US2510322A - Selenium rectifier - Google Patents

Description

Patented June 6, 1950 SELENIUM RECTIFIER Robert E. Shearer, Edgewood, Pa., assiznor to The Union Switch & Signal Company, Swissvale, Pa., a corporation of Pennsylvania Application September 22, 1945, Serial No. 618,003

2 Claims.

My invention relates to selenium rectifiers, and particularly to an improved method for electrically forming selenium rectifiers.

According to one well-known process for manufacturing selenium rectifiers, a layer of selenium which may or may not contain other substances orcompounds of selenium is first applied to a suitable support or base plate, and the selenium layer or coating is then subjected to an annealing or heat treatment to convert the selenium into the metallic electrically conducting state. Following this conversion, a counterelectrode layer preferably composed of tin, cadmium or an alloy of these metals to which a small amount of thallium has been added is applied, for instance by a Spraying or sputterin operation, to the outer surface of the selenium layer, and a voltage is applied to the rectifier in the reverse direction to electrically form the rectifier. The result of this electric forming process is to increase the reverse resistance of the rectifier and thus cause it to inhibit the well-known rectifying characteristic of transmitting more current in one direction than the other.

Heretofore, in forming selenium rectifiers it has been the usual practice to apply the forming voltage for a predetermined period of time, allow the rectifiers to cool, and then test them. For reasons which have not been determined the rate at which the rectifiers form varies considerably even between units which have all been made at the time, and as a result with this method of forming it is frequently necessary to form the rectifiers a second and a third time in order to obtain the desired reverse resistance characteristics. I

One object of my present invention is to provide a method for forming therectifiers which eliminates the necessity for reforming and retesting the rectifiers, and which permits automatic control of 1 the process.

According to my inventionl' form the rectifiers by applying the forming voltage through suitable means which permits the forming voltage toincrease in a controlled manner during the forming process, and in interrupting the forming process when the voltage drop across the rectifiers has increased to a predetermined value which depends upon the current flowing and the temperature of the rectifier, and which is determined in accordance with the characteristic desired in the finished rectifier.

My present invention is an improvement upon that described and claimed in Letters Patent of the United States No. 2,279,187, granted to Thompson et al. on April 7, 1942, for Alternating electric current rectifiers of the selenium type.

Other objects and characteristic features of my invention will become apparent as the description proceeds.

I shall describe one forming process for selenium rectifiers embodying my invention, and shall then point out the novel features thereof in claims.

The accompanying drawing is a diagrammatic view showing a forming circuit which is particularly suitable for carrying out the forming process embodying my invention.

Referring to the drawing, the forming circuit here shown comprises two forming electrodes i and 2 between which the rectifier elements which are to be formed are placed. As illustrated in the drawing, only one such rectifier element A is shown, but it will be understood that in actual practice when absolute uniformity is not required a number of such elements may be placed between the forming electrodes in such manner that these elements will all be connected in parallel during the forming process. The elements may be manufactured up to the forming step by any of the well-known processes, and each element comprises as usual a seleniumlayer 3 firmly attached to a suitable support or base plate 4, and provided on the side opposite to the base plate with a counterelectrode layer 5.

To facilitate the forming, the lower electrode l is preferably maintained at a temperature of between and 6., but any temperature may be employed which is high enough to permit rapid forming yet not sufficiently high to cause the forward resistance to deteriorate or to melt the counterelectrode alloy; This temperature may be obtained in a variety of ways, the particular means shown for obtaining this temperature being a heating element 5 which is supplied with current from a suitable source not shown under the control of a thermostat which is likewise not shown.

During the forming process a suitable potential is impressed across the electrodes I and 2 in such manner that the resulting current which flows through the elements will traverse the elements in the high resistance direction. The source of potential is not shown, but the positive and negative terminals thereof are designated B and C, respectively. Any convenient source of direct current may be employed such for example as a full wave rectifier, a motor generator, or the like. For reasons which will appear hereinafter, it is essential that the current which flows through the rectifier at the completion of the forming period should have a predetermined magnitude, and for convenience in accomplishing this result I prefer to impress on the forming circuit a potential which is considerably higher than the desired forming potential and to limit the current flowing through the rectifier by means of a ballast resistance BR connected in series in the forming circuit. The current which is permitted to fiow in the forming circuit upon the initial closing of the circuit, and hence the value of the ballast resistance BR, will vary with the size of the elements and the number of elements in parallel, the ballast resistance being so chosen that this current will be insuificient to impair the rectifying properties of the rectifier. In actual practice, current densities in the range of 150 to 500 ma. per square inch of rectifier surface have been found to give satisfactory results, a preferred value being 300 ma. per square inch of rectifier surface. By a proper choice of the power source voltage and the limiting resistance BR, it is possible to have the desired current at both the start and the end of the forming period.

With a total voltage of 65 volts impressed across the forming circuit, and the current in the forming circuit initially limited to 300 ma. per square inch of rectifier surface, the initial potential across the rectifier elements will be found to be from 4 to 8 volts, and as the elements form, the resultant rise in resistance of the elements will cause this voltage to gradually rise. According to my invention, I permit the forming voltage to.

continue to rise until the potential drop across the rectifiers has increased to a predetermined value which depends upon the current flowing through the element, the temperature of the element at the time the measurement is made, and

the characteristics desired in the finished rectifier. That is to say, I have found that the forming voltage s measured will be an accurate indication of the ultimate reverse resistance characteristics of the rectifier. When the desired voltage is reached, the forming circuit-is interrupted. For example, tests show that to obtain a completely formed rectifier passing 22.5 ma. per square inch at a test voltage of volts, in the reverse direction, the forming circuit should be interrupted when the voltage drop across the rectifier has reached approximately 21 volts at 500 ma. per

matically, and to this end I connect a relay D,

which is adjusted to pick up its armature at the voltage at which it is desired to interrupt the formingcircuit, in multiple with the forming electrodes l and 2 in such manner that the voltage impressed across the winding-of this relay will be-the voltage drop across the rectifier ele-v ments being formed. I also connect a first back contact I or this relay in series with the forming electrodes in the forming circuit in such manner square inch, or 16 volts at 290 ma. per square inch for a rectifier temperature of 60 to C. For the same forming temperature, a rectifier which will pass 22.5 ma. per square inch at 20 volts may be obtained by interrupting the forming circuit when the drop across the rectifier has increased to 22 volts at 290 ma. per square inch, etc. If the forming temperature of the rectifier is above or below 60 to 70 C. when the measure ment is made, then the forming circuit should be interrupted at a slightly lower orhigher voltage, as the case may be, for the completely without departing that the picking up of this relay will automatically interrupt the supply of forming current to the rectifier elements. Furthermore, in order to prevent damage to the relay due to voltage surges when the relay picks up I connect a protecting resistance PR in series with the relay winding, and in order to nullify the effect of this resistance on the remainder of the forming circuit prior to the picking up of the relay I connect another back contact 8 of the relay D in multiple with this resistance in such manner that this resistance will be short circuited by the contact 8 until the relay picks up.

With the forming circuit constructed in the manner just described it will be apparent that when the forming voltage impressed on the rectifier elements increases to the desired voltage, the relay will automatically pick up and will interrupt the supply of forming current to the rectifier. Furthermore, the picking up of this relay will connect the current limiting protectiveresistance PR in series with the winding of the relay to thereby prevent damage to the relay due to any high voltage surge which might otherwise be impressed on the relay winding.

One advantage of the method of forming selenium rectifiers described above is that any rectifier which can be formed will be automatically formed in a single operation thereby eliminating any necessity for repeated forming operations and repeated testing.

Another advantage of the method of forming selenium rectifiers embodying my invention is that the uniformity of the resulting product is materially improved.

Although I have herein shown and described only one method of forming selenium rectifiers embodying my invention, it is understood that various changes and modifications may be made therein within the scope of the appended claims from the spirit and scope of my invention. I

Having thus described my invention, what I claim is:

1. The method of forming a selenium rectifier which consists in maintaining the rectifier at a predetermined temperature, in applying continuously to the rectifier a voltage in the reverse direction through a ballast resistance so that as the reverse resistance increases the applied voltage will also increase, and in interrupting said voltage when it has increased to a predetermined value which depends upon the current flowing through the rectifier and the temperature of the rectifier when the voltage measurement is made.

2. The method of forming a selenium rectifier which consists in applying continuously to the rectifier while maintaining the rectifier at a predetermined temperature a voltage in the reverse 5 direction through a ballast resistance so that as the reverse resistance increases the applied voltage will also increase, and in automatically interrupting said voltage by means responsive thereto when it has increased to a predetermined value which depends upon the current flowing through the rectifier and the temperature of the rectlfler.

ROBERT E. SHEARER.

REFERENCES CITED The following references are of record in the file of this patent:

Number 6 UNITED STATES PATENTS Name Date Geiger May 1, 1934 Wilson 'Mar. 12, 1940 Geisler Apr. 16, 1940 Hoppe et al Jan. 28, 1941 Hoppe Nov. 4, 1941 Mershon Aug. 18, 1942 Becker et al. Jan. 18, 1944 Williams May 1, 1945 Certificate of Correction Patent No. 2,510,322 June 6, 1950 ROBERT E. SHEARER I It is hereby certified that error appears in the printed specification of the above numbered patent requlring correction as follows:

Column 1, line 22, for the word inhibit read exhibit;

and that the said Letters Patent should be read as corrected above, so that the same may conform to the record of the case in the Patent Ofiice. Signed and sealed this 23rd day of January, A. D. 1951.

' THOMAS F. MURPHY,

Assistant Commz'ssz'oner of Patents.

Images