US1648690A - Method of making long crystal tungsten filaments - Google Patents

Description

Nov. 8, 1927. 1,648,690

R- JACOBY METHOD OF MAKING LONG CRYSTAL TUNGSTEN FILAMENTS Filed April 19. 1924 Inventor: Ric hard JacobH;

His Abbornex Patented Nov. 8, 1927.

UNITED STATES I 1,648,690 PATENT OFFICE.

RICHARD JACOBY, or BERLIN, GERMANY, AssIoNoR T0 GENERAL ELECTRIC COMPANY, A CORPORATION or NEW YORK.

METHOD OF MAKING LONG CRYSTAL TUNGS'IEN FILA'MEN'IS.

Application filed April 19, 1924, Serial No. 707,759, and in Germany May 17, 1923.

M invention relates to the method of making long crystal helical tungsten filaments for electric devices as for example incandescent lamps.

Heretofore, long crystal wire or filament for electric lamps has been made from such refractory metal as tungsten by passing the wire through a zone of high temperature. In most cases this is effected by feeding the wire between two contacts which are connected to a source of electric current and thus raised to the high temperature necessar It is also well known that tungsten wire can be made into helical filaments by winding it on a mandrel and that it is advantageous to use it for this purpose in the long crystal form, known also as the uni-crystal or macro-crystal'form, in which form the filament is stiff and strong.

The object of the present invention is to simplify the making of the helical filaments from long or macro-crystal filaments and at the same time to produce these filaments in such a manner that their crystalline structure will be better than that obtained by following previously known methods.

According to the invention, this aim is obtained by winding the wire filament into the helical form immediately after its transformation into the long crystal structure or during this transformation. By so doing, the wire is changed into the long or macrocrystal state and simultaneously brought into helical form in one continuous working process and in this manner the making is simplified considerably. Furthermore, the wire can be fed through the heating zone without jerks. This is of particular importance since the creation of long crystals is dependent upon the condition that the wire shall not, in any time unit, enter into the heating zone at a greater speed than that of transformation of the small crystals of the wire into the long crystals. When the wire is passed through the heating zone according to the former methods, the cold piece of tungsten wire between the feeding device and the high temperature zone, will always be somewhat elastic and will enter this zone of high temperature at irregular speed. Such irregularities which disturb a uniform creation of long crystals are comparatively small but their great eflect is readily realized when we stop to consider that the growth of the crystals is a process between micro-crystals which is hardly detectable, even microscopically. Therefore, the regular feeding of the wire through the high-glow zone will help to make the wire positively assumea structure which consists of long crystal elements and.

a perfectly uniform structure of this kind is especially favorable as regards the behavior of the helical filament in the lamp.

The method may be carried out in the followin manner: A mandrel on which the heate long crystal wire is wound is placed immediately beyond the glow-zone and this mandrel is moved transversely to the direction of travel of the wire and rotated on its axis, relative to the tungsten wire. An additional advantage is obtainable by making-the winding mandrel serve at the same time as the second contact of the high-glow zone in that the wire will pass onto the mandrel in a highly heated plastic or semiplastic state and its crystalline structure will be better adapted in this state to assume the helical form than it would otherwise be. The wire when thus wound has less tendency to straighten out during burning of the lamp than when wound otherwise and hence has less tendency to become distorted and break.

Tests have shown that though the tungsten wire becomes heated in the process of making the long crystal to very nearly its melting point in the heating zone it withstands being wound around a mandrel, and in fact the process can be carried out at a reasonable speed.

This method has nothing in common with the method of winding tungsten wire into helical filaments while it is hot, such as is commonly used in the manufacture of incandescent lamps, in which cas the temperature of the wire never rises above a red heat. This temperature is far below the recrystallization temperature which is used in the method of making long crystal filaments in the present invention.

In the accompanying drawing one form of device for carrying out the process is shown in schematic form.

The process and apparatus may be described more in detail as follows: The tungsten wire b which is fed in the direction of the arrow a enters the high temperature zone 03 at the terminal 0, which high temperature zone lies between the contact 0 and the mandrel 0. With the arrangement shown it will be seen that the mandrel is in the circuit over which current is supplied for heating the tungsten wire and serves as one of t e terminals of the circuit to which one of the current leads is connected. The other current lead s own is connected to the terminal a. The mandrel, during the operation is rotated and at the same time moved in its longitudinal direction as indicated by arrows h and c'. As this takes place the wire d is wound on the mandrel. The winding of the long crystal wire on the mandrel can also be effected by moving the wire around the mandrel-and laying it on in helical turns.

The heating and winding of the wire may be advantageously carried out in a non-oxidizing atmosphere for instance in hydrogen or in a gas containing hydrogen. The terminal c for making contact with the wire 6 ma be of any suitable material provided t e friction it offers is not too great. For example, mercury may be used as well as an easily fusible alloy or other similar material. The raising of the temperature of the wire to the necessary high point can also be effected in steps. Also the temperature of a portion of the high temperature zone can be changed by means of an auxil iary contact. The Winding of the wire on the mandrel is best done close to the high temperature zone. By choosing suitable dimensions for the mandrel and the proper arrangement for conducting current to the roper heat conducting under which thewire can be maintained at the most favorable temperature for winding about the mandrel.

It will be understood that I do not wish t be limited to the specific illustration and description in the application of my inventions inasmuch as variations will readily suggest themselves in view of the disclosure embraced within the scope of the claims appended hereto.

What I claim as new and desire to secure by Letters Patent of the United States, is,-

1. The method of making a coiled long crystal filament of a refractory metal for electric devices, which consists in coiling the wire while the long crystal filament is in the plastic state.

2. The method of making a unicrystal tungsten filament which consists in heating a tungsten wire to a high temperature until it is converted into unicrystal form and coiling it While still at the high temperature.

In witness whereof, I have hereunto set my hand this 31st day of March, 1924.

RICHARD JACOBY.

Images