US3351740A

US3351740A - Punching apparatus

Info

Publication number: US3351740A
Application number: US299448A
Authority: US
Inventors: Heuer Hans
Original assignee: Olympia Werke AG
Current assignee: Olympia Werke AG
Priority date: 1962-08-09
Filing date: 1963-07-25
Publication date: 1967-11-07
Anticipated expiration: 1984-11-07
Also published as: DE1242020B

Description

Nov. 7, 1967 H. HEUER PU NCHING APPARATUS 5 Sheets-Sheet 1 Filed July 25, 1963 INVENTOR HANS fizz/E FIG. 4

ATTORNEY 5 Q H. HEUER 3,351,740

PUNCHING APPARATUS Filed July 25, 1965 s Sheets-Shed a INVENTOR By I HANS HAY/ER ATTORNEY Nov. 7,1967 7 Filed July .25, 1963 HEUER 3,351,740

PUNCHING APPARATU$ 5 Sheets-Sheet 3 From 30, 32

. I I INV/ vmk.

HA/ys HEUEA ATTORNEYv United States Patent Office 3,351,740 Patented Nov. 7, 1967 Filed July25, 1963, Ser. No. 299,448 Claims priority, applicgtitsm9 1G9ermany,Aug. 9, 1962,

14 Claims. 61. 219-384) The present invention relates to a new punching apparatus.

Known punchingappara tus operate with punching rods usually called punches to be moved up' and down to perforate a web, tape, sheet, card etc. Necessarily, the speed of operation of such an apparatus, is severely limited by the required physical movement of mechanical parts which have to be accelerated, moved, braked, reversed for retraction, accelerated again, moved, braked and stopped just for punching one hole. I

It is an object of the present invention to provide a new parts to be. accelerated.

It is another object of the present invention to provide for a new punching apparatus in which the time for punching of one hole is reduced to 40 ,usec.

According to one aspect of the'presen't invention in a preferred embodiment thereof it is suggested to provide a spark gap by way of two electrodes preferably having ring-shaped edges of similar diameter and being coaxially disposed. The web or tape traverses the correspondingly ring-shaped spark gap in a direction transversely to the electrode and spark gap axis andpunching is carried out in a two-fold manner. First, high frequency, A.C. pulses are applied tothe electrodes to produce a series of sparks in alternate directions weakening the web structure in an annulus; second by way of air pressure the rondelle or disc encircled by the weakened annulus is pushed out. The voltage to be applied to the electrodes is, for example, controlled in dependence upon the successful completion of a previous punching step and in synchronism with the web transport.

While the specification concludes with claims particularly pointing out and distinctly claiming the subject matter which is regarded as the invention, it is believed that the invention, the objects, and features of the invention and further objects, features and advantages thereof will be better understood from the following description taken in connection with the accompanying drawing in which:

FIGURE 1 illustrates a cross-sectional view through a punching station in accordance with the invention, showpunching apparatus using no mechanical and physical ing a pair of electrodes, the spark gap and a web defining the basic elements of a single punching station unit;

FIGURE 2 illustrates in perspective and exploded view electrodes and electrode carriers for a punching station capable of punching a row of holes or a series of coded holes simultaneously;

ing, there is first shown a running web or tape 1 (FIG- URES 1, 3, 4, 5) running in and through a flat horizontally extending guiding channel 2 defined between two stationary block type frame or

support elements

2a and 2b. An electrode carrier 13 is integral with block 211 carrying a bar assembly composed of

bars

9 and 9a (see FIG- URES 1, 2 and 3) consisting of electrically'insulating material but having also a high thermal resistivity such as asbestos. There is a further electrically insulating layer 13a interposed between

bars

9, 9a and carrier 13.

The particular shape of bar 9 with supplementing bar 1 9a can best be seen from-FIGURE 2. Bar 9 is provided 10a in carrier 10 are of larger diameter. The latter bores with a plurality of holes or cylindrical bores in axis parallel relationship and being orientated along a line extending across the path of traverse of web 1 through gap or channel 2. Each hole consequently has an axis extending tranversely to the surface of web 1 when in this flat channel 2. Each such hole in bar 9 receives a cylindricalltube 3 constituting a first electrode having anyannular sharp cutting edge 4. The inner space of each tube 3 is denoted as 3a. Each tubular electrode 3 when inserted in bar 9 is aligned with a corresponding tubular channel 5a in. carrier 13 and all such channels communicate with a conduit 5 to be supplied with high pressure air. There is a continuing channel portion 5b for each tubular channel 5a axially aligned therewith and terminating ina photo-electric receiver 15.

' Beneath bar 9 with tubular electrodes attached to carrier 13 a second electrode carrier 10 is attached to block 2b. Carrier 10 is mounted on a flange 16a on a waste outlet duct 16 forming part of the stationary frame.

Guiding channel 2 is continued between

bars

9, 9a and carrier 10. On carrier 10 is positioned a filling 8 of insulated material similar to that of

bars

9 and 9a. Filling 8 and carrier 10 together have a plurality of cylindrical receive tubular electrodes 6 terminating upwardly in annular inner cutting edges 7, also called electrode edges.

FIGURE 5 illustrates in perspective view web driving 7 means transporting the web through the aforementioned punching station;

FIGURE 6 shows a circuit diagram of a control circuit network for a single unit in the aforementioned punching station; and

' margin define the relations among the severalpulses.

Proceeding now to the detailed descriptionof the draw- The inner space of each electrode 6 is denoted with ref 'erence character 6a.

The electrode edges 4 and 7 are of similar diameter but in such a manner that the outer diameter of a tubular electrode 3 is the same as the inner diameter of a tubular electrode 6(see FIGURE 1). Electrode edges 4 and 7 are made of electric conducting material having a high resistivity against burning, such as platinum, platinum iridium, etc.

. Each pair of aligned electrode edges 47 defines a spark gap traversed by web 1 as stated. The purpose thereof is to have an electrode system to which are applied high frequency voltage pulses capable of producing sparks in alternate directions between electrode edges, 4-7 and through web 1. j 4

Upon sparking, the web 1 is structurally weakened in a thin annular zone so that when high air pressure is applied to conduit 5 through the tubular electrodes 3, the air pushesthe severed rondelle into electrode 6 and waste outlet channel 16 feeding and into a waste basket or container 31. This will be described more fully below.

The web to be perforated or punched is preferably of a type in which any spark between a portion of edge 4 and the adjacent portion of edge 7 prefers to pass through the web 1 at a location not yet' having been passed through by a spark. This is important, since normally sparking will not occur uniformly at a ring-shaped spark.

It has to be observed that any electric field between such sharp electrode edges 4 and 7 is very inhomogeneous, and naturally the web 1 when passing through channel 2 .will not traverse the ring-shaped spark gap 4-7 uniformly, nor is it to be expected that the web is exactly coplanar with either electrode edge nor will it be of precisely uniform thickness. Thus, any spark will start to develop at a particular small spot as between a small portion or section of edge 4 and the adjacent portion of edge 7. Such spark traverses a particular spot of web 1. In view of the above stated requirement for the web material, sparking will not continue through that particular spot where sparking occurred initially but adjacent thereto, so that a spark that has once started anywhere will then run around edges 4-7 like a cutting knife. This ensures that with every sparking step only a small spot on the web is attacked initially which is enlarged to an annular portion enveloping a disc or rondelle which is then driven out.

A web meeting such requirements preferably consists of a paper which does not char and which is interspersed with very finely divided metal powder (aluminum or magnesium powder). This metal in fact shortens the nonconductive air gap between a pair of edges 4-7 prior to sparking and thus offers a preferred area for spark penetration.

Any area or spot having been traversed already by a spark is depleted from metal powder particle vaporization thereof. Thus, the effective air gap at such spot is larger as compared with an undisturbed web area. In particular, any web area just being subjected to a spark or a series of sparks and being depleted thereby from metal powder particles becomes less susceptible to further spark penetration than adjacent areas. Thus, upon applying a voltage across a pair of electrode edges 4-7 when traversed by a paper web as aforedescribed, there first will be a particular spot of preferred sparking due to minute inherent inhomogeneity of the edge-paper relationship and due to the great inhomogeneity of the spark pr0ducing electric field. Once sparking occurs at such spot, this spot becomes less susceptible to further sparking than adjacent spots, so that the sparks will tend to run in a circle around the ring-shaped gap, weakening the paper in an annular arc thereof outlining a punched out disc.

As will be described more fully below and with particular reference to FIGURE 4, high air pressure is applied to duct 5, effective in the spaces 3a of hollow electrodes 3 directly above web 1. Low air pressure is established through outlet channel 16 to bores a and the individual spaces 6a of hollow electrodes 6 to be effective beneath web 1. The web portion traversing the ringshaped spark gap 4-7 is thus subjected to a downwardly directed pressure differential.

Normally, the pressure difference thus applied across Web 1 is insuflicient to cause any ripping or perforating thereof. However, when by sparking between electrode edges 4 and 7, the web 1 is weakened in an annular space, the disc enclosed by such weakened annulars can and will be pushed downwardly into outlet duct 16. The weakening of the web in an annular area by sparking has already been described.

Basically, it is unimportant weather the pressure gradient as defined is directed upwardly or downwardly, or whether web 1 runs vertically so that the pressure gradient is orientated horizontally. It is advantageous, however, to have gravity assist the removal of the disc so that the pressure needs only to loosen the disc in the web around its weakened circumference. Thus, the device as shown with high pressure being applied from above is the preferred structural arrangement.

Proceeding now with the detailed description of the inventive construction one can best see from FIGURE 2 that

electrodes

3 and 6 are standardized or standardizable elements. Tubular electrodes 3 are embedded in bar 9 having one side abraded so that there are longitudinal slots exposing a small metallic surface portion 12 of any tubular electrode 3 inserted therein. These surface portions 12 serve as electric connecting surfaces respectively engaging contact springs 14 which are inserted into channels of body 2a and positioned centrally therein by insulators 14a. Springs 14 furthermore traverse a slanted, elongated slot in the supplementing bar 9a suitably attached to bar 9. Electric connection is made with each individual tubular electrode 3, 4 by means of these contact springs 14 insulatedly introduced into block 2a to individually engage the exposed contact surfaces 12 of electrodes 3. It is thus apparent that tubular electrodes 3 with their sharp electrode edges 4 are, in fact, completely insulated from each other and from block 2a, and there is further provided the insulating layer 13a to insulate electrodes 3 from metallic carrier 13.

Electrodes 6 are inserted directly into metallic bar 10, thus these electrodes are all kept at equal potential, which may be ground or mass potential. However, each electrode edge 7 is surrounded by the insulating material of filler 8 so that, in fact, each electrode edge 7 appears in channel 2 as insulated from the other electrode edges thus enabling individual sparking with the respectively adjacent electrode edge 4 Without influencing neighboring electrode pairs.

Bar 10 has its cylindrical bores 10a communicating with a space defined by frame structure portions 16 constituting a common outlet channel for the removed discs. The flange 16a on frame structure 16 is provided to support bar 10 as stated. A glass window 17 in frame structure 16 permits light from light source 18 to enter each individual bore 10a of bar 10, and the space 6a in each tubular electrode 6.

Whenever a disc is being removed from web 1 the light can pass on into the space 3a of the individually aligned tubular electrode 3, and the

bores

5a and 5b thereabove in carrier 13.

Aligned bores

5a and 5b serve as light channels for the individual photoelectric receivers 15, the purpose of which will be explained below. It is apparent that light from source 18 will reach a photoelectric receiver 15 whenever a perforation below it in web 1 has been completed. Whenever Web material covers the ring-shaped electrode edge 7, either because no perforation was initiated or the perforation was faulty and the disc adheres to the web, no light will be received by photocell 15.

A lens 19, preferably a cylindrical lens and a mirror 20*, preferably a cylindrical mirror serve to equalize the light for each light channel 10a-6a-3a-5a-5b. Lamp 20 is preferably of elongated form.

Web 1 is transported by a motor (not shown) connected to a clutch 22 having on its driving side a shaft 26 on which is mounted a transport roller 23. Shaft 26 is suit-ably journaled in the stationary framework of the device. Roller 23 cooperates with a pressure roller 24 to positively advance web 1 without slippage. Pressure roller 24 is mounted on a lever 25 which is engaged by a spring 25a urging lever 25 in such a direction that roller 24 is urged toward roller 23. Clutch 22 is preferably of the controlled, electromagnetic type so that upon de-energization shaft 26 with roller 23 can be stopped thus stopping transport of web 1.

A slotted disc 27 is mounted on shaft 26 rotating therewith. There is provided .a light source 29 and a photoelectric receiver 28 defining a light path. The disc 27 is positioned so that upon rotation thereof the slits temporarily open this light path and thus light pulses are received by photoelectric receiver 28 and converted therein into electric pulses of limited duration and frequency as compared with the light-pulses. It is apparent, that the frequency of these light pulses is directly proportional to the speed of web 1 and these pulses thus constitute clock pulses used for control purposes in a manner described more fully below.

The motor also drives a pump 30 pumping high pressure air into a pressure chamber 32 (see FIGURE 4) feeding conduit 5. This pump also reduces the pressure in the container 31 collecting the punched out discs. In particular, the low pressure is to be effective in the tubular electrodes 6 below the passing web 1 as aforedescribed. A sieve 34 prevents the pump from sucking up punched out discs. A basket of wire mesh 31 to serve as a waste collecting container could also be used so that no paper will enter the pump and air circulation. Since some air will escape from between the open electrodes 6 and web 9, a valve 33 serves. to replenish the amount of air pumped into pressure chamber 32. V p p The slots of disc 27' aredimensioned so that the Web moves by 0.1 mm. during the punching of one hole. The slits are spaced apart to correspond to a web transport distance of between two successive punchings. The diameter of the punched hole is about half that size.

The diameter of transport roller 23 may for example be about 16 mm. and rotating with a speed of 3000 revolutions per minute thus advancing the web by 2.54 m. per second. The distance from punch hole to punch hole is to be about so that there are a thousand punching steps carried out per second. i 1

The punching time of one hole measured as the time of sparking to cut the disc to bepunched out is about 40 sec. so that the web advances 0.1 mm. during the cutting. Assuming further that thepunching of one hole is performed by about 100 sparks (per 40 asec.), the necessary frequency for the production of sparking will be: 2.5 megacycles per second. Since sparking is carried out by AC. using voltages between electrodes 4 and 7 in both directions, a high frequency generator of 1.25 meg-acycles per second produces this rate of sparking.

Having thus developed the conditions for the electric control device as being set by desired operating conditions as well as the mechanical provisions outlined above, I now shall proceed to describe the electric control circuit.

In FIGURE 6, there is shown a pulse generator 40 producing output pulses at the rate of 1.25 megacycles per second. These pulses are preferably unidirectional and are fed to the main input side of an electronic gate 41. Assuming gate 41 is open, the pulses are then passed on to an amplifier 42 and from there to a step-up transformer 43 converting eachpulse into a highvoltage A.C. oscillation. One side of the secondary winding of step-up transformer 43 is connected to oneelectrode edge 4. The opposite side of this secondary transformer winding is grounded, so is the corresponding electrode edge 7.

It will be remembered that the device illustrated in the figures described above shows aseries of punching arrangements (pairs of electrodes 3, 4-6, 7) arranged transversely to the direction of web movement to punch holes across the entire web width at desired locations. For each pair of electrodes there is an individual gate 41, an amplifier 42, and a transformer 43. However, for the entire device only a single generator 40 is needed.

The gate 41 is controlled by two independently produced signals A and B, but the presence of both such .signals is necessary to open the gate 41.

Signal A is the main input signal representing a bit information drawn from a data processing device 50. Signal A basically consists of a voltage (or current) level to be applied to one control input terminalof gate 41 as one gate opening voltage and representing a punch hole command signal for one hole, thus being in fact a bit of information. Absence of such signal A means pause or no hole.

The second gate opening signal B is drawn from photoelectric'receiver 28 (see also FIGURE As was described above this photoelectric receiver 28 produces electric clock pulses at a rate determined by the web speed. In fact, these clock pulses determine (at a given web speed) the distance (and time intervals) between successive punchings.

Whenever both pulses A and B are present, a'train of H.F. pulses from generator 40 is permitted to pass through gate 41 and ultimately cause sparkings between the associated pair of electrode edges 4 and 7.

- 6 nized clock pulses. Returning briefly to FIGURE 5 it will be observed thta thev photoelectric receiver-28 is energized as between two levelsbright and dark. Bright occurs when light is permitted to pass through a slot of disc 27. Moreover, bright will occur as long as this slot determines the time of light energization of receiver 28 which in turn means that this determines the length of pulse B. It is thus apparent that by properly selecting the slot width of the slots in disc 27 the pulse duration for each pulse B is determined. This holds true of course only for a given speed of disc 27, but constant speed thereof and constant web speed. can readily be assumed during normaloperation. Y

The duration of each pulse B thus determines the number of HF pulses. permitted to pass through gate 41. as well as the duration of sparking to produce one punched hole. The number of slots in disc 27 determines the B-pulse. frequency. Thus, ultimately the disc 27' and its specific design alone determines the duration and frequency of the punching process. In case it is desired to change the punch hole distance on the web, or to adapt to a different brand of paper for web it is only necessary to exchange disc 27. This facilitates and simplifies greatly the electric circuitry.

It should be mentioned further that the frequency of p the HP. generator 40 and the frequency of punching are of course different and to some extent unrelated. The only requirement isthatea'ch punching step is to be carried out by a large number of individual sparking pulses, here by one hundred. Thus, as a general rule, the frequency of generator and frequency or. rate of occurrence of the B pulses should be apart by several orders of magnitude.

In this connection it is also assumed that pulses A is of relatively long duration. Appearance of pulses A can be controlled as follows:

In FIG. 6 there is further provided a coincidence type network 44 to which is also applied the command pulse or bit information A. Network 44 in its simplest form is primarily comprised of a logic and circuit 45 and a delay element. 46 both having one input each connected to adelivery line for pulses A.

The'and circuit 45 of network 44 has a second input terminal to which is connected photoelectric receiver 15 feeding pulses C to this second input terminal. It will be recalled that photoelectric receiver 15 (see FIG. 3) produces a pulse C whenever a single punching step had been completed successfully; The output of and circuit 45 is connected to one side, called the on-side, of a flip flop.

. duce any output at all.

Assuming now, a short pulse C arrives at the said second input terminal of and gate 45 indicating that for that bit information pulse A still standing at the first input the command pulse A. Pulse C is of short-duration because pulse C is produced in receiver 15 after a punched out disc has dropped out, and pulse C remains on only until the advaneing web interrupts again the light path 18-15 (see FIG. 3). The output of and circuit 45 is correspondingly short but suflicient to trigger flip flop 51.

Pulse D remains on after pulse C has disappeared. Thus,

pulse D can be used (1) to turn 01f the particular pulse which was the punching command signal and (2) to call on the next pulse A. There are now two possibilities.

First, there now is another bit representing a new pulse The pulses B however are more than merely synchro- A appearing again at gate41, then the next pulse B will open gate 41 as aforedescribed, and another hole is being punched. The other possibility is that no new pulse A is present which means that upon occurrence of the next pulse B no punching is carried out leaving a particular spot on web 1 unimpaired.

Complete utilization of pulse D thus is as follows: Flip flop 51 hasits output side connected to a first control terminal 52a of an and gate 52. Thus, this first control terminal 52a is activated upon occurrence of a pulse D which particular activation stays on until flip flop 51 is turned off.

And gate 52 has a second input terminal 521) which is connected to the output on-side of another flip flop 53 which is being turned on by the output of a differentiating stage 54 which in turn is fed with the pulses B from receiver 28. Differentiating stage 54 is coupled to flip flop 53 in such a manner that only the differentiated trailing edge of each pulse B performs the switching action effectively turning on flip flop 53 to activate input terminal 52b.

The output of and gate 52 is being used to call on station 50 in synchronism with the occurrence of pulses B. Flip flop 53 is being reset by feeding back the output of gate 52 as turning-off pulse for flip flop 53.

Input terminal 52 remains activated after production and continuation of a pulse D until a bit pulse A is being produced next, which next pulse A turns off flip flop 51. Preferably, the off-side input of flip flop 51 for pulse A and for turning flip flop 51 off is a differentiating one, responding only to appearance of an A bit pulse, but not to disappearance thereof. A next following pulse A can be called upon only after occurrence of a new following pulse D, that is to say, that any particular pulse A having turned off the flip flop 51 must be properly evaluated in the punching station before another pulse A can be called on. Also, the differentiated respective trailing edges of the B pulses provide the necessary synchronism between calling on the next A pulse and the desired rate of punching.

In case a punching step has not been carried out properly, no pulse C will appear and flip flop 51 remains turned off, thus no pulse D is being produced by network 44. Correspondingly, no new activation occurs at input terminal 52a so that in fact no new A bit will and can be called upon from station 50 regardless of further occurrence of B pulses. However, non-occurrence of pulse C results in the production of a pulse E which in effect is a delayed pulse A. As can be seen from the drawing, delay element 46 is also connected to the A pulse delivery line and feeds its delayed output to an and gate 47. Delay line 46 may be constituted by a monostable flip flop which produces output pulse E upon reversing to its stable state after having been placed into the unstable state by a pulse A. The second input for and gate 47 is the output of the off-side of the flip flop 5-1. Thus, whenever flip flop 51 has been turned on, and gate 47 is blocked (pulse D).

The output of and gate 47 is the pulse E produced (1) when flip flop 51 is being turned off and (2) when a delayed A pulse has been transmitted through delaying element 46. And gate 47 is a necessary element because element 46 delays every A pulse regardless of evaluation, but pulse E is to be produced only when no C and thus D pulse was being produced so that production of a pulse E is to be made dependent upon non-occurrence of C and/or D.

Pulse E activates a relay 55 turning off the current supply to a coil 22a normally energizing clutch or coupling 22 so that clutch 22 is being de-activated upon occurrence of pulse E. Such de-activation in fact stops the transport of web 1, and now the necessary steps can be taken to remedy the reason why the last hole was not properly punched.

The circuit network as described uses circuit elements of conventional design such as flip flops, bistable multio vibrators, gates, and circuits, monostable multivibrators, differentiating stage.

Elements of this type are described for example in Ledley-Digital Computer and Control Engineering, McGrawHill; Richards Digital Computer, Components and Circuits, Van Nostrand; Henney Radio Engineering Hand Book; Us. Letters Patent, Hense, 3,048,330 and others.

Considering the above definitions of the various pulses, the following equations in Boolean terminology will facilitate the orientation: The call pulses for A are -=C-52b. The cut-off control pulses E=46 -D, with 46 and 52b here denoting the output pulses of the correspondingly denoted elements.

The entire arrangement as described thus far operates in a manner best to be described with reference to FIG. 7.

Assuming for the sake of brevity that web 1 is already running, is. that by way of a manually operated push button switch it was assured that clutch or coupling 22 is energized and that terminal 52a is activated. It is thus simulated that a previous punching was carried out successfully. There are already produced a train of pulses B and generator 40 produces HF. pulses at desired rate, but gate 41 is still closed.

The next pulse B is transmitted through

elements

54, 53 and 52b, 52 as aforedescribed, to call on the first pulse A, called A and to turn flip flop 53 off. Occurrence of this pulse A turns off flip flop 51 so that the input line 52a reverts to the nonenergized state. In order to allow sufficient time for this pulse A to appear, the .trailing edge of the previous pulse B was used to control this calling, and only the succeeding pulse B undifferentiated is used as a whole to open the gate 41 for the duration of a B pulse. When gate 41 opens, a train of HZF. pulses passes through elements 41-4243 to produce a series of sparks at the corresponding electrode pair 4-7. Sparking occurs at a rate as determined by the frequency of generator 40 and for a duration determined by the length of the pulse B.

With the occurrence of pulse A and the resulting turning off of flip flop 51, the pulse D (non-D) as derived from the off-side of flip flop 51 potentially enables gate 47 to function.

The sparking occurs between electrode edges 4 and 7 and first traverses a minute spot in web 1 and then weakens an annular area thereof for cutting out a disc in paper 1; the air pressure difference between the corresponding

cavities

3a and 6a (see FIG. 1) causes the disc to drop into basket 31. Upon removal of this disc, light is permitted to pass from lamp 18 (see FIG. 3) onto and into the photoelectric receiver 15 producing a pulse C indicating that punching was duly carried out. The occurrence of C pulse activates network 44 to produce coincidence pulse A.

It is presumed that completion of punching of bit pulse A occurs slightly after termination of the HF. sparking, so that the differentiated trailing edge of pulse B appears slightly before pulses C and D.

Pulse D energizes terminal line 52a of gate 52, also and gate 47 is blocked again. Monostable flip flop 46 still produces a delayed A which however remains ineffective.

The trailing edge of that pulse B which caused the punching just completed, turned on flip flop 53 slightly before occurrence of pulse D. Thus, gate terminal 52b is already activated when flip flop 51 is turned on and now an output is produced at the output side of gate 52 to call on the next bit pulse which is A Flip flop 53 is reset by this call pulse as derived from the output side of gate 52. When A appears, flip flop 51 is turned off again and gate 47 is potentially enabled.

In the pulse diagram of FIG. 7 it is assumed that upon occurrence of the next pulse B this pulse A is also being properly evaluated, and in fact the situation is shown in which the punching (cutting of a disc and removal thereof is completed even prior to the termination of this particular B pulse. Thismeans, that now C and correspondingly D will appear prior to the difierentiated trailing edge of B. There now will only be a short output at flip flop 53, because D (output of flip flop 51) appeared first and immediately upon occurrence of the output of flip flop 53, and gate 52 responds, turning offagain fiip flop 53 and calling on the next pulse A, which may be called A In the meantime a delayed pulse A appeared at the output side of monostable vibrator 46 but this remains ineffective since gate 47 was blocked immediately upon occurrence of C and D.

Assuming that this last mentioned call pulse from flip flop 52 results in no pulse A (A then the next pulse B (photocell 28-disc 27) still prepares gate 41, but the A input thereof remains de-activated, gate 41 remains open but no punching occurs. There will'now be no pulse C since no paper was removed obstructing the light path from lamp 18. However, since there was no A bit pulse, flip flop 51 still remains in the on-state continuing to produce D. This is indeed correct, because pulses C and D are to indicate that the respectively last punching step has been carried out properly, and this particular result remains true regardless for how long no new A will appear. Thus, the effect of a C must remain as a continuous D until another A to be punched appears. And gate 47 remains blocked as long as D stays on and as long as no D appears, no E pulse will result even though there is no new C pulse.

Coupling 22 thus remains activated. The pulse B which caused opening of gate 41 without initiating H.F. transmission because there was no A pulse, still has its trailing edge differentiated. This differentiated trailing edge of B causes flip flop 53 to be turned on so that input terminal 52b is newly activated. Since 52a did remain activated, the output of and gate 52 can call on the next pulse A, if any.

Assuming there is a next pulse A both

flip flops

51 and 53 will then be reset (closed) as aforedesrcribed, and the A terminal of gate 41 will be activated, and the and gate 47 opened. The next B pulse opens gate 41 and sparking occurs again at edges 4-7.

Assuming now, that for some reason the disc thus cut in the web is not removed, or not properly being cut or the like, no pulse C will occur.

Absence of pulse C has a twofold effect, first, is the resulting absence of a pulse D which in turn means that terminal 52a remains de-activated. Since there did appear the pulse A flip flop 51 is turned off so thatno output leaves gate 52 and no new A bit pulse is being called on. Since this cutting step was actually initiated, the trailing edge of the B pulse prepares terminal 52b, but terminal 52a remain unactivated because no C and no D- appeared. Thus, pulse D stays on at gate 47. And gate 47 is still open because absence of D means presence of T). Second, since gate 47 is open, a delayed pulse A will pass on through element 46. A pulse E is produced by the output of flip flop 46 which was activated upon occurrence of A E turns off the drive for the web 1 by de-activating the coupling 22.

The invention is not limited to the embodiments described above but all changes and modifications thereof not constituting departures from the spirit and scope of the invention are intended to be covered by the following claims:

What is claimed is:

1. An apparatus for perforating coded holes in a tape comprising means for advancing the tape, first electrodes disposed on one side of the tape and second electrodes disposed on the opposite side of the tape for perforating the tape with an electric spark, the improvement comprising said first electrodes having first ring-shaped edges and said second electrodes having second ring-shaped edges of substantially the same diameter of said first ring-shaped edges; said first ring-shaped edges disposed coaxial to said second ring-shaped edges defining a gap whereby a spark 10 which forms at onepoint" between a pair of said edges moves around the peripheries of said edges in a closed path severing a rondelle from the tape by burning away the periphery of the rondelle, in combination with pneumatic means for removing rondelles and waste material from said edges and photoelectric means responsive to optical conditions in said gap actuating said electrodes and said means foradvancing whereby the tape is advanced when the rondelle is removed and the advancement of the tape is stopped when the rondelle is not removed.

2. The apparatus of claim 1, wherein said electrode have hollow passages connecting said edges and said pneumatic means.

3. The apparatus of claim 1, wherein said edges are metal having high resistance to oxidation under sparking conditions. 7

4. The apparatus of claim 2, wherein said pneumatic means comprises a closed conduit for air circulation.

5. The apparatus of claim 4, wherein said closed cond-uit includes a blower 30 for maintaining a positive air pressure on one side of the tape and a reduced pressure on the other side.

6. The apparatus of claim 5, wherein said closed conduit includes an air inlet valve 33 connected to the low pressure side of said blower for replacing air lost from the high pressure side.

7. The apparatus of claim 2, wherein said photoelectric means comprises a first light source 18 adapted to project a light beam through said hollow passages of a pair of first and second electrodes to a first photoelectric cell 15 whereby an impulse C is produced after each perforation.

8. The apparatus of claim 7, wherein said means for advancing includes a transport roller 23 mounted on a drive shaft 26 and said photoelectric means further comprises a disc 27 having slots therein mounted on said drive shaft, a second light source 29 disposed on one side of said slotted disc adapted to project a light beam, and a second photoelectric cell 28 disposed on the other side of said slotted disc whereby an impulse B is initiated by said light beam passing through said slots and said electrodes are actuated in synchronism with the forward movement of said tape.

9. The apparatus of claim 8, wherein said electrodes are cylindrical tubes connected to an impulse transformer 43, said impulse transformer is connected to an electronic gate 41, said electronic gate is connected to a high frequency generator 40 and said electronic gate is connected to a source of perforation command pulses A and said second photoelectric cell.

10. The apparatus of claim 9, wherein said source of perforation command pulses A is connected to a coincidence network 44 and said coincidence network is connected to said first photoelectric cell 15 whereby upon coincidence of said command pulses A and said impulse C said network produces an output pulse D initiating the next perforation command pulse, A for the next punching operation, and upon failure of coincidence of said command pulses A and said impulse C, an impulse E is generated and the forward movement of the tape is stopped.

11. The apparatus of claim 1, wherein said first electrodes 3 are cylindrical tubes seated in a first heat stable electric insulating body 9 having a slot 11 for exposing a portion 12 of said first electrodes.

12. The apparatus of claim 11, wherein a second heat stable electric insulating body 9a abutting said first insulating body supports contact springs 14 in contact with said portion 12 of said first electrodes.

13. The apparatus of claim 12, further comprising an electrode carrier 13 having a conduit for air therein, said carrier adapted to be connected to and readily detachable from said first and second insulating bodies, said contact springs and said first electrodes.

14. The apparatus, of claim 13, wherein said second electrodes 6 are cylindrical tubes seated in a recessed metallic body 10 connecting a carrier member 16- adapted to remove rondelles and waste material.

References Cited UNITED STATES PATENTS 12 Strickland 219-502 Meaker et a1. 219--384 Dresser 219-384 Dixon et a1. 219-69 Looschen 83-365 X Henry et a1. 219384 X RICHARD M. WOOD, Primary Examiner.

R. F. STAUBLY, Assistant Examiner.

Claims

1. AN APPARATUS FOR PERFORATING CODED HOLES IN A TAPE COMPRISING MEANS FOR ADVANCING THE TAPE, FIRST ELECTRODES DISPOSED ON ONE SIDE OF THE TAPE AND SECOND ELECTRODES DISPOSED ON THE OPPOSITE SIDE OF THE TAPE FOR PERFORATING THE TAPE WITH AN ELECTRIC SPARK, THE IMPROVEMENT COMPRISING SAID FIRST ELECTRODES HAVING FIRST RING-SHAPED EDGES AND SAID SECOND ELECTRODES HAVING SECOND RING-SHAPED EDGES OF SUBSTANTIALLY THE SAME DIAMETER OF SAID FIRST RING-SHAPED EDGES; SAID FIRST RING-SHAPED EDGES DISPOSED COAXIAL TO SAID SECOND RING-SHAPED EDGES DEFINING A GAP WHEREBY A SPARK WHICH FORMS AT ONE POINT BETWEEN A PAIR OF SAID EDGES MOVES AROUND THE PERIPHERIES OF SAID EDGES IN A CLOSED PATH SEVERING A RONDELLE FROM THE TAPE BY BURNING AWAY THE PERIPHERY OF THE RONDELLE, IN COMBINATION WITH PNEU-