US2605307A - Temperature equalizer for mercury tube switches - Google Patents

Description

y 29, 1952 J. OSTLIND ET AL 2,605,307

TEMPERATURE EQUALIZER FOR MERCURY TUBE SWITCHES Filed Dec. 12, 1949 2 Sl-iEETS-SHEET 1 INVENTORS 7 Joa 05mm) 9 BY h/ILLMM L. WERNER S/ HE Z;- K"

HTTOENEVS.

y 29,1952 J. OSTLIND ETAL TEMPERATURE EQUALIZER FOR MERCURY TUBE SWITCHES 2 SHEETS-SHEET 2 Filed Dec. 12, 1949 4 w M may Vwl D)? m M a am M w 3 \W/ E Q E 4 1 W 5 w w, q y @Nw? 5 M .w a n 4 Y m 5 E E /I\\ m I v m w I Patented July 29, 1952 TEMPERATURE ,asoaso'z EQUALI'zER'FoR MERCURY TUBE SWITCHES Joel am, Oakland, and William L.

Palo Alto, Calif.

Werner,

Application December 12, 1949, S erial N orl3 2,605

3 Claims. (01; 171-97) This invention relates to power relays of the delayed action type, andparticularly to relays adapted for sequential operations, such as may be connected in cascade to provide for the operation of groups of street-lighting circuits.

At the presenttime street lighting is largely provided by lamps of the incandescent type. The difference between the cold resistance and the hot resistance in such lamps is very large, the hot resistance being many times that which the lamps exhibit when cold. Ordinarily all of the lamps in a single city are lighted at dusk, and if all are lighted at the same time the initial surge of current, lasting until the lamps have come up to approximately their final temperature, may be such as greatly to overload the circuits, throwing undue strain upon the generating equipment and the protective devices appurtenant thereto. better if the distribution system for such lighting be divided into groups or sections which are excited successively, so that the transient currents or surges consequent upon the excitation of one group may have an opportunity to die away before the next section is excited. If the entire system is divided into a suflicient number of groups an initial current of up to ten times the steady value in one section is not sufficient seriously to disturb the system as a whole, whereas if the entire system were forced to carry a current bearing this ratio to the steady value it would amount to a very serious overload indeed.

Sequential operation of a large number of sections from a central station, particularly in a large city, would require an extremely expensive distribution system, all the circuits being terminated at the station, or, in the alternative, a large number of separate control lines being required. If, on the other hand, the excitation of one section can be made to accomplish the delayed excitation of a succecdng section, and this, inturn, cause the delayed excitation of thenext section in the series, an extensive subdivision of the system can be accomplished at a reasonable cost and with a very simple type of control network. a,

, Such operations as just mentioned presuppose Accordingly, it is much that the control relays be mounted. inexposed positions; on the poles supporting the light, in manholes, or the like. This implies that in many cases the operating relay be exposed to extreme variations in temperature, and it is well known that equipment which will operate perfectly at average temperatures-6O degrees Fahrenheit or thereabout-may not operate satisfactorily" under sub-zero ambient temperature conditions, or'that their lives may be seriously shortened by operating undersuch conditions.

With these facts in mind; among the objects of this invention are to provide arelay which will successfully make and break circuits carrying substantial amounts of power; to provide a relay whichwill control such'circuits and which will delay their operation in so' controlling them by predetermined amountswhich may vary between a very few seconds andia'minuteor even several minutes; to provide a type of relaywhich may be operated in cascade, utilizing a control current ofthe same voltage and frequency characteristics as that in the circuits'to be controlled; to provide a relay which will operate under extremes of temperature with complete reliability; and to provide a relay which will withstand manythousand cycles of operation, under such extremes of temperature, without electrical or mechanical failure. i

In its broad aspect, this invention comprises a power switch of the mercury-tube type, wherein the make and break'operationsare accomplished by tilting the tube so that the fluid mercury either bridges or does not bridge a gap between two contacts sealed into the tube."- 'The drive mechanism for the relay comprises a motor, preferably a synchronous motor of the type familiar in electric clocks, which, in the most ordinary applications of our invention, .is providedwiththe usual step-down gear. The motor isof the self-startingtype and is provided with means for establishing two opposing electromagnetic fields, so that one tends to rotate the common armature and shaft structure in a clockwise direction while the other tends to cause rotation in. a counterclockwise direction. One of these fields is excited from a control'circuit, while the other is connected for excitation from the controlled circuit when the latter is closed through the mercury switch. Rotation of the motor tilts the switch-tube to make or break the circuit. For short delays the tube may bemounted for: direct operation by the motor shaft. Where. longer delays are required there is preferably mounted uponthe motor shaft acam,convenient1y in the form of a'disk having. a portion of its periphery cut awayso as to form. a slot or gap. The width of this gap determines the time delay of the relay; it may be very narrow, in which case the delaywill be very short, or it may'extend around nearly the entire periphery of the disk leaving merely a projecting tooth in which case a delay as long as 59 seconds (assuming a one-minute period of rotation for the shaft) will be the result. The faces at the edges of the slot engage a tiltable support which holds the mercury tube, rotation of the disk in one direction tilting the support and tube in such sense as to close the contacts, while rotation in the other tilts it back to open the contacts. To operate the relay, current is supplied to the motor field which rotates the shaft in the proper direction to tilt the mercury switch to the closed position. Rotation in this sense continues until the cam disk engages the tiltable structure and closes the switch, energizing the line to be controlled and the other motor field structure which is bridged across this line. This tends to rotate the armature structure in the opposite direction, balances the torque produced by the control field and brings the device to rest. The fields remain in this balanced condition until the voltage is removed from the control winding, whereupon the second winding assumes control, tilts the switch into the open position, and deenergizes the motor.

It will be apparent that the controlled line for one relay may become the controlling line for another, which may have either the same or a different delay constant depending upon the formation of the cam disk. As many relays as may be desired can therefore be connected in cascade, closing successive circuits at successive predetermined intervals.

Mercury switches of the type here considered have relatively long lives and will stand a large number of cycles of operation when carrying the loads for which they are designed. We have found, however, that when failure does occur as a result of continued use, the failure is caused by the flaking off of minute chips of glass from the interior of the tube. This flaking or sloughing action takes place very gradually and is apparently caused by intense local heating of the glass envelope due to the arcing or sparking which occurs, usually at the time of the break, but which may also occur at the time that the circuit is closed. Actually it appears that the sparking which occurs upon the make action is more destructive than that occurring at the break, even though it occurs less frequently and is apparently less severe.

It has been found, moreover, that the action referred to is much more serious at low temperatures than it is at high.

The reason for the phenomena just mentioned appears to lie in the nature of glass. Glass is frequently referred to as a supercooled liquid the reason being that when a true glass is cooled from a liquid state there is no point at which it can be definitely said that solidification occurs. The cooling curve of the material is practically continuous, no discontinuities occurring as in the case of a crystalline substance wherein energy is librated upon the transition from a liquid to a solid phase. Moreover, the viscosity curve of glass indicates that it is one of the most rigid and elastic substances known, but its rigidity increases with decrease in temperature. Local heating, causing expansion of the glass, is therefore very much more destructive when the glass is at greatly reduced temperatures than it is when it is relatively warm, not only because of the mere difference in temperature but also because the glass has more give when it is warmer.

A very important part of this invention, therefore, is the fact that there is provided, adjacent to and preferably surrounding the mercury tube, a heating winding of relatively high resistance which is bridged across the operating contacts and which carries current when the switch is open, thereby maintaining the glass at a temperature which is materially above the ambient temperature. To accentuate this, the relay is preferably surrounded by a complete enclosure.

When the mercury switch is open the temperature is therefore maintained by the heating coil. When the switch is closed the heating coil ,isshorted out by the mercury itself, but since the latter is carrying a relatively heavy load, the heat generated would be of the same order of magnitude, or even higher, so that it will maintain the glass in its more malleable condition. Furthermore, when the relay is closed, additional heat is generated in both of the actuating coils and this is retained by the enclosure. The total temperature rise under both conditions will be generally of the same order of magnitude, although somewhat higher, usually, during the closed condition of the switch. This is not disadvantageous, however, as it keeps the glass in its most malleable condition at the time of severest sparking, occurring at the break.

The invention will be more clearly understood from the following description of a preferred embodiment, taken in connection with the accompanying drawings, wherein:

Figure 1 is a plan view of a relay embodying this invention, with the cover removed;

Figure 2 is a front elevational section through both relay and cover, the plane of section being taken along the front wall of the enclosure of Figure 1;

Figure 3 is an end elevation of the relay mechanism, the base and cover being shown in section;

Figure 4 is a wiring diagram showing a plurality of sections of a lighting distribution sys tem actuated by relays in cascade and showing the wiring of the relay and heating coils.

Figure 5 is a front elevation of a modified form of relay of this invention which may be used where time delays of from one to two seconds only are required.

The embodiment of the relay of this invention illustrated in the first three figures comprises a baseplat-e l whereon there is mounted the frame 3 of the relay motor mechanism. In this case the frame comprises an L-shaped strip, the vertical arm of the L being of a length which extends from the baseplate i to the full height of the structure so that a surrounding enclosure or cover 5, when in place, rests upon the top of the frame. Th horizontal branch of the L is secured to the baseplate, and is provided, at its end, with a short vertical riser I. A motor mechanism, generally indicated by a reference character 9, is mounted upon studs H and I3 respectively, the lower of these studs being supported between the riser portion 1 and the vertical arm 3 of the L, while the upper one is secured to the vertical arm alone.

The two field structures of the motor, comprising the cores l5 and I5 and the windings l1 and 11' are mounted on the studs just described. The armature structure of the motor is enclosed in a cylindrical housing [9. This structure is not illustrated in detail as it comprises essentially simply two clock motors of known type mounted in bucking relationship, and various types of such motors are well known and are commercialany .portion of the periphery ma be cut 1y available. The usual step-'downfgear case? I, of thesa me type as is used inmost clock motors,

is mounted on the end of the h'ousingm. The gearing within this housing reduces '"the 'speed Thelower of the two-studs supporting the motorgte, stud H, projects'forwardofthesupport'frame 3"t'o frm' a stationary-shait on "which is mountedabearing bushing 21. The lower end of a vertical arm 2 9-oif-a T shaped rocker is secured to "this bushing. The horizontal arm 3 I of the rockeris provided with spring clips 33 for 'holding a mercur'y "switch-tube or 'capsul'e 35 ofknown type.

'I'he vertical arm of the rocker is'slotted, as shown at the reference character 31', to permit the passage of the motor shaft.

b'y bendinga tab out of the material of the 'rok'er itself) projecting outii ardly at 'its median line The periphery of this disk character '40, and the time delay bf the relay is determined by the width "of this slot, which may *be varied as desired, as has alread been stated. ln the present "instance the "slot is'shown of such proportions as would give ade'layo'f. approximately six seconds'with ashaft spee'dof one revolution per minute. It should be "clear "that in. this manner, or to almost the entire per-i iery of the disk, in which case "the cutaway portion losesits slot-like character and the disk becomes more comparable to a single-tooth gear. The relativeposition of the parts as shown tang-um 2 is onewhich would never 'be assumed in actual, operation but'which is shown in the drawing for purpose of clarity. As shown in the drawing the rocker 29 would be in unstable equilibrium and this position would be assumed only when one or the other of the two edges of the slot 46 were in contact with th pin 38 on the rocker arm. Clockwise motion of the cam disk 39 causes the right hand edge of the slot to engage the pin and tilt the rocker to the left, causin the mercury in the tube 35 to bridge the two contacts connecting two leads 43 and 45 (which'ai'eshown only in part) and thus making the circuit. counterclockwise rotation of the cam disk will move the rocker in the opposite direction, breaking the circuit. The instability of the rocker arm when it is nearly in the position shown is in creased by the fluidity of the mercury, which, as the device is otherwise nearly in. balance, starts to flow in the direction of motion and speeds up the make or break operation.

A terminal block 49 is provided on the forward portion of the baseplate and the power and control leads 5| and 53 are connected to this block. The switch leads 43 and 45 connect with the power lead 5|, but this is not shown in the first three figures in order to prevent confusion in the drawings, the actual connections being indicated in the circuit diagram of Figure 4.

Surrounding the switch tube 35 is the heating coil 55. This coil is wound of fine resistance wire insulated by heat-resisting material such as asbestos or fiberglass. The coil terminates at the same contacts as the leads 43 and 45, also as indicated in the circuit diagram. In the par- The "arm -2 9 is also'iprovided wi-th a pin "'38 -(whi'ch ina-y be formed if designed for operation on 240 volts.

and the operation-repeat.

ticular relays shown "in thejdrawings' the coil 55 is-designedlto dissipate about sixwatts, having 2,440 ohms resistance whenb'uilt for operation on a 120-volt circuit or nearly a thousand ohms When the switch is closed the dissipation by the heating elementis of the order of one one-hundredth or 3, watt, while a 30-amp. current through the mercury generates about 3.6 watts. To this,however, must be added the heat generated inv the coils of the motor so that actually the heat liberated within the casing is very nearly'the same under both conditions. I v

Theoperation of the device can best be appreciated by considering the circuit diagramjof Figure 4,. The control leads 53 connect to the coil ll and'nothing else. Excitation of thiscoil causesclockwiseoperation of the motor of relay Aftiltingthe switch tube35 into theposition shown, and closing the circuit between the input leads 5| to the switch. These leads are in series with the lighting circuit 5 which is a branch of a distribution network '59, thus-lighting the lamps 6| excitingcoil l1, and thus bringing the motor of relay A to rest. At the same time control circuit 5330f relay B is energized, startingthis motor in turn in the clockwise direction so that eventually the switchtube 35B will close Relay A will remain closed until the control voltage is removed from the leads 53, at which time coil ll, no longer balanced by an opposing torque, will takecontrol and open the relay, thus, in turn, removing the control voltage from coil 53B and again the operation will repeat, 3 as many circuits as desired being opened or closed in successionafter "that the device will work satisfactorily with Iull voltage on the coil 111i. e., 120 volts, even though the control voltageacross coil ll be dropped to less -than volts. Similarly, operation wasjstill satisfactory with control voltage on coil 11 "of volts and a'controlled circuit voltage of 90 or less. When the voltages of both coils were reduced together, satisfactory operation could still be obtained through about the same range of voltages.

Because of the use of the heating coil, operation under extreme conditions is still satisfactory; the device has been cycled repeatedly at temperatures of minus one hundred degrees Fahrenheit, the freezing temperature of mercury being minus forty. Furthermore, the life of the device, using the heating coil is indefinitely long. Operated at normal ambient temperatures the life of the device even without the use of the coil and considering a cycling operation of once per day as normal for this type of use, indicates a life expectancy of approximately forty years. The etching or flaking off of the inner surfaces of the tube when used at greatly reduced temperatures without the coil indicated a greatly decreased life under these circumstances, but using the heating coil an approximately equal life may be expected even under these conditions. The durability of the relay indicates that it may be expected to last longer than the circuits on which it is operated.

Where time delays of short duration only are required, i. e., delays of from one to two seconds, the cam and rocker mechanism may be omitted.

This modification of the device is illustrated in Figure 5, wherein the motor mechanism itself is identical with that described above and the parts therefore are identified by the same reference characters. In this instance, however, a crank arm 60 is mounted directly upon the motor shaft. This arm carries a clip 6! for holding the mercury switch tube. The switch tube in this case is supplied with the heating coil 55 as before.

A stop 63 may be provided for limiting the motion of the crank arm in the counterclockwise or off position. The reason for the stop is that there is one specific set of conditions wherein the current through the heating coil may become a sneak current which tends to cause rotation of the relay. This condition occurs only when through some accident the load is removed from the work circuit, as, for example, by the burning out of all the lights in the circuit. Under these circumstances the potential across the work circuit divides between the off" coil IT and the heating element, in which case enough current may flow through the coil to produce sufiicient torque to actuate the motor in the off direction when the coil H is not excited. The stop blocks any such motion. When normal load is connected to the work circuit this effect does not occur, since, in a 120-volt, 30- amp. circuit, the resistance of the load, brid ing the coil [1, is of the order of from one-half to four ohms, and this effectively shorts out the coil l'l' so that practically the entire voltage drop takes place across the heating coil =55. In the form of the device shown in Figures 1 to 3, no additional stop is necessary since the slot 49 serves the same purpose in blocking rotation of the motor.

Street-lighting circuits require complete reliability. Ordinarily the relays here described have shown such extreme durablity that no especial precautionary measures appear necessary to take care of the situation where some relay early in the cascade system may fail, causing those later in the sequence to turn to the off position. If, however, some such difficulty is feared, the relays described are sufficiently inexpensive to permit two of them being operated in parallel with respect to both their control and work circuits, so that should one fail to operate through switch-tube failure or otherwise the other will carry the current. A slightly less, but often adequate degree of protection may be obtained merely by the use of two mercury switch tubes in parallel, both operated by the same actuating mechanism.

We claim:

1. A power relay comprising connections for a control circuit and a power circuit, a mercury switch connected in series with said power circuit connections, said switch comprising a pair of contacts and a quantity of mercury within a sealed tube, an actuating coil for closing said switch connected to said control circuit connections, and a heating element connected across said switch contacts and adjacent said switch tube, whereby said tube and mercury are maintained at a temperature elevated above ambient by said heating element when said switch is open and by current through said mercury when said switch is closed and said element shorted out.

2. In a relay for controlling power circuits under extreme conditions, a switch comprising a sealed tube having a pair of contacts passing through the walls thereof and a quantity of mercury within said tube for making connection between said contacts, a power circuit including said contacts, and a heating element connected across said contacts, whereby said tube and mercury are maintained at a temperature above ambient by said heating element when said contacts are open and said heating element is shorted out when said mercury is heated by power current therethrough.

3. In a relay in accordance with claim 2, a control circuit, a closing coil for said relay connected in said circuit, power and load connections for said mercury tube contacts, an opening coil for said relay connected to the load side of said connections and bucking said closing coil to block the action thereof when said contacts are closed but inefiective to open said relay when said closing coil is excited, and a heat retaining enclosure surrounding said coils and said switch.

JOEL OSTLIND. WILLIAM L. WERNER.

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

UNITED STATES PATENTS Number Name Date 1,272,444 Holliday June 6, 1918 1,272,447 Jacobs June 6, 1918

Images