US2939977A - Electrical brush - Google Patents

Description

June 7, 1960 G. A. FORSTER ET AL ELECTRICAL BRUSH Filed July 14, 1955 INVENTORS GEORGE A. FORSTER BY CORD H. SUMP ATTORNEYS States Patent ELECTRICAL BRUSH Filed July 14, 1955, Ser. No. 522,015 2 Claims. (Cl. 310-248) The present invention is directed to improvements in electrical brush assemblies of the type employed in motors and generators to establish electrical contact with a rotating element.

The present invention has particular applicability to the field of electrical equipment intended for use in aircraft. One of the difiiculties previously encountered in the design of motors and generators for aircraft use has been the tendency of brush assemblies to generate noise at their contact with slip rings or commutators. Since the generated noise extended over a wide band of frequencies, it frequently caused serious disturbances in electronic equipment, particularly radio, radar and the like.

A brush assembly to be useful in the types of electrical 1 equipment mentioned must not only produce a minimum of noise but should still have a low contact resistance, a resistance to oxidation, and should be conformable to the surface of the rotating element which it engages. Furthermore, brushes of this type should provide a reasonable wear rate, i.e., they must be sufiiciently tough to resist being ground down by friction between the brush and the slip rings or commutator.

Accordingly, an object of the present invention is to provide an proved electrical brush assembly in which the noise production is held to a minimum.

Another object of the invention is to provide im-' proved electrical brush assemblies having long wearing characteristics, flexibility, and strength making them particularlya daptable for use in high speed motors and generators.

A still further object of the invention is to provide an improved electrical brush and slip ring combination having excellent electrical characteristics, and being relatively free of generated noise.

We have now found that particularly good electrical brushes can be made by matting and compacting together a plurality of discrete wires into a substantially coherent form. We particularly prefer to employ mo1ybdenum wires because of their strength, toughness, and electrical properties. While molybdenum fiber brushes of this type result in a lower noise level with many different types of contacting surfaces, particularly good results are obtained when the molybdenum wire brushes are combined with a slip ring assembly or commutator assembly in which the surface which the fibers contact is composed of electrodeposited silver.

In order to achieve the maximum benefits of the invention, certain parameters in the design of the brushes should be observed. Probably the most important is the" apparent density of the brush, the latter being a term which is defined as the ratio of the weight of a brush body of given dimensions to the weight of a block of solid molybdenum of the same dimensions. We have found that the brushes should have an apparent density between about 25% and 50% for best results. At densities substantially lower than 25%, the brushes tend to wear rapidly and in some'cases to shred. At densities over 50%, the brushes tend to bounce i.e., fail to con- 2,939,977. Patented June 7,1960

form exactly with the surfaces of the slip rings, and cause high ring wear and in some instances seizure of the ring. It was also found that brushes having densities greater than about 50% produce larger variations in noise voltages.

The diameter of the individual wires making up the matted mass of the brush may vary, broadly, from about 0.001 to 0.010 inch, with a diameter in the range from 0.003 inch to 0.007 inch being preferred.

The brush assemblies can be manufactured in any of a variety of processes. The fibers can, for example, behand laid in the desired orientation or they can be deposited on a foraminous screen or filter from suspension in a liquid medium. Whatever the means for laying the fibers, it is desirable to have the predominating fiber lay, i'.e., the direction of the greatest number of fibers in compact, in line with the longitudinal axis of the compact.

The felted mass is then cold pressed to the desired density. Pressures of about 20,000 to 50,000 p.s.i., are typical for molybdenum with 30,000 p.s.i., being preferred. The pressure is applied in a direction perpene dicular to the contact force when the material is used'as a brush. The fibers need not be sintered, except where particularly strong compacts are required.

A further description of the invention will be made in conjunction with the attached sheet of drawings which illustrates a preferred embodiment thereof.

In the drawings:

Figure 1 is a somewhat schematic, fragmentary view with parts in elevation of a brush and slip ring assembly employing the principles of the present invention; and

Figure 2 is an enlarged, exploded view, partly in crosssection, illustrating a typical brush holder assembly of the type shown in Figure 1.

As shown in the drawings:

. In Figure 1, reference numeral 10 indicates generally a shaft of an induction motor to which a rotor 11 is keyed, or otherwise secured. An insulating collar 12 is rigidly secured to the shaft 10, the collar 12 serving to insulate a pair of slip rings 13 and '14 from the shaft 10 and from each other. A lead '16 connects the slip ring 13 to one end of the rotor windings, and a lead 17 connects the slip ring 14 to the other end of the windmgs.

The brush assemblies, illustrated generally at numeral 18 in Figure 1 are best illustrated in the exploded view of Figure 2. As seen in that view, the brush assemblies 18 may include a hollow cylindrical brush holder 19 ofv insulating material having a reduced diameter base portion 21 and an internally threaded upper end portion 22..

The brush element proper consists of a compacted mass 23 of molybdenum fibers, preferably oriented so that the predominating direction of fiber orientation islengthwise of the brush. A ferrule 24 is rigidly connected to the matted mass 23 and serves to anchor a lead-in wire 26.

One end of the matted body 23 is tightly received within the reduced diameter base portion 21 of the brush holder :19, with a sufiicient amount of the matted mass 23 protruding beyond the end of the reduced diameter portion 21 to be pressed into conforming relation with the surfaces of the slip rings 13 or 14. A helical spring 26 received within the holder 19 urges the brush assembly outwardly into good running contact with the surfaces of the slip rings. The top end of the spring 26 is bottomed against a cap 27 which is received in threaded engagement within the threaded portion 22 of the holder 19.

Numerous experimental tests, both under ordinary conditions and under conditions simulating high altitude flight have shown that the brush assemblies of the present in-- vention have excellent wear properties and provide only a very lo level of noi tpu n making he oise level tests, a high input impedance vacuum tube volt meter was connected across the slip ring brush assembly, and a ea h de Y SCi OSCQI Q was connected in p rallel with the volt meter to observe the typ f n is voltage bein enerated. Measurements of noise voltage and brush wear were made only after allowing a certain amount of run-in time for each brush tested. The amount of run-in time necessa y was considered to be the time required for the noise voltage \to reach a stable point. The brush wear rate was determined by accura ely weighing the brushes by means of a pan balance before and after running the brush for one. ho r- T e difference in the two weights Was taken as the amount of Wear and the percentagechauge per hour was then calculated. A brush pressure of 5 lbs. per square high was used for all measurements.

The following table illustrates the results obtained at various brush densities, fiber diameters, slip ring speeds, and brush currents;

NOISE VOLTAGE AND WEAR-RATE Molybdenum-wool brushes on electrotlepnsiled silver and brass rateof increase of noise voltage with increase of speed is much less apparent here silver seated slut rin s are employed.

In connection with the selection of slip ring materials,

it was previously noted that electrodeposited silver forms the best commutating surface for the molybdenum fiber brush. Other metals, however, .ican be employed but at a sacrifice in noise level. For instance, the noise voltages observed when brass slip rings were employed were ug y t i e hese btained w h' ver Rhod umslin r ngs exh ted noise factors f about 2 to. 3 times that obtained for silver. Rings made of coin silver.(90% silver copper) exhibited noise voltages not substantially higher than for pure silver, but exhibited greater brush and ring wear than pure silver. Slip rings made of stainless steel gave rather high noise voltages and caused considerable heating of the brushes at high speeds.

It appears that the best slip ring materials are those m tals o alloys uch as silver or brass which exhibit su stantia ly no mutual solubility with y u C n equ a yi these combinations lack the tendency to Brush Parameters Ring Brush R.M.S.

Speed, Current, Noise Wear Ring Feet M i- Voltage, Rate, Material Apparent Density, Fiber Diam- Per ampercs Millivolts Percent Percent eter, Inches Minute 110. 7 Per Hour 10, 000 19.0 1. 13 Silver. 10,000 100 0 l4 0. 51 Do. 10, 000 100 0. 8 0. 96 Do. 10, 000 1, 000 .1, 1. 22 Do. 10, 000 .1, .000 l. 7 0. 88 Do. 10, 000 1,000 8. 0 0. 28 D0. 20, 000 100 0. 3 0. 14 Brass.

20, 000 100 5.0 v Do. 20,000 100 10. O O. Do. 20,000 500 t 1. s 0. 31 Do. 20,000 500 25 0. 32 D0. 29, 000 500' 300 0. Do. 20, 000 0. 32 0. 48 Do. 20, 000 100 0. 28 0.21 Do. 20, 000 100 0. 18 0. 41 D0.

T he wires employed in manufacturing the brushes employed in the tests described above had an average length of about 0.5 inch. Succeeding tests indicated that longer fibers, specifically those having an average length of about 1.5 inches g ave lower noise voltages, all other conditions being equal. Accordingly, we prefer to employ fiber lengths at least 0.25 inch in length and preferably at about 1.5 inches. The following illustrates the effect of changing the fiber length upon the noise voltage generated, and the wear rate observed.

CHARACTERISTICS OF BRUSHES OF 35% DENSITY, 0.0035 INCH WIRE SIZE I Additional tests were made with brush currents up to. one ampere and peripheral ring speeds of 30,000 feet 7 per minute. The noise voltage was found to he proportional to increases in brush current and ring speed. The wear rate of the brushes was proportional to the ring speed'but rp ly was not significantly effec d y i creased brush currents. A typical value of noise voltage ob ain y op ra ing a b u h on. a bras ring t 3.0.0 feet per minute with a brush current of 100 milliarnperes DC. is, 0.58 rnillivolt, as QQrnpaIed to 0.18 millivolt for the same brush r nning at 10,000 feet per minute. The

alloy, weld, or gall at the contact surfaces. The rather poor results obtained with stainless-steel slip ring ma. terials bears out this requirement, because stainless steel and molybdenum are capable of substantial alloying.

The brushes of the present invention also exhibited excellent characteristics at simulated high altitude condl tions. When tested at conditions of 8 .5 inches: mercury pressure and a temperature of minus, 59 R, correspond ing to conditions at an altitude of approximately'30,000 feet, there was little change noted from the values ob.- tained for room temperature and sea ieve'l pressures. Noise voltages and wear rates were either unchanged" or in some cases reduced. At these conditions, the typical values obtained for a brush current of 100 rnilliamperes DC. at a speed of 20,000 feet per minute were as to]- lows; for electrodeposited silverQa noise voltage of 0.18 millivolt with a brush wear rate, of 014%; for thebrass slip rings, a noise voltage of 0.26 millivolt and awear brush rate of 0.6%. The i t n acte st c f riou time mate ials w r als t te h tes asmade by new. the p e nput t h r ve motor h Power req ired to rotate the slip ring at a constant speedbeing recorded with and without brush oading. The iffer nce betwe n he e two values was ta en a n ndica i n of the power loss due to friction. Such measurements show that electres d posited l r h t e lowest oss due to friction and that brass was slightly higher. Stainless steel; had a much higher loss due to friction than either brass or silver. I f 1 F o he fo e o it: wil be epua ien lthat the all a sembl s o t e p e ent nve tion pro ide lo gswefire' iug. tough brush elements, generating only .a. minimum ameunt f noise when in use. s

It will be understood that modifications and variations may be effected without departing from the scope of the novel concepts of the present invention.

We claim as our invention:

1. An electrical brush comprising a matted mass of 5 denum wires matted together into an elongated brush assembly wherein the predominating direction of the fiber lay occurs lengthwise of said brush, said mass having an apparent density in the range from 25 to 50% of the theoretical density of molybdenum, and said wires having diameters in the range from 0.001 to 0.010 inch.

References Cited in the file of this patent Electrical Contacts,"

UNITED STATES PATENTS Chaplin Mar. 10, 1896 Chaplin Aug, 31, 1897 Corey Sept. 5, 1911 Cooley Nov. 15, 1921 Merrill May 20, 1924 Lindenblad Sept. 14, 1948 Pandapas Dec. 7, 1954 FOREIGN PATENTS Great Britain of 1911 OTHER REFERENCES N. Chicago, Ill., page 8.

by Fansteel Metallurgical Corp., W

Images