US 3492612 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
Jan. 27, 1970 s o. AYREs ETAL FLUID-OPERATED ELECTRICAL SWITCH 3 Sheets-Sheet 1 Fig.2.
Filed Feb. 28, 1967 Fig. 1.
1970 s. D. AYRES ETAL FLUID-OPERATED ELECTRICAL SWITCH a Sheets-Sheet 2 Filed Feb. 28, 1967 1970 s. D. AYRES' ETAL FLUID-OPERATED ELECTRICAL SWITCH 3 Sheets-Sheet 3 Filed Feb. 28, 1967 United States Patent 3,492,612 FLUID-OPERATED ELECTRICAL SWITCH Sidney Dennis Ayres, John Christopher Hammond Davis, and Kenneth Hutton, Taplow, and David John Craft, Chandlers Ford, England, assignors to ,British Telecommunications Research Limited, Taplow, England, a British company Filed Feb. 28, 1967, Ser. No. 619,263 Claims priority, application Great Britain, Mar. 3, 1966,
9,450/66 Int. Cl. H01h 29/28 US. Cl. 335-112 12 Claims ABSTRACT OF THE DISCLOSURE A mercury contact switch consists of a tube having two fixed contacts protruding through the walls near one end. The bead of mercury, disposed in the tube, is movable by fluid, preferably air pressure between one end of the tube and the other. When the bead of mercury is at one end of the tube it makes connection between the two fixed contacts. Preferably the ends of the tube are closed by porous plugs.
The present invention relates to electric switching devices and is particularly concerned With devices suitable for use in light current engineering, particularly telecommunications.
In automatic telephone systems, which may be consldered as a typical example, connections are generally made by the selective closure of some of a group of contacts, and since these contacts will often be carrying speech currents, it is important that the contact should be a good one and made reliably even after a large number of operations which may amount to millions. An obvious possibility is to make use of a highly conductive liquid and for this purpose mercury is very suitable. A generally similar use is already Well known in connection with contact thermometers where contacts which need to be closed at a specific temperature are fused into the bore of a mercury thermometer at the appropriate place. For such use, however, speed of operation is not likely to present any problems, and clearly the mercury is not likely to move very fast. Where the operation of a switch is effected by tilting a tube containing mercury as is also well known, the arrangement tends to be somewhat cumbersome for small currents and is quite slow in the context of the telecommunications art. The chief object of the present invention is to provide an improved mercury switch which occupies only a small space, requires little power for its operation, gives reliable contacts, and can be operated at high speed.
According to the invention, an electrical switching device comprises a closed channel of small cross-section containing a bead of an electrically conductive liquid and having walls of an electrically non-conductive material through which first and second contacts protrude into said channel, said contacts being so disposed that the bead of electrically conductive liquid makes contact therebetween when in a first position in the channel and does not make contact therebetween when in a second position in the channel, in which said bead is arranged to be moved between said first and second positions by pressure exerted thereon by a non-conductive fluid, said pressure being arranged to be generated by means external to said fluid.
Preferably the movement is controlled by means of a flexible diaphragm which may be spaced away from the mercury and may itself be operated by fluid pressure on the opposite side or may be moved mechanically or electromagnetically or piezoelectrically. Such a diaphragm can readily be arranged to control a plurality of beads of 3,492,612 Patented Jan. 27, 1970 mercury simultaneously, thereby giving the same effect as a multi-contact electromagnetic relay. This facility is of importance in automatic telephone systems where switches in the speech circuits are usually required to operate at least three contacts simultaneously.
In practice, it is found desirable to limit the movement of the mercury by a buffer of sintered or other suitable material which is impervious to mercury but permits fluid pressure to be readily transmitted therethrough. Such buffers are provided at each end of the travel of the mercury and generally the design will incorporate a fluid cushion at the end remote from the operating diaphragm.
For most control purposes, it will not be possible to isolate the switch from ambient temperature and pressure variations, which will affect the internal pressure and hence the operation of the switch. To deal with this, it may be desirable to provide a bleed channel between the opposite sides of the mercury bead to enable pressure equalisa tion to take place at a slow rate after a switching operation has taken place. If the tube in which the bead is located is of circular cross-section, it will generally be advisable to provide a separate bleed channel which may, however, be common to a number of separate switches which operate simultaneously. If, however, the section of the channel is rectangular or D-shaped as might be produced by moulding a channel in a block of material and closing it with a cover plate, the mercury will not readily move into the corners of the section owing to surface tension effects, and this may provide a suflicient leak to obviate the necessity for a separate bleed channel.
The contacts themselves, which are bridged by the mercury, may be of any suitable contact metal, but preferably one which is wetted by the mercury, for instance silver.
According to the method of switching required, it may be desirable that the movement of the mercury bead should accurately follow the movement of the diaphragm, but for some purposes it might be desirable to produce a bi-stable characteristic, and this may readily be attained by using what may be termed a waisted capillary, that is to say one in which the bore is narrower in the middle than at the ends. With this arrangement, once the mercury has been pushed past the waist, it will tend to remain in that position when the fluid pressure difference across it which caused its movement decays or is removed. Thus, as suggested above, the leak or bleed will cause the pressure difference to disappear even if the operating input pressure is maintained.
The switch according to the invention leads itself to grouping on a cross-point basis where the switches are arranged in rows and columns to form a matrix, and a switch is operated at the point common to a chosen row and column. This concept may be extended to the design of a switch arranged to simulate the operation of a standard cross-bar switch in which contacts are closed at a particular cross-point in response to the selection of appropriate rows and columns and the connection is retained after one of the original operating forces is removed, conditions then being returned to normal when the other operating force is also removed.
The invention will be better understood from the following description of several methods of carrying it into effect, which should be taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a partially broken away plan of one embodiment arranged on a cross-point basis and using commoning conductors connecting the respective rows and columns;
FIG. 2 is an exploded sectional view taken on the line IIII in FIGURE 1;
FIG. 3 shows a possible electro-magnetic method of operating the actuating diaphragm;
FIG. 4 shows a modified section of the tube in which the mercury bead operates;
FIG. 5 indicates how the use of a tube of square section automatically produces narrow bleed channels owing to the surface tension of the mercury;
FIG. 6 is a plan of a modified form of switch according to the invention with a top cover plate removed;
FIG. 7 is a sectional view taken on the line VII-VII in FIGURE 6 but showing the cover plate;
FIG. 8 is a diagrammatic plan view of a modified construction in which the mercury bead is normally in two parts which are united when operating pressure is applied to the diaphragm;
FIG. 9 is a sectional view taekn along the line IXIX in FIG. 8; and
FIGS. 10-14 show diagrammatically a construction from which regular cross-bar operation may be obtained, these figures indicating different stages in the operation.
Referring first to FIGURES 1 and 2, it is assumed that the switch is constructed on a row and column basis and that at each cross-point four separate contacts are obtainable as may be required in an automatic telephone system. The conductors for the rows are indicated at 1 and the conductors for the columns at 2, and these conductors are located in separate parallel, e.g. horizontal, planes spaced slightly apart. They are assumed to be embedded in glass or plastics material, preferably of the transparent type, by suitable moulding operations. Thus the column conductors may be disposed on a base member 3 of plastics material and covered with a thin sheet of material. The row conductors are then put in position and are covered with a further sheet of material, and finally the whole is fused together, by the application of suitable heat and pressure. At any one cross-point, four holes 4 are drilled through the composite sheet so as to pass through corresponding row and column conductors, the holes thus forming a diagonal of the 4 x 4 matrix formed by the two sets of parallel conductors. These holes are preferably arranged to be convergent/divergent as shown in FIGURE 2; in other words, to have a waisted portion in the middle. Each of the holes contains a bead of mercury 5 which fills approximately half of the hole, that is to say up to the waisted portion as indicated in FIGURE 2.
The holes are covered at each end by the stops 6 and 7,
On the lower side, as sohwn in FIGURE 2, the various cross-points are enclosed by a similar plate which however, provides a chamber 9 individual to each cross-point.
Each chamber 9 includes a diaphragm10 opposite which is a hole through which operating fluid is supplied.
As mentioned above, in order to cater for changes in ambient temperature and pressure, it is desirable to provide a bleed, and this is effected in the construction shown by the device 11 which is arranged to offer high resistance to fluid flow. It will be understood that the bleed should provide equalisation of pressure in a time -which is small compared with the rate of such pressure changes but is large compared with the operating time of the switch. Since it is visualised that the switch would operate in a time of the order of one millisecond, while ambient changes, even if produced by the heating effect of associated electrical equipment, would almost certainly require a matter of minutes to produce their effect, it is contemplated that the time for equalisation of pressure to tak p a by the b e d m ght a abes wa d- One form of construction which might be adopted is shown in FIGURE 5, in which it is assumed that the bleed compresses a tube of square section which is also filled with mercury and extends between the two stops 6 and 7. Owing to the surface tension effect, the mercury will not extend fully into the corners of the section, and this permits very slow movement of operating fluid so as to produce the necessary equalisation effect. Of course, the cross-section of the channel need not be square. It can have any angular shape provided that one of the internal angles thereof is less than the complement of the angle of contact of the electrically conductive liquid.
It will be understood that when pressure is applied to deflect the diaphragm 10, a pressure pulse will be effective through the stop 7 to cause the mercury beads to move to the other end of their channels, thereby completing contact between the row and column conductors. Owing to the waisted section of the channels, there will be a tendency for the mercury beads to stay in the operative position owing to surface tension effects and in any event reliable contact will be maintained even if the device is subject to some degree of vibration which otherwise might cause the mercury to restore. The opening of the switch can then be effected by moving the diaphragm in the opposite direction so as to exert a suction effect.
FIGURE 3 shows how, if the diahpragm is provided with an armature or is in fact made of flexible magnetic material, the necessary switching effect may be produced by a coil, so that if the control of the switch is electric, no separate transducing operation is required.
FIGURE 4 shows another form of tube which may be used in which the section in which the mercury makes contact is parallel-sided, while the lower section which accommodates the mercury in the open position of the switch is convergent. Further, at least part of the bi-stable action can be produced by mercury-Wetted inserts at each end of the tube.
Considering now the second embodiment shown in FIGURES 6 and 7, this is of the type in which the tubes in which the mercury operates are not in a vertical plane as is the case with the FIGURES 1 and 2 construction but in a horizontal plane, and they are formed by moulding channels in the surface of a base member and closing these channels by means of a cover plate. As in the previous arrangement, the base member is assumed to be of glass or plastics material and is of the circular shape shown at 14. It is provided with moulded channels 15 of double-conical waisted section, which are terminated at one end by a stop 16 and at the other end by a similar stop 17, these being of sufficient size to cover the ends of all the four channels shown. Each of these channels contains a bead of mercury 18 which approximately half fills it. The channels are closed by a cover plate 19, which carries pairs of contacts 20, which projects sufficiently below the plate to dip into the mercury when it is moved to the other end of the channels, and above the plate provide connections for external wiring. The lower portion of the base 14 accommodates the operating diaphragm 21 which defines a chamber 22 communicating by means of a suitable orifice 22a with a chamber 24 formed behind the stop 16. A similar chamber 23 is formed behind the stop 17 and this communicates via a channel 25a with the annular chamber 25, which may if desired be made common to a plurality of switching points. This chamber is closed by a lower cover plate 26 so as to form a chamber 27 beneath the diaphragm to which the operating fluid at pressure P is applied by way of the pipe 28. As before a number of individual cross-points may be combined to form a matrix and commoning between the different conductors may be effected by means of ribbon cable or the like connected to the upper ends of the contacts 20. These cables might, for instance, extend diagonally in two directions at right gl s. in l cn. o F GU E 6. t wi l be pp i ted that with this arrangement the section of the channels is not circular and thus the mercury does not enter completely into the corners, thus allowing minute channels through which pressure equalisation can take place. Accordingly, a separate bleed will not be necessary as the constructionin effect provides a built-in bleed.
Referring now to FIGURES 8 and 9 these show an alternative arrangement in which two beads of mercury 41 and 42 are normally located at opposite ends of a dumb-bell-shaped groove 43 with a hump 44 in the middle portion, each of the end portions 45 and 46 having a contact (not shown) extending into the mercury. The groove is covered by a domed circular diaphragm 47 so as to prevent escape of the mercury, and operation is effected by applying mechanical pressure to the centre of the diaphragm 47. The pressure generated thereby in the mercury acts against the surface tension of the two mercury beads 41 and 42 which are forced along the groove 43 so that they join and thus complete a circuit between the two contacts. This condition will be maintained until the pressure on the diaphragm 47 is removed, when the mercury will again separate into two beads 41 and 42. Obviously, with this construction from the diaphragm 47 must be made of material which, does not react undesirably with the mercury. Moreover, in the simple form shown, the switch is unlikely to be satisfactory if it is subjected to serious vibration or movement involving appreciable acceleration.
Consider now the embodiment of the invention shown in FIGURES 10-14, which is suitable for providing a method of operation similar in most respects to that of a cross-bar Switch. In these figures, which are intended to be only diagrammatic, the mercury 29 is contained within a vertical tube 30, and is prevented from escaping by the stop 31 at the lower end. At the other end, the tube is separated by a stop 32 from a cavity 33 of large size compared with the displacement of the mercury 29. The contacts 34 and 35 to be bridged are located in the upper portion of the tube, just below the stop 32. The tube is provided with a side tube 36 and this is closed by a stop 37 which prevents the mercury moving into it.
In the normal open position, the mercury 29 is maintained in the lower part of the tube by gravity or surface tension effects if this part is of larger section or of wetted metal, and the contacts are not made and there is no excess pressure in the cavity or applied to either of the control inlets. When the switch is to be set, pressure P is applied to the tube below the stop 31 corresponding to the operation of a select magnet in a normal cross-bar switch and this has the effect of moving the mercury bead to the other end of the tube, thereby closing the contacts as shown in FIGURE 11. Pressure equal to 2P is now applied to the side tube 36 corresponding to the operation of a bridge magnet, and this has the the effect of dividing the mercury into two heads, one 39 of which fills the upper portion of the tube and maintains the contacts closed, while the other 40 rests on the stop 31 as shown in FIGURE 12. The original pressure applied to the lower end of the tube may now be removed and the contacts will remain closed until pressure is removed from the side tube corresponding to the release of the bridge magnet of a normal cross 'bar switch. When release is to take place and this :pressure also is removed, the upper bead 39 which is bridging the contacts now falls by gravity and joins up with the lower beads 40 as indicated by the arrow A in FIGURE 13. In order to ensure that this shall take place satisfactorily, a passage 38 is provided which permits fluid to escape from between the two beads so that they are able to coalesce.
A further facility offered by the switch is shown in FIGURE 14 whereby, if pressure 2P is applied to the side tube before pressure P is applied at the lower end, no movement of the mercury will take place when both pressures are effective, and the switch will remain in the 6 OFF position. Obviously, in these circumstances a pressure of 2P will also be developed in the cavity 33.
It will be understood that though it is suggested that the buffers of stops would be of sintered or ceramic material, other material of a porous nature may be used and the plugs produced may be fused in position or retained mechinically, for instance by spring loading. Similarly, the operating diaphragm may be made of any suitable material which will provide the necessary flexibility, will not react with mercury, vapour or liquid mercury, and can be readily sealed into its support. While the operation of the diaphragm can generally be most advantageously effected by fluid pressure, it can, however, be effected by direct mechanical action, electro-mechanically or piezoelectriqially. Furthermore, in embodiments where the diaphragm is not required to provide resilience, it may in fact take the form of a free mercury bead sealing a surrounding tube. Such an arrangement may be extended to incorporate a mercury-floated inductive needle which can be moved by a magnetic field to operate the switch.
The invention accordingly provides a novel switch arrangement which, by virtue of its compact construction, rapidity of operation, and possibility of fluid control, offers important advantages in the field of telecommunications.
1. An electrical switching unit comprising first and second chambers, a plurality of closed channels interconnecting said first and second chambers, each of said channels having walls of an electrically nonconductive material, respective pairs of electric contacts projecting into each of said channels, a respective bead of an electrically conductive liquid disposed in each channel, means communicating with one of said two chambers for applying a fluid pressure differential between said two chambers to move each of said beads of electrically conductive liquid between a first position in proximity to said first chamber in which it establishes connection between the corresponding pair of electrical contacts and a second position in proximity to said second chamber in which it does not make such connection.
2. A unit as claimed in claim 1, including a pair of buffers for each channel, said buffers being made of a material which is impervious to the electrically conductive liquid but which permits fluid pressure to be transmitted therethrough, one of said pair of buffers being adapted to close the end of its respective channel adjacent said first chamber and the other of said pair of buffers being adapted to close the end of its respective channel adjacent said second chamber.
3. A unit as claimed in claim 1, in which one of said chambers contains a diaphragm adapted to divide said one chamber into a first portion and a second portion with all said channels communicating with said first portion, whereby deflection of said diaphragm is effective to generate said pressure differential.
4. A unit as claimed in claim 3, including means for applying fluid pressure to a second of said two portions.
5. A unit as claimed in claim 3, including an electromagnet disposed confronting said diaphragm and means associated with said diaphragm whereby said diaphragm is movable under the influence of the magnetic field generated when said electromagnet is energised.
6. A unit as claimed in claim 5 in which a member of magnetic material is mounted on said diaphragm, said diaphragm being made of flexible non-magnetic material.
7. A unit as claimed in claim 5, in which said diaphragm is made of a flexible magnetic material.
8. A unit as claimed in claim 1, in which a bleed duct is arranged to connect said first chamber to said second chamber.
9. An electrical switching arrangement comprising a plurality of switching units as claimed in claim 1, said units being arranged in a matrix comprising rows and columns,
a respective first group of electric commoning connections for each row, each commoning connection of said first group connecting a first contact of the contact pair in a respective channel of a device in the row to first contacts of the contact pairs in corresponding channels in the other devices in the row, and a respective second group of electrical commoning connections for each column, each commoning connection of said second group connecting a second contact of the contact pair in a respective channel of a device in the column to second contacts of contact pairs in corresponding channels in the other devices in the column.
10. An electrical switching device comprising a body portion of an electrically non-conductive material and having a dumb-bell shaped groove in the upper surface thereof, the central part of said groove being humped, a pair of electrical contacts each disposed in a respective end portion of said dumb-:bell shaped groove, two beads of an electrically conductive liquid disposed in respective end portions of said groove and a domed diaphragm secured to said body portion and covering said groove whereby, on depression of said diaphragm, the two beads of electrically conductive liquid are forced into the humped central part of the groove and join together to establish connection between the two electrical contacts.
11. An electrical switching device comprising a closed channel, two buffers made of a material which is impervious to liquid but which permits gas pressures to be transmitted therethrough disposed in respective ends of the channel a closed cavity connected to one end of the channel, a side duct connected to the channel between the two buffers, a further buffer of a similar material to said two buffers closing the side duct at the point of connection with the channel, a pair of electrical contacts projecting into the channel between the side duct and the butter adjacent to the close cavity, and a bead of an electrically conductive liquid disposed in said channel and having a length greater than the distance between the side duct and the buffer adjacent to the cavity but less than the distance between the side duct and the buffer remote from the cavity, where by application of gas at a first pressure to the end of the channel remote from the cavity causes the bead to take up a position in contact with the buffer adjacent to the cavity and subsequent application of gas at a second pressure greater than the first said side duct causes said bead to divide into two parts at the point of connection of said side duct, a first part remaining in contact with the buffer adjacent to the cavity to maintain electrical connection between the contacts and a second part returning to the end of the channel remote from the cavity.
12. A device as claimed in claim 11, in which a by-pass duct extends from the channel at a position adjacent to the end of the bead remote from the cavity when said bead is in the position taken up after application of said first pressure and prior to application of said second pressure to the side duct, said by-pass duct being separated from the end of the side duct remote from the channel by a buffer similar to those at the ends of the channel.
References Cited UNITED STATES PATENTS 2,611,057 9/1952 Slonneger 200 8l.6 3,271,543 9/1966 Schonfeld 200-81.6 3,328,543 7/1967 Johnston 20081.6 2,744,980 5/1956 Bellamy 33549 BERNARD A. GILHEANY, Primary Examiner H. BROOME, Assistant Examiner US. Cl. X.R. 200'8l.6