US3285293A

US3285293A - Filament forming

Info

Publication number: US3285293A
Application number: US520036A
Authority: US
Inventors: Wilfrid G Matheson
Original assignee: Sylvania Electric Products Inc
Current assignee: GTE Sylvania Inc
Priority date: 1966-01-03
Filing date: 1966-01-03
Publication date: 1966-11-15
Anticipated expiration: 1983-11-15

Description

Nov. 15, 1966 w. G. MATHESON 3,285,293

FILAMENT FORMING Original Filed July 26, 1957 WILFRI D G. MATHESON INVENTOR.

ATTORNEY United States Patent 3,285,293 FILAMENT FORMING Wilfrid G. Matheson, Marblehead, Mass, assignor to Sylvania Electric Products Inc., a corporation of Delaware Continuation of abandoned applications Ser. No. 674,364, July 26, 1957, and Ser. No. 257,207, Feb. 8, 1963. This application Jan. 3, 1966, Ser. No. 520,036

8 Claims. (Cl. 140-715) This application is a continuation of my copending application, Serial No. 674,364, filed July 26, 1957, now abandoned, and copending application, Serial No. 257,207, filed February 8, 1963, now abandoned.

This application relates to tungsten wire, ribbon or sheet, such as is used in incandescent and other lamps and devices, and in particular it relates to methods of forming such a tungsten piece to a definite and predetermined shape which will be retained on final assembly into a device.

At present such tungsten wires, ribbons or sheets are heated prior to assembly, but only at temperatures such as about 1900 K., which are too low to completely relieve the residual stresses in the tungsten. Hence when the tungsten piece is later raised to a higher temperature causing more complete crystallization, for example, on flashing the tungsten filament in an incandescent lamp after manufacture, the tungsten piece often changes its shape and position.

Such change in shape and position is generally undesirable in incandescent lamps and other devices and is especially disadvantageous in projection lamps, where the filament must be in a particular position for focusing purposes. In fact, prior to my invention, it had been impossible to make certain types of projection lamps with internal reflectors, where the filament must act as a concentrated source maintained in a precise position.

I find, however, that the size and shape of the final filament can be confined within precise limits if the filament is heated to a temperature high enough for complete recrystallization, for example to a temperature between 2200 K2800 K., while the filament is held precisely in its desired final shape by a highly refractory form. The form itself is previously relieved of all stresses by a suitable heat treatment and has a higher melting point than the temperature required for complete recrystallization, of the tungsten piece, that is, for the formation of long, interlocking crystals in the tungsten. The refractory material of the form can be molybdenum, tantalum, rhenium or some other suitable material, for example, even tungsten itself.

For straight coils, the refractory form can be a mandrel wire on which the filament is wound, and for coils bent into various shapes, it can be a mold fitting around the filament from the outside, the filament having previously been bent to approximately the proper shape to fit the mold. When the coil used is a coiled coil, a mandrel can be used in both the primary and secondary coil 1 during the heat treatment.

The heating is performed in a controlled atmosphere. An atmosphere such as oxygen, which reacts with tungsten, will be undesirable. However, a vacuum is satisfactory, or an atmosphere of gas inert with respect to the filament, for example, nitrogen or helium gas, or a reducing atmosphere such as hydrogen. For example, an atmosphere containing a mixture of hydrogen and nitrogen, say about 25% hydrogen to nitrogen has been used. In fact, a hydrogen atmosphere will be helpful, in other respects also, because it will clean the tungsten piece while the latter is being fixed in shape.

I find that temperatures about 2000" K. cannot be obtained with the furnace customary in wire processing, and that moreover, heating at temperatures above such a value would be ineffective unless the wire was held in precise shape desired by a form, because without the form the wire would sag to an uncontrolled shape. By the use of a refractory, electrically-conduction form for fixing the shape of the coil, and the further use of that form as an electrical resistance element through which current is passed to heat the coil to the desired temperature, I use the refractory form for a double purpose.

In this manner, the heat gets to the coil directly and rapidly, and the process is simplified. Other suitable means for heating the wire to the proper temperature can be used, however.

Temperatures between about 2200 K. and 2800 K. are generally suitable for heating the tungsten according to the invention. The lower temperatures in that range are generally best for large wire, for example, of 50 mils diameter, whereas the higher temperature is best for Wires of about 2 mils diameter.

The coil needs only to be maintained at the desired temperature for a few seconds, generally from about 5 seconds to 50 seconds, depending on the temperature used and the wire size. The lower the temperature, the longer the time required. A wire of the largest size in which tungsten is presently drawn, that is, about mils, requires a heating time of about two minutes, but such a large wire size is not ordinarily used for filament coils.

The same processing described above for wire can also be used for tungsten ribbon and sheet.

The process of the invention greatly extends the useful range of coiled-coil design, especially by permitting a radical change in the ratio of secondary to primary coiling diameter. The process permits the use of much larger secondary coiling diameters for a given primary diameter, thereby providing more compact light sources using smaller size wire. The invention eliminates the need for low voltage filaments in small projection lamp design; by allowing the proper type filament to be made for -volt circuits, it eliminates the need for transformers and allows operation of the lamp directly from the usual house lighting power lines.

Filaments formed by the process are also useful as heating elements in the'vapor deposition of metals, a gain in life of from to 300% for the tungsten heating element being obtained in such use. In addition, the aluminum deposited by vaporizing with one of the filaments has a more uniform reflecting surface, and there is therefore less product scrap due to filament distortion and failure. The greater uniformity of deposition is due to the fact that the filaments of the present invention do not sag.

Improved maintenance of light output throughout life has been obtained in high pressure mercury lamps using cathodes in which the tungsten coil which holds an electron-emissive thorium piece is treated by the process.

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

FIG. 1 is a side view of a coiled-coil filament made according to the invention;

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

FIG. 3 is a top view of the forming device for said coil; and

FIG. 4 is a side view of the same device.

In FIGS. 1 and 2, the tungsten filament wire 1 is coiled about a molybdenum mandrel wire 2, and the minor coil 3 so formed is itself wound into a larger coil 4, generally by being wound around a larger mandrel, not shown.

In FIGS. 3 and 4, a series of coils such as 4, each having a small-diameter mandrel 2 in its minor coil, are shown held by the larger diameter mandrel 5 which fits inside the major coils of the coiled-coils 4. Each end of the mandrel 5 is fastened to a fixed support 6 by the clamping piece 7 which is held to piece 6 by the threaded machine screw 8. Lead-in wires 9, through which a voltage can be applied across the ends of mandrel 5, are attached to the support 6 by machine screws 10. One of the supports 6 is fixed to a base plate 11 and the other support 6 is fixed to a rod 12, which passes through a hole in the support 13, the latter being fixed to the same base plate 11. The outer end of rod 12 is threaded, and nut 14 is screwed thereon. The spring 15 bears on one end against the support 13 and at its other end against the nut 14, biasing the two apart with a force depending on the tension to which spring 15 is compressed by the nut 14.

In operation the tungsten wire 1 is first wound on the molybdenum primary mandrel 2, and the stresses in the winding are then relieved :and the wire cleaned by heating in an atmosphere of hydrogen of about l900 K. The wound coil 3 with its mandrel 2 is then Wound around a larger mandrel and the larger mandrel removed, leaving the coil as shown in FIGURES 1 and 2. That much is common practice in the art, the second coiling being generally done on a well-known type of coiling machine in which the coil is wound on a steel mandrel which is retracted after winding to leave the coil as in FIGURE 1.

Several coils such as that of FIGURE 1 are slipped onto a close-fitting tungsten mandrel 5 which is then fired between the supporting electrodes 6, the mandrel being held taut by the action of spring 15. An atmosphere of substantially 100% hydrogen is placed around the coils 4 and mandrel 5, preferably by placing the unit support on base member 11 in a bell jar. An electrical current is then passed through the mandrel 5 from the lead-in wires 9, which can extend outside the bell jar. The voltage across the mandrel 5 is increased from zero to the level necessary to bring the tungsten to a temperature of about 2800 K. for complete recrystallization. The filaments are kept at this temperature for several seconds, after which the voltage is lowered to zero. After the filaments have cooled to about room temperature, the mandrel 5 is taken from supports 6 and the filaments 4 slipped 01f. A filament so formed will maintain its shape very precisely when later placed in an electric lamp bulb and operated at the usual lamp temperatures. In a particular example, the wire 1 was of 17.78 milligrams weight per 200 millimeters, that is, it had a diameter of about 3.02 mils. The mandrel 2 was of molybdenum, of 8 mils diameter, and the coil 3 had 196 turns per inch. The larger coil 4 was of 50 mils inside diameter and 56 turns per inch. The tungsten mandrel 5 was of 48-mil diameter and ten inches long between the supports 6. The coil 4 on the mandrel was brought up to a temperature of about 2800 K., which is just below the melting point of molybdenum at 2860 K. The voltage across the molybdenum mandrel 5 was brought up to attain this temperature in 14 seconds and maintained at said voltage for 10 seconds after which it was brought down to zero in 14 seconds.

In general a temperature range of 2200 K. to 2800 K. is satisfactory for treating a tungsten filament according to my invention and a time of from about 5 seconds to about 50 seconds is satisfactory. The longer times would ordinarily be used with the lower temperatures and vice versa. The heavier the mass of the wire, the lower the temperature required within the limits given.

After the heat treatment above, the filaments are placed in a solution of acid which dissolves away the primary molybdenum mandrel in the usual manner, but which leaves the tungsten. The finished filament then remains and is mounted between lead-in wires in an electric lamp bulb which is exhausted and filled in the usual manner to form a sealed lamp. A filament so processed will maintain its position and shape during the usual further process steps of lamp manufacture. This is in sharp contrast with filaments treated in the manner previously used, which tend to sag greatly during the usual manufacturing process, and also in later use. In order to insure that the tungsten mandrel 5 is itself properly relieved of stresses and recrystallized, a preliminary run as above can be made and the coils 4 shaped on it can be discarded. The tungsten mandrel 5 will then be in proper condition for use in subsequent runs to fix the shape of coils for use in various devices.

The tension in spring 15 should be sufiicient to keep the mandrel 5 straight as the latter expands on heating.

Various modifications can be made by persons skilled in the art without departing from the spirit and scope of the invention.

What I claim is:

1. The method of fixing the shape of a tungsten wire coil having ends disposed therein, the steps which comprise: placing said coil on a tightly fitting refractory metal mandrel with the ends of said coil being free, heating said coil to a temperature between about 2200 K. and 2800 K. while on said mandrel, cooling said coil, and thereafter removing said refractory mandrel and mounting the coil in a lamp bulb to form a sealed lamp.

2. The method of claim 1 in which the heating of the coil is done in a non-oxidizing atmosphere.

3. The method of claim 2, in which the coil is a coiledcoil having a major and a minor coil, with a mandrel in each coil during the heating, both mandrels being removed after cooling the coil, and the coil then mounted in a lamp bulb to form a sealed lamp.

4. The method of claim 1, in which several coils are placed on the same mandrel and heated together, cooled, the mandrel then removed, and each coil placed in a lamp bulb to form a sealed lamp.

5. The method of fixing the shape of a tungsten piece, said method comprising holding said piece by an electrically conductive refractory metal form to confine it to a predetermined shape, and providing a non-oxidizing atmosphere around said piece, and passing an electric current through said electrically conductive metal form to bring said piece to a temperature of between about 2200 K. and 2800 K., and thereafter removing the form.

6. The method of fixing the shape of a tungsten piece, said method comprising holding said piece by an electri cally conductive refractory metal form to confine it to a predetermined shape, and passing an electric current through said electrically conductive refractory meta] form to bring said piece to a temperature of between about 2200 K. and 2800 K., and thereafter removing said metal form.

7. The method of fixing the shape of a tungsten wire coil by placing said coil on a tightly-fitting electrically conductive refractory metal mandrel and passing an electric current through said metal mandrel to heat the coil to a temperature between about 2200 K. and 2800 K.

8. The method of fixing the shape of a coiled coil of tungsten wire, said coiled coil having a major and a minor coil, said method comprising: placing a second tighly-fitting electrically conductive refractory metal mandrel in said major coil along the longitudinal axis thereof While there is a first tightly-fitting mandrel in said minor coil, and passing an electrical current through said second electrically conductive refractory metal mandrel to heat said coil to a temperature of between about 2200 K.

and 2800 K.

References Cited by the Examiner UNITED STATES PATENTS FOREIGN PATENTS 5/ 1927 Great Britain.

CHARLES W. LANHAM, Primary Examiner.

W. H. JUST, Assistant Examiner.

Claims

1. THE METHOD OF FIXING THE SHAPE OF A TUNGSTEN WIRE COIL HAVING ENDS DISPOSED THEREIN, THE STEPS WHICH COMPRISE: PLACING SAID COIL ON A TIGHTLY FITTING REFRACTORY METAL MANDREL WITH THE ENDS OF SAID COIL BEING FREE, HEATING SAID COIL TO A TEMPERATURE BETWEEN ABOUT 2200*K. AND 2800* K. WHILE ON SAID MANDREL, COOLING SAID COIL, AND THEREAFTER REMOVING SAID REFRACTORY MANDREL AND MOUNTING THE COIL IN A LAMP BULB TO FORM A SEALED LAMP.