US1107135A - Telephone-exchange system and apparatus. - Google Patents

Description

.E. E..GLEMENT. TELEPHONE EXCHANGE SYSTEM AND APPARATUS.

-APILIOATION FILED JULY 11, 1905.

1,107',1 35. v Patented Aug. 11,1914.

9 SHEBTS-SHBET 1.

I E. I CLEMENT.

TELEPHONE EXCHANGE SYSTEM AND APPARATUS. I AAPBLIOATIOH FILED JULY 11, 1905. 1,1 07 1 35 Patented Aug. 11,1914

I a SHEETS-SHEET 2.

E. CLEMENT. TELEPHONE EXCHANGE SYSTEM AND APPARATUS.

APPLICATION FILED JULY 11, 1905. I 1510731 35 Patented Aug. 11, 1914;,

' 9 SHEETS-SHEET 3.

dllllllll lllllllllllll E. E. GLEMENT.

TELEPHONE EXCHANGE SYSTEM AND APPARATUS} APPLICATION FILED JULY ll, 1905.

Patented Aug. 11, 9 14* 9 SHEETS-SHEET 4.

6 3 .I 3 2: :1 7d r n v 7 Z w M 4 w 4 x m 4 7 M 7 w 4 W n 51 F ll v T! 4 0 IO! 11:! I fl 0 F M f a o WZZM JJeJ:

B. E. CLEMENT. TELEPHONE. EXCHANGE SYSTEM AND APPARATUS APPLICATION FILED JULY 11 1905.

I SHEETBS HBET 5.

Patented Aug. 11, 1914.

E. E. ULEMENT. TELEPHONE EXCHANGE SYSTEM AND APPARATUS.

APPLICATION FILED JULY I J 1905.

Patented Aug. 11, 1914 8 SHEETS-SHEET 6.

mu E ew v M T W B 11. CLEMENT. TELEPHONE IQX U H AN GE SYSTEM AND APPARATUS. 51 1 516111011 IILED'JULY 11, 1905.

'Patehted Au 11, 1914 l SHEETS-SHEET 7.

E.-E. CLEMENT. TELEPHONE EXOHANGBSYSTBM AND APPARATUS.

APPLICATION FILED JULY 11, 19 05.

Patented Au fn, 1914; I

Wifiwsseai E. CLEMENT. TELEPHONE EXCHANGE SYTEM AND APPARATUS.

APPLIUATION FILED JULY 11, 1905.

Patented Aug". 11,1914.

9 SHEETS-SHEET 9.

UNITED STATES PATENT OFFICE.

IEDXVARD E. CLEMENT, OF VTASHINGTON, DISTRICT OF COLUMBIA, ASSIGNOR, BY MESNE ASSIGNMENTS, TO FREDERICK C. STEVENS, 0F ATTICA, NEW YORK.

TELEPHONE-EXCHANGE SYSTEM AND APPARATUS.

To all whom, it may concern:

lie known that I, EDWARD E. CLEMENT, a. citizen of the United States, residing at \Vashington, in the District of Columbia, have invented a certain new and useful Improvement in Telephone-Exohange Systems and Apparatus, of which the following is a specification, reference being had therein to the accompanying drawing.

My invention relates to telephone exchange systems and apparatus, and has for its object the improvement of what are known as semi-automaticcircuits and apparatus. By semi-automatic, I mean those systems and that apparatus in and by means of which the subscribers are interconnected through the conjoint action of operators and automatic apparatus, the latter being controlled either by the subscribers or by the operators. It will be apparent after reading the following description that my appara tus and much of the circuit arrangement will be equally useful in full automatic or semi-automatic systems. I shall describe specifically a semi-automatic arrangement but I do not wish to limit myself thereto.

My invent-ion herein set forth makes use of the circuit described in my prior application filed June 5, 1905, S. Number 263,7 84, but is not confined thereto, as will sufficiently appear from the following detailed description. Briefly stated, the circuit scheme in the present case is for a two. wire metallic circuit, with quick working magnets and very slow control magnets, preferably solenoids with movable cores. I shall call this for convenience a two-wire circuit with time element control.

The apparatus described herein comprises:

1. An automatic switch suit-able for any, circuit. 2. A controller or side switch for the same. 3. A. sluggish relay by which the time elementl is made available. 4. An improved subscribefls sender, or variable signal transmitter.

' The switch consists of a frame or base plate having 100 contacts in separated groups of ten, mounted on a straight strip along ope edge, with a parallel spindle carrying {in-arm or stud for each grou of ten, theseflainis being at different ang es in a helical line around the spindle. The latter has a reciprocating motion through a space equal to the'length of one group of contacts, also a rotary motion whereby one stud after Specification of Letters Batent.

- Application filed July 11, 1905 Serial No. 269,157.

-erence letters an Patented Aug. 11, 1914;.

The sluggish relay comprises a solenoid preferably arranged vertically, with a solid core sliding up and down within its spool,

which is preferably lined with copper. When down the core rests upon the cndof a contact spring, to close it upon another conform a check valve at the upper end of the I spool, so that the core will move in adashpot.

I find however that this is not necestact. The core is intended to pull uprapidly, 1 but sink down slowly, andto this end I may sary in ordinary practice, as the application of current energy brings the-core up very sharply, whereas its own weight has to carryit down, and it moves comparatlvely slowly. Thev subscribcrs sender comprises a disk orwheel having fivesets of teeth, four sets 'havmg ten teeth each and the last one an indeterminate number, depending upon the character of the system in which it is to he used. Each set of teeth has a segment underlying it, all pivoted on the same center,

each segment movable by hand to cover one or more of the teeth, one tooth in each set being preferably exposed normally. An opening in the disk over each segment exposes a number which changes as the seg-" "nent is' moved, to indicate the number of teeth exposed or covered as the case may be. When set, the entire disk may be revolved by pulling a handle which windsa spring train that is then detained by a trigger until a magnet lying beneath the movement is energized from central to pull the trigger.

My invention is illustrated in the, accom-' panying drawin s, wherein the sameref- .viewo the contro1ler orside'switch, Fig. 6

figures point out thesame segments.

is a section thereof on the line 6 of Fig. 5. Fig. 7 is a back view of the switch. Fig. 8 is a vertical section thereof on the line 8 of Fig. 5. Fig. 9 is a side view of the sluggish relay, with the spool and winding in section.

Fig. 10 is a face view of my subscribers,

sender, and Fig. 11 a top view of the same inclosed in its case. Fig. 12 is a face View of themechanism with the case removed, and Fig. 13 is a side view of what is shown in Fig. 12, the contact springs being broken away to show the manner of pivoting the Fig. 14 is a bottom view of the sender; showing the polarized release magnet. Fig. 15 is a rear view of the dial or disk of the sender, removed. F ig. 16 is a diagram of a subscribers line, with sub-station and central station terminal apparatus and an operators cord circuit. Fig. 17 iwa similar diagram of a switch trunk and automatic switch, with cooperating parts.

Referring first to Fig. 16, A is a subscribers station equipped with the usual transmitter T, receiver R, switch-hook H, condenser C and ringer Q. This station is connected to the central oflice where linewires 1-2 terminate in relay contacts n, a of the cut-oii relay N. These contacts are normally in engagement with other contacts nected 'through a, a the first connected by Wire 5 through relay M to the main battery B,- and the second to ground. The contacts, n, a, are taken away from their normal resting contacts when the relay N is excited, and brought into engagement with'other contacts n n connected to the line jack J, in the spring and sleeve sides of the same, through the wires 3-4, and also through wires 78 to the multiple-contacts J which are disposed in automatic switches.

The cord-circuit shown terminates in an answering-plug P and a calling-plug P, each with tip and sleeve contacts 19, p, concord-conductors 9 11, 10-12, broken by condensers C, C. The main battery .13 is bridged across each end of the cord through a pair of relays; On the answering end the control relay S is between the battery and the sleeve side, and the supervisory relay\S between the battery and the tip side, the former being a front contact relay and the latter a back contact relay. These two relays jointly control the local circuit of the supervisory lamp 8. On the calling end of'the cord the arrangement is the same, the control relay S and the super visory 'relay S jointly governing the circuit of the supervisory lamp s the calling end of the cord is also included the ringing key K, which may be provided with listening contacts, if desired, but in the present system is not so shown.

So far, the circuits and apparatus are all of a well-known type of two-wire multiple,

" and as I describe my additions thereto it will appear that this much, which is the essential and distinguishing part of the manual system, remains unchanged. I add to but do not alter the cord or line-circuit.

In a multipleswitchboard the wires 3-i are connected to the multiple-jacks through the branches 7-8; and here the first difiercnce occurs, for I discard the multiple-jacks, and substitute the terminals J in the banks of the switches.

At the subscribers station it is necessary to provide some means for sending impulses in predetermined order. These all go over the No. 1 side of line, hence it is possible to make and break the circuit manually, or to use an automatic sender such as that shown in Figs. 10 to 15. It is entirely possible to use the ordinary switch-hook without any change whatever. This is manipulated to send successive short breaks in groups, leaving the line closed for an appreciable time between the group s,the number of breaks in each group corresponding to a digit of the number wanted. Thus in sending the number 4436, the subscriber would remove his receiver from the hook, and listen. In Fig.

16, I have shown the special variable tone generator G connected through a condenser C to the wire 5, so that as long as the line is connected to the line-relay M a characteristic sound will be produced in the subscribers receiver. Even without this, however, the subscriber can tell when the plug is inserted, because (if the circuits are traced. it will be found) the battery is reversely connected to line when the plug is inserted. Having waited until he hears .the plug inserted, knowing this either by the co. sation of sound from generator G or by hearing the loud click due to the battery reversal, the subscriber taps his hook sharply and rapidly four times, then after a short interval during which it remains up he taps it four times more. As soon as he has done this he will hear the battery cut off by the opening of contacts e 2 in the side switch, and he will then wait until battery comes on again, if the generator G is employed the tone signal becoming audible in the meantime, innsmuch as the cut ofi'-relay N has lost its current. When the plug P is inserted in the trunk jack J the relay S pulls up, the side switch Z moves forward a step, and the contacts 2 2 are again closed, whereby magnet N again takes cl'lrrent, relay M and the generator G or reversed battery B are cut off, and the subscriber knows he can go ahead. He then taps the hook three times, and after an interval six times The description of this operation sounds somewhat involved, but really it is simple in the extreme. The subscriber has only to remember that when he hears the musical note of '{Icnerator G p'he must wait until it-ceescs before he can signal. It almost goes without saylilj Cir

ing that a phonograph might be substituted for generator G giving oral instructions.

My sender is illustrated in Figs. '10 to 15, inclusive. The function of this sender is to transmit makes and breaks, or more properly speaking to send over the metallic circuit long and short current impulses in predetermined order. The sender herewith presented is a device having five groups of teeth,

four of which are adapted to be varied as to number from two to ten by covering segments, the fifth being invariable and con sisting of one or'more teeth or notches, which control the testing, the ringing, etc. The four segments are set by hand to uncover the desired teeth or notches, and then a train is wound up which turns the whole disk, bringing the successive teeth successively into position to engage the line-contact springs.

Fig. 10 is a face View of the device in its case. The central spindle carries a dial W, around'which' the edges 20 of the case are rounded so as to conceal the edges of the disk and the teeth thereon. In this dial'are out four slots @0 w, w, w". In the first named works the handle to", in the second the handle w, in the third the handle w, and in the fourth the handle 10 These handles protrude from the face of the instrument a suflicient distance to enable them to be grasped conveniently, as best shown in Fig. 11. In Figs. 12 and 13 I have shown the mechanism without the case, and in Fig. 13, particularly, there appear below the disk W the four segments 10", 10 w and 10 These are all pivoted on the central spindle W and have the sameradius as the disk, the result being that when a handle, as for instance to", is moved over to the extremity of its slot 10, all the teeth, to, in its set except one are covered up as indicated in the dotted line in Fig. 12. In order to be able to set these. segments accurately, I provide four openings in the face of the disk W, as shown at 10 to, w, @0 On each segment I mark 'in a curved line concentric with its edge the numerals from 0 to 9. If a segment is left with its numeral 0 showing, only one tooth of its set will be uncovered; if the nupiber 1 is showing, then two teethwill be uncovered, and so on. The reason I add this is that in order to make the successive selections in an automatic exchange system it is necessary to send four sets of impulses, each of which may consist of any number from 1 up. Obviously, therefore, it

y is necessary to regard the 0 as the one im-- pulse figure, otherwlse no number containing a naught could be transmitted. 'By moving the handles and thereby the seglnents'yon the disk W, they may be set so that any tiombination'of teeth appear uncovered. i

- Referring p-articularlv to Figs. 12 and .13,

"the central spindle W below the disk W passes through a bridge w, which extends panion gear w and the heavy fly-wheel W".

Below the fly-wheel is fixed the angle arm w by which the rotation. of the spindle and the disk WV is stopped. Attached rigidly to the companion w is an arm w,

carrying a pawl w taking into the ratchet M22. Uprising from the fly-wheel W is a post upon which is carried a pawl w, which engages the ratchet @0 To one side of the movement thus described is a post w carrying at its upper end the contact springs w", w, and 10 separated and securedby the insulation to. 'Pivoted upon the post below the springs is a winding arm W carrying a handle 10 and a stop-pawl w, the latter, however, not being essential, but provided to take into the latch w when the arm is up or idle, so as to positively prevent any back movement of the disk, This arm V carries a toothed segment or are w", engagingthe pinion w and adapted to turn the' same when the arm is moved. The arm is pulled down by hand, and is moved back to the position shown in Fig. 12 by a suitable spring to attached to a post to indicated in dottedlines. As the arm is pulled .down the pinion 'w is turned, the arm w and pawl w turn with it and the latter moves back over the teeth of ratchet w, .which it finally engages with a tendency to turn as the spring exerts an upward tension on the arm. This turning is prevented, however, by the stop-arm to which normally engages the detent w on the end of the horizontal arm of the bell-crank or trigger w, which is piyoted at its angle in a slot in the plate W and iscontrolled by the armature w of the polarized controlling magnet W. This magnet'is.supported on the back plate as indicated at W9", and car'- ries a yoke w, in which the armature w is pivoted by means of screws w. This soft iron armature lever 10* carries the magnet winding w, the field being a laminated permanent magnet 10 whose component parts are secured together in any desired manner, as by boltsw. The advantage of having the winding carried onv an armature is that the permanent magnetism is never reversed but is always strengthened. After vthe 'box is wound up, when the magnet is energized by current of proper polarity, the

armature lever w. is thrown ov er to trip the trigger and release the. spindle. It will be observed that no retracting is requiredfor this trigger, by reason or the fact that it isanoved positively in bqthdirections by the al 'inature. The lower end f the trigger is slotted to take a. pin or screw w? tapped into the end of the-armature. The train being thus released, the spring w draws up the arm V and with it the companion w, the awl w and ratchet 'w, the twinratchet w-i, which is fast to w, the pawl w and finally the heavy wheel W. This balance wheel prevents the. too rapid movement of the train, which as it turns the spindle, turns also the disk W. The turning of this disk, which is in the direction of the arrow in Fig. 12, makes and breaks contact between the springs to, w, as the groups of teeth \pass them. It should be stated that when the arm W is first moved down, supplementary arnrw" moves down with it, carrying a in 10 away from the spring L029, permitting the latter to leave the spring 10 with which it is normally in contact and come into position to be engaged by the spring w. hen the. arm Vi again reaches its uppermost position, the pin /10 returns the spring to to the position shown in Fig. 12. Owing to the double arm 7.12 the train is'only permitted to run a little more than one-third of an entire revolution, when owing to a reversal of current inthe line after the second group of teeth has passed the springs, the trigger to again engages the .arm at the end w 'whercupon the spindle is instantly stopped, but the wheel turns idly on, its pawl 10 riding over the ratchet teeth in. Upon a second reversal of current in the line the trigger is again pulled and the arm is. again released, when the disk V and the spindle turned through the balance of their entire revolution, until the parts again resume the positions shown in Figs. 12 and-l3.

Associated with the cord-circuit is an indicator shown as comprising two stcp-bytep disks, each adapted to be set around by a magnet I or I, a retaining pawl being provided for each disk, and a restoring device, 1 The indicator magnets, I, I are each grounded on one side, and are connected back through wires 22 and 220 to the contacts a and a of the control or pilot switch, Z. This indicator may be of any well known or desired form. This shifting or -pilot switch Z-consists of a spindle with a ratchetwhcel and a magnet to turn the same, the latter being shown at Z, and the movement as shown in Figs. 5 to S, which will now be described. Referring first to Fig. 5, it will be seen that the stepping and release magnets and the working parts of the side switch are mounted on abuse plate 2". The stepping magnet 2 is fastened to the plate 2"" by means of the angle support This magnet tvhcn energized attracts the armature 2", one end of which is pivoted at .2 The other end terminates in the fork 2-", which engages the pin The manner in which this fork engages the pin is well to the plate .2 by the shown in Fig. 8. The pin 2 is fastened to the arm 2 which rides freely on the shaft 2" and carries the pawl 2 which engages the ratchet wheel 2 when the magne a is energized. In Fig. 5, the apparatus is shown. in its normal position, and it will be noticed that one end of the pawl when in this posi-- tion rests against the sto 2 which throws itout of engagement with the teeth of the ratchet wheel. This is to allow the ratchet wheel to resume its normal position when the release magnet is energized. The ratchet wheel is fastened firmly to the shaft .2 as are also the wipers e, .2 shown in Figs. 7 and 6. This shaft extends through the plate and is supported by the bridge 2 and the plate itself. In this form of construction I am. enabled to mount the magnets and operating parts on one side of the plate, while the other side is used exclusively for the wipers and their banks of contacts. These wipers are insulated from one another b the bushings 2,'and are held on the shaft by the nut, a. The contacts 2, a", 2*, etc., Fig. 7, are separated from one another by the insulation 2 as shown in Fig. 8, and are held in place by the screws 2 passing through the brass clamping p.

base plate, .2.

and threaded into the the-a contacts are shown in Fig. 16 in pairs, mp posed to be opened and closed sncc r as the arms or cam drum are turn .2. the first step the pair 2 a is close to cut in the thousands indicator magnet, l; at the next step this pair is opened and the next pair, a, 2 is closed to cut in the hundreds indicator magnet, I; at the next step this pair is broken, and the next paiiyz e 5-,: closed to connect the battery wire 2.4.0 by way of wires 17 and 21 to the conductor it and so through the tip of the plug to the tip side of the switch trunk, so as to repeat the impulses that follow, around the condenser C, into the trunk circuit; this same step opeps the battery circuit through wire 14.- by opening the contact pair a .2 which is normally closed. There is then another step by which the pair .2 2- are again closed and the pair .2 2" left closed. The final step or restoration then cuts off the wire 21., leaving the pair 2 2 closed and all the other pairs open, so as to havea clean cord-circuit. The wipers are restored to their normal position by the watch-spring .2 shown in Figs. 6 and 8, one end of the spring being fastened to the shaft, while the other end is fastened in 2. The release magnet .2 is fastened to the plate 2 by the angle support 2'. ,This magnet has the armature e pivoted at To this aianaturc there is fastened an arm .2, which it rmally engages the teeth of the ratchet wheel, 2", and when dci ncrgized the lllg' lltt attracts the armature and the arms releases the wheel 2, which allows the shaft carrying the wipersto return to its normal position. The movement of the pawl 2 controlled by the armature 2 is regulated by the setscrew a, which is mounted to the plate 2 by the post a. This screw engages the lower end. of the arm e which carries the pawl, and by adjusting this screw the move ment of the pawl as well as the movement of the wipers is readily adjusted. Through the end'of the shaft 2 projecting above the bridge 2 a pin .2 is driven, which engages a similar pin 2 projecting from the lorldge itself. These pins on coming 1nto engagement serve as a stop for the ratchet and wipers when released by the magnet a al- .ways causing them to come to rest in their normal or starting position. The operation of this switch is as follows: When the step ping magnet 2 becomes energized, the armature .2 is attracted and causes the pawl .2

' to engage the ratchet wheel 2 causing itto move the space of one tooth. The wipers z,

'2 being attached firmly to the shaft, but

separated from one another by the insulating material 2 are caused to make a move ment corresponding to the movement of the, ratchet. 'The contacts a, a, 2 etc., are

spaced corresponding to the movement of one tooth. When the stepping magnet 2"? 1S energized the-first time the wipers a and e engage the first pair of contacts, and when tracted an the arm a releases the wheel 2 the shafts carrying the wipers start-s back to itsnormal posit-ion under the tension of the spring 2 When the pins 2 and a come into contact the shaft is brought to rest and the wipers are now in their normal or starting position.

In Fig. 9 I have shown a sectional plan of the solenoid which controls the foregoing pilot or' side switch. The solenoid'consists of the winding .9 wound on the copper sleeve shaving insulated ends .9 s. A

soft iron cores' moves freely in this sleeve, and in its normal position the lower end having the insulating block 8 rests on the contact spring s, causing it to make connection with the spring .9. The solenoid is fastened to a base plate 8 by the angle sup port .9, this angle support being fastened to the solenoid by means of a screw a threaded into the soft iron pole piece 8. This pole piece becomes magnetized when the circuit of the winding 8 is closed and "serves to reinforce the pull on the sliding core .9 The contact springs s and s are mounted on a support 8 and separated from one another by the insulating material 5 The frame of this support has a projeci'ion s which serves as a stop or rest for the contact springs s and 8", when the core 5 of the solenoid is in its normal position. Referring now to Fig. 17, J is one of a group of trunk jacks located in front of each operator, each trunk line, 31 and 32, preferably appearing on a jack J in multiple at every section of the switch-board. The trunk comprises two wires 31 and 32, the former normally 0 Jen everywhere, the latter connected only t rough the relay S to ground. \Vhen a plug is inserted in a trunk jack the magnet S gets current through the sleeve side of cord and magnet S, and pulling up closes the tip side 31 of the trunk t-o relay'S and ground. impulses coming over the tip side thereafter affect this relay, which corresponds to the supervisory relay S ofthecord-circuit. Each trunk is provided with a side or pilot switch by manipulation of which the different pieces of apparatus are successively brought into proper relation with the relay S and other parts of the circuits and apparatus, as will be fully explained in a moment. -This switch is here designated for convenience Z", and is in every respect similar to the switch Z in the cord-circuit. Each of those switches has an operating magnet for stepping its arm around a stepat a time, this marked Z in.

the cord-circuit, Fig. 16, and Z" in the trunk circuit, F 1g. 17. v Fach switch ls'controlled by a solenoid or other slew relay S, wlnch is quick to be energized and pull up its arma-- ture or core, but very slow to fall back and connection initiated by the operator with This I plugs'P and P may becompleted. switch is of' the type designed by myself and showninFigs. 1 to 4. Fig. lshows a face view of the switch containing the straight row of contacts, the spindle-carryin'g'the wipers or .studs, the rotary, nertical and-release magnets with their nec'ssary operating parts. Fig. 2 is a side view of the switch showing-the magnets and their operating parts, the, row of contacts being omitted forclearnessl Fig. 3showsa form of contacts used where the studs in place of the ordinary wip'er's 'are; -mounted on the spindle. The contacts this figure are arranged sop'la'tinum points can be used,

Fig. 4 shows a formo'fcontacts'used where "the ordinary ipers' are mounted ,Ol the has contacts are mounted as shown in Fig. 3.

7'; j, represent a pair of I line-contacts there being a pair of these contacts for every line connected to the switch. In this case there would be a hundred pairs of these contacts. j and 1' are bosses common to all v the-contacts of the switch, so that, when 'these 'busses are spread by one of the studs gonqtheispindle they will'c'ome into contact with any pair of line contacts picked out. The busses in this switch are made of a continuous piece of metal preferably Ger- 1 man lsilver made in the form of combs, each haying a hundred teeth, one tooth opposite eachline-'icontact. The line-contacts and the bosses. are "separated from one, another by insulating material as shown in Fig. 3. The whole strip thus made is fastened rigidly to the plate by means ofthescrews j, 3 1' etc.,passing through the heavy brass strip J. These screws enter the spacings between the various groups and thereby avoid shortcircuiting any of the contacts. 1' The spindle 79 preferably made of steel, is mounted on the p'l'ate'P by means of the supports p and 72. Fig. 2 shows the form of these supports, 79 being a steel bearing-piece mounted on the brass angle support p for rigidity. The spindle moves freely through theopenings in the steel bearings 19, p. The studs p, p, etc., are mounted on the spindle by-means of a locking device, such as a set-screw; each stud is nothing more than a pin with an insulating sleeve. Commencing with one end these arms or studs are set so that each will correspond to, one of the groups of contacts andwill lie opposite the spacing at the beginning of that group. They are arranged helically, that is the arms are set around by successive steps so that only one of them can ever be brought into line withthe strip of contacts to engage the same at, any one time. p is a broad face pinion with ten teeth, which is engaged by the pawl of'the rotary magnet to cause the spindle to rotate. p bis a circular ratchet or a ratchet with circular teeth, which upon being engaged by the pawl of the vertical magnet causes the spindle to step up in a vertical position. A fixed detent- 79 is secured to the frame, which as soon as the spindle steps up in a vertical position 'encannot turninntitit is slid back it its normal -or startingposition. :At the upper end of the'spindle'a ooil'spring is provided housed in a cap, p, having one end connected to the spindle and its other end connectedto an arm, 72 which slides on the post, p; which is fastened to the plate P. The, pair of rotary magnets Stare fastened to the frame by the angle support 8, their armature 8 being pivoted at the points 8 and s on the supports 8", 8 which are screwed to the base-plate. This armature has a pro-' jecting arm a carrying a pawl, which engages the teeth of the pinion p The tension spring .9 with one end fastened to the armature and the other end fastened to the support. .9 tends to keep the armature away from the pole-pieces of the magnets S.

The. pair of vertical magnets S, bettershown in Fig. 2, are fastened to the plate by the angle support 8, their armature s is pivoted at s to the support 8 which is fastened to the base-plate P. p The armature s is bent so as to form the arm 8 which carries the pawl s, which engages the circular ratchet 7). The tension spring 72 Fig. 2, one end of which is fastened to the nrmature s, the other end being fastened to, the base-plate P, tends to keep the armature away, from the pole pieces of the vertical magnet. The pair of release magnets X are fastened directly to the base-plate by means of screws passing through the plate and entering thecores of the magnets themselves. The armature w is pivoted to the support w which is fastened to the basepl'ate. A screw 13' passes through one end of the armature m and rests upon the detent piece :2, which normally engages both the pinion, p and the circular ratchet, p. This detcnt'piece ispivoted to the angular supports, m. and m which are fastened to the base-plate. \Vhen the releasemagncts are energized the armature r11 isnttracted and the screw m pushes on the detent piece m which then disengages the ratchet and the pinion and allows the spindle to resume its normalposition. The operation of this switch is as follows: The first impulses coming in actuate the rotary magnet S, which upon attracting 1ts armature according to the number of impulses, actuates the pawl which engages the pinion p and moves it accordingly. Suppose the first would have stepped up six spaces and 'the tacts of the group J, thereby connecting wipers themselves would have come into the ratchet.

reached its resting position and the detentstud 72 would pusha pair of teeth of the combs i 7' against the sixth pair ofconthe busses to pair 36 in the bank of contacts. It will be noticed by referring to Fig. 3 that only one stud can enter the contacts at any one time. If the type of wipers shown in Fig. 4 had been used, in which the arms or studs are provided with conducting wipers, the

contact with the springs leading to the de sired line but by using the type shown ln Fig. 3 I am enabled to use latinum contacts which in itself is a desirable feature. The method of operations, no matter which type were used would he the same. At the end of the conversation when the release magnet is operated, the armature a is at tractcd and the screw a2 presses down the detent piece as, releasing both the pinion and The spindle drops by its own weight and then turns under tension of the spring inthe case p. The spindle is kept from turning when first released by means of the detent 77", which enters one of the long teeth on the pinion p and prevents it making a rotary movement until it has 7) is disengaged.

Referring now again to Fig. 17, I have shown the terminal contacts of the switch at 72 p and have indicated the spindle at 72 The multiple line-contacts are shown at J and this illustration, taken with the foregoing description will enable any one skilled in the art to practice my present invention. The trunk wires 31 and 32 pass to contacts g and y, being also connected to contacts e 2 These contacts are normally open, the former pair being controlled by the test relay Y and the latter by the pilot switch Z. The wipers 12 ,12 are connected back by wires 33 and 34 to the contacts 3 y of the test relay, normally touching the generator contacts y), 1/; and wires 33 and 34 are also connected to the contacts a and 2 0f the pilot switch. At a point between the generator G and the sleeve side of wiper p, the wire 45 is connected through a choke coil to the main battery B, this being for the purpose of holding up the cut-oii relay N while ringing. A. busy-back generator G preferably working through a condenser, C is connected across the contacts 3 51 of the testrelay, these contacts being crossed together with the contacts 3 and y when the relay Y is energized, but insulated from the contacts g and y. X is the release magnet for the switch, and isconnected from ground by wire 42 to the contact a of the pilot switch. The generator. circuit 45-46 also passes through a pair of contacts 2 2, and the test relay circuit 4344 passes through a similar pair 2 2 of the pilot switch. At the right-ii and end of Fig. 17 is shown a subscribers line-circuit and its connected ap- PpIXtuS, for a sub-station A, similar to that o The operation of my system will now be stated:

Assume. that subscriber A desires to be connected to subscriber A. He first sets his sender to the desired number, which we will assume to be 3364. He then removes the receiver from the switch-hook and we find the following circuit conditions: The subscrib'ers talking circuit is broken at the contacts w and w by the operation of setting up the desired number. It will be noticed that while the talking circuit remains open the line is closed by the contacts w and to, through the polarized magnet w, in series. The closing of the line by this circuit encrgizes the line-relay M, which attracting its armature lights the line lamp m. The current furnished through the line-relay M does not affect the polarized magnet 20* as it is of the wrong polarity. The operator perceivingthe line-signal inserts the plug 1, thereby putting battery on the line of an opposite polarity to that furnished by the line-relay M. The polarized magnet w is now energized and releases the sender at the subscribers station, to send in the impulses in their proper order. lVhcn the plug P was first inserted the cut-off relay pulled up, disconnected the line-relay M and extinguished the line lamp m. The train 20 runs down, sending the first number in a seiies of short breaks, the relay S is caused to vibrate its armature, opening 'and closing acircuit which may be traced as follows: B, 24, s 8 249, a, s 17 2", z, I, ground back to battery. Each short break puts battery onto this cir-,. cuit, that is each short break in the line-'cir cuit producing a corresponding make in the local circuit just traced, and-causes the magnet I to advance its indicator one step. The number of impulses in the first sot deter-- mines the number to which the first indicator is stepped around.

At the same time that magnet I gets its first pulsation of current, the relay Z which is connected in parallel therewith through the wire 17, also takes current, and instantly pulls up its armature, which is made very sluggish in returnii'ig, so that on the succession of short breaks in the local circuit due to the short makes in the line, the armature will not fall back far enough to close the circuit ofmagnct Z. At the end of the first set of impulses, however, the wheel Vi at the subscribers station pro due-es a long make (or if the subscriber manipulating the switch-hook he lets it stay up 'for a moment). The effect of this is to cause relay S to holdup its armature and thereby cut off current from the relay Z long enough to permit the armature of the latter to. fall all the way back and close the step around the si switch, openingitacts 2, 2 and The 1" Water Ina I, I, are thus cut oil the. battery supplied through the circuit 18 of the magnet Z as follows: B, ill a, s 19, 18, Z, ground. Magnet Z then pulls up and steps the side switch around, opening the springs 2*, 2 and closing the pair 2, z. lhe first break in the "inc of the second set, as the subscribers wheel W continues to run, produces a corresponding make in the circuit of magnet I, and of relay Z the second indicator disk thus being started and the circuit 18 being asi'n broken. h l this second set oil eted, a long me. :e in I) the armature 3" and mature oi Ur;

also cut. oil from line-wire 2 and the cutoff relay N, the latter permits its armature to fall lack. At the same time current reversal in the line throws over the armature of magnet to, and stops the substation sender. The line-lamp on may now glow, and if the subscriber is using his switchhook for a sender he will hear the tone of generator G and the operator, perceiving the number which has been set- "up on the "indicatorb'y the magnets I, 1, takes up the -of lines which will be in at once.

plug P and inserts it in-the trunk jack J of a group bearing that number. There may be anywhere from 2 to trunk-jacks in a group, depending upon the percentage order to determine which of dress a the operator may test them in the ordinary way, hence my reference to listening contacts in the'key K. The test circuit is sim- -ple. It will be observed that the sleeve side of the trunk, 32, is normally toground, but when a plug is inserted in any one of the trunk jacks the potential of the sleeve side is raised by the connection of the battery I through the sleeve side of the cord. '1 may also use lamps or other busy signals instead of a test. The ordinary test 1s considered preferable, however.

As soon as the plug P is in the trunk .l the following circuit is established: B i the cord), 24, S 16, 12, p, j, 32, 41,

I 2 35, S and ground. The two rethe following circuit: B, 24, s", s", T59, Z, ground. Magnet Z thereupon steps the cord side switch around one more step, closes the contacts 2 2 and also closes the contacts 2 and The former closure put; battery back on e sleeve side of sulr scribers line, 2, again energizes magnet to, and lets the second set of impulses start. if the subscriber is manipulating his switchhool: he hears the generator Gr go off d the battery come on. The relay theotherhand, pulls up its two armatur clo ethe circuit 1 c, S thereupon pull up. Theformer I lino, tl'lcy are repeated as bci'mxire" end closessthe 'Sollowing two circuits;

memes battery B (in Fi 1?), wire 40, 8 a, s, and if the relay S be deenergiraed to :1 magnet to ground; also from wire 3 36, s, 8 37, 2 z, 39, S and ground. Now, since the wire 21 in Fig. 16 has connected the point 8 to the tip conductor ll ot' the cord, it is obvious that the. breaks c ining in over the line 1 and causing the armature of relay {'5 to fall back, will produce a {corresponding series of current zulsmio'is over ii-,1 H. T T" ,1 the rolle .ng patl i. l) mg. lo 2 t, 249, s, 9, ll, ill, 11. .51 the pupil of conne tion with wire 11 .ll-

the iidltlll. vides. a portion going by ich the for ring paths: fir. wire 15. rel

M an

wiie 1?) to ground and bard; to l sec ond, i'jhrough conductor ll, to p, n v 2 ll, .5, 37, relay 5, and to ground, also 31?,

.. 2' and S to ground. None of thew, parallel paths will allcct the others lllw'rllfll they are fed directly from lmttcry through wire 21 without any intervening r '--.E =l'.1ucc,

but care shouldbe taken that all the contacts and wire connections have a sulliciont current carrying capacity ,to supply the three paths without burning out or undue heating. Thus as the impulses come from the subscribers station they are repeated by the relay S around the condenser U into the tip side of the trunk. These impulses coming over the wire 31 affect the magnet S, which is one of the two stepping magnets of the automatic switch I. As it takes the impulses the. wipers p p, are moved, vertically if it is a Strowger switch, but in a rotary direction if my own type of switch, in either case the tens group of line-contacts J being selected in which the terminal o the wanted line are to be found. At the -.l of this set of impulses there comes a .lmake, as before, whereupon the armature a remains back long enough to make contact with the point .9 and by energizingl'bo magnet Z to step the side switch Z around one step, thereby opening the springs a, z? and closiu .2 2 thereby putting wire 37 through wire 38 to thesecond magni 5 of the switch I", this being the magnet which in the Strowger switch moves the wipers around the row, and in my switch moves the selector up the row, in both case. to select the unit contacts wanted.

hen the side switch in the trunlmvorl. the last time, the switch Z of the cord we moved a step; but no effect was prodm. thereby, the contacts .2 being mniniv 510 closed d urii'ig both steps. As the. 1:. )1. or uu'l impulses now come iu as locals ow 1} i iii, to energize magnet 8 as stated. A the end of a set of impulses the long; umhc comes, armature .s" in the cord stays up and ike ays e and of the side swicluof l and trunk, fall'sll'the way back, 1 side switches are advanced a step.

re switch Z opens the contacts 2*, a and re xains in idle position. The switch 2 opens he contacts .2 2 thereby disconnecting he magnets S and S, and closes the springs .2 thereby connecting the test relay Y o the wiper p by the following path: Y, l3, e 44, 34s. lis this wiper rests on the sleeve contact of the wanted line, if the sleeve side of that line has its potential in raised because of another connection existing, the magnet Y will instantly take current and will pull in its armature n, 3/ thereby disconnecting the generator contacts y and and bridging the busy-back G 15 across the wires 3]., 32 through the insulated contacts 3 3/ 3/ and 3 The calling subscriber then hears the buzz of the busy-back, or if desired the voice of a phonograph say- -ing Line is busy The subscriber thereupon hangs up. A wire 47, it will be noted, connects branch of the wire 4: 2 to the front contact 3. of the relay Y. When the relay pulls up therefore, as an ordinary automatic switch is used, this 3 puts the release magnet X on the wire 82 through the contact 3 and the magnet immediately takes current and pulls up. its commonly used in Strowger automatic switches thiswould be what is called a two-step release, that is the magnet when it pulls up would simply hook its lever onto the release end of the detent pawls oi the switch, and when it lets go the retractile swing of its armature would pull oil" the awis so let the switch return to normal or inoperative position. Hence in such case the switch remains set after the subscriber hangs up. When he hangs up, however, the operator perceives by the lighting of the supervisory lamp 8 that he has done so, and thereupon pulls the plug 1 out of the jack This cuts off the current supply from llilgl S and X, and as the latter then s go its armature, the switch is forthwith released. By winding the magnet Y high d making it of coi'isiderable impedance, o iiniount of current abstracted from a y circuit and the consequent annoyance to busy subscribers is no more than in any tin operation in systems now in $1156. tch shown in Fig. l is used as a connective switch, the release magnet must r arranged to operate when a release circuit iS closed under the control of the relay if line wanted is not busy, the magnet Y no current when the switch wipers come t on the line terminals d", and the proure somewhat dillercnt. it should be stated here that after the last teeth of the for number transmitting on the sub i eth which are alwaysoperative for the so setting theside switch Z". The witch Z is unafiected after it makes the l movements stated, either because its memes wheel, W there are two, or more arms have reachedv a stop which prevents further movement, or because the last steps go over tied contacts or solid contacts so as to maintain conditions unchanged. The side switch Z- has five movements, however, and if we cause the insertion of the plug P to give the first movement and out in the mag net S simply by putting battery on at J and at the same time through the closure at 8 putting battery current on magnet Z before relay S" is energized, which is the proper and economical course,) still the subscriber only gives two long closures after that, by which means we are brought to testing, but the wantedline must still be rung and the circuit then put through. So two more invariable breaks of short duration are provided just before the wheels W run down. If the line wanted is not busy, the next break comes in, in due course, and the 5 side switch Z makes another step, closing the generator contacts 2 and 2 which may be on one side as shown, or double-pole if desired. This puts generator to the wanted line. After any desired and predetermined interval the last break comes, followed by the final make as the arm w closes the talking circuit 50--52-53, at the sub-sta-. tion, and the switch Z makes its final step, opening the generator contacts, leaving the .5 test contacts open, and takingspring 2 off spring 2 and putting it on 2, thereby cutting oil the relay S and cutting on the release wire 42, which remains on the sleeve side of line during the conversation. Everything else is disconnected,however, including magnets 8* and S and wire 86. The same step of the side switch closes the contacts 5 21 and 2 2 whereby the trunk is prolongedto the wipers, with the same clear talkingcircuitas in a manual system, and with no more contacts than any, but less than most, to talk through.

After the trunk is completed, the two supervisory relays S and S in the cord-circuit are responsive to the two subscribers, as usual, and if the wanted subscriber does not answer within a reasonable time, tho'opera- 'tor can ring him again by'means of the key ii, also as usual. This will not occur'except in a small percentage of cases, and in fact automatic repeat of the ring can be arranged by merely giving the subscribers sender more ringing teeth and the switch Z more steps or a set-back, that is to say a means for setting the wheel back so that the last llllplllSQSWlll be repeated. lfhe finally. fails to answer the operator Wlll withdraw the plug P and insert it in a busy back jack of usual pattern and connection, or can put the caller onto a manual or chiefs desk to be talked with and soothed if necessary. The subscribers operator under no circuxn stances attempts to talk, and in fact she cannot, having no equipment for the pur- 1:30

pose, unless speed of service is to be sacrificed. nary course of business, for if the steps in She does not need to talk in the ordi- .that attention is required to insert the plug P. The supervisory lamps flash as the signals go through, and respond to all the movements of the subscribers afterward. When the connection has been completed and the subscribers finish conversation, they hang up, and the supervisory lamps announce that they desire disconnection. T he, operator then pulls the plugs and the mech-' anism restores itself. In case the calling subscriber desires a recall he flashes his lamp, or if the wanted subscribers lamp continues to glow while the calling supervisory remains dark, it may denote the same thing, although it is dilficult to distinguish this from the condition when the subscriber has not answered and the operator then pulls the plug P, leaving the plu P in the jack. The indicator I, I and the side switch Z being at the same time restored to normal or zero position, the subscriber can get another number at once. He must set up his sender, of course, or manipulate hi hook as before.

In the manner of'showing the circuits in Fig. 16, it is to be observed that the generator and busy-back circuits are both tied onto one side of the trunk duringconversation. It will be understood, of course, that doubleole cut-ofi's or switches are to be used throughout so these connections will not exist in practice, but are herein shown because of a desire to simplify the drawing as much as possible. Such connections would make -the lines noisy.

' The sender WV can be entirely omitted, with all its special circuit connections, leaving an ordinary common battery telephone only atthe substation.

In practice the magnets Z and S are solenoids with plunger armatures, which when at their lowest point rest by their own weight on springs to close them together. The first impulse 0t a'somewhat rapid series pulls up such a plunger core or armature and relieves the springs of its weight, so that they open, and the inertia of the core, cranbined with the dash-pot efiect of its fitg in the spool, vibrate it but do not permit to fall far except on along break.

in all mechanism of the above type herefore described it has been found that -lerzfole wear and deterioration is due a successive, shocks caused by the dropof the switch spindles and the violent a ct other parts to their zero or normal ;iris'l2'i"i15. In, the 'cas'eloi my switch, shown in. "l I, the spindle and its eonneclcd parts are light, but in order to obviate any possible damage due to shock I have designed a simple take-up or check, consisting as shown, of a single element bellows supported on the frame and receiving the thrust of the lower end 6f the switch pindle in dropping. This comprises a pair of flexible disks clamped together around their edges by a metal ring crimped over upon them. Between these disks is a spring tending to force them apart, and the upper diskhas a small hole in the center. A bracket fastened to the frame extends out and carries the device. The lower end of the spindle has a little leather washer secured to it, large enough to cover the hole in the upper disk, or this washer may be attached to the disk. As the spindle rises the disks spread apart, but when the spindle falls its end comes down upon the upper disk and closes the hole, the contained air thereupon taking up the shock and as it escapes slowly letting down the spindle quite gently upon its bearing brackets. The same check may be applied to the rotary stop lever, p, and to other parts, as the side switch.

I do not claim herein broadly a combination between the magnet S, which is of the slow relay type, the side switch, and the automatic selector switch, but wish it undcrstood that I relinquish none of my rights to this subject matter, since I have made claims thereto in my copending application Serial No. 291,097, filed December 9, 1905.

I do not herein claim automatic switches controlled by relays having different time constants or through impulses of varying electrical characteristics transmitted over the subscribers metallic line cir uit (including the two sides of the line in series or other circuit normally free from ground connections) inasmuch as the same is claimed in my application Serial N 0. 291,097, filed December 9th, 1905.

Having thus described my invention, what I claim and desire to secure by Letters Patent is:

1. In a telephone exchange system, a pinrality of lines, central oiiice link circuits for interconnecting them, automatic line switches for said link circuits, and controlling means therefor, each switch comprising line-contacts arranged in a straight row, a common contact, a spindle mounted to turn upon an axis parallel to said row and capable of a longitudinal movement, and a plurality of arms or studs on said spindle arranged helically' around the same, and adapted to suecessively connect said line-contacts with said common contact at each longitudinal movement of said spindle, substantially as described.

2. 111 a telephone exchange system, a plurality of lines, central office link circuits for interconnecting them, automatic line switches for said link circuits, and controlling means Lies : row of iine contacts and a spindle carrying a plurality of angular-1y placed arms or studs, means to rotate the spindle so as to bring any desired arm into alinement with contact in the row means to, move the spindle iong'itudinaily and thereby bring the stud. into operative connection with succes sive contacts, substantially as described.

3. in a telephone exchange system, a plurality oi lines, central ofice link circuits for interconnecting them, automatic line switches for said link circuits, and controllin means thereioneach switch comprising a p urality of line contacts composed of subordinate groups or divisions, a spindle in juxtaposition to said contacts, a plurality of selective arms or studs on said spindle arranged in a helical line around the same, one stud for each division of contacts, means to rotate the spindle so as to bring any desired stud into alinernent with its division or grou i of contacts, and means to move the spindle longitudinaily to bring; the studs into operative connection with the successive contacts, substantialiy described.

' 4%. In a telephone exchange system, a plurality of lines, central office link circuits for interconnecting them, automatic line switches for said link circuits, and controlling means therefor, each switch comprising a plurality of line contacts composed of subordinate groupsor divisions, a spindle in JHXtflPOSltion to said contacts, a plurality of selective arms or studs on said spindle arranged in a helical line around the same, one stud for each division of contacts and means to lo tats the spindle so as to bring any desired stud into alinement with its division 1 or group of contacts, together with means to move the spindie endwise soas to bring the selected stud intooperative connect-ion with any desired contact in its subdivision or group, substantially as described.

5. in a telephone exchange system, a plurality of .l s, central oii'ice link circuits for intcrcc nesting them, automatic line switches for said Einlz circuits, and controlling means therefor, eacl. switch comprising a row o no contacts a member common thereto having individual contacts corresponding to and in juxtaposition with'the several line contacts, a snindle capable of longitudinal movement and carrying a seie'ctive' arm or stud, and-m L8 to move said spindle longito cross or interconnect any d sired unit line cont-act with the common n ember through its corresponding contact, :Lbstaniaily as described.

."In a teienhone enchan e system, a plurality of lines, central -o co link "circuits for interconnecting them, automatic line switches forsaid link circuits, and controliing means therefor, each switch comprising trains or support, a straight row. of conshirt or moving said in its group, substantiaily as described.

7. in a telephone exchange system, a line circuit, an automatic selective switch adapt ed to be associated w th linewhen $0 sailing and containin' iii-ls oi other lines, cont-roll We Loniagnets and iocai branch circuit connections for said swtcl, a step by step ciectroinag' 'tically actuated side switch controlling said local branch cir- 85 cuits, and controlling means for side switch responsive to distinctive current changes in the line 0,: more than a minate length only, substantially dc; scribed. v I

8. In an automatic telephone exchai system, an automatic switch and a con o ling or side switch. therefor comp the following instrumentalities: a .-.e or'hase plate, a seriesof contacts tl a switch or; arm adapted to sweep over cents-stern ratchet connected with the switch arm, a magnet controlling the ratchet, connections from the contacts to than pective parts to 'rality of lines, central cites link circuits for interconnecting them, automatic line switches for said link circuits, and controlling means therefor, each switch comprising a plurality of line contacts divided into 11o groups or divisions, a common spindle in juxtaposition to said contacts having a pin rality of contact rins corresponding to the line contact arr-oops, n :7 ring the spindle to b o any arm incl ually into 115 operative position relatively to particular roup of contacts, and other means for ter into operative correlation with is 4. group successively, suhstantialr scribed,

10. in a telephone or plurali'y 0, lines, cent v for interconnecting em, automatic witches for said "linh one s, and. ling means therefor, eac switch co a plurality of line-c tacts groups, a spindle hearing :1 arms arran i in helical iitne r e same, means to rotate spindle so as to 130

Images