EP0386489B1 - Method of controlling dampening in an offset printing machine - Google Patents

Description

The invention relates to a method for regulating the amount of dampening solution or dampening solution layer thickness on the printing plate of an offset printing press according to the preamble of claim 1.

In offset printing technology, the quality of the print depends crucially on the dampening solution supply. The problem here is to adapt the amount of dampening solution to the changing production conditions. The amount of dampening solution has a very high influence on the dot gain, so that its optimal setting, which corresponds to the respective operating state, is of great importance.

From the reference "Deutscher Drucker" No. 13, 1986 w 235 ff., A dampening solution control method in offset printing is known, in which the amount of dampening solution on the printing plate is detected and regulated to a predetermined target value by adjusting the dampening roller speed. The nominal value for the moisture is taken from an empirically determined nominal value table, which is based on empirical values for various plate-paper combinations, in particular from previously printed editions. This known control method is not yet optimal with regard to the process control and the work result.

From US-PS 3946297 a special controller or a special control method has become known. An optimal comparison between a measured actual value and a predetermined target value is to be achieved without instability problems arising as a result of the target value being exceeded or undershot. It is proposed that the controller or the control method work in two modes: in the event of large deviations between a measured actual value and a predetermined target value, the controller becomes a P controller operated while a changeover takes place when the actual value approaches the setpoint, and the controller now works as a PI controller.

The invention has for its object to provide a method of the type mentioned, which leads to excellent control behavior, in particular to optimal speed without instability problems. In addition, the amount of dampening solution should be kept largely constant even when attacking strong disturbances.

This object is achieved by the features in the characterizing part of claim 1. Because of the method according to the invention, it is possible to select the setpoint value for the amount of dampening solution relatively close to the so-called lubrication limit. The smear limit is the state of a quantity of dampening solution at which ink-free areas of the printing plate begin to take on color, which leads to the production of waste. Despite the setpoint adjustment close to the lubrication limit, no undesired state is approached due to the control method according to the invention. There is no danger of water noses, that is, of a condition in which the amount of dampening solution is too large. By changing the amplification factor according to the invention, on the one hand a strong reaction of the controller to the critical state of insufficient amount of dampening solution is achieved, which reduces the above-mentioned risk of lubrication. On the other hand, the use of the lower gain factor in the case of too much dampening solution also avoids the risk of instability in the control loop. Experiments with the control method according to the invention have shown that the control is of excellent quality in the case of a feedforward control. A water film thickness that decreases due to the feedforward control does not lead to process limit parameters being exceeded due to the very fast response of the controller caused by the relatively high amplification factor. As soon as the feedforward control is eliminated, a somewhat longer state transition is uncritical due to the gain factor which is now smaller according to the invention, since there is no risk of lubrication or water noses. At the same time, however, an overreaction (overshoot) of the controller is avoided, and there is also the risk of continuous vibrations banned. The control behavior of the control in the event of sudden changes in the setpoint (upward or downward) is extraordinarily good due to the configuration according to the invention. Overall, the control according to the invention enables the setpoint to be set to a setpoint which is favorable for the print quality, and at the same time the control is carried out in critical areas with an extremely rapid reaction and yet is operated in such a way that stable conditions are established. The slower controller response due to the smaller gain factor (basic gain factor) - in the event that the setpoint is smaller than the actual value - also leads to a stable, uncritical situation.

According to a development of the invention, the arrangement is such that the larger amplification factor is selected to be approximately 1.2-2, preferably 1.5 times as large as the basic amplification factor. These values have proven to be particularly favorable.

It is preferably possible to proceed in such a way that a control process which brings about a change in the manipulated variable only takes place when the setpoint / actual value difference lies outside a predetermined tolerance band. Because of this configuration, the controller only emits a control voltage change when the predetermined tolerance band is exceeded. The controller works with a hysteresis. This mode of operation has the advantage that stochastic fluctuations in the measurement signal (actual value or a value dependent thereon) are not taken into account, so that there is no unnecessary intervention by the controller and, moreover, the adjustment devices are protected by the lower stress. Due to the aforementioned, sign-dependent selection of the gain factor (setpoint greater than actual value: gain factor large; setpoint smaller than actual value: gain factor small) is despite the described tolerance band regulation a timely and well-dosed reaction of the controller can be achieved without causing an unacceptable drift of the relevant values.

The process control can be further improved by superimposing a control for the compensation of disturbance variables of known effect on the control. The consequence of this procedure is that the effects of known types of disturbance variables do not have to be compensated for in full by means of the control loop. Rather, the compensation of the disturbance variable or disturbance variables ensures that all essential parameters adjust to a new operating state. The procedure is such that the control always remains active even during the compensation process. The described compensation of compensated disturbance variables can be, for example, changes in the printing speed, the ink flow, the temperature or even air flow. The mentioned disturbance variable compensation is preferably implemented by superimposing compensation variables on the controller output variable, which are determined from the measurement of disturbance variables with a known effect on the dampening solution guidance.

It can be done in such a way that a compensation variable is superimposed on the controller output variable, which results from the measurement of only one disturbance variable. If several disturbance variables are to be compensated, a corresponding number of measurements are carried out and there is a corresponding number of compensation variable superimpositions.

Additionally or alternatively, however, it is also possible to calculate the compensation size or to determine it from stored tables or from characteristic curves. Such tables or characteristic curves can, for example, be stored in corresponding process control computers. Even the calculation the compensation size can be made by a computer.

To achieve excellent print quality, the process is carried out with a setpoint setting that is close to the lubrication limit.

The use of a PI controller has proven to be particularly advantageous.

For the determination of the actual value, it has proven to be advantageous if it is detected optically by light reflection measurement. The actual value represents a measure of the amount of dampening solution or the thickness of the dampening solution layer.

Fig. 1

eine schematische Darstellung der Regelanordnung zur Durchführung des erfindungsgemäßen Verfahrens,

Fig. 2

ein Blockschaltbild des Regelkreises,

Fig. 3

ein Diagramm, das den zeitlichen Soll- und Istwertverlauf in Relation zur Schmiergrenze zeigt,

Fig. 4

ein Diagramm des zeitlichen Soll- und Istwertverlaufes bei Störgrößenaufschaltung,

Fig. 5

ein Diagramm des Führungsverhaltens bei einem Sollwertsprung,

Fig. 6

ein Diagramm von Soll- und Istwert bei Störgrößenaufschaltung und Berücksichtigung eines Soll-Istwert-Toleranzbandes,

Fig. 7

ein Blockschaltbild und ein Diagramm zur Ausgestaltung mit Soll-Istwert-Toleranzband,

Fig. 8

ein Diagramm betreffend das Führungsverhalten bei einem Sollwertsprung unter Berücksichtigung des Soll-Istwert-Toleranzbandes und

Fig. 9

ein Diagramm des zeitlichen Verlaufes von Soll- und Istwert bei durch überlagerte Steuerung erfolgender Kompensation einer bekannten Störgröße.

The invention is explained in more detail with reference to the following drawings; it shows:

Fig. 1

2 shows a schematic representation of the control arrangement for carrying out the method according to the invention,

Fig. 2

a block diagram of the control loop,

Fig. 3

1 shows a diagram which shows the course of the setpoint and actual values over time in relation to the lubrication limit,

Fig. 4

a diagram of the temporal setpoint and actual value curve in the case of feedforward control,

Fig. 5

a diagram of the management behavior in the case of a setpoint jump,

Fig. 6

a diagram of the setpoint and actual value in the case of feedforward control and consideration of a setpoint / actual value tolerance band,

Fig. 7

2 shows a block diagram and a diagram for configuration with a target / actual value tolerance band,

Fig. 8

a diagram regarding the management behavior in a setpoint step taking into account the setpoint-actual value tolerance band and

Fig. 9

a diagram of the time course of the setpoint and actual value when compensating for a known disturbance variable by superimposed control.

The control method according to the invention is explained in more detail below with reference to the structure according to FIG. 1. An offset printing press 1 is shown, which has a printing unit 2 and a further printing unit 3. In the following, only the printing unit 2 will be discussed and the control arrangement will only be described here. However, the printing unit 3 is also provided with a corresponding control device.

A number of inking rollers (not shown) and, moreover, several dampening rollers 5 are assigned to the plate cylinder 4 of the printing unit 2 which has the printing plate. The dampening roller 5 'cooperates with a dampening solution reservoir 6. The dampening roller 5 'thus formed as an immersion roller can be changed in its speed. This is indicated by the arrow pointing to it. The speed is taken from the pressure cylinder 7 via an operative connection 8 and transmitted to a tachometer generator 9. The pressure cylinder 7 is provided with a rotary encoder 10 which detects the angular position.

1, the following peripheral devices are also provided: an evaluation circuit 11, a floppy disk drive 12, a computer 13 and a measurement data acquisition unit 14. To determine the dampening solution layer thickness, the surface of the printing plate arranged on the plate cylinder 4 is optically scanned by a sensor 15, which is connected to the evaluation circuit 11 via a line 16. The evaluation circuit is connected to the computer via a connection 17. This is connected via lines 18 and 19 to the floppy disk drive and the measurement data acquisition unit. The rotary encoder 10 is connected to the measurement data acquisition unit 14 via a line 20 and the tachometer generator 9 is also connected to the measurement data acquisition unit 14 via a line 21. Finally, the line 22 leads from the measurement data acquisition unit 14 to the drive, not shown, for the dampening roller 5 '.

Roughly speaking, the following functional sequence results: The sensor 15 determines the layer thickness of the dampening solution on the printing plate and feeds it to the evaluation circuit 11, which carries out appropriate data processing. The actual value of the dampening solution layer thickness is compared with a predetermined target value, which is selected such that it is relatively close to the so-called lubrication limit. This smear limit denotes the dampening solution quantity value or dampening solution layer thickness value at which ink-free areas of the printing plate begin to take on color. Specifying such a target value has the advantage that excellent printing results are achieved. The peripheral devices of the offset printing press 1 now form a control circuit, which will be described in more detail below, such that a deviation of the dampening solution layer thickness (actual value) detected by the sensor 15 compared to the predetermined target value leads to an adjustment of the speed of the dampening roller 5 '(regulation of the dipping roller speed nT) . According to the invention In the control process, proceed so that the controller - depending on the state - works with different amplification factors. Furthermore - as can also be seen from FIG. 1 - an influence depending on the printing speed is provided; This is done, for example, by supplying the tachometer voltage of the tachometer generator 9 via the line 21 to the measurement data acquisition unit 14. The details of this are explained below. The measurement request is sent when a certain machine position (counted in degrees or pulses) is reached, so that the measuring head measures at the designated place on the printing plate. The zero pulse is only used to reset a counter, the pulse count to determine the current machine position.

2 shows a block diagram of the control loop described above. The setpoint value of the dampening solution quantity (or dampening solution layer thickness) Sw soll is fed to a summing point 23. At the same time, the actual value of the dampening solution quantity (or dampening solution layer thickness) Sw is also fed to the summing point 23 with a negative sign. The difference (control deviation Xd) Sw soll - Sw ist resulting from these two values is fed to a controller 24. The controller output is connected to the dampening unit 26 of the offset printing press 1 via a summing point 25. Depending on the output variable of the controller 24, the immersion roller speed nT of the dampening unit 26 is set. The block shown next to the dampening unit 26 represents the printing plate 27, which is located on the plate cylinder 4 of the offset printing press 1. This is followed by a further summing point 28, the output variable of which embodies the real dampening solution layer thickness on the printing plate. The arrow 29 pointing to the summing point 28 indicates disturbance variables influencing the dampening solution layer thickness. These disturbances can, for example caused by air movement, color tracking or temperature changes.

The dampening solution layer thickness on the printing plate 27 is detected by the sensor 15, which feeds its data to the evaluation circuit 11, which is represented, among other things, in FIG. 2 by block 30 (measurement value processing, averaging). The output value of the measured value processing is the actual value Sw ist which is sent to the summing point 23.

The control deviation Xd is fed to a block 32, the circuit of which influences the controller 24 by influencing the gain factor KR. According to the invention it is provided that in the case of a larger setpoint Sw soll compared to the actual value Sw ist a greater amplification factor is used than in the case of control deviations in which the setpoint Sw soll is smaller than the actual value Sw ist. The higher amplification factor is preferably approximately 1.5 times as large as the lower amplification factor (basic amplification factor). The controller 24 is preferably designed as a PI controller.

2 also shows a compensation circuit 33. For example, this can carry out a speed compensation, that is to say an influence on the printing process parameters, caused by changing the printing speed. However, variables other than disturbance variables can also be taken into account in terms of compensation. All the disturbance variables whose effect is known are possible, so that they can be measured effectively by means of measurement, calculation or extraction from tables or characteristic curves. Returning to the example according to FIG. 2, the compensation circuit 33 is supplied with a possible change in speed of the printing process. This is called Delta UM. Via a dampening system compensation characteristic the quantity Uk is determined, which represents a voltage for speed compensation. This voltage is supplied to the summing point 25. Such an arrangement means that changes in the printing speed do not have to be corrected by the control loop, rather this speed compensation is superimposed on the control loop as a control. This combination of regulation by the compensation device (compensation circuit 33) results in a substantially more constant dampening solution guidance than in the event that one would not intervene in a controlling manner but would compensate for such disturbance variables in a regulating manner. According to the invention, a deviation from the target value is corrected considerably faster. Compensation devices known from the prior art, which operate as a pure control, have the disadvantage that additional disturbance variables are not compensated for by the lack of regulation. This does not apply to the arrangement according to the invention due to the superimposition with the control.

According to the invention, a different control characteristic for control deviations greater or less than 0 is provided by using different gain factors KR according to the above explanations. Such a regulation can preferably be superimposed on the described compensation method for disturbance variables with a known effect as a controller. The larger gain factor is preferably chosen to be 1.5 times as large as the basic gain factor.

With regard to the design of the sensor 15, the following should be noted as an example:
This has an illuminating device which throws light directed onto the surface of the printing plate via illuminating optics. The light reflected from the printing plate is detected via a line of photodiodes. By using With the photodiode line, it is possible to measure the entire scattered light curve over the length of the line. A characteristic variable Sw corresponding to the amount of dampening solution or dampening solution layer thickness is calculated in the downstream evaluation circuit 11 from the radiation intensities determined on the individual diodes of the diode row.

FIG. 3 shows a diagram on the abscissa the time t and on the ordinate the dampening solution layer thickness Sw is shown. The lubrication limit is shown in dash-dotted lines. As desired according to the invention, the setpoint for the dampening solution layer thickness Sw soll extends above this lubrication limit. The actual value of the dampening solution layer thickness Sw ist fluctuates in time over the nominal value Sw soll, whereby it can be seen that, for example, control deviations greater and less than 0 occur due to disturbance variables, that is, where the nominal value is greater than the actual value, If there is a control deviation greater than 0, which leads to the application of an increased gain factor and where the setpoint is less than the actual value, the control deviation takes on a value less than 0, with an amplification factor being used that is smaller than in the case described above . These different states are identified in FIG. 3. It can also be seen from FIG. 3 that the critical range is formed between the target value Sw soll and the lubrication limit, that is to say that regulator vibrations located in this section run the risk of spoiling the printed product by lubrication. The so-called uncritical range lies above the setpoint Sw soll. If the actual value is in this zone, there is more water than required on the printing plate, but this condition is not so critical for a deterioration of the printed product. In order to leave the critical area as quickly as possible, according to the invention there is an opposite in the critical zone the gain factor of the controller is increased. If the uncritical area is approached, it is permissible for the process control to use a lower amplification factor, so that leaving the uncritical area takes place more slowly, but - as already described - this does not have any damaging effects on the pressure.

4 shows the course over time of setpoint Sw target and actual value Sw actual in the case of a feedforward control. At time t1, to carry out the experiment, for example, an air flow generated by a fan is directed onto the surface of the pressure plate, that is to say the evaporation of the dampening solution is accelerated. As a result, the water film thickness decreases with the risk of driving into the lubrication area. Due to the increased amplification factor, however, this tendency is counteracted quickly in accordance with the invention, so that the actual value Sw act moves back to the target value Sw desired within a very short time and then runs approximately parallel to it in the further course. The disturbance variable is switched off at time t2; that is, in the experiment described: the air flow is interrupted. Accordingly, the layer thickness of the dampening solution film on the printing plate will initially increase, so that the actual value Sw actual exceeds the target value Sw target. The controller accordingly operates in the non-critical range (cf. FIG. 3), only the basic gain factor KR being used. With a corresponding time constant, the actual value Sw act again moves towards the setpoint Sw target in the further course.

5 shows the temporal actual and setpoint curve for the representation of the control behavior of the control. A sudden change in the setpoint is specified here. When the setpoint Sw jump to the larger value SW soll2 follows the controller very quickly, since it operates with the larger gain factor. However, the area in which the setpoint Sw soll2 jumps back to the setpoint Sw soll1 is critical. Here, for a short time, there is the case that the target value is smaller than the actual value, that is to say the operation is carried out with a gain factor KR which corresponds to the basic gain factor. Accordingly, the processes take place correspondingly slower, although it is nevertheless possible due to the choice of the size of the basic gain factor that the unavoidable overshoot (which is marked in FIG. 5 as a dashed area) does not go so far that the lubrication limit is exceeded. In practice, the risk of lubrication can be reduced even further by not allowing such setpoint changes to become effective immediately, but only gradually. The suitable means for this are known to the control engineer.

6 to 8 correspond to an operating mode of the control according to the invention using a target / actual value tolerance band. This means that the controller works with a kind of hysteresis, because its control voltage only changes when the setpoint-actual value difference lies outside a predetermined tolerance band. This means that the controller only reacts with manipulated variable changes when the specified tolerance of the control deviation is exceeded. In particular, stochastic fluctuations in the measurement signal are disregarded, which leads to a homogenization of the overall process.

7, the procedure is such that a hysteresis element 34 is connected between the controller 24 and the summing point 23. The control deviation Xd is supplied to the hysteresis element 34 as input value and a modified control variable Xd * leaves the arrangement as the output voltage. The The functional relationship between Xd and Xd * can be seen in the diagram in FIG. 7. Starting from the coordinate origin, the modified control deviation Xd * remains at 0 when the positive value Xd increases. Only when the threshold S1 is exceeded does the modified control deviation Xd * jump to a specific value and then increase linearly from there with Xd. If the control deviation is reduced from there, the modified control deviation Xd * linearly decreases to 0. If the control deviation Xd exceeds the value 0, that is, if it becomes negative, there is a corresponding mirror image behavior. The value -Xd * increases suddenly only when a threshold value S2 is exceeded.

Overall, this configuration results in the effect that not even slight deviations between the setpoint and actual values lead to a controller intervention, but that the latter only takes place if the corresponding threshold values S1 or S2 are exceeded. This has a positive effect on the overall system.

6 and 8 correspond to FIGS. 4 and 5, but the controller operates with the hysteresis behavior described above. Because of this hysteresis operation, it can be seen in the comparison of the respective figures that fundamentally larger control deviations occur, but the process is nevertheless carried out with sufficient speed and stability, with the advantage that the controller intervenes less frequently and consequently the adjusting devices are less in use Be claimed.

FIG. 9 shows the previously described effect of the actual and setpoint values due to the compensation control superimposed on the controller structure. For example, at time t1 the printing speed increases from 4,000 sheets per hour to 6,000 sheets, while at time t2 the printing capacity is reduced from 6,000 prints per hour to 4,000 prints per hour. As described above, the speed compensation only results in a relatively short, damped oscillation of the actual value Sw around the setpoint Sw target, the controller needing to intervene only slightly, since the essential parameter changes are already intercepted by the higher-level control. Overall, an optimal process control can be achieved with the method according to the invention.

All new features mentioned in the description and shown in the drawing are essential to the invention, even if they are not expressly claimed in the claims.

Claims (7)

1. Method for regulating the quantity of dampening agent on a printing plate (27) in an offset printing machine (1), the quantity of dampening agent on the printing plate being recorded and being regulated to a predetermined desired value by the adjustment of dampening-agent metering elements, the desired value (S_{W Des}) being selected so that print-free locations of the printing plate begin to take up ink ("desired value S_{W Des} near the lubrication limit"), characterized in that, in the event of control deviations (Xd) which lie outside a predetermined tolerance band around the desired value (S_{W Des}) and in which the desired value (S_{W Des}) is higher than the actual value (S_{W Act}), a controller (24) is operated with a higher amplification factor (KR) than in the event of control deviations (Xd) which lie outside a predetermined tolerance band around the desired value (S_{W Des}) and in which the desired value (S_{W Des}) is lower than the actual value (S_{W Act}).
2. Method according to Claim 1, characterized in that the higher amplification factor (KR) is selected 1.2 to 2, but preferably 1.5 times as high as the basic amplification factor (KR_basic).
3. Method according to Claim 1 or 2, characterized in that the dependence of the quantity of dampening agent on further printing parameters, such as, for example, the printing speed, ink conveyance, temperature or airflow, is predetermined, and in that a corresponding compensation control is superposed on the regulation.
4. Method according to Claim 1 or 3, characterized in that a compensation variable resulting from the measurement of a disturbance variable (Z) is superposed on the output variable of the controller (24).
5. Method according to Claim 4, characterized in that the compensation variable is calculated automatically or is determined automatically from stored tables or characteristics.
6. Method according to Claim 1, characterized in that a PI-controller is used as a controller (24).
7. Method according to Claim 1, characterized in that the actual value (S_{W Act}) representing the quantity of dampening agent is recorded optically by light-reflection measurement.

Images