot will thus provide a quick-acting, high efliciency lamp at high power inputs.

To put all the indium at the hot spot, for example, on the electrode, would not achieve the same effect for all conditions of lamp operation, not even if the proportions of the indium and mercury were adjusted to give the correct vapor pressure at the temperature of the hot spot. For if any part of the lamp operates at a temperature below that corresponding to the correct vapor pressure for the mercury alone, the mercury will condense on that part.

Instead of having the second deposit of indium on the electrode, it can be at some other spot near the electrode, so that it will be heated by it. The indium can, for example, be placed on the usual stem press near the electrode, so it will receive. heat from the electrode, although the heating of the indium in such a position may not be as rapid as when the indium is actually. on a part of the electrode itself.

However, if the anodes get hot enough .in position to vaporize not only the mercury, but also an appreciable portion of the indium itself, then the indium is preferably placed on a somewhat cooler portion such as the'stem press.

In addition to its use as a source of indium in a lamp using indium to fix the mercury vapor pressure, the pressure of mercury on or near the electrode can also be used to facilitate starting and bringfluorescent lamps of other types up to brightnesses more quickly at low temperatures.

Other objects, advantages and features of the invention will be apparent from the following specification, taken in connection with the accompanying specification, in which:

. FIG. 1 is a side perspective of a lamp mount according to the invention;

FIG. 2 is an end view of the same mount;

FIG. 3 is a top view; and

FIG. 4 is a slanting view, showing the lamp tube and base'around the mount in phantom.

"In FIG. 1, the mount 1 has the glass flare 2, or skirte portion, with the glass press 3 extending out of the narrower portion of the flare 2. An exhaust tube 4 extends from the opening 5 in the press 3 in the usual manner, and the customary lead-in and

support wires

6, 7 are sealed through the glass of the press 3 in the manner standard in the art. The flag anodes 8, 9 are attached to the lead-in

wires

6, 7 onto nickel support wires 10, 11 as is often done in mounts designed for high power input.

The anodes are pr ferably of iron, because the indium 3 drop at the electrode. The filament 12 is shown from a different angle in FIG. 3.

On a portion of the surface of the flag anodes 8, 9, that is, on the anode surfaces facing, or facing away from, the filament 12, a small coating or piece 13 of indium is placed. The weight of indium present can be quite'small; about 1.8 milligrams has been found effective for the usual IOU-watt very high output lamp in a T-l2 glass tube, that is a tube about 1 /2 inches in diameter, and about four feet long. The inditun can cover an area about /8 inch diameter, although the exact dimension is not critical.

FIG. 4 shows the mount 1 sealed into a fluorescent lamp tube 14 in the usual manner, the outside edge of flare 2 being sealed to close the end of the lamp tube 14. A fluorescent phosphor coating is present over the inside surface of tube 14, as is customary in fluorescent lamps. The sealed lamp tube 14 contains a filling of a rare gas, for example a mixture of 85% argon and 15% helium, anda small quantity of mercury, say .50 mg.

A ring or band 15 of indium is present on the inside of glass tube 14, as shown in copending application Serial No. 220,714, filed August 30, 1962, by Richard Menelly. This indium ring 15, and the indium coatings 13 on anodes 8, 9 absorb the mercury present in the lamp and form an alloy therewith, reducing the mercury vapor pressure in the lamp at a given temperature.

If all the indium were in the ring 15, or in some other suitably cool spot, the lamp would not emit much light for a period of several minutes, in some case not for as much as 12 or 15 minutes after starting. The delay is due to the time necessary to heat the indium ring 15 to a temperature suflicient to give high enough mercury vapor pressure, since the indium reduces the pressure. However, on flag anodes 8, 9 the indium is quickly heated by the radiation from filament 12 and by the electric current picked up by the anodes 8, 9 themselves, so the mercury vaporizes and leaves the indium for the interior of the lamp. The indium, however, is not readily volatilized, because of its lower vapor pressure and higher melting temperature, so it remains on the anodes 8, 9. As the lamp cools on being turned off, the indium 13, absorbs mercury, the amount of mercury present being shared with the indium ring 15 during the 011 equilibrium state, that is when the power is not connected to the lamp. The indium coating 13 is made small so that the amount of mercury present in the lamp will not have to be greatly increased to inaintainthe proper amount in band 15, and so that too much mercurywill not be. vaporized immediately into the discharge from the indium coatings 13 on the anodes 8, 9 with undesired reduction of light output by temporarily increasing the vapor pressure above the desired 3 to 6 microns amount.

At equilibrium in the off condition, that is, with no power to the lamp, the proportions of indium and mercury will be the same in indium coating 13 on the anodes 8, 9 as in the indium. ring 15 if at least a part of the alloy or mercury and indium is in the liquid state, and so the total proportions of mercury and indium have to be those necessary for the proper operation of indium ring 15. At equilibrium in the on condition, with power connected to the lamp, the proportions of indium and mercury will of course be different in the anode 9 and the ring 15 because of the dynamic condition.

The main quantity of indium does not have to be in a single ring 15, but can be distributed over two or more rings, if desired, and the indium does not have to be present as a ring, but can have other forms or shapes, and can be at a cool spot other than the middle of the tube.

All of the indium can be placed on the flag anodes 8, 9, with the proportions of indium and mercury being such 'amalgamate with the mercury as to give the desired vapor pressure at the operating temperature of the anodes 8, 9. However, the anodes 8, 9 being warmer than some other parts of the tube during normal operation, the mercury will leave the anodes and deposit on the coolest part of the tube, which will gen erally be near the middle. If the indium-mercury proportions are such as to give a vapor pressure corresponding to that of mercury alone at 32 C., for example, then the mercury will condense out at any spot on the tube which drops below 32 C. (about F.). The lamp will operate properly as long as the lamp temperature, at its coolest spot is about 32 C., but not when it falls below that value. At lower temperatures, the mercury vapor pressure will be fixed by the temperature of the coolest spot in the lamp.

In some cases, the use of indium on a flag anode or other warm part of a lamp may be desirable for reasons other than fixing the vapor pressure at a desired value for a given bulb wall temperature. In very cold weather, many fluorescent lamps are slow to start or to come up to temperature, because of the low mercury pressure. If there is mercury present on the anode, however, for example mercury held at an indium spot, it will be quickly vaporized into the gaseous atmosphere by the heat from the filament or other electrode at starting. The lamp will therefor start more readily and give more light while it is still in the cool stage.

Although in the foregoing specification we have used the flag anodes as an example of a place to put the indium, it can be used in other Warm positions in the tube also, for example as a coating on the stem press, care being taken so that the coating will not cause short-circuit in the lead-in Wires.

What we claim is:

1. An electric discharge lamp comprising a sealed enclosing envelope, electrodes between which a discharge can be produced in said envelope, a filling of inert gas at low pressure and a quantity of mercury in said envelope, a quantity of indium in a portion of said lamp away from said electrodes, and a quantity of indium on at least one of said electrodes.

2. An electric discharge lamp comprising a sealed enclosing envelope, electrodes in said envelope, a stem press in said envelope, at least one of said electrodes being near said stem press, and a quantity of indium on said stem ress.

p 3. An electric discharge lamp comprising a sealed envelope containing a filament and mercury which when vaporized should be maintained at a constant pressure within the lamp, and a flag electrode adjacent the filament within the lamp having indium thereon adapted to in the lamp to provide a reservoir for the mercury, in order to facilitate rapid starting of the lamp and the maintenance of the constant pressure therewithin at thermal equilibrium.

7 4. The structure of claim 3 wherein the lamp is elongated and has a filament at each end thereof, with the flag electrode at least at one end of the lamp.

References Cited by the Examiner UNITED STATES PATENTS 2,991,386 7/1961 Szigeti et al. 313-109 3,007,071 10/1961 Lompe et al. 313-17s 3,013,175 12/1961 Waymouth et al. 313-219 X 3,152,278 10/1964 Dziergwa et al. 313-109 x GEORGE N. WESTBY, Primary Examiner.

DAVID J. GALVIN, RQBER'I SEGAL, Examiners.