EP1372965B1 - Register control method - Google Patents

Abstract

A method for register correction in machines for processing webs of material includes providing at least one transport shaft and at least one processing shaft cooperating with the at least one transport shaft, driving the shafts synchronously with one another each by an individual drive mechanism, obeying by at least one shaft a chronological guide shaft function which corresponds to an instantaneous position of a guide shaft, correcting a plurality of register-tracking shafts formed by the transport shafts in accordance with a scan of register marks or the web of material relative to the guide shaft function, effecting only one common scanning for one group of the register-tracking shafts which correspond to one another in term of the register correction, and deriving from the only one common scanning a common correction function that all of the register-tracking shafts of the group obey.

Description

The invention relates to a method for register control on processing machines of material webs according to the preamble of claim 1. Such a machine has transport and processing stations, for example, with driven, corresponding rollers. In this context, for the sake of simplicity, reference will be made here only to the axes thereof.

Such methods are used, for example, in rotary printing presses, paper processing machines or sheetfed presses if an already processed or printed paper web is to be further processed or printed (insetting), so that the subsequent processing steps must take place on a longitudinal position which is precisely aligned with respect to an imprint already present on the paper web, for example. This ensures that, for example, two printing motifs applied in succession coincide in a predetermined relative position on the paper. To achieve this, cooperating transport and processing axes are corrected relative to one another by register control.

In the case of processing machines for material webs, the principle has now largely prevailed to equip the axes of a processing machine or a machine part with mutually synchronized individual drives and thereby replace a mechanical kingshaft (see eg documentation SYNAX 6, 2000, Rexroth Indramat GmbH). This is followed by the respective axes (through the synchronization of the associated drives / higher-level controllers) of a higher-order, leading-edge time-axis function and are synchronized by it. Consequences in such a context means that the movement at the corresponding axis is derived directly or via an (electronic) conversion from the master axis function. The leading axis function corresponds to an instantaneous position of, for example, a virtual, ie electronically generated, or real leading axis. It can, for example, reflect the time profile of the instantaneous position, ie the angular position of the leading axis; However, it can also include the time profile of the rotational speed or other parameters corresponding to the instantaneous position of the master axis. In particular, it is an electronic, temporal setpoint sequence.

In addition, a plurality of register-following axes are corrected relative to the leading axis function in accordance with a scan of register marks of the material webs. They are corrected for their current position, their current speed of rotation or corresponding parameters. The amount of correction is given by the sampling of register marks. The register marks can - as customary in the art - for example, be printed and scanned optically.

It is known to regulate each axis to be corrected with its own register controller. This results in the need to individually parameterize each axis and its controller and to optimize it with regard to the correction movements and the synchronicity with the other axes. The effort during commissioning is therefore high; the provision of a correspondingly large number of individual register regulators is additionally associated with high equipment costs and leads to high costs. Nevertheless, the synchronicity of the axes to be corrected is not always satisfactory, since naturally mechanical and electronic deviations between the individual register regulators can occur. This can lead to web tension fluctuations.

Furthermore, it is known to allow a register controller to act on several axes simultaneously. This is done on each axis - ie the corresponding drive / to the appropriate control of the corresponding element, such as the roller - a transmitted individual correction signal and there converted into the corresponding individual correction movement. The effort for this increases sharply with the number of axes to be controlled, so that this method for a large number of axes to be controlled - as is common practice - is not applicable or only to a limited extent. Also in this case can occur synchrony problems due to excessive cycle times in the transmission of the correction signal.

It is an object of the present invention to provide a method of the type mentioned above, which ensures - especially with a large number of axes to be controlled - a higher degree of synchrony of the axes to be corrected and at the same time allows easy start-up with comparatively low outlay on equipment.

This object is achieved by the features of claim 1.

The invention has the advantage that with just one register controller any number of axes can be controlled synchronously. As a result, the expenditure on equipment is reduced and commissioning much easier. A method according to the invention for register control automatically leads, while maintaining these advantages, to a maximum of synchronicity of the correction movements.

These advantages are achieved by deriving from a common scan a common, in particular temporal correction function for a plurality of axes to be corrected. This correction function is followed by all the axes of a group of register-following axes which correspond to the register correction. Consequently, the entire information of all correction movements is included in the uniform correction function with respect to all the axes of the group. A group of register-following axes which correspond to each other only comprises axes which are to be controlled by a common register controller, for which the same register correction and the same sampling are decisive. These are axes on a coherent / uninterrupted web. In rotary printing presses, some or all axes of a processing tower, For example, be printing tower or axes of different processing towers, between which the web is not cut / not interrupted.

By using a uniform correction function, which is calculated in accordance with only one register controller and is uniform for all axes of the group, a multiplicity of register regulators can be dispensed with in comparison with the prior art. Nevertheless, a high degree of synchronicity is achieved, so that the invention has a double benefit.

Even using only one register controller for a group of axes - which can also include a plurality of axes - automatically ensures a high degree of synchrony, since only one correction function - and thus only a correction signal - can be used for all axes of the group. Therefore, only one signal is to be transmitted to the axes of the group. The once determined correction function can be used for all axes practically simultaneously and uniformly and thus provides by itself a high degree of synchrony of the adjustment movements based on the correction, without that any further precautions would have to be taken.

The invention further makes it possible for the first time to adjust a plurality of axes in accordance with only one register regulator while maintaining a maximum of synchronicity. The adjustment movements are to be determined for a plurality of axes only with a register controller and are then used for all these axes and can be transmitted to these axes practically simultaneously.

Preferred embodiments of the present invention are described in the subclaims.

The correction movement can be provided directly and thus quickly at the corresponding axes if the correction function essentially contains only the corrections to the master axis function and as such for Register correction is used. Due to the virtually immediate use of the correction signal this is to be determined with relatively low computing capacity, in particular with low computational complexity.

If the correction function is linked with the leading axis function to form an additional, time-series sequential control axis function, this linkage can be performed centrally and uniformly within the framework of a register control and transmitted to the corresponding axes as register sequence leading axis function; The individual axes can then practically and directly follow such a sequence of register control axes without having to carry out decentralized derivations - which are associated with an increased amount of computation - on the individual axes. The register trace master function then contains virtually all data for each axis in a uniform signal. Since in any case the technical provisions for the provision and transmission of a master axis function must generally be made, this is a natural solution for the method according to the invention, which can be readily integrated into the existing drive structures / controller structures. There are then two Leitachsfunktionen - namely the unmodified and the register sequence Leitachsfunktion - for which usually the computing and transmission capacities are already present.

Depending on the type of expected or registered deviations (ie the amount by which the material web "leaves the register", ie the amount of deviation from the specification by the register marks) is the type of deviation Select correction function. The invention is already suitable for a multiplicity of applications in which the deviation is practically constant when the correction function comprises a position offset determined by the scanning of the register marks relative to the instantaneous position of the leading axis. The correction function then consists essentially of a constant or in accordance with the sampling of the register marks changing position offset. A register sequence leading axis function in this case has a correspondingly either constant or - preferably temporally comparatively slowly changing - deviation from the master axis.

Additionally or alternatively, it can be provided that the correction function comprises a function which is determined by the scanning of the register marks and corresponds to a gear ratio with respect to the leading axis. This corresponds in the case of a correction function that includes only the correction movements, a pure gear ratio, which may also be constant or changes in time in accordance with the sampling. In the case of claim 3, this corresponds to a register sequence leading-axis function, which is derived with a gear ratio of the (higher) Leitachsfunktion.

As a result of the aforementioned embodiments, a simplification, namely a possible restriction to only two methods of the derivatives of the correction function / the chronological sequence of register sequence leading axes, is provided by which the invention, however, is suitable for practically all occurring applications.

Any remaining deviations between the axes of a group are minimized by the fact that the scanning takes place practically in a central area-in relation to the longitudinal direction of the material web-of the register-following axes. The possibly remaining deviations generally have a continuous course, ie seen in the longitudinal direction of the material web. H. they are practically equal to zero at the sampling point or at the sensor location since the register control is related to this sensor. Measured in the longitudinal direction they are usually strictly monotonous and change their sign at the sensor location. In this case, the said sampling point is the location where the sampling leads practically to the smallest possible maximum amount of the individual deviations on the axes of the group and at the same time to the smallest sum of the amounts of deviations of the individual axes from the corresponding setpoint.

In particular, in insetting applications on rotary printing machines, it is proposed that a group is provided which only comprises transport axes. Then the correction function / the register sequence leading axis function is decisive for all transport axes of the group, so that according to the invention it has a large size Synchronicity be corrected. This leads to an extremely precise, common correction of the transport axes relative to the machining axes.

In order to achieve an increased accuracy of the processing in a register control according to the invention, it is proposed that in addition a predetermined, with respect to the computational effort / computing capacity simple correction of processing axes takes place, which correspond to the register correction of the group of transport axes. The o.a. Definition of corresponding axes accordingly. This measure introduces an additional degree of freedom that is easy to implement into the control system. A simple correction in this sense should be sufficient for most cases, to compensate for any remaining deviations. Particularly in the case of register control, the requirements for the coincidence of the processing with the positions specified by the register marks are very high. The mentioned embodiment makes it possible in a simple way to improve this coincidence even further. In this case deviations can be eliminated, which are virtually the same for a variety of processing axes; but it can also be eliminated deviations that may be different for different processing axes. The latter relates, in particular, to a possibly remaining deviation which may arise due to the distance of a machining axis from the sensor location (corresponding to the above stated embodiment).

A simple and effective correction in the above sense can be achieved by determining the longitudinal error per unit length of the material web and for each machining axis to be corrected their longitudinal distance to the sampling and the correction of the respective processing axis is essentially formed by the product of longitudinal error and longitudinal distance. Since the material web is usually divided into individual products after processing, it is proposed that the material web is subdivided into individual products of predetermined product length, whereby the longitudinal error per product length is determined and the correction of the relevant processing axis essentially by the product of longitudinal defects per product length and quotient: longitudinal distance / product length is formed. This procedure is in terms of required computing power / computing capacity simplified. It naturally also leads to a better coincidence (see above), as the deviation is related to the product length. In any case, the product length is the relevant quantity for the machining axes, so that the calculation and implementation of the corresponding correction can be carried out simply and precisely.

The above-mentioned additional correction can be realized in that a plurality of machining axes to be corrected form a group according to claim 1. This reduces the number of required correction calculations - usually by the number of axes that are grouped or grouped by the number of such groups. This summary into a group with corresponding corrections creates a central structure with all the advantages of the invention; this central structure may be subordinate to other groups. Then machining axes can be divided into several groups. It is essential that the deviation within a group remains relatively small.

The possible capacities of the method are fully utilized if at least one register-independent axis is provided, which follows the time-leading axis function. Then two or more master axis functions are provided, which are integrated into the system and used by respective axes, i. This means that the associated axes follow the respective master axis function (or register sequence master axis function).

• Fig. 1a eine schematische Darstellung einer Bearbeitungsmaschine mit einem Registerregler und einem Antriebssystem zur Durchführung des erfindungsgemäßen Verfahrens,
• Fig. 1b eine Ausschnittvergrößerung aus Fig. 1a mit den Einzelheiten des Registerreglers,
• Fig. 2 ein Diagramm einer Leitachsfunktion, einer Registerfolge-Leitachsfunktion und einer Korrekturfunktion.

The invention will be explained in more detail with reference to exemplary embodiments illustrated in the figures. Show it:

• 1a is a schematic representation of a processing machine with a register controller and a drive system for carrying out the method according to the invention,
• 1b shows a detail enlargement from FIG. 1a with the details of the register regulator,
• 2 shows a diagram of a leading axis function, a register sequence leading axis function and a correction function.

Unless otherwise stated below, all reference numbers always refer to all figures.

1 shows - schematically simplifying - a processing machine 1 for processing a material web 2. It is a rotary printing machine, consisting of several driven rollers 33, each with associated pressure rollers 34th

The processing machine 1 has an input transport station, which is essentially formed by the transport axis 3 with its two rollers 33. At the other end (seen in the longitudinal direction 23) is an output transport axis 4, consisting of also two cooperating rollers 33. Between the transport axes 3,4 are four processing stations 5,6,7,8, hereinafter for the sake of simplicity, only as Machining axes designated 5,6,7,8.

The term of the axis is used here for the corresponding station with the associated rollers 33, their motors M and the associated drive 9. The term of the axis is in particular to be distinguished from the physical axis of rotation 35, 36 of the respective rollers 33,34.

The transport axes 3, 4 and the processing axes 5, 6, 7, 8 cooperating therewith are each driven by an associated individual drive 9. This replaces a continuous, mechanical shaft (vertical shaft). For this purpose, it is necessary that the individual drives 9 are synchronized with each other. For this purpose, the individual drives receive 9 Leitachssignaldaten (see below) via a data bus 28. For synchronization, the axes follow 5,6,7,8 a time Leitachsfunktion 12, which is fed into the data bus 28 and transmitted via this to the individual drives 9. Deviations are compensated by register control by first register marks 14 (here symbolized by crosses at the corresponding longitudinal positions) are scanned by an (optical) sensor 29. From the sampling, a correction is then calculated with respect to the leading axis function 12 in the register controller 30, which initially acts only on the register-following axes 3, 4. First of all, no register correction of the other processing axes 5, 6, 7, 8 is provided (but this can additionally take place, see below), so that the register correction corresponds to a relative correction between the transport axes 3, 4 and the processing axes 5, 6, 7, 8.

The (unaffected by the registration correction) leading axis L is symbolized here only by a circle. It is irrelevant for the invention, whether this is a virtual master axis, whose instantaneous position is generated in a purely electronic way, or a so-called real master axis, whose instantaneous position by scanning a physically physically present mechanical shaft or by a feedback of a Drive is given.

According to the invention, a group 15 is formed from the register-following transport axes 3, 4, which correspond to the register correction, as explained in greater detail above. For this group 15 of register following axes 3,4 only a common scan takes place. This takes place at only one scanning point 44 by the sensor 29, which may be, for example, a photodiode or a CCD camera with a downstream evaluation to detect the register marks.

From the common scan, a correction function 16, which is likewise common with respect to the group 15 of the register-following axes 3, 4, is derived. This can be formed from the fact that from a target-actual comparison in accordance with the sampling of the register marks, the local deviation whose derivative (that is the speed) or corresponding functions are formed. The correction function is formed in the embodiment shown by the comparison of the scanning result with the desired value S and / or the Leitachsfunktion 12, which is to - fed together with the scanning signal from the sensor 29 - in a computing unit 31. The setpoint S contains the information at which relative position with respect to the leading axis function 12 and / or the processing axes 5,6,7,8 on the material web, the register marks to be located at the sampling point 44.

From the control deviation formed in the computing element 31 (see FIG. 1b) (corresponding to the correction function 16), a register sequence leading axis function 17 is derived. This is shown schematically for clarity with greatly exaggerated by the slope of the leading axis function 12 slope. The leading axis function 12 is fed to the register controller 30. The combination of the correction function 16 with the master axis function 12 also takes place in the register controller 30 of the invention. Since the communication line is a data bus 28, the (unchanged) master axis function 12 as well as the register sequence formed from the correction function 16 can be used on all individual drives 9 -Leitachsfunktion 17 are provided, wherein the respective drive 9 is driven or addressed / addressed only in accordance with a variable setting of the predetermined corresponding Leitachsfunktion 12 / register sequence leading axis function 17. Thus, the freedom of choice is ensured that virtually every axis 3,4,5,6,7,8 in accordance with the (pre-) setting any of the proposed Leitachsfünktionen 12,17 or the correction function 16 after processing / adaptation - for example, in the relevant Drive controller 10 can follow.

The respective master axis function 12,17 or the correction function 16 is then processed in the drive controller 10 and the respective motor M according to its specification synchronized / corrected driven by the power electronics 11.

• Generell ist zur Synchronisation der vorhandenen Achsen eine Leitachsfunktion 12 vorgesehen, welche über den Datenbus 28 an jeden der Einzelantriebe 9 einzeln übertragen/adressiert werden kann und den jeweiligen Antrieb 9 übergeordnet synchronisiert. Auf der linken Seite der Ausschnittvergrößerung ist der Registerregler 30 im Detail gezeigt. Dort wird aus dem Sollwert S und dem Abtastsignal A die Korrekturfunktion 16 gebildet und nach Maßgabe der Korrektur mit der übergeordneten Leitachsfunktion 12 zu einer Registerfolge-Leitachsfunktion 17 verarbeitet. Aus der Detailansicht ist zu entnehmen, dass im Einzelnen zunächst aus Sollwert S, Leitachsfunktion 12 bzw. Leitachse L und einem Abtastsignal A in dem Rechenglied 31 eine Funktion f (A,S,L) berechnet wird. Dies könnte die Korrekturfunktion 16 sein. Im vorliegenden Fall ist es eine (vorzugsweise momentane/aktualisierte) Vorgabe, nach Maßgabe derer über die Parameterleitung 42 aus der Leitachsfunktion 12 die Registerfolge-Leitachsfunktion 17 abgeleitet wird. Wie in der Detailansicht gezeigt, sind/ist für die Ableitung der Registerfolge-Leitachsfunktion 17 lediglich ein Offset-Addierer 20 und/oder ein Getriebeglied 21 vorgesehen, die von dem Rechenglied 31 über die Parameterleitungen 42 angesprochen werden/wird. Dies bedeutet, dass nach Maßgabe der Abtastung entweder ein reiner Positionsoffset 19 oder eine Getriebeableitung oder beides zur Ableitung der Registerfolge-Leitachsfunktion 17 verwendet wird. Für die Bildung der Korrekturfunktion 16 / der registerfolgenden Leitachsfunktion 17 wird aus dem Abtastergebnis, dem Sollwert (diese kann auch eine zeitliche Sollwertfunktion sein) und der Leitachsfunktion 12 das Maß des Positionsoffsets 19 und/oder die Getriebeübersetzung für das Getriebeglied 21 berechnet und vorzugsweise im Rahmen der beteiligten Taktrate und der erwarteten Zeitkonstante für das Reglersystem aktualisiert. Über die Parameterleitung 42 werden somit die zur Bildung dieser Funktion erforderlichen Parameter an die Glieder 20,21 geleitet.

The operation of a register control according to the invention is shown schematically in the detail enlargement in FIG. 1b:

• In general, a master axis function 12 is provided for synchronizing the existing axes, which can be individually transmitted / addressed via the data bus 28 to each of the individual drives 9 and synchronizes the respective drive 9 in a higher order. On the left side of the detail enlargement register register 30 is shown in detail. There, from the setpoint S and the sampling signal A is the Correction function 16 is formed and processed in accordance with the correction with the parent control axis function 12 to a register sequence leading axis function 17. From the detailed view, it can be seen that, in detail, a function f (A, S, L) is first calculated from setpoint value S, leading axis function 12 or master axis L and a sampling signal A in the computing element 31. This could be the correction function 16. In the present case, it is a (preferably instantaneous / updated) specification according to which the register sequence leading axis function 17 is derived via the parameter line 42 from the leading axis function 12. As shown in the detail view, only one offset adder 20 and / or one gear element 21 is provided for deriving the register sequence leading axis function 17, which is / are addressed by the computing element 31 via the parameter lines 42. This means that either a pure position offset 19 or a transmission derivative or both for the derivation of the register sequence leading axis function 17 is used in accordance with the sampling. For the formation of the correction function 16 / the register-following leading axis function 17, the degree of the position offset 19 and / or the gear ratio for the gear element 21 is calculated from the sampling result, the desired value (this can also be a time reference function) and the leading axis function 12, and preferably in the frame the clock rate involved and the expected time constant for the controller system is updated. Via the parameter line 42, the parameters required for the formation of this function are thus passed to the members 20, 21.

If no control deviation is present or no control is desired, all the parameters can also be dimensioned or specified such that the elements 20 and / or 21 are indifferent and the register sequence leading axis function 17 is substantially equal to the leading axis function 12. Both existing Leitachsfunktionen 12,17 are passed on the respective Leitachsgeneratoren 40,41 (eg software in the calculator) to the data bus 28 with the appropriate addressing. The addressing is not discussed here; However, it is done selectively for each individual drive 9 in accordance with its parameters, namely the distance of the associated axes 3,4 of the sampling point 44, etc. This will be discussed in more detail below.

Additionally or alternatively, a correction function 16 may be provided, which contains essentially only the corrections to the leading axis function 12 and the - for the axes 3,4 of the group 15 - directly as applied to the global synchronization clock of the leading axis function 12 correction - namely the respective drive 9 - acts.

In addition to the transport axes 3, 4, processing axes 5, 8 can also be combined to form a group 43. These have their own, e.g. additional, register-following leading axis function. It could also be combined all processing axes 5,6,7,8 to a group. Here are the most distant from the scanning 44 remote processing axes 5.8 combined to form a group 43, as this is a possibly (residual) deviation above said particularly large. Concerning. the registering axes 3,4; 5,8 of the groups 15; 43, the scanning is carried out practically in a central region 22 relative to the longitudinal direction 23 of the material web 2, i. practically in the middle between the mentioned axes. As a result, any remaining (register) deviations of the register-following axes with one another are minimized, as explained above.

On the processing axes 5.8 of the group 43 has a simple in terms of computational effort correction. This is formed by dividing the material web into products 25 of a product length 26, which in the present case coincides with the spacing of the register marks 14 (not necessarily the case). By means of register control, the longitudinal error 27 (shown exaggerated here) per product length 26 is determined. For each machining axis 5, 8 to be corrected, its longitudinal distance 45 to the scanning point 44 is determined and the correction of the machining axes 5 is formed by the product of longitudinal error and quotient: longitudinal distance 45 / product length 26.

Finally, FIG. 2 shows a diagram of different master axis functions 12, 17, 37 and of a correction function 16. The current position is plotted in angular degrees over time t. The register sequence leading axis function 17 and the Register sequence leading axis function 37 are examples of correction leading axis functions derived from the unchanged leading axis function 12. The register sequence leading axis function 37 consists of only one position offset 19 with respect to the leading axis function 12. The register sequence leading axis function 17 has a transmission derivation from the leading axis function 12; As a result, the register sequence leading axis function 17 has a different slope than the leading axis function 12 and therefore also a different period duration 39 with respect to the period duration 38 of the leading axis function 12. Due to the greater slope of the register sequence leading axis function 17, the associated period duration 39 is shorter.

In addition, a correction function 16 is shown in FIG. This merely reproduces the corrections with respect to the leading axis function 12, by which the register-following axes 3, 4, 5, 8 are possibly corrected. Instead of the instantaneous position a in angular degrees, an angular velocity could, for example, also be provided as the encoder signal for the corresponding leading axis functions / correction functions.

LIST OF REFERENCE NUMBERS

1

processing machine

2

web

3

Transportachse

4

Transportachse

5

machining axis

6

machining axis

7

machining axis

8th

machining axis

9

individual drive

10

drive controller

11

power electronics

12

guide shaft

13

Current position of the leading axis

14

registration mark

15

Group of register-following axes

16

correction function

17

Register sequence-guide shaft

18

-free-

19

position offset

20

Offset adder

21

transmission member

22

Central area

23

Longitudinal direction of the material web

24

-free-

25

single product

26

product length

27

Longitudinal error per product length

28

bus

29

sensor

30

register controller

31

computing element

32

-free-

33

driven roller

34

pressure roller

35

Rotary axis of the driven roller

36

Rotary axis of the pressure roller

37

Register sequence leading axis function with only position offset

38

Period of the leading axis function

39

Period of the register sequence leading axis function

40

master axis

41

master axis

42

parameter line

43

group

44

scanning point

45

Distance of the processing point from the sampling point

L

master axis

S

Setpoint generator

Claims (12)

1. Method for register correction of processing machines (1) for material webs (2), in particular rotary presses, paper processing machines and sheet-fed presses having at least one transport axis (3, 4) and at least one processing axis (5, 6, 7, 8) interacting therewith, which are driven by a respectively associated individual drive (9) so as to be synchronized with one another, and of which at least one axis (6, 7) follows a timed master axis function (12) which corresponds to an instantaneous position (13) of a master axis (L), and a plurality of axes (3, 4) following the register are corrected with respect to the master axis function (12) in accordance with scanning of register marks (14) of the material web (2), characterized in that for one group (15) of register-following axes (3, 4) which correspond with regard to the register correction, only common scanning is carried out, from which a common correction function (16) is derived which is followed by all the axes (3, 4) of the group (15).
2. Method according to Claim 1, characterized in that the correction function (16) contains substantially only the corrections with respect to the master axis function (12) and is used as such for the register correction.
3. Method according to Claim 1, characterized in that the correction function (16) is combined with the master axis function (12) to form an additional, timed register-following master axis function (17).
4. Method according to one of Claims 1 to 3, characterized in that the correction function (16) comprises a position offset (19) with respect to the instantaneous position (13) of the master axis (L), determined by the scanning of the register marks (14).
5. Method according to one of Claims 1 to 4, characterized in that the correction function (16) comprises a function corresponding to a transmission ratio with respect to the master axis (L), determined by the scanning of the register marks (14).
6. Method according to one of Claims 1 to 5, characterized in that the scanning is carried out in practice in a central region (22) - relative to the longitudinal direction (23) of the material web (2) - of the register-following axes (3, 4).
7. Method according to one of Claims 1 to 6, in particular in the case of insetting applications on rotary presses, paper processing machines or sheet-fed presses, characterized in that one group (15) comprises only transport axes (3, 4).
8. Method according to Claim 7, characterized in that in addition a predetermined correction of processing axes (5, 8) which correspond to the group (15) of transport axes (3, 4) with regard to the register correction and is simple with regard to the computational effort is carried out in accordance with the scanning.
9. Method according to Claim 8, characterized in that the longitudinal error (27) per unit length (26) of the material web (2) and, for each processing axis (5, 8) to be corrected, its longitudinal spacing (45) from the scanning point (44) are determined, and the correction of the relevant processing axis (5, 8) is formed substantially by the product of longitudinal error (27) and longitudinal spacing (45).
10. Method according to Claim 9, characterized in that the material web (2) is subdivided into individual products (25) of predetermined product length (26), the longitudinal error (27) per product length (26) being determined and the correction of the relevant processing axis (5, 8) being formed substantially by the product of longitudinal error (27) per product length (26) and the ratio: longitudinal spacing (45) / product length (26).
11. Method according to Claim 8 or 9, characterized in that a plurality of processing axes (5, 8) to be corrected form a group (43) according to Claim 1.
12. Method according to one of Claims 1 to 10, characterized in that at least one register-independent axis (6, 7) is provided which follows the timed master axis function (12).

Images