WO2023110436A1

WO2023110436A1 - Calibration method for a belt drive, belt drive, and sheet-metal processing plant

Info

Publication number: WO2023110436A1
Application number: PCT/EP2022/084076
Authority: WO
Inventors: Martin Schober; Bernd Renz
Original assignee: TRUMPF Werkzeugmaschinen SE + Co. KG
Priority date: 2021-12-14
Filing date: 2022-12-01
Publication date: 2023-06-22
Also published as: DE102021133034A1

Abstract

The invention relates to a method for calibrating a belt drive (10) having a belt (12) and a drive pulley (14), wherein a value for an effective circumference of the drive pulley (14) is stored in a control device (34), the method comprising the steps of: B) rotating the drive pulley (14) between a first rotational position and a second rotational position through a rotational angle being detected by a rotary encoder; C) detecting an actual longitudinal displacement of a point on the belt (12) during rotation of the drive pulley (14) between the first rotational position and the second rotational position by means of a length-measurement system (30); D) correcting the stored value for the effective circumference of the drive pulley (14) on the basis of the actual longitudinal displacement and the rotational angle that was covered.

Description

Calibration method for a belt drive, belt drive and

sheet metal processing plant

Background of the Invention

The invention relates to a method for calibrating a belt drive with a belt and a drive pulley. The invention also relates to a belt drive having a belt, a drive pulley with a rotary drive, a tensioning pulley, a rotary encoder for detecting a rotation of the drive pulley and a length measuring system for detecting a longitudinal displacement of a point on the belt. Furthermore, the invention relates to a sheet metal processing system with a processing tool and a belt drive. Belt drives are used in different areas to transmit movements or to move components of a system. It is basically known to attach automation components to a belt and to move them to their respective place of use by means of the belt. In particular, such an automation component can be used for handling workpieces or residual parts in sheet metal processing. It is necessary here that the automation components can be positioned precisely.

Position control can be done using a rotary encoder on a driven pulley. In order to be able to precisely determine the position of the component moved by the belt, precise knowledge of the relationship between the angle of rotation of the drive pulley and the longitudinal movement of the belt is required. This is made more difficult by the fact that this relationship depends not only on manufacturing tolerances but also on the pretensioning of the belt. An effective circumference of the belt pulley (also referred to as “spindle pitch” in this context) can be used to describe this relationship.

After a positioning process, the position of the component moved with the belt can be determined directly. If this position is not sufficiently exact due to inaccuracies in the underlying relationship between the angle of rotation and the longitudinal movement, repositioning may be necessary. The resulting sticking and sliding effects, such as stick-slip, can make repositioning difficult.

Until now, the position control of belt drives has been calibrated manually, which requires maintenance personnel to access the belt drive. The often poor accessibility makes this even more difficult. In addition, known methods for determining the pretension via sound measurement on the belt are very complex.

A method for monitoring a belt drive is known from WO 2016/177883 A1. The belt drive includes a rotatably mounted disc and a Belt that is deflected on the pulley, the belt and the pulley each being equipped with a marking and the belt and pulley being positioned relative to one another in such a way that when the belt runs around the pulley, the markings on the pulley and belt are directly opposite one another in a release position. The release position is detected by a monitoring device, the monitoring device emitting a signal when the markings are in the release position. From this, the number of signals triggered over a specific period of time is determined. Based on the number of signals triggered, the number of revolutions of the belt, and taking into account relevant influencing factors, the loads to which the belt or pulley has been exposed over the past period of time are determined. A prognosis for the remaining service life of the belt or pulley can be derived from this.

object of the invention

It is an object of the invention to simplify the precise calibration of belt drives.

Description of the invention

This object is achieved according to the invention by a method having the features specified in claim 1 and a belt drive according to claim 14 and a sheet metal working system according to claim 15. The dependent claims and the description specify advantageous embodiments.

According to the invention, a method for calibrating a belt drive is provided. The belt drive has a belt, in particular a toothed belt, and a drive pulley. The drive pulley can be driven via a rotary drive. The belt guided over the drive pulley can be moved as a result. A value for an effective circumference of the drive pulley is stored in a control device. The effective scope of the sheave corresponds to the distance a point of the belt moves in a longitudinal direction in one complete rotation of the sheave. In this context, the effective circumference of the drive pulley is also referred to as the so-called "spindle pitch". The effective circumference can be at least 100 mm and/or at most 500 mm.

The method according to the invention comprises the following steps:

B) rotating the drive pulley between a first rotational position and a second rotational position by a rotational angle with detection by a rotary encoder,

C) detecting an actual longitudinal displacement of a point on the belt when rotating the drive pulley between the first rotational position and the second rotational position using a length measuring system,

D) Correction of the stored value for the effective circumference of the drive pulley based on the actual longitudinal displacement and the swept angle of rotation.

In step B) the drive pulley is rotated. For this purpose, the rotary drive can be controlled accordingly. The angle of rotation covered when rotating the drive pulley between the first rotary position and the second rotary position is detected by a rotary encoder.

The rotation of the drive pulley moves the belt. The actual longitudinal displacement of a point on the belt that occurs is recorded in step C) by a linear length measuring system. Longitudinal displacement is always measured along the belt. The point on the belt typically does not run over a pulley, particularly not over the drive pulley.

In general, the effective scope value stored will not exactly match the actual effective scope value. In other words, the actual longitudinal displacement of the point on the belt may differ from an expected longitudinal displacement based on the rotation angle and the stored value for the effective scope can be determined differ. Reasons for this deviation can be manufacturing tolerances. On the other hand, the (toothed) belt running off the drive pulley or running onto the drive pulley lengthens with increasing tension, so that the effective extent can appear larger with increasing tension.

In step B), the stored value for the effective scope is corrected. This correction is based on the measured actual longitudinal displacement and the corresponding swept angle of rotation. The newly stored value then corresponds to the current conditions on the belt drive. In principle, the control device carries out the correction itself.

The method according to the invention enables the stored value for the effective circumference (the so-called spindle pitch) to be checked and, if necessary, corrected quickly and with little effort. Physical access by maintenance personnel to the belt drive is typically not required. In particular, the method can preferably be carried out by remote maintenance or fully automatically. Due to the reduction of manual activities, the process works very precisely and reproducibly.

In principle, the method can be used for all types of belt drives. The method can be carried out simultaneously with several, in particular all, belt drives of a system with several belt drives. In particular, the method can be carried out parallel to a main time of the system. No set-up times, or at most short set-up times, are required to carry out the method.

A time gradient for the respective belt drive can be determined and documented by recording and tracking the value stored in each case for the effective scope over time. Measures for preventive maintenance and repair can be derived from this. In step D), a difference between an expected longitudinal displacement, which was determined using the angle of rotation and the stored value for the effective extent, and the actual longitudinal displacement recorded in step C) is determined and the stored value for the effective extent is changed accordingly. The accuracy of the method can be increased by taking into account the currently stored value for the effective scope when calculating the difference.

An iterative correction of the respectively stored value of the effective scope is particularly preferred. Steps B) to D) can be carried out again for this purpose. The accuracy of the method can be increased further as a result. With an automated implementation, the effort for this is extremely low.

Also preferred is a method variant in which steps B) and C) are each carried out several times, and in step D) a method of linear regression is used. In this way, the value for the effective scope can be determined or corrected particularly precisely. In addition, the application of linear regression can allow conclusions to be drawn about the friction conditions, in particular sliding friction, in the belt drive.

The angle of rotation by which the drive pulley is rotated in step B) can be predefined. This simplifies the implementation of the method. The angle of rotation is preferably chosen so large that the point on the belt is moved over as large a straight line as possible.

The angle of rotation covered in step B) is preferably an integral multiple of a complete rotation. Out-of-roundness of the drive pulley and geometric effects due to the meshing of a toothed belt on the drive pulley can be compensated in this way.

The rotation of the drive pulley preferably begins before the first rotational position.

Also preferably, the rotation of the drive pulley ends after the second rotary position. In other words, the rotation angle can be measured between rotation positions that are reached after the belt drive has been started and/or before the belt is stopped. This reduces the influence of friction effects and dynamic effects when starting or braking, so that the method works more precisely.

The point on the belt may be defined by a component attached to the belt. The component attached to the belt can be an automation component. In particular, the component can be releasably clamped to the belt. The automation component can, for example, have pins or suckers for handling sheet metal parts. In particular, the component can have a part of the linear length measuring system. The part can be a locator. An existing component or part (the position transmitter) of the length measuring system can thus be used to carry out the method according to the invention. This reduces the effort involved in carrying out the method.

A target value for the effective range is preferably predefined. The target value for the effective amount corresponds to a desired tension of the belt. Maintaining the correct belt tension is important for the belt drive to function properly. The tension of the belt can be changed to bring the effective range closer to the target value. This adjusts the belt tension to the desired value. This exploits the fact that the effective circumference of the drive pulley depends on the tension of the belt. Within the scope of this variant of the method, the tension of the belt can be determined in a particularly simple manner, compared with the desired value and adjusted if necessary. A direct measurement of the voltage is not required for this. In particular, knowledge of the longitudinal stiffness of the belt can suffice.

To change the tension, a distance between the drive pulley and a tension pulley can be changed. This distance is for belt drives typically adjustable, so the effort required to adjust the voltage is low. The change in tension causes a corresponding change in the effective circumference of the drive pulley.

In a step A), a total pretensioning path for a tension-free belt can be calculated on the basis of the target value and a nominal value of the effective extent. The nominal value describes the effective range when the belt is applied without tension. Starting from a state in which the belt rests without tension on the drive pulley and the tensioning pulley, the distance between the drive pulley and the tensioning pulley can be increased by a fraction of the pretensioning path. The fraction is generally a maximum of 100%. The fraction can be 100%. The fraction is preferably at most 90% and/or at least 70%. In particular, the fraction can be 80%. This procedure simplifies the tensioning of a newly fitted belt without the risk of over-tensioning it.

In a step E), a remaining preload path can be calculated using the setpoint and the stored value for the effective circumference. The stored value can be the one after the correction in step C). The distance between the drive pulley and the tensioning pulley can be changed by the remaining preload distance. The remaining bias travel can be positive (voltage must be increased; spacing is increased) or negative (voltage must be decreased; spacing is decreased). In this way, the tension of an already pre-tensioned belt can be corrected, in particular if the belt has already been in operation for some time or after pre-tensioning in step A).

Steps B) to D) are preferably carried out again after the distance has been changed in step E). In this way it can be achieved that the effect of re-tensioning is taken into account in the stored value for the effective scope. If the corrected value of the effective scope deviates from the target value, in particular beyond a predefined extent, step E) can be repeated be performed. The tension of the belt can thus be iteratively approximated to the desired value. The cycle of carrying out steps B) to D) and optionally E) can be repeated several times.

A belt drive also falls within the scope of the present invention. The belt drive shows

- a belt, in particular a toothed belt,

- a drive pulley with a rotary drive,

- a tension washer,

- a rotary encoder for detecting a rotation of the drive pulley, and

- a length measuring system for detecting a longitudinal displacement of a point on the belt. The belt is guided around the drive pulley and the idler pulley. Typically, a distance between the drive pulley and the idler pulley is adjustable. According to the invention, the belt drive also has a control device which is set up to carry out a method according to the invention as described above. The control device simplifies the implementation of the method according to the invention. In particular, the control device can be set up to carry out the method according to the invention automatically. To change the distance between the drive pulley and the clamping pulley, an actuator can be provided, which is controlled accordingly by the control device.

A sheet metal processing system with a processing tool also falls within the scope of the present invention. The processing tool can be a processing head that can be moved relative to a sheet metal support. The sheet metal processing system can in particular be a laser cutting system. The processing tool can be a laser processing head, in particular a laser cutting head. According to the invention, the sheet metal working installation has at least one belt drive according to the invention as described above. The belt drive can be used to move an automation component. Preferably, an automation component attached to the belt includes part of the linear length measurement system. In particular, the automation component can be detachably clamped to the belt. The part can be a locator. An already existing component or an already existing part (the position transmitter) of the length measuring system can thus be made usable for the invention.

Further advantages of the invention result from the description, the claims and the drawing. According to the invention, the features mentioned above and those detailed below can each be used individually or collectively in any desired, expedient combination. The embodiments shown and described are not to be understood as an exhaustive list, but rather have an exemplary character for the description of the invention.

Detailed description of the invention and drawing

The invention is illustrated in the drawing and is described using exemplary embodiments. Show it:

1 shows a belt drive according to the invention when carrying out a calibration method according to the invention, a component clamped to a belt being located with a position transmitter in a starting position, in a schematic side view;

FIG. 2 shows the belt drive from FIG. 1, wherein the component clamped to the belt has been transferred into a first longitudinal position by rotating a drive pulley into a first rotational position, in a schematic side view; 3 shows the belt drive from FIG. 1, in which the component clamped to the belt has been transferred into a second longitudinal position by turning the drive pulley into a second rotary position, in a schematic side view;

FIG. 4 shows the belt drive from FIG. 1, with the component clamped to the belt being in an end position, in a schematic side view;

5 shows a schematic flow chart of a calibration method according to the invention;

6 shows a sheet metal processing system according to the invention with two belt drives according to the invention for moving automation components, in a schematic side view.

FIG. 1 shows a belt drive 10. The belt drive 10 has a belt 12, here a toothed belt. The belt 12 is guided around a drive pulley 14 and an idler pulley 16 . The drive disk 14 can be rotated by means of a rotary drive 18 . A rotary position of the drive pulley 14 is detected by a rotary encoder 20 .

The belt drive 10 has at least one, in this case three, components 22, 24, 26 that can be fixed to the belt 12. The component 22 is clamped to the strap 12 . In Figure 1, the component 22 is in a starting position xo with respect to the drive pulley 14. The components 24, 26 are detached from the belt 12 and are each in a parking position, here near the tensioning pulley 16. The components 22, 24, 26 can are also referred to as shuttles and can be automation components, for example.

A position transmitter 28 of a linear length measuring system 30 is arranged on each of the components 22 , 24 , 26 . The position sensor 28 interacts with a scale 32 of the length measuring system 30 to determine the position of the respective Component 22, 24, 26 to determine. In particular when the belt 12 is stationary, the position can be determined using the length measuring system 30 .

The belt drive 10 also has a control device 34, see Figure 6. The control device 34 records the signals from the rotary encoder 20 and the length measurement system 30. In addition, the control device 34 controls the rotary drive 18. Furthermore, the control device 34 can clamp or loosen the components 22 , 24, 26 on and off the belt 12 respectively.

A value for an effective circumference of the drive pulley 12 is stored in the control device 34 . When controlling the position, the controller 34 assumes that the components fixed to the belt 12 move by the amount of the effective circumference per revolution of the drive pulley 14 . The movement of the belt 12 or the components 22, 24, 26 is controlled during their movement on the basis of the signals from the rotary encoder 20 using the stored value for the effective circumference. The signals of the length measuring system 30 are typically not evaluated during the movement of the belt 12 or the components 22, 24, 26, i.e. they are not used in particular to regulate the movement of the belt. This allows for much simpler control based on drive pulley rotation. While the method according to the invention is being carried out, the signals of the length measuring system are preferably continuously recorded and then evaluated as described below.

In order to calibrate the value for the effective circumference of the drive pulley 12, the component 22 is set in motion from the initial position xo. For this purpose, the drive disk 14 is rotated by means of the rotary drive 18 .

In FIG. 2, the component 22 is in a first longitudinal position, which, starting from the starting position xo, is advanced by an approach distance Xstart. This first longitudinal position corresponds to a first rotational position of the drive pulley 14. The first longitudinal position is passed over in motion. The The starting distance Xstart corresponds to at least half of the stored value for the effective circumference of the drive pulley 14.

In FIG. 3, the component 22 has Xstart starting from the first longitudinal position

5 covered a measurement distance Xmess. This second longitudinal position corresponds to a second rotational position of the drive pulley 14. The second longitudinal position is passed over in motion. The component 22 is moved further beyond the second longitudinal position shown in FIG. 3 by a stopping distance x _s to _P , compare FIG. 4. The stopping distance x _s to _P is preferably at least 1 cm.

10

The angle of rotation of drive pulley 14 swept between the first rotary position and the second rotary position is detected by rotary encoder 20 . The angle of rotation between the first and the second rotary position is preferably specified as a plurality of complete rotations. The turning of the drive pulley

15 14 by the (predefined here) angle of rotation and its detection by means of the rotary encoder 20 correspond to a step 104 in FIG.

The amount of the measuring distance Xmess, i. H. those between the first and the second

20 rotational position of the drive pulley 14, the actual longitudinal displacement of the component 22 is also recorded by the length measuring system 30, compare step 106.

Based on the actual values determined by the length measurement system 30

25 longitudinal displacement of the component 22 and determined by the rotary encoder 20 swept angle of rotation is in a step 108 in the

Control device 34 stored value for the effective scope of

Drive pulley 14 corrected.

During the movement from the first rotary position to the second rotary position, the values of both the rotary encoder 20 and the length measuring system 30 are preferably recorded several times at the same time. The acquisition takes place preferably in a fixed time frame. For each pair of values from rotary encoder 20 and length measuring system 30, the difference (1)

be calculated. there is

the (actual) travel distance measured with the direct length measuring system 30 and

the travel distance calculated from the swept angle of rotation using the stored value U _eff for the effective circumference.

This difference can be represented by a straight line (2)

be approximated. The coefficients a and b can be determined using linear regression. In particular, it follows

where n denotes the number of measurement data values and i denotes the measurement data index.

About the equation

where ΔU eff _denotes the average deviation from the currently stored value of the effective circumference and Nu denotes the number of revolutions of the drive pulley 14 over the measuring section (and thus between the first and second rotational position), the average deviation ΔU _eff can be calculated by which the stored value U _eff must be corrected.

Steps 104 and 106 and the respective linear regression are preferably carried out several times and the results are averaged, compare the dashed arrow in FIG. 5. This improves the accuracy of the correction. Steps 104 and 106 are particularly preferably repeated several times with different angles of rotation between the first rotary position and the second rotary position Drive pulley 14 performed. The different angles of rotation are preferably multiples of a complete revolution of the drive pulley. As a result, irregularities in the drive pulley are not included in the calculation.

The correspondingly corrected value U _eff ,new=U _eff +ΔU _eff is stored in the control device 34 .

The value of the coefficient b is irrelevant for the method according to the invention.

In order to adjust the tension of the belt 12, a distance 36 between the drive pulley 14 and the tensioning pulley 16 can be changed, compare FIG. 4. This is indicated by a guide 38 for moving the tensioning pulley 16.

The tension of the belt 12 can be calculated from equation (6)

are calculated, where

the pre-tension path starting from a tension-free belt 12,

Fvspg the tension (preload force) of the belt 12,

Lr the length of the belt 12 and

Cr,spec denote the specific belt stiffness.

The effective circumference U _eff depends linearly proportionally on the elongation of the belt 12 and is therefore a measure of the belt pretension. The deviation from the nominal circumference of the drive pulley 14 U _eff , nom (the nominal spindle pitch) is thus a measure of the pretensioning of the belt with linear specific belt stiffness. Consequently, the actually effective circumference of the drive pulley 14 is also a measure of the pretensioning of the belt 12 and vice versa. In particular, the preload path Δlvspg can also be calculated as

(7)

with

Δlvspg : preload path,

AL- : elongation of the belt due to the pre-tension,

Lr : length of the belt in the slack state, U _eff ,nom : nominal effective circumference of the drive pulley 14, ΔU _eff : (average) deviation of the effective circumference from the nominal value.

From equations (6) and (7) a target value for the effective circumference U _eff ,gew. Calculate what needs to be adjusted to get the desired preload F vspg,wt. of the belt 12 to obtain. The result is: U _eff ,wt. ⁼ U _effective ,nom • ( 1 + Fvspg,wt. / Cr,spec ) (8)

With Equation (8) it finally follows by inserting ΔU _eff ⁼ U _eff ,gew. “ U _eff ,nom in Equation (7) also the preload path Δlvspg, wt., to be set, with which the desired preload is obtained. The effective scope then corresponds to the setpoint.

A belt 12 to be newly applied can be pretensioned in a step 102 . For this purpose, the required preload path can be calculated using equations (7) and (8). After the belt 12 has been applied without tension to the drive pulley 14 and the tensioning pulley 16, the distance 36 can be increased by a fraction of this total pretensioning path. Preferably, the distance 36 is increased by about 80% of the calculated preload travel to avoid overtightening while ensuring adequate preload. The effective circumference of the drive pulley 14 that results after the pretensioning in section 102 can be checked with steps 104 to 108 and stored in the control device 34 .

A fine adjustment of the tension of the belt 12 can take place in a step 110 . As already explained above, equation (7) results in the desired effective scope. From steps 104 through 108, the current value of the effective range is known. The remaining preload path to be set can can be calculated from the difference between the desired and the current effective range using Equation (7). The distance 36 between the drive pulley 14 and the clamping pulley 16 is adjusted accordingly.

Preferably, the stored value for the effective scope is then corrected again by repeating steps 104 to 108. If the set effective scope does not yet correspond to the target value with sufficient accuracy, step 110 (and then steps 104 to 108 again) can be repeated become.

FIG. 6 shows a processing system 40, here a laser cutting system by way of example. A processing tool 42 , here a laser processing head, can be displaced with respect to a workpiece support 44 . The processing tool 42 can be controlled by the control device 34 .

A blank 46 can be arranged on the workpiece support 44 . A state is shown here in which a sheet metal part 48 has been cut out of the blank 46 so that a residual skeleton part 50 remains. The residual skeleton part 50, in which the sheet metal part 48 is arranged, can be moved on the workpiece support 44 from a processing position to a removal position.

The processing system 40 has two belt drives 10a, 10b. The belt drives 10a, 10b are used to move automation components 52, 54 and 56, 58, 60. The automation components 52, 54 and 56, 58, 60 can be guided on the belt drives 10a, 10b in a manner that is not shown in detail. The belt drives 10a, 10b can be moved horizontally in the plane of the drawing relative to the workpiece support 44. The automation components 52, 54 and 56, 58, 60 each have a position transmitter 28 for interacting with a scale 32 of a respective length measuring system 30 of the belt drive 10a or 10b.

The automation components 52, 54 of the belt drive 10a can have suction cups 62 for transporting the workpiece 50. The automation components 56, 58, 60 of the belt drive 10b can have pins 64. Extending the pins 64 can support the detachment of the sheet metal part 48 from the residual skeleton part 50 . Due to the small size of the sheet metal part 48, only the automation components 52 and 56 are used in the example shown.

The design and function of the belt drives 10a, 10b correspond to the belt drive 10 described above in connection with Figures 1 to 4. The calibration of the position control and tension of the belts 12 can be carried out in the case of the belt drives 10a, 10b in each case with a one described above in connection with Figure 5 described procedures. In this way it can be achieved that the automation components 52, 54 and 56, 58, 60 can be positioned sufficiently exactly (for example with an accuracy in the tenth of a millimeter range) using the rotary encoder 20 of the belt drives 10a, 10b for handling cut sheet metal parts 48. Repositioning using the respective length measuring system 30 can then generally be omitted.

belt drive 10; 10a, 10b

strap 12

Drive pulley 14

Tension washer 16

Rotary drive 18

Encoder 20

Components 22, 24, 26

Locator 28

Length measuring system 30

Scale 32

control device 34

distance 36

leadership 38

Processing plant 40

Editing tool 42

Workpiece support 44

blank 46

Sheet metal part 48

Remnant skeleton part 50

Automation components 52, 54, 56, 58, 60

sucker 62

pins 64

Starting position xo

Approach path X _start

measuring section X _meas

stopping distance x _stop

Preload path Δl _vspg

102: Pretensioning a belt

104: Turning the drive pulley and detecting the swept angle of rotation 106: detecting an actual longitudinal displacement on the belt

108: Correct the stored value of the effective scope

110: fine adjustment of the tension of the belt

Claims

patent claims

1. A method for calibrating a belt drive (10) with a belt (12) and a drive pulley (14), a value for an effective circumference of the drive pulley (14) being stored in a control device (34), with the steps

B) rotating the drive pulley (14) between a first rotational position and a second rotational position by a rotational angle with detection by a rotary encoder,

C) detecting an actual longitudinal displacement of a point on the belt (12) when rotating the drive pulley (14) between the first rotational position and the second rotational position using a length measuring system (30),

D) Correction of the stored value for the effective circumference of the drive pulley (14) based on the actual longitudinal displacement and the swept angle of rotation.

2. The method according to claim 1, characterized in that in step D) a difference between an expected longitudinal displacement, which was determined based on the angle of rotation and the stored value for the effective extent, and the actual longitudinal displacement detected in step C) is determined and the stored value for the effective scope is changed accordingly.

3. The method as claimed in claim 2, characterized in that an iterative correction of the respectively stored value of the effective scope is carried out.

4. The method according to any one of the preceding claims, characterized in that steps B) and C) are each carried out several times, and in step D) a method of linear regression is used.

5. The method according to any one of the preceding claims, characterized in that the angle of rotation by which the drive pulley (14) is rotated in step B) is predefined.

6. The method according to any one of the preceding claims, characterized in that the angle of rotation covered in step B) is an integer multiple of a complete rotation.

7. The method according to any one of the preceding claims, characterized in that the rotation of the drive pulley (14) begins before the first rotational position and / or ends after the second rotational position.

8. The method according to any one of the preceding claims, characterized in that the point on the belt (12) is defined by a component (22, 24, 26) fixed to the belt (12), in particular the component (22, 24, 26) has a position transmitter (28) of the length measuring system (30).

9. The method according to any one of the preceding claims, characterized in that a target value for the effective circumference is predefined and that a tension of the belt (12) is changed in order to adjust the effective circumference to the target value.

10. The method according to claim 9, characterized in that a distance (36) between the drive pulley (14) and a clamping pulley (16) is changed to change the voltage.

11. The method according to claim 10, characterized in that in a step A) a total pretensioning path for a tension-free belt (12) is calculated on the basis of the target value and a nominal value of the effective circumference, and the distance (36) starting from a state in which the belt (12) rests without tension on the drive pulley (14) and the tensioning pulley (16), is increased by a fraction of the pretensioning path.

12. The method according to claim 10 or 11, characterized in that in a step E) a remaining pretensioning path is calculated based on the target value and the stored value for the effective extent, and the distance is changed by the remaining pretensioning path.

13. The method according to claim 12, characterized in that after

Changing the distance in step E), steps B) to D) and preferably E) are carried out again.

Having 14 belt drive (10).

- a belt (12), in particular a toothed belt,

- a drive pulley (14) with a rotary drive (18),

- a tension washer (16),

- a rotary encoder (20) for detecting a rotation of the drive pulley (14),

- a length measuring system (30) for detecting a longitudinal displacement of a point on the belt (12),

- a control device (34) which is set up to carry out a method according to one of the preceding claims. Sheet metal processing system (40), in particular a laser cutting system, with a processing tool (42) and with a belt drive (10) according to claim 14, in particular wherein an automation component (52, 54, 56, 58, 60) fixed to the belt (12) has a position sensor ( 28) of the length measuring system (30).