DE2439123C2 - Printing device for alphanumeric data - Google Patents

Description

Γ 40 The invention relates to a printing device according to the preamble of the main claim.

A printing device of this type is known (DE-OS 22 58 744}. Here, a continuous web is on

■ 'Moves past a print head and the correct point is displayed by means of a corresponding light scan

the paper web is scanned. A control circuit controlled therewith ensures that changeable

_t data are always written in the correct position on the paper. With this recorder you can

that is, new data are printed at locations on the recording medium that are defined by data that have already been recorded

will.

It is also known that the character clock signals in a mosaic printer as a function of the relative

[to generate speed between recording medium and print head (GB-PS 12 44 651).

It is the object of the invention to provide a printing device of the type mentioned, in which

a form pattern can be printed on a continuous path by means of a main printing cylinder and which simultaneously displays variable data at predetermined locations by means of an ink pot ejector predetermined positions can be printed exactly.

This task is carried out on the basis of a printing device according to the preamble of the main claim

'solved its distinguishing features. Advantageous further training results from the company claims.

ι 55 With a printing device according to the invention, it is possible at the maximum printing speed of the

'' Main printing cylinder to print the variable data in exactly the right place, even if

this print speed should fluctuate for any reason. For the first time business forms in one operation on a fast-working machine with any form text or form template or an advertising print intended for direct mailing at the same time with variable data, for example an address.

The invention is explained in more detail below with reference to schematic drawings of exemplary embodiments. It shows

F i g. 1 shows the block diagram of the individual structural units of a printing system with features according to the invention,
F i g. 2 a schematic perspective illustration of the printer unit of the printing system,

F i g. 3 is a schematic perspective illustration to illustrate the arrangement and assignment of the Printing units and cylinders of a printing press used in the printing system according to the invention can be,

F i g. 4 the matrices or print format distribution for the individual characters to be printed, F i g. 5A and 5B the block diagram of the coupling electronics of a control device for the format head, 6 shows the block diagram of the coupling electronics of head control units for adaptation to the usual Address circuits of computers,

F i g. 7 the circuit diagram of a connection unit for the main head control of the system, F i g. 8 the block diagram of the coupling electronics for the head controls 2, 3, 4 and 5 of the system, F i g. 9 the block diagram of the head controls for the end of the information to be printed, F i g. 10 the block diagram of the common coupling electronics for the nozzle control, F i g. 11A or 11B the gate circuit arrangement which is provided for the computer connection unit, F i g. 12 the operational assignment and link between the units according to FIG. 5A, 5B and 6-10,

F i g. 13 the side view of a plate cylinder to illustrate the format depths and the decoder or the decoding provided during the timing of the pressure nozzles of the system,

14 is a diagram showing various control signals in the computer connection unit as a function a certain printing speed,

F i g. 15 the combination of a block diagram and a schematic representation to clarify the Relationships between electrical signals for actuating an ink nozzle and the ratio of the Flow of ink droplets with respect to the nozzle elements and the advanced web,

16A to 16G a schematic representation to illustrate the working principle of the nozzle control electronics,

17 shows the schematic representation of an ink feed and vacuum arrangement which is present in several units for a printing unit with five nozzles, which is part of an ink supply system, and

F i g. 18 shows the schematic representation of an ink supply regulator which is part of the printing ink supply system

According to the in F i g. The block diagram arrangement shown in Figure 1 will be variable and, for example, based on business forms data to be printed out in alignment with certain fixed data in a magnetic tape unit 30 on two Magnetic tapes are stored, which are designated as unit 0 and unit 1. The variable data is on brought a readable format and read into this under the control of a computer 31. The computer 31 arranges the variable data for transmission to a computer connection unit 33 in an alpha-numerical format Character format around. The computer 31 can read the data stored on the magnetic tapes and takes care of the basic data control and the provision of the program sequence functions for the definition of the variable data at the input of the computer connection unit 33. The computer 31 also has the necessary input-output system and the data lines for establishing a connection between the Computer connection unit 33 and a cathode ray tube (CRT = Cathode Ray Tube) for operational monitoring and display 32.

The CRT operational monitoring and display 32 has at least one CRT display unit for viewing Information display and an assigned control panel or keypad via which the operator in Dialogue with the computer can enter into the necessary data and information, for example various system parameters that are required for the operational process. The CRT display provides that They also operate the information generated by the computer relating to the operation of the system.

The unit 34 for the form position and the web speed has the necessary transducers, for example optical encoder, to provide the data related to the desired format depth, the web speed and refer to the speed of the master cylinder. The web speed and rotational speed of the main printing cylinder provide the necessary timing or clock pulse inputs for the computer connection unit » 33, so that these the necessary clock and control signals for coordinating the computer 31 and the nozzle control electronics 35 for controlling the ink spray nozzles for printing the variable information in the form of alpha-numeric characters that can be used in a certain assignment and alignment the moving web is to be printed.

The computer connection unit 33 has the necessary circuitry for receiving the converter signals from the unit 34 for the form position and web speed of the data control and time sequence information from the computer 31 and the status signals from the nozzle control electronics 35 and the nozzles 36, to enable synchronization and operational control of the whole system. The computer connection unit 33 essentially has five basic assemblies, each of which has specific control and synchronization signals for internal use in the connection unit itself and for operating the nozzle control electronics, hereinafter referred to as nozzle controls, deliver. As related to education and the Precipitation of the ink droplets in a certain position on the moving web, their speed If necessary, it can be changed, later it becomes clear, it is necessary to adjust the time of the color droplet release to control or regulate as a function of the path speed. The computer connection unit 33 has a circuit unit for controlling and monitoring the form head in order to release the ink droplets as a function of the belt speed and the head of the form on the main plate cylinder to be able to. The output signal of this connection unit, a corrected form head pulse (CTÖF pulse), represents a basic control signal in the connection unit and serves as a reference from which all character marking and clock signals in the connection unit for the subsequent control of the nozzles in the printing unit or generated or derived in the printing units. The CTÖF signal is also dependent set by a desired format depth, which is selectable by the operator, whereby the changing information to be printed by the color nozzles with regard to the changing format depths can be displayed or saved on the master cylinder.

The computer terminal unit 33 has a main head controller for each printing unit. This main head control receives the head distance information from the computer and generates the required time

or clock signals for controlling the release of ink droplets through the first nozzle of this printing unit to which the respective main head control is assigned. These time control signals are generated by counting clock pulses whose pulse sequence is variable in direct proportion to the web speed. The operation of the main head control is controlled by the CTOF pulse.The timing signals contain character marking pulses (STR 1 pulses) to provide a reference frame or reference limit within which five further character marking pulses (STR 2 -Impulse), whereby each of the STR 2-impulses defines the cycle for the release of a respective column of a 5 x '7 matrix or matrix within which all alphanumeric characters of the entire system are generated. The main head control also generates additional time control signals which ensure control or monitoring functions for head controls 2, 3, 4 and 5 of the associated printing unit. The main head controller also generates a print request signal for the nozzle controller which is prior to the actual droplet release time, so that nozzle No. 1 of the printing unit can be prepared for printing.

The computer connection unit 33 also has a common head control circuit for reception an address signal and control functions for actuating the main head and head control 2, 3, 4 and 5 a printing unit. These control functions include the signals INACTIVE, EFFECTIVE, STOP BALANCING and START BALANCING as well as a start signal from the computer that controls the operation of the main head controller for each print unit and the operation of the remaining four head controllers for each print unit synchronized

The computer connection unit 33 also contains identical circuit units for the second, third, fourth and fifth head control of each printing unit for generating timing signals for the respective droplet release at each of the associated nozzles 2, 3, 4 and 5 of this printing unit (STR ! Pulses) and further character marking pulses (STR 2 pulses) which fulfill the same function as the correspondingly designated pulses generated by the main head control. The STR 1 and STR 2 pulses from the respective head controls 2, 3, 4 and 5, however, are generated to count the displacement of the nozzle along the direction of movement of the web within a printing unit. This shift is fixed during a certain printing process, but can change within the mechanically predetermined limits of the nozzle structure itself and the desired print format. In other words: the distance between the nozzles in the direction of web movement can be changed, as can the respective distance between the nozzles along the transverse direction to the movement of the web. Each of the head controls 2, 3, 4 and 5 has a circuit for the clock-correct control of the generation of the STR 1 and STR 2 pulses in order to count or compensate for the distance between the nozzles in the direction of web movement. Each of the head controllers also includes circuitry for generating a print request signal which is applied to the associated nozzle in order to prepare the nozzle in question for printing.

The head control circuit, which is identical for all head units, is located in the computer terminal unit 33 to count the number of characters printed by each nozzle so that a control signal indicating the end of the message to be printed can be generated so as to produce the STR 1 and STR 2 pulses for the main head controller and head controllers 2, 3, 4 and 5 of each printing unit, as well as to indicate to the computer that printing of a certain set of variable data has been completed. This circuit unit also generates memory marker signals for controlling the output of the encoded alpha-numeric character data to the nozzle control electronics.

A set of memory and print buffer memories is assigned to the nozzles of each printing unit address respective memory marker signals from the common master head controller to mark the character data from the computer collective output line onto a seven-line output, which correspond to a particular character in a seven-bit representation of an ASCII code. Each of the nozzle controls has an addressing, a scan or interrogation line and a data control circuit Monitoring the reception of the information from the computer as well as the facilities to make the connection each of the nozzles so that the computer can determine their respective printing status. The print format in the embodiment of the invention described here has a length of thirty-eight characters in each print line. In addition, a character spacing of ten characters per inch (approx Characters per cm) of the web movement. The line spacing between the printing nozzles of a certain printing unit and between the printing units themselves is variable and only due to the mechanical structure the press, the mounting of the mechanical part of the printing units, etc. are limited The nozzle controls 35 are supplied with the coded alpha-numerical output data as well as with the STR 1 and STR 2 pulses of the computer connection unit 33, whereby the printing of the alpha-numerical characters can be controlled by each of the nozzles within a printing unit each of the nozzles generates a train of pulses which are synchronized with respect to the generation of the ink droplets in the nozzles themselves. This pulse sequence is clocked by the character-marking pulses from the computer connection unit so that each column of the 5 × 7 matrix for releasing the color droplets is time-correlated in a specific assignment and orientation on the moving web, regardless of its speed

The pulse train output from each matrix generator is provided by a digital to analog converter circuit implemented with a separate circuit responsive to each of the matrix generators thereby producing a low level video sawtooth or staircase signal which is at seven different voltage levels can be set for each column of the matrix. These low-level video signals are amplified and used as control voltages for a charging tunnel through which each of the successive droplets in the Stream of droplets ejected through each nozzle according to the desired displacement in the

is loaded in the direction transverse to the web movement. The video amplifier is adjusted according to the excitation of a piezoelectric crystal that forms the droplets in each of the nozzles, so that the Loading of the droplets is precisely matched to the generation of the droplets.

The nozzle controls 35 also include high voltage deflection circuitry to control the loading plates a deflection tunnel to impose a static charge through which each of the charged droplets passes, thereby deflecting each droplet a distance or distance that is directly proportional is to the charge imposed on each droplet during passage through the charging tunnel. Uninvited Droplets are not deflected and are collected by a collector before they are released onto the web so that they do not play a role in the printing of the alphanumeric characters.

The nozzle controls 35 also have well-known balance and droplet sensing circuits to the Matching or balancing the ink droplets to be detected and corrected if such a correction is necessary will.

Finally, the system has a unit for monitoring and controlling the paint supply and the paint flow to each of the respective nozzles. This part is shown in FIG. 1 identified by reference numeral 37. Each of the Printing units has an ink supply in multiple designs, whereby each of the five nozzles one Printing unit is supplied with ink from ink tanks at the same time and in parallel operation. The uninvited Ink droplets intercepted by each of the respective collectors associated with each of the nozzles are withdrawn through a multiple vacuum return connected to each of the headers. The inking unit 37 further includes appropriate filters and pressure regulators to provide proper ink supply to each the spray nozzles.

F i g. 2 shows a typical embodiment of a printing unit 38 in operational assignment to a movable web 39. Each of the five color ejection nozzles 38a, 386, 38c, 38c / and 38e of the printing unit 38 is used to print a corresponding print line 40a, 406, 40c, 40 </ and 4Oe. As the F i g. 2 shows, the print lines 40a-40e are arranged equidistantly from one another, but the distance between the lines can be changed in a direction transverse to the movement of the web 39 by a suitable setting of the holder for a desired or all color nozzles 38a-38e. In the system described here, each of the nozzles 38a-38e is arranged at a distance D from the adjacent nozzle in the printing unit 38. The distance D for the device described here is 6.35 cm (2V2 inches). This dimension for the assignment of the respective nozzles in the printing unit is only to be understood as an example, and it should be emphasized that the distance D between each nozzle can be changed if desired by a suitable modification of the connection unit circuit, as can also be seen from the following Description of the construction and operation of this unit will show.

In an operational embodiment, the paint spray nozzles lie in a horizontal plane while the web 39 moves in a vertical plane. However, the positions of the printing unit and the moving web be swapped in terms of horizontal and vertical plane, if it turns out that only unsatisfactory Results can be achieved if the ejection of the paint droplets through the paint spray nozzles is contrary to the Gravity occurs. Each of the fax spray nozzles 38a-38e lies in a plane normal to the plane of the moving ones Lane 39.

F i g. 3 illustrates a paint spray printing system with three printing units 38, 38 'and 38 ". Each of the printing units 38, 38 'and 38 "has five paint spray nozzles, each labeled 38a-38e, 38a'-38e' and 38a" -38e ", respectively. are designated. In the embodiment shown, the movable path 39 is driven by drive rollers 41a, S

41 b driven in the vertical direction, as shown in FIG. 3 shows the holder for each of the printing units 38, 38 '|

or 38 "is not shown in detail in FIG. 3 in order to avoid overloading the drawing. The j |

Printing units can be fastened by a suitable bracket so that they are in the correct spacing

with respect to the moving web 39. 45 ff

In FIG. 3, the main printing cylinder 40 is operationally assigned to the printing and drive roller 41a ff

The assignment relationships of the printing cylinder 40 to the printing roller 41a and the printing units 38, | J

38 'or 38 "is only to be understood as an exemplary embodiment. The main pressure cylinder 40 can also be shown in fi

The direction of advance of the moving web 39 can be arranged further down than in the illustration in FIG. 3. It ρ

it should also be emphasized that the main printing cylinder 40 in the web guiding direction also in front of the printing units 38, 38 '50 | §

and 38 ". The mechanical format head or format start on the main pressure cylinder 40 is j |

in Fig. 3 specially indicated. A slot 41 is provided, offset from the mechanical format head, through which |

electrical format head pulses generated by a suitable optical coding circuit in a known manner s |

can be. A number of slots 42 are provided around the periphery of the master cylinder 40 in order to

to generate a fixed number of clock or timer pulses with each revolution of the master cylinder ». In the In the embodiment described here, 2500 of the slots 42 are provided. The slots 42 is for creation appropriate electrical impulses assigned to an optical encoder. The electrical format head impulse as well as the series of 2500 pulses per revolution of the printing cylinder 40 arrive as input signals on the Connection or connection circuit to provide the necessary timing functions for the operation of this circuit to ensure. There is also a converter for generating clock pulses for the connection unit for each has a fixed number of pulses per inch of web advance. This converter is in F i g. 3 not shown, but it can be implemented as a known speed converter, as is normally the case used for drive and translation units in printing presses to indicate the speed will.

The distance between the printing units 38, 38 'and 38 "can be used to ensure a desired changeable data printing format on the forms to be printed by the main printing cylinder changeable be. It should also be emphasized that the lateral spacing of the printing units 38, 38 'and 38 ", if desired, in a transverse direction to the movement of the web 39 can be adjusted, whereby the by the respective pressure

,! » units in relation to the form or the forms or the templates on the main printing cylinder

: i generated pressure can be adjustable.

I ■; With each of the printing units 38, 38 'or 38 ", which are held in a fixed assignment to the main printing cylinder 40

b _; the system has a format depth selection for the operator, the format depth being the extent

j? 5 of the shape or format over the circumference of the main printing cylinder (Fig. 13) and wherein the connection

\ d unit automatically stops generating the character-marking pulses in order to achieve that the pressure

ι - is selected in the correct assignment according to the format depth.

;; F i g. 4 shows an example of a matrix or matrix division for a total of sixty-four alpha-numeric

.;. · See characters. As this figure reveals, each alpha-numeric character goes through and within a

|| 10 5 χ 7 matrix is generated, which is explained in more detail in the following illustration. Every spray of paint is in the

Position, each of the sixty-four alpha-numeric characters shown in FIG. 4 to be recognized. It should be noted that the Matrix distribution in FIG. 4 is only to be understood as an example and of course other subdivisions as well

ψ can be provided.

Rj The theory on which the determination of the release of paint droplets is based will first be briefly explained

ι $ before with the detailed representation of the circuit for correcting the format head pulse in adaptation to S

the path speed based on FIG. 5A and 5B is started: A first counter counts B continuously

Pulses that have a pulse sequence that is directly proportional to the web speed. Another, by one free-running stable oscillator operated counter serves as monitoring or control for the first counter, to set a specific counting period of the first counter. The first counter is then reset to 0 and triggers a cyclical counting process for consecutive fixed counting periods. The through the The number of pulses counted in the first counter corresponds directly to the speed of the moving web and is continuously detected for each cycle of the counting process. The counting period is set so that the The interval in question is shorter than the time it takes for the speed of the moving web to change could. One of the factors that play a role in changing the web speed is: the inertia of the press, which is considerable and this means that the web speed is in a short amount cannot suddenly change its time period. Another factor is the accuracy of the press equipment, and the following circuit also corrects errors in the speed of the press drive mechanism.

The format header control is explained below first:

With the printing device according to the invention, it is possible to use the paint spray nozzles on different Web speeds in the range of approx. 214 m / min, to be adjusted to essentially 0 m per minute. It is necessary to vary the release time for the ink droplets so that it is precisely adapted to the different web speeds

When printing using paint spray nozzles, there is a certain amount of time between exits the color droplets at the individual spray nozzles and the moment at which the droplets appear hit the web. It can be assumed that the distance between the nozzle and the web and the speed of the paint droplets emitted from the paint spray nozzles are substantially constant. The time period for the ink droplets to transfer from the nozzles to the moving web is short, occasionally on the order of a few milliseconds. At high printing speeds, the web movement during the droplet transfer time is decisive, with the web traveling a distance of about 2.5 cm (1 inch) covers the actual advance movement of the Orbit can be negligible. It is therefore evident that the triggering is the color droplet emission the spray nozzles a critical factor in securing the touch contact in the correct position represents on the web, depending on the web speed, if at the same time precise alignment of the spray paint is desired at all web speeds

The data recorded per counting process by the first counter unit are converted into a line control signal Discharge of the ink droplets against the web implemented in order to ensure their correct contact to the desired one To ensure points on the track. This transfer is effected by a third counter Output signals arrive at a logic and digital decoding circuit, which only has a predetermined Number of pulses registers, for example, the calculated maximum number in cm or inches that the path the droplet travels during flight, multiplied by the number of pulses per cm or inch of the Path feed, üie are detected by the first counter. Such a calculation leads to a maximum number of pulses that can occur during the flight of the droplets.

This circuit is also designed in such a way that the time or the point in time of the droplet release can be changed according to the desired format depth, in order to thereby adjust the droplet release to the respective to be able to adjust the printing cylinder. This format header circuit is also described in detail below However, it should be noted at this point that the format head circuit sends a pulse The third counter initiates the counting process up to a preset maximum count The web speed data obtained from the first two counters are used to preset the third counter to an initial value that is directly proportional to the web speed. The third counter then starts and starts the counting process with the trigger pulse from this initial value and counts the clock pulses dependent on the web speed until the preset maximum count is reached At this point in time it triggers a control signal that is used to determine the triggering of the color droplets from the Paint spray nozzles are used

Suppose the first counter counts for 12.5 milliseconds. It is further assumed that pro Inch (= 2.54 cm) of the web feed at a web speed of 213 m / min. 120 pulses occur so

the first counter counts a number of pulses represented by the following equation during the aforementioned Time period can be determined:

213 m / min. = 355 cm / sec.

355cm / sec x 120 pulses / 2.54 cm χ 12.5 ms χ = 2100 pulses

From knowing the speed of the paint droplets and the distance between the paint nozzle and the web it can be calculated that at a maximum web speed of, for example, 213 m / min, the To release ink droplets approx. 14.8 mm in front of the desired printing point on the web, d. H. the release time corresponds to a distance of 14.8 mm from the relevant one at the specified path speed Spray nozzle. The calculation can be made by an impirical determination, i. H. by actually measuring the Displacement of the ink droplets at a web speed of 213 m / min assuming the above Confirm the fixed parameters mentioned. This makes it possible to use the aforementioned information, namely the 120 Pulses per 2.54 cm web advance and a displacement of the web droplets of 14.8 mm the number of To determine impulses up to which the mentioned third counter must count to the color droplets in the correct Time to release at a web speed of 213 m / min. The following results from the calculation:

Number of pulses = 14.8 mm χ 120 pulses / 25.4 mm = 70 pulses

In the embodiment described here, an optical encoder is arranged with respect to the plate cylinder in such a way that it emits the format head pulses 14.8 mm before the actual mechanical position of the format head (on the printing cylinder) . From the previous description it is already known that the paint droplets from the spray nozzles have to be released at a web speed of 213 m / min 14.8 mm in front of the web point that is aligned with the nozzle in question Counter initially preset to the pulse count that corresponds to the maximum web speed, i.e. 70 pulses in the example given above, which means that there is no delay in the release of the respective ink droplet the droplet release can be determined for each web speed. With reference to FIG. 13, the distance Υϊύν the format head position of the encoder for web speeds of less than 213 m / min can be determined as follows:

,,,. - web speed (m / min.)

Ya *** = 14.8 mm χ 213 (m / min)

From the discussion above, the number of pulses to be initially preset in the third counter can be determined determine as follows:

_v -, mi web speed (m / min)

Yunpu ^ = 70 pulses χ ^L

With the device according to the invention, printing with color nozzles can also be carried out with regard to the format depth (i.e. the size or dimensions of the shape or format on the impression cylinder) and relay a message. The plate printing cylinder is illustrated in FIGS. 3 and 13. Plate cylinders have different Dimensions and different scope. For the purposes of this description, let Assume that the press cylinder was 17 inches in circumference. A common format size When fully unfolded, for example, corresponds to a 21.6 cm format

A format depth of two can thus be achieved by using one on two separate halves of the cylinder plate display arranged 21.6 cm format so that 2 forms are pressed with each revolution of the plate cylinder will. A formation of three would accordingly subdivide the student into three of such formats, a format depth of four corresponds to a division into four such formats, etc. the Format depths are illustrated in Figure 13. As previously mentioned, the format header control provides one Impulse to trigger the counting process at the third counter mentioned, in order to display the different format depths to be charged so that the color spray printing can also be adapted to different sized formats can.

In order to be able to carry out these functions, the format header control needs certain time control data. According to FIGS. 3 and 13, as mentioned, an electrical format head pulse (ETOF ) offset in time from the actual (mechanical) format head on the plate cylinder 40 is generated by a suitable transducer for each revolution of the plate cylinder, for example via a slot 41, which is part of an optical one Encoder is through other converter devices, such as the slots 42, which are part of a further optical encoder, additional timing pulses are supplied. For the purposes of the embodiment described here, the plate cylinder 2500 has equally spaced slots 42, distributed over its periphery, so that each Circulation of this plate cylinder 2500 pulses are generated. Depending on the speed of the web being moved forward, clock pulses are also generated. For the purposes of the following description, it is assumed that the clock pulse sequence, which is dependent on the web speed, amounts to 120 pulses per 2.54 cm of web feed. Such pulses can be generated in a known manner in various ways. For example, the web speed can be set directly via a gear to E in the printing address

Drive mechanism detected and converted into pulses by a suitable electrical converter. The structure of arrangements for generating the aforementioned electrical format head pulses, the 2500 pulses per revolution of the printing cylinder and the 120 pulses per 2J54 cm of the web feed are known, so that a detailed description of these units can be dispensed with here to explain the invention.

In the following, the format header control is explained with reference to FIGS. 5A and 5B explained in detail:
The electrical £ TOF pulse of the optical encoder assigned to the plate cylinder 40 is delivered to terminal 60. The £ TOF pulses are used as trigger pulses for the speed correction counter and as reset pulses for the format depth counter. These pulses pass through an OR gate

ίο 62 (AND gate with four common inputs), a Schmitt trigger, an OR gate 64 (NAND gate with two common inputs) and inverter 66, 68. The pulses generated by the circuit described below to trigger operation the format head control circuit are supplied by the computer command SyR7 "via a NOR element 70. The £ 7ÖF pulses at terminal 60, together with the selected outputs of a digit or number decoding circuit (which is described below), are

Controlled via format head switch (TOF switch) 1, 2, 3, 4 in the lower area of FIG. 5A, these switches being selectable by an operator. The TOF pulses reach a contact of the TOF switch 1 via a NAND element 72; to the TOF switch 2 via an inverter 74, the NOR gate 76 and the NAND gate 78; to the TOF switch 3 via the inverter 74, a NOR gate 80 with four inputs and a NAND gate 82, and finally to the TOF switch 4 via the inverter 74, the NOR gate 84 with four inputs and the NAN D- Link 86.

The plate cylinder circulation pulses (PCÄ pulses) run via the terminal 88 sin to provide counter clock pulses via an OR gate 90 (AND with four common inputs), a NOR gate 92 (NAND with two common inputs), which has a Schmitt Trigger includes, and l.iverter 94, 96 to get to the clock pulse inputs of binary counters 98, 100 and 102 connected in series. The outputs of the counters 98, 100, 102 are decoded by a digit decoding circuit which has the following components and operates as follows:

Certain outputs of the binary counter 98, 100, 102 are inverted by inverter 104, and the outputs are provided along with the selected output values of the binary counter 98,100,102 as inputs for digit decoding circuits 106,108,110,112 and 114 for jointing The input signals for the Ziffem-Dekodierschahung IOS are such that the number 1250 is decoded; the input of the digit decoding circuit 108 is designed to decode the number 833, and the inputs to the digit decoding circuits HO, 112 and 114 are also designed to decode the digits 1666, 625 and 1875, respectively. Since the printing cylinder is distributed over 2500 pulses, it can easily be seen that the decoders 106-114 each decode pulses corresponding to V2, V3, ² Z ^, V4 and V4 of the circumference of the printing cylinder. The outputs of the digit decoys 106,108, 110,112 and 114 come as an input to NOR elements 115,116,117,118 and 119. The output of the NOR element 115 (corresponding to 1250 pulses) is given as an input to the NOR elements 76 and 84. The output of the NOR element 116 (corresponding to a count of 833) is applied as an input to the NOR gate 80. The output of the NOR gate 117 (corresponding to a count of 1666) is given as an input to the NOR gate 80. The output of the NOR element 118 (corresponding to a pulse count of 625) is applied as an input to the NOR element 84. Finally, the output of the NOR element 119 (corresponding to a pulse count of 1875) is connected as an input to the NOR element 84

From the description given so far it can easily be seen that the outputs of the NOR gates 72, 78, 82 and 86 supply pulse trains which correspond to format depths of 1, 2, 3 and 4, respectively. The corresponding pulse trains at the outputs of the TOF switches 1, 2, 3 and 4 are inverted by an inverter 116a and are used for monitoring and control functions, which are explained in more detail below.

The circuit, the aforementioned number of decoding circuits including the inverter 104 and the connection of the binary outputs of the binary counters 98, 100 and 102 for performing the required number or digit decoding is so well known that a single circuit diagram in connection with the description of the invention is unnecessary, since the person skilled in the art If necessary, a suitable number decoding circuit can be set up in order to obtain the known decoding functions mentioned above.

In order to set the counters 98, 100 and 102 correctly so that an exact correction compensation takes place, if for example the format depth two is selected, it is necessary to send the computer SKÄT pulse from the NOR element 70 to a phase control flip-flop 118a, in order to prevent the possibility of a print release on the second half of the plate cylinder. The SYÄT control pulse resets the flip-flop 118 , while the £ TOF pulse switches the flip-flop 118 , which in turn resets the counters 98, 100, 102 via the output of an AND element 120. The other input of the AND element 120 is acted upon by the SyÄT pulse or the £ TOF pulse, which are input to the corresponding inputs of the NOR element 124 via the inverters 122 and 74, respectively.

The SVÄT computer command as well as other data and control signals are input from the computer to a trigger circuit 130 in the following manner. Format header control is performed using the / L // 4A "control signal and the first six bits, ie the EO bits ES of the computer collective output line t is addressed. The address bits mentioned are sent to NOR gates 132, 134, the outputs of which, together with the control signal IUAX, go to a NAND gate 136 with three inputs inverts the inverter 138 and thus supplies an input for a NAND element 140 with four inputs. The other three inputs of this NAND element 140 are acted upon by external control commands (EX commands) from the computer, via the computer Datenammelleitu rig, through the bits EB6, EB1 and EBS. The output of the NAND element 140 arrives - together with the bit EB 11 and the control signal TRYX from the computer - as an input to a NOR element 142, the output of which is the Input for the NOR element

70 forms, together with the computer control signal SYRT, in order to trigger the format depth switch and the circuit for correcting the .ETOF pulse (modified by the format depth switch). This is also explained further below

The clock pulses CP for the connecting unit by the clock pulse generator 150 generates a Wandlereingangssigna! of 120 pulses / 2 ^ 4 cm is delivered to the terminal 152 and arrives at an OR element 154 (AND with four inputs connected together), a NAND element 156, which contains a Schmitt trigger, and the inverters 158, 160 . in order to avoid difficulties in reading of the circuits showing the logic circuit of the terminal unit, the Takiimpulse in the entire circuit diagram of the connection unit with CP are designated

With reference to FIG. 5A and 5B, the counter circuit for generating the corrected CTOF pulses is described in detail below: The clock or CP pulses arrive at the clock pulse inputs of flip-flops 162,164,166 in a divide-by-five circuit 160. The (? - The outputs of the flip-flops 162 , 164 form the inputs for OR elements 168, 169 and the outputs of these elements reach the / input of the flip-flop 166 via a NOR element 170. The (^ output of the flip-flop 162 controls the flip -Flop 164. The (^ output of flip-flop 166 is connected to the / input of flip-flop 162. The (^ output of flip-flop 166 corresponds to V _{5 of} the CP input on divide-by-five Counter 160. The flip-flops 162, 164 and 166 are reset by the SY7 "7" command

The output of the flip-flop 166 is the input for the CP input of the binary counter 168, which is part of a counter 169. Another counter 161 comprises the binary counters 170, 172, which cyclically count the pulses of a 10 KHz oscillator 174 so that one period are of 12.5 ms for sampling of the divided pulses from the flip-flop 166 to the counter 169 for disposal, which in the following manner in each case takes place after 12.5 ms produce the output of the binary counter 170,172, the inverter 176, the NAND gate 178 , the inverter 180, the NAND gate 182 and the inverter 184 have a pulse output from the gate 182, which is also controlled by the TOF pulse from the TOF switches described above, which is adapted to the format depths. The counter 169 is reset by a pulse output which is generated by the NAND element 186 and the NAND element 188 . At the other input of the NAND gate 188, the command signal is Syrt. The inputs of the NAND gates 178 and 186 are correctly selected by the inverter 176 , so that the NAND gates 178 and 186 deliver a pulse every 12.5 and 12.7 ms, respectively. As already mentioned, this number decoding circuit is known. The value of 12.7 ms is useful because the binary counters 170, 172 each count up to 128; and then fed back (to the new counting process). This gives the necessary time delay for the counters 169 and 171 to be reset.

The last counting stage of the counter 168 is switched to the CP input of the counter 188. The binary count values stored in the binary counters 168, 188 of the counters 169 are keyed into four-digit permanent or locking memories 190, 192 by the pulse output of the inverter 184 after every 12.5 ms Information about the web speed stored in these retention memories 190, 192 is reloaded into the binary counter 194, 1% of the counter 197 by the same pulse every 12.5 ms. This means that the binary counters 194.196 are preset to the respective web speed data every 12.5 ms.

The web speed CP pulses are scanned onto the incremental binary counters 194, 196 by means of the NAND gate 198 (FIG. 5A), which is controlled by the (^ output of the flip-flop 200 , and which the adapted format depths ETO / ⁷ pulses from the aforementioned TOF switches via line 201. If the binary counters 194,196 have reached a maximum preset count, a pulse appears on the NAND element 202. The maximum preset value corresponds to that for the correct color droplet— Release required number of pulses at maximum path speed.

The maximum preset count is determined by the number decoding circuit which includes inverter 204 and NAND gate 202 and which receives selected inputs from binary counters 194,196. The pulse output at NAND gate 202 corresponds to a speed-corrected CTOF pulse which is used in the connection unit as a basic timer pulse to generate STR 1 and STR 2 character print control signals as well as other control signals, which is explained in more detail below will. The binary output 64 at the counter 196 is used to switch the flip-flop 200 (FIG. 5A) in order to interrupt the CP input to the third counter 197.

The Head Control - Common Address Control Circuit

The system requires external control commands from the computer to make the head controls ineffective or effective

sam and to pause or start equalization for the paint spray nozzles. The external 55 ^

Control commands INEFFECTIVE, EFFECTIVE, STOP COMPENSATION and START COMPENSATION arrive in the following way via the in F i g. 6 shown circuit on the connection unit. The response or device address for the head control is decoded by NOR gates 202, 204 and a NAND gate 206. Each of the commands INACTIVE, EFFECTIVE, STOP COMPENSATION and START COMPENSATION is reproduced by a special code or key 0, 1, 2 or 3. The device or device address at the output of the NAND gate 206 activates a binary-decimal decoder 208 , so that the EY command bits from the computer output line enable a key 0, 1, 2 or 3 to be written into the decoder 208 enable. The respective keys 0, 1, 2, 3 go to inverters 210, 212, 214 and 216 and their outputs are in turn passed to NAND gates 218, 220, 222 and 224. The NAND gates 218-224 are activated by a pulse from the NAND gate 226 via an inverter 228. The input of the NAND gate 226 has the computer control signal FR YX and data from the data bus EB 11 . This means that the external computer commands INEFFECTIVE.EFFECTIVE.STOP ADJUSTMENT and START ADJUSTMENT appear as outputs of the NAND gates 218, 220, 222 and 224 when the address signal IAUX together with the control signal! FRYX from

Computer is received The symbols D and E correspond to INEFFECTIVE - DISABLE and EFFECTIVE = ENABLE as previously described, and the symbols in FIG. 6 are the same for the entire connection unit.

The head controllers explained in more detail in the following, triggered by the EFFECT signal E, which at the clock pulse input of the control-enable flip-flop 236 is applied the flip-Flo p 26 is set by the output signal of the AND gate 230, which at the Computer control signal SYRT and the EFFECTIVE signal D responds The (^ output of the flip-flop 236 is the EFFECTIVE signal E 1, which is used to switch on a special circuit within the connection unit, which will be explained in more detail below The device address signal (GA signal) is available as an output via an inverter 234, i. E. as Control function for the main head controller, which will be explained in detail below

The STOP COMPENSATION and START COMPENSATION signals are sent to the main head control (Fig. 8) and the head controls 2, 3, 4 and 5 (FIG. 8), their respective functions also in more detail below are explained

The requirements for the commands STOP COMPENSATION and START COMPENSATION are clarified in connection with the description of the electronic circuit of the nozzle control. Shortly before the actual printing process, however ύζτ control end determines that the program is running correctly, and that using a issuesi checklist that is played by the computer on the CRT display to the operation monitoring. The operator answers the questions of the computer in order to give it the necessary information to complete the program. During this set-up of the computer program, the software causes the nozzle operating functions to run automatically via start and stop compensation. This control function and its structure for carrying out the monitoring and control of the system is also explained in detail below. The commands STOP and START BALANCING include a check of the nozzles to ensure that they are operating correctly. The computer ignores the nozzle control electronics and activates a tactile or sensing device in the collector of each of the nozzles, and if the compensation (for later consideration of the web speed) is correct for each of the nozzles, the nozzle control electronics give the computer a message so that it is informed that the nozzles are ready for printing.

Sampling the nozzle SERX response

The scanning or checking of the nozzles takes place via the connection unit in the following with reference to FIG Fig. 10 explained manner: The print ready signal (PRT-RDY) is through the Nozzle control via the inverter 240 as an input signal to a NAND gate 242 given for the following Description Assuming that there are fifteen nozzles, there are also fifteen lines to each

the respective nozzles provided. For the sake of a clear description, however, only one of these lines is examined in more detail below. A computer code signal is available on the computer connection units EB 6, EB 7 and EB% as an input signal for a NOR element 244. The remaining input of the NAND element 242 receives a suitable code key signal to identify each of the fifteen nozzle controls. This code signal is transmitted to the computer data lines EB 3, EB 4, EB 5 and EB 0,

EB 1 and EB 2 delivered, these signals each representing inputs for the NOR elements 246,248.

The address decoder also has a NAND element 250 to which the respective outputs of the NOR elements 246, 248 and the computer interrupt signal JAUX are applied. The output of the NAND element 250 is reversed by the inverter 252 and is applied as an input the NAND gate 242 for the particular nozzle for which the relevant scanning or interrogation line is intended

stated that he can query the condition of the nozzle in question to check that the compensation is correct correct is. If the compensation signal is correct, the computer sends a command to the nozzle control electronics interrupt further compensation or readjustment for this nozzle.

This command is required because it is not possible to use the paint spray nozzle to print while it is running the alignment or adjustment continues. A alignment error occurs in any of the fifteen nozzles

is discovered, the computer identifies the nozzle in question via the aforementioned scanning line. Will be such a If calibration errors are detected for one or more of the nozzles, the computer emits an operating alarm signal. This makes the operator aware of the error and can switch off or interrupt the device in order to check the nozzles.

The main head controller

The aforementioned external effective / ineffective computer commands can also be used to activate / deactivate the connection unit, for example during periods of time when the printing press is running but is not to be printed because the operator is currently making a setting at the press

undertakes or other operational functions are to be carried out. In these cases the computer switches the print head controls ineffective. The last thing the computer did before printing is concerned that all controls are activated. The terminal unit has a main head controller which controls the first nozzle head in a printing unit, one of these control circuits is shown in FIG. 7 to Clarification of the description shown. The main head controls for print units 2 and 3 are

those of F i g. 7 similar.

The device address (GA signal) as described above with reference to FIG. 6, is reversed by an inverter 234 (FIG. 6) and is passed as one of the inputs to a NAND gate 260 with four inputs. The computer sets the NAND gate 260 with the control signal FRYX and the data line £ ß14. the

Outputs of the NAND gate 260 set a flip-flop 262 which is switched by the computer signal DR YX to a 12-bit counter 264 (which has three 4-bit counters 266, 268 and 270 connected in series) via a LADf pulse from a NAND gate Ϊ72, which in turn is conditioned by the (^ output of flip-flop 262 and the computer signal DR YX . A head clearance count from the computer is then entered into counter 264 via the computer -Datenanschlußleitungen EBO to EBU written in. the aforementioned CTÜF pulse is by an AND gate 274 (is set 306 whose other input is approximated by the output of zero decoder, as discussed below) via an inverter 276 keyed to a flip-flop 278 The (^ output of the flip-flop 278, together with the web feed clock pulse CP, is applied to the input of a NAND gate 280 to increment the contents of the counter 264 from the head distance count value stored by the computer to dismantle them.

It is necessary to provide a print request signal in the nozzle control electronics, which occurs a little before the actual triggering of the printing process, for example in order to build up the high voltage for charging the droplets. This print request signal is given to the nozzle control electronics 240 pulses a 5.08 cm web advance before the desired point to be printed on the web has reached the relevant nozzles. A decoder 282 decodes the octal number 360, which corresponds decimally to the number 240, by means of the octal-decimal decoder logic, the NOR elements 284, 286 and 288. The inputs to the NOR element 284 come from the counter 266. The inputs of the NOR element Elements 286 and 288 come from the output of the counter 268. The correct inverted signals are generated as inputs for the NOR elements 286 and 288 via inverters 290, 292, 294 and 296. The output of the NOR element 284 comes together with the outputs of the NOR Elements 286 and 288 as an input to the decoder 282. The output of the decoder 282, together with the computer command STOP EQUALIZATION, is an input to an AND element 298. A message end signal (EOMS = End of Message Signal), the generation of which is explained below is, together with the computer command START BALANCE as an input to an AND element 300, the output of which is given together with the switch-on pulse E 1 as an input to an AND element 302. The output of the AND element 302 sets t a flip-flop 304, and the output of AND gate 298 switches flip-flop 304 to provide the print request signal to the nozzle control electronics two inches or 240 pulses of web advance before the dot to be printed on the web by the first nozzle. The pressure request signal to the nozzle control electronics is the command to go to deck readiness, to end the calibration mode and to switch to printing mode. The value of 5.08 cm (2 inches) is selected to provide the nozzle control electronics with the necessary time to build up the high voltage level and to interrupt the calibration, in which case a maximum web speed of 213 m / min. (700 feet per minute) is used.

Simultaneously with the above-mentioned decoding, the counter 264 further reduces its content by means of the web speed clock pulse CP , which is supplied via the NAND element 280. A zero counter 306 determines the count value zero in the counter 264 via a decoding logic which has the NOR gates 308, 310 and 312 which are connected to the respective outputs of the counters 266, 268 and 270. The output of the zero decoder 306 represents an artificially generated single key pulse (ASTR 1), which triggers the head controls 2, 3, 4 and 5 of the printing unit connected to the main head control. The ASTR 1 pulse arrives at the input of a NAND gate 314. This Pulse is also an input pulse for a flip-flop 316 and via an inverter 320 for a divide-by-twelve circuit 318. The flip-flop 316 is set by an AND gate 444 (described below), and the ASTR 1- Pulse flipped, and its (^ output reaches an AND gate 322 together with the web speed clock pulse CP , whereby the clock pulse sequence enters the divide-by-twelve counter 318. The output of the counter 318 is through NAND gates 324,326 and 328. Ten characters per 2.54 cm (10 characters per inch) are provided in each line, ie there are 12 pulses per character if - as provided here - 120 clock pulses are available per 2.54 cm. Fig. 14 ve shows the relationship of the first character marking pulses STR 1 and the second marking pulses STR2. For each STR 1 pulse there are five STR 2 pulses. This means that a total of 1 STR 1 and 5 STR 2 pulses are available for each character. The output of the NAND gate 228 is inverted by an inverter 330 and, together with the Y signal (the generation of which is explained in detail below), is input to a NAND gate 332. The output of the NAND gate 332 provides, together with the ASTR 1 pulse, the necessary STR 1 pulses at the output of the NAND gate 314 to produce each character in a printing unit through the nozzle under the control of the main head controller. The STR 2 pulses are generated at the output of the NAND gate 326 and each consist of five pulses that determine the monitoring for the columns X \, X2, ΧΊ, Χ4 and X5 in the 5 χ 7 character matrix of the nozzle control electronics.

The mentioned operational sequences of the main head controller for each printing unit run cyclically in the specified one Way off. The computer provides the necessary head clearance pulses to counter 264, if any A change in information is required, for example if the form is printed in different locations on the Printing cylinder 40 is to take place.

The following are the print controls for print heads 2, 3, 4 and 5 of each print unit described:

For the sake of simplicity of description, only such a head controller will be described with reference to FIGS. 8 described. Four such head controls are provided in each printing unit, so that in the presently selected example of an embodiment of the invention with three printing units, a total of 12 of these head controls are present. Each of the controls 2, 3, 4 and 5 of each printing unit generates ASTR 1, STR 1 and STR 2 pulses. Part of the circuitry in the head controllers 2, 3, 4 and 5 is similar to that of the main head controller (FIG. 7), and therefore the corresponding parts are given the same reference numerals. In this case, the web feed clock pulses CP arrive at the input of the AND element 340, which is set on the input side by the (^ output of the flip-flop 342, which in turn is set by the output of an AND element 344 and

is switched by the CP input pulses. The AND gate 344 receives an EFFECTIVE input pulse E \, together with an £ OM signal (the generation of which is explained below). By setting the AND gate 340, the divide-by-twelve counter 318 is activated in order to divide the signal CP. A decoding logic comprises the NAND gates 324, 326, 328, the inverter 338, the NAND gate 332 and the NAND gate 314 generates the STR 1 and STR 2 pulses in a similar manner to the generation of the pressure control pulses the main IC head control has been explained. The aforementioned // signal, which is generated in connection with the EOMS circuit for the main head control, is input to the NAND gate 314 in order to interrupt the STR 1 pulse generation. The 577? 1-pulses are delivered at the output of the NAND gate 326 in a similar way as before when the 577? 2 pulses for the

ίο main poll tax was explained. The main difference between the head controls 2, 3, 4 and 5 and the main head control for a particular printing unit is the counter circuit. The head controls 2, 3, 4 and 5 each have a counter 350, the series-connected binary counters 352, 354 and 356 includes. There is a fixed spatial relationship between each of the nozzles and each printing unit Aus For descriptive purposes, the distance between each of the nozzles is 63.5 mm (2.5 inches) - see FIG. 2.

63.5 mm χ 120 pulses per 25.4 mm of the web feed results in 300 pulses. This corresponds to the nozzle lag or distance between adjacent nozzles in a printing unit. So that the 577? Generate 1 and STR 2 pulses of a respective head control 2, 3, 4 and 5 of a certain printing unit by counting up to 300, this counting process being triggered by the ASTR 1 pulse of the immediately preceding head. That is, for the head control 2, the ASTR 1 pulse of the main head control of this relevant printing unit reaches an AND gate 358a as an input. The other input of the AND gate 358a is set positive because the NAND gate 274 has not yet been keyed and thus the flip-flop 360a is toggled by the output of the AND gate 358a via an inverter 362a. Flops 360a clears and triggers the counting process for the CP pulses by counters 352,354 and 356. Counters 352,354 and 356 are then incremented to a binary-coded octal count of 454, which corresponds to a decimal number of 300. The output of the counters 352,354 and 356 is converted from the binary-coded octal representation into a decimal value by a known octal-decimal converter circuit, which is represented in the circuit diagram by NAND gates 358, 360 and 362, inverters 364, 366 and 368 and by inverter 370 The actual wiring of the decoding circuit is not shown in order to simplify the drawing. In addition, such number decoding circuits are known, as already mentioned. The output of the NAND gate 274 corresponds to the ASTR 1 pulse of the special head controller 2, 3, 4 or 5 and is reversed by the inverter 374 to reset the divide-by-twelve counter. The ASTR 1 pulse from each head controller 2, 3, 4 and 5 appears as the output of a NOR gate 376. The ASTR 1 pulse from the head controller is input to the head controller 3, whereupon the same operating sequence takes place as described above. Similarly, the ASTR 1, STR 1 and 577? 2 of the head controls 3, 4 and 5 are generated

The print request from the head controls 2, 3, 4 and 5 of a certain printing unit must also be sent before the STR 1- and 577? 2 pulses can be generated in a manner similar to that previously discussed with reference to the main head controller. In this case the binary octal-decimal conversion takes place via the NAND gates 378, 380, 382, the inverter 372 and the inverters 384, 386 and 388. This number decoding circuit decodes the octal number 74, which corresponds to the decimal number 60. The output of the NAND -Gate 390 comes as an input to an AND gate 292, together with the command STOP COMPENSATION, in order to control the flip-flop 294 corresponding to the flip-flop 304 in the main head control in order to receive a print request output signal from each of the head controls 2, 3, 4 and 5. The flip-flop 294 is set by the output of the AND gate 296, which is conditioned by the command START COMPENSATION and the fOM signal described below.

Common head control - end of message (EOMS \ gn & \)

Each nozzle prints a line consisting of thirty-eight characters (including spaces) around the

So to determine the printing process by each nozzle, it is necessary to monitor what number of characters have been printed by each nozzle. The in F i g. The circuit shown in FIG. 9 ensures this function for each of the nozzle heads. Since the circuits are similar, only one will be described in detail to show how the characters are counted.
The above-mentioned STR 2-pulses pass to the clock pulse input of a flip-flop 400. The K-Em is gear connected to ground, and the / input is connected to +5 volts angeschlossea The control pulse fern Q output of flip-flop 400 passes as Input to a NAND element 401, the output of which is connected to the clock pulse input of a flip-flop 402. The / input and the K input of this flip-flop 402 are each connected to +5 volts. Flop 402 causes it to toggle, so that output signals appear at terminals Q and Q. The (^ output of the flip-flop 402, together with the (^ output of the flip-

Flops 400 and the STR 2 pulses from the assigned head control arrive at a NAND element 404 as an input. The output of the NAND element 404 and the C7W pulse from the format head control go as an input to a NAND element 406 to serve as a register tactile shift pulse for purposes which will be described below in connection with the nozzle control circuit of FIG. 10 will be explained. The Q and (^ outputs of flip-flop 402 are also available at terminals 409 and 411, respectively, in order to serve as additional register key pulses, the purpose of which is also explained below in connection with FIG. 10.

Binary counters 408,410 count the number of characters printed by each nozzle in the following Way. However, before explaining the counting operation in detail, it is necessary to point out that two characters in each memory buffer of the nozzle control of each nozzle in the case of the one provided for this system

Computer device can be saved. The data for two additional characters is transferred from the computer to the memory buffers of each nozzle controller by the aforementioned CTOF pulse. The CTOF pulse also shifts the two previously stored characters from the memory buffers of each nozzle control into the associated print buffers or memories. This means that four characters are stored in each of the nozzle controls before the actual printing process at each nozzle. Therefore, in order to determine the character pressure from each nozzle, it is necessary to count the aforementioned two character shift operations seventeen times. In the head controller 1 of each printing unit, the CTOF pulse is transmitted to an input of an AND element 412. The other input of this AND element 412 receives the control pulse E 1. The output of the AND element 412 serves as an input for the room input of the flip-flop 414. The (^ input of the flip-flop 414 is the input of the NAND element 416. The other input of the NAND gate 416 supplies the output signal Q which occurs after the flip-flop 402 has flipped and which is inverted by an inverter 418. The output of the NAND gate 416 is connected to the clock pulse trains of counters 408, 410 Counters 408, 410 are binary counters, ie the individual outputs of counter 408 correspond to the values 1, 2, 4 and 8, and those of counter 410 correspond to the values 16, 32, 64 and 128. A number decoder 420 is set to be seventeen STR 1 pulses decoded For every STR 1 pulse, five STR 2 pulses occur, and therefore the decoder 420 is set to decode thirty-four STR 1 pulses to count the number of characters in Fig. 9 by an I. nverter 422 reproduced. Here, too, the actual connections from the respective outputs of the binary counters 408, 410 are not shown in order to keep the drawing clear. The decoding circuit and the individual connections are known. After the number decoder 420 has determined the thirty-fourth STR 1 pulse (the seventeenth character shift pulse), its output signal triggers the clock pulse input of the flip-flop 414. This triggers a switching operation of the relevant flip-flop 414 at the Q and Q outputs the end. The Q output then closes the NAND gate 416, so that the clock pulse inputs (subdivided STR 2 pulses from the (^ output of the flip-flop 402) to the binary counters 408, 410 are suppressed. The NAND gate is disabled 416 also terminates the generation of the data print request commands (Print Data Request Instructions = PDRI) to the PIM-Em gang of the computer via the chain of NOR elements 424, 426, 428 (NAND elements with two connected inputs).

The binary counter 430 and the associated circuit supply two end-of-message signals (EOM signals), which are generated in the following manner and for the following purpose: In the main head controls, the pulse output is at the clock pulse input of the counter 430 of the inverter 330 (FIG. 7), which generates the STR 1 pulses for the main head control in the chain of logic circuits. For head controllers 2, 3, 4, and 5, the clock pulse input at counter 430 corresponds to the outputs of inverter 338 in each of head controllers 2, 3, 4, and 5, which are illustrated with reference to FIG. 8 were explained. The aforementioned output of the number decoder 420 shifts the Q output of the flip-flop 414, which reaches the room input of the binary counter 430. This activates a corresponding counter 430 for the STR 1 pulses from each of the associated main head control or head controls 2, 3, 4 and 5. As long as four characters remain in each of the memories and print buffers for each associated nozzle control (after seventeen Character shift operations) it is necessary to count four 5TK 1 pulses in order to stop the generation of the STR 1 and STR 2 pulses, i.e. to end the print line generated by each nozzle. An end-of-message number decoder 432 detects count four in binary counter 430, and number decoding circuit 434 shown in block form provides the necessary inputs to end-of-message decoder 432 from binary outputs 1, 2, 4 and 8 of binary counter 430 to perform this function. Numeral 434 denotes the inverter circuit used for this number decoding operation. The end-of-message number decoder 432 is connected to the output of a NOR gate 436 (a NAND gate with through-connected inputs) via an inverter 438. The output of the NOR gate 436 provides an EOM signal which is transmitted to the PIM input of the computer.

An end-of-message number decoder 440 provides an additional control function which will be explained below. The number decoder 440 is connected to the corresponding outputs of the binary counter 430 (using the inverters 434) in order to be sampled by the third STR 1 pulse counted by the counter 430. The output of the end-of-message decoder 440 is connected to the clock pulse input of a flip-flop 442. This flip-flop 442 has been reset by the ASTR 1 signal at its room input from its assigned main head controller or, depending on the case, from the head controllers 2, 3, 4, 5. The clock pulse signal from end-of-message decoder 440 to the input of flip-flop 442 causes its (^ output to toggle and the aforementioned ^ signal produces the J signal from each of the fifteen head controllers' common ends The message circuit is transmitted in the following way to the three respective main head controls and the twelve head controls 2, 3, 4 and 5. For the main head control, the Y signal is input to the NAND gate 332 in F i g.7, which is thus disabled in order to terminate the generation of the 577? 1 pulses. Similarly, the / ^ signal is on the input of the NAND gate 332 in the head controllers 23, 4 and 5, respectively switched as explained above with reference to Fig. 8, whereby the STR 1 pulses from each of the head controllers 2, 3, 4 and 5 are terminated.

The EOM output at an associated end-of-message number decoder 432 is connected in the case of the main head controller for each printing unit to the input of the AND element 444 in this main head controller (FIG. 7) . At the other input of the AND gate 444 is the activation signal E 1. The pulse output from the AND gate 444 is switched to the room input of the flip-flop 316 in order to end the operation of the divider 318 and the subsequent decoder circuit, whereby the generation of the 57R2 pulses (Fig. 7) is interrupted.

In each of the head controls 2, 3, 4 and 5 (FIG. 8), the £ OA / output of a respective end-of-message number decoder 432, together with the activation signal Ei, reaches the input of the AND element 244. The pulse of AND gate 344 is applied to the room input of flip-flop 342, thereby enabling the operation of divider 318 and the subsequent ones and shown in FIG. 8 is terminated in order to also terminate the generation of the STR 2 pulses from each head controller 2, 3, 4 and 5 of each printing unit.

The nozzle control Supply of the coded alpha-numeric data to the nozzle control circuit

In the embodiment of the invention described here, there are five nozzles per print head, and in total three print heads are provided. The nozzle control circuit fulfills the function of receiving that to be printed Data from the computer as well as the transfer of the data to each of the nozzles in each print head so that each Line of the desired variable information in a fixed orientation with respect to the format on the Printing cylinder 40 is printed out. In the system described here there are fifteen nozzle control circuits intended. A typical circuit of this type is shown in FIG. 10.

Each alpha-numeric character is represented by a seven-bit code in ASCII code format. Two characters at a time are written in each nozzle control circuit. One character is transmitted on the computer busbars FSO and EB 6, and another character via the busbars EBS to EB 15. The character data on buses EBO to EB 6 are written into buffer memories 450 to 452. The other character is cached in buffers 454-456. The character information is keyed into the buffer memories 450 to 456 by a flip-flop 458 which is cleared from the output signal of a NAND gate 460 by the computer signal FRYX, the signal on the ^ busbar e £ 514 and the output of the previously described address register. The computer signal DRYX is fed to the clock pulse input of flip-flop 458, the (^ output of which samples the data from the buses EBO to EB 15 into the buffer memories 450 to 456. The shift pulse at terminal 407, which is triggered by the CTOF pulse of the NAND -Glem 406 (previously explained with reference to FIG. 9) is generated, the character information for two characters at a time from the buffer memories 450 to 456 in the respective pressure accumulators 462, 464, 466 and 468 via the gate 470, the inverter 472 and the Gate 474.

The character data in the print memories 462 to 468 are alternately shifted out of these memories in the following manner: The previously explained memory or register strobe pulses appear on terminals 409, 411 in FIG. 9 and are fed to the respective inputs of gates 476 and 478, as explained , the impulses at terminals 409, 411 correspond to alternating STR 2 impulses of an assigned main head or head control. As also in FIG. 9, a PDRI-lmpvAs is additionally transmitted to the / YM input of the computer in order to request new character data. This takes place every time the character data is moved between the pre-storage or temporary storage and the printing storage.

The pulses at the terminals 409,411 from the corresponding Q or (^ output of the flip-flop 402 in FIG. 9 alternately control a series of AND gates 480,482,484 ... 486 etc. in order to alternate the character output data from the Print memories 462, 464 and the print memories 466, 468 via additional AND gates 488, 490 etc., OR gates 492, 494, 496 etc. and line amplifiers 498, 500, 502, 504, 506, 508 and 510 to scan through seven information bits which correspond to each character on the data lines L 1 to L 7. This information on the lines LX to L 7 is then input to the nozzle control circuit for generating the necessary control voltages in the charging tunnel of the associated nozzle for each character Not the entire logic switching chain at the output of the pressure accumulator 462 to 468 and the lines L 1 to L 7 is shown in the drawings. Only a sufficient number of ports and connections is shown in each case indings between the outputs of the pressure accumulators 462 to 468 and the lines LX to L 7, so as not to impair the clarity of the drawing. A person skilled in the art can add the additionally necessary logic gates for the respective outputs of the pressure memories 462 to 468 in order to obtain the necessary character data on the lines L 1 to L 7 . The character data output at each nozzle control is in ASCI 1 code.

In the case of the head control circuit explained with reference to FIG. 9, the logic circuit 309, whose The output of the flip-flop 400 clears and which receives the command INACTIVE, and the output of the inverter 303 Jf in the respective main head controller and head controllers 2, 3, 4 and 5 (Figs. 7 and 8, respectively) the correct one

§J 55 State of flip-flop 400 safe. The logic circuit 309 differs in the main head controller and || head controllers 2, 3, 4 and 5. Logic circuit 309 is for the main head controller in Figure 11A

Il shown, while F i g. 11B the logic circuit 409 showing head controls 2, 3, 4 and 5

β The links between the individual figures of the connection unit circuit described above

fl are from FIG. 12 can be seen.

U 60 In Fig. 13 are the relationships of the depths of form on the plate cylinder, along with the electrical "Si Format head fCTOF pulse) and the mechanical format head switches (TOF switches) and the clock slot

[3 zen shown

| J F i g. 14 illustrates the relative timing of the occurrence of the PRINT REQUEST, des signal

The PRINT READY signal, the STR X and STR 2 pulses, the status of the data lines L 1 to L 7 and the

,;! there path feed clock pulses CP. As far as the CP pulse changes with the path movement, the ■ 'i The signals mentioned above and shown in Fig. 14 are shifted in time, but the order and timing

Si; Sequence in which they are generated relative to one another, the same and for all web or press speeds

i || corresponds to the representation according to FIG. 14th

14th

The nozzle control electronics

15 shows in a block diagram the essential components of the nozzle control electronics and the application of the signals generated by this control circuit to various components of a nozzle, which are shown in the lower part of FIG. 15 are shown. A matrix generator 550 receives the character setting command in ASCII code on the data lines L 1 to L 7 of the connection unit (FIG. 10). Timing pulses STR ! Are also sent to the matrix generator 550. STR 2 for the character generation of each of the main head controls and the head controls 2, 3, 4 and 5 of the connection unit To print a line of alphanumeric characters - to provide fifteen matrix generators 550, each of which is supplied with individual character data on separate data lines L 1 to L 7 and character data clock control signals STR 1 and STR 2 from the head controls assigned to the respective nozzles.

Each matrix generator 550 generates a sequence of digital pulses, each pulse corresponding to a color droplet. The matrix generator 550 also generates clock control pulses and the sequence of digital pulses and the clock control pulses are converted by a digital-to-analog converter circuit 552 (video generator). recorded, which converts the sequence of digital pulses into an analogue using the clock control pulses Video signal converts The output of the video generator 552 corresponds to a low-frequency video display for each of the characters received from the matrix generator 550 concerned.

The low-frequency video character data from the video generator 552 reach a final amplifier circuit 554, which has the following basic components: a nozzle driver circuit, a video amplifier circuit device and a high voltage deflection circuit. The nozzle driver circuit provides a 66 KHz sine wave Excitation of a piezoelectric crystal in the nozzle 556, the purpose of which is explained in more detail below. That The 66 KHz sinusoidal signal passes from the video generator 552 to the nozzle driver circuit in the final amplifier stage 554. The video generator 552 transmits the signals PRINT READY and receives the signals PRINT READY REQUIREMENTS mentioned above in connection with the explanation of the connection unit.

The video amplifier section of the final amplifier stage 554 amplifies the low level of the video character data from the video generator 552 and the output is fed to a charging tunnel 558 into the selected one One of seven possible charging voltages can be applied to color droplets (or no charging voltage is obtained), in accordance with the video character information generated by the video generator 552 from Matrix generator 550 was generated. The loading of the droplets to generate the alphanumeric characters will be discussed further below, in connection with FIG. 16 explained

The KHz-excitation of the nozzle 555, together with the pressure of the color, which the nozzle over the opening 557 of is supplied to an ink supply (which will be described in detail below), causes the ink in discrete small droplets are torn apart, in such a way that about 66,000 droplets per second be generated.

The high voltage from the deflection circuit of the final amplifier stage 554 is fed into a deflection tunnel 560 (the one shown in Fig. 15 from its actual position shown with respect to droplets Rotated 90 °), thereby deflecting each of the color droplets a distance that is directly proportional the imprinted charge that the ink droplet receives as it passes through charging tunnel 558. In the The illustrated embodiment of the nozzle shows the color droplets from the plane of the drawing in FIG. 15 on the itself Moving web 39 deflects those ink droplets 559 that were deposited during passage through the loading tunnel 558 are not loaded with a charge, are deflected in the deflection tunnel 560 and are through the Collector 564 received and returned through a vacuum return line 566 to the paint supply system

A sensor 568 is provided in the collector 564 in order to detect the scanning of the ink droplets during the To ensure that the nozzle can be calibrated. Generally speaking, it is necessary to sense whether the droplets are in the correct phase or in the correct alignment in order to be charged with the correct charge. If the correct charge is not impressed on the color droplets, the adjustment of the Color droplets adjusted by correcting the phase of the 66 KHz vibration signal applied to the nozzles will. The technical procedure for this comparison of the color droplets and a suitable device are disclosed, for example, in US Pat. Nos. 34 65 35ö, 34 65 351 and 35 62 76 i and are assumed here and only to improve the understanding and the function of the color droplet adjustment mentioned how it is applied in the system according to the invention. However, it should be emphasized that the circuit and the method for droplet sensing and alignment itself do not form part of the invention. the Color droplet scanning and matching is done by a matching and droplet sensor 570 in FIG. 15, which determines the alignment signal for the video generator 552, which generates the PRINT READY signals can be.

It should be noted that the nozzle spray opening has a diameter of a little more than Άοο mm (V2000 inch) and that is the distance between collector 564 and the moving web in this case system described is substantially 12.7 mm (1/2 inch)

The creation of characters by charging ink droplets

The Fi g. 16A to 16G illustrate the principle or the operating theory for the generation of the video signals provided here, with the generation of sixty-four alpha-numeric characters in the matrix display. division (see Fig. 4) are to be generated by the nozzle control electronics. As already with reference to F i g. 15 determine the amplified video signals representing the charging voltages special symbol and are fed to the loading tunnel 558 F i g. 16B shows the relationship of the thirty-five

Pulses that are available for generating a given alpha-numeric character within the 5x7 character matrix. As already described in connection with the connection unit, the STR 1 pulses in the nozzle control electronics are used to display each alpha-numeric character in a print line above each of the fifteen nozzles. The STR 2 pulses generated by the respective main head controllers and the head controllers 2, 3, 4 and 5 are each assigned to a specific column Xi, X2, X3, XA and XS of the 5 × 7 character matrix Rows are indicated as Yi, Y2, Y3, YA, YS, Y6 and Yl in Fig. 16G, respectively. Within the time interval of each 5TR 2 pulse, which determine the intervals X 1 to XS , corresponding voltage levels are generated, each of which represents a row or row Yi to X 7 of the matrix in the case of the FIG. 16A correspond to the representation selected. So were - if thirty-five pulses through the matrix genes rator 550 (FIG. 15) can be generated - as a result, a complete rectangle is printed out by the relevant pressure nozzle, since in the charging tunnel 558 charging voltages for each droplet within the droplet stream, as in FIG. 16C, and would be laid out for each column X i to XS of the matrix.

16D, E, F and G illustrate how the letter E is generated: The matrix generator in the nozzle control electronics generates, in accordance with the STR 1 and STR 2 pulses and the character data

ij th, a pulse train on lines 1 to 7 from the connection unit during the X 1 interval in order to generate successive voltage levels for the line locations YX to Y1. These voltages are fed to the charging tunnel by the video amplifier.As far as the droplets passing through the charging tunnel are balanced or occur synchronously with the iripulses, each of the seven successive droplets that pass through the charging tunnel during the interval Xi is correspondingly higher Voltage impressed A dotted line is generated for column X 1 on the advancing web where the charged droplets strike at different intervals along an axis transverse to the feed movement of the web when the droplets pass through the deflection tunnel. In the interval X2 , only the first, fourth and seventh pulse in this interval is generated by the matrix generator and leads to excitation levels in the video amplifier which correspond to the line positions Yi, YA and Y1. This impulse generation by the matrix

Generator is repeated for the intervals X 3 and XA . In the interval XS , only the voltages corresponding to the line locations Y 1 and Yl are generated by the pulse sequence according to FIG runs through this interval, is applied with a charge. The remaining droplets, namely droplets 2 to 6, are not charged and therefore not deflected when passing through the deflection tunnel. Uncharged droplets are, independently gig from the interval X 1 to X 5, not diverted and get into the in F i g. Collector 564 shown in FIG.

The printing ink system The F i g. 17 shows the side view of the printing unit 38 and illustrates the mechanical structure and the

Allocation of the ink supply and the vacuum ink return system for a typical printing unit Die Paint is injected under pressure into a multiple feed line 600 from an electromagnetically controlled supply arrangement 602, which supplies the paint at a paint inlet 406 from a paint supply unit which is described below. For the sake of simplicity, there is only one Color system shown and described insofar as the individual color systems are identical for each color nozzle unit is. The paint for the paint spray nozzle 38c enters the paint multiple supply 600 via a connection 606 and passes through a manually operated shut-off valve 608. Before the pressurized paint enters the When nozzle head 610 enters, it is filtered through a filter 612 and passes through a line 614 to the nozzle head 610 via the connection unit 616. The paint droplets generated by the vibrating transducer pass through a in Fig. 17 loading tunnel, not shown, and then pass the deflector plate 612 onto the moving one Path 39 or into collector 614, depending on the deflection due to the charge of the droplet in question. As mentioned earlier, the uncharged droplets are collected by collector 614, and the im Collector 614 is collected paint via internal lines and a vacuum discharge line 616 returned, which is connected to the multiple line 618 for the vacuum return. The one in every collector Paint accumulated on each paint spray nozzle is evacuated via vacuum return line 620

The paint supply regulating system essentially consists of a double connection pump, 2 paint containers, two containers for additional solution, a vacuum pump, various caliber tubes, pipe connections and the associated electrical control. The pump supplies the ink to the print heads with a pressure of 336 kg / cm ² p! Us or minus 0.0175 / cm ² . The pump draws color from one or both of the color reservoirs, using either one or both of the pump connections. The one contained in the paint container Air is sucked out by the vacuum pump. The remaining vacuum is used to restore the paint from the print heads to the memory. The color memories are automatically taken from the memory for the Additional solution replenished. The paint system can be provided with suitable low-pressure and low-vacuum intermediate locks in order to avoid large-scale waste of paint. An exemplary embodiment of the paint supply system is shown in FIG. 18 622 is under a pump 624 through a filter 626, a paint trap 628 and a vacuum line 630 generated vacuum, the vacuum line 630 having an opening 632 into the paint container 622, like the Figure shows. A storage 634 for the additional solution is under atmospheric pressure, and that of this Color to be supplied to memory 634 is automatically added to color memory 622 via a line 636, a controllable solenoid valve 638 and a line 640. The solenoid valve 638 is by a not shown Level sensing device in paint tank 622 is controlled. The paint pump 642 pumps the paint through a valve 644 and a filter 646 from the color memory 622. With 625 a switch for high vacuum is designated, which is set to about 635 mm Hg is set, while a vacuum switch for minimum vacuum is designated by 629, which is on is set to about 127 mm Hg. A vacuum gauge 631 is also provided to measure the vacuum in the

Vacuum line 633 to be able to read a

The paint is pumped by means of the pump 642 via the line 648 to the pressure regulator 650, which ^{is variable between 0 and 4.20 kg / cm 2.} A pressure regulator 652 is set to about 2.45 kg / cm ² , so that only then paint to the assigned nozzles 38a to 38e is released when the color pressure in the line 648 the value of 2.45 kg / cm ^? A bypass regulator 654 is provided as an additional device for controlling and monitoring the pressure of the paint in the line 648. In order to be able to read the pressure, pressure gauges 656 and 658 are provided at suitable points on the line 648. For example, the pressure gauge 656 shows pressures from 0 to 7.00 kg / cm ² , while the pressure gauge 658 shows pressures from 0 to 4.20 kg / cm ^{2 which} are under regulated pressure of about 2.4 kg / cm ² Paint reaches the multiple feeder 600 via the previously described supply and solenoid valve arrangement 602 (FIG. 17) and then through the manually operated shut-off valves 608a to 608e.

The paint from the collectors 614a to 614e is accordingly conveyed into the vacuum collecting line 618, as FIG. 18 shows, and from there via the vacuum return 620 and through the filter 660 into the Recirculated paint container 622 FIG. 18 shows only part of the paint supply system. Also is There is still a reserve system, not shown, which is constructed similarly to the illustrated ink feed regulating system.

The ink usable for the printing device can be a water-soluble ink. Your critical value is the viscosity, which should normally be 1.8 cP. The paint must be electrically conductive in order to use the charge can be applied when the color droplets pass through the explained color tunnel. However, the electrical conductivity of the paint is not critical, since it is compensated by changing the Charging voltages at the charging tunnel! can be provided. The paint dries by absorption and not by evaporation, so that the paint is prevented from drying on in the nozzles.

15 sheets of drawings

Claims (4)

Patent claims: 1. Printing device for printing changeable alphanumeric data on a printing speed device moved web at positions which are opposite fixed, by a main pressure cylinder on the Web-printed data are aligned, with a device for generating timing signals and encoded data corresponding to the variable data of a selected multi-line format, a device for printing out the variable data by ejecting independently controllable streams of ink droplets onto the web and with a controllable by the timing signals and with the coded data-acted control device for generating character clock signals corresponding to the sequential position of the alpha-numeric data in the multi-line format, characterized by a device (34) for automatically setting the character clock signals to different web speeds with a device (41) for generating a format head pulse (ETOF) with each revolution of the main printing cylinder (40) in front of the mechanical format head of the printing cylinder, a pulse generator (42) for generating first pulses (CP) corresponding to the web speed and a device for correcting the temporal position of the format head pulses (ETOF) as a function of these first pulses, the character clock setting device (34) being controlled as a function of these corrected format head pulses (CTOF) 2. Printing device according to claim!, Characterized in that a control device (33) has a Has a unit for setting a number of different format depths of the main pressure cylinder (40), and that the setting device for the character clock signals a switching unit (F i g. 5A, 5B) for the temporal Correction of the format head impulse depending on the desired format depth 3. Printing device according to claim 2, characterized in that a number unit counts the first pulses (P) corresponding to the construction speed during a fixed period of time, a first counter 169 for detecting the first pulses and a device controlling this counter for counting during a fixed period of time ( 171) , and that the device for correcting the format head pulse contains a counter (197) which holds the count value of the first counter, and counts the first pulses until a predetermined count value is reached and generates a signal which generates the corrected format head pulse (CTOF) represents 4. Printing device according to claim 3, characterized in that the control device has a further device for generating a second pulse sequence with a fixed number of pulses per revolution of the main printing cylinder that the device for automatically setting the character clock signals a further counting unit (98-102 ) for these second pulses that a digit or number decoding device determines the number of pulses corresponding to the number of format depths and that a gate scans the number of decoded pulses as a function of the format head pulse (ETOF) on the correction device and the excitation of the correction device accordingly the desired format depth is delayed.