EP2374551A1 - Method and device for producing coil springs by means of coiling machines - Google Patents

Abstract

The method involves defining a target geometry of a helical spring by a numerically controlled (NC) control program, and measuring an actual position of a structural element of the spring relative to a reference element at a measuring time point. The actual position is compared with a target position of the structural element at the time point for determining actual position difference that represents difference between the actual position and the target position at the time point. A position of a pitch tool (130) of a shaping device (120) is controlled based on the position difference. Independent claims are also included for the following: (1) a computer numerical controlled spring winding machine comprising a feeding device for supplying a wire to a shaping device (2) a computer program product for executing a method for manufacturing a helical spring.

Description

The invention relates to a method for producing coil springs by spring winds by means of a numerically controlled spring coiling machine according to the preamble of claim 1 and to a suitable for performing the method spring coiling machine.

Coil springs are machine elements that are required in numerous applications in large numbers and different designs. Coil springs, which are also referred to as twisted torsion springs, are usually made of spring wire and designed depending on the load in use as tension springs or compression springs. Compression springs, in particular suspension springs, are needed for example in large quantities in the automotive industry. Among other things, the spring characteristic can be influenced by designing sections of different pitch or gradients. For example, there are compression springs often a more or less long central section with a constant pitch (constant section), at the end of both ends of the spring contact areas with decreasing slope to the ends. The spring diameter is constant with cylindrical coil springs over the length of the springs, but it can also vary over the length, such as conical or barrel-shaped coil springs. Also the total length of the (unloaded) spring can vary widely for different applications.

Coil springs are nowadays commonly manufactured by spring winches using numerically controlled spring coiling machines. In this case, a wire (spring wire) is supplied under the control of an NC control program by means of a feeder a forming device of the spring coiling machine and formed by means of tools of the forming device to form a coil spring. The tools typically include one or more wind pins that are adjustable in position to define and, if necessary, alter the diameter of spring coils and one or more pitch tools that determine the local pitch of the spring coils at each stage of the manufacturing process.

Spring wind machines are generally intended to produce many springs with a specific spring geometry (nominal geometry) within very narrow tolerances at high unit output. Among the functionally important geometry parameters u.a. the total length of the finished coil spring in the unloaded state. Through the total length u.a. the installation dimensions of the spring and the spring force determined.

With regard to high quality requirements, for example in the automotive sector, it is customary to measure certain spring geometry data, such as the diameter, the length and / or the pitch or the gradient of the spring, after completion of a spring and the finished Depending on the result of the measurement, automatically sort springs into good parts (spring geometry within the tolerances) and bad parts (result outside the tolerances) and, if necessary, into further categories. This procedure is very uneconomical, especially for long springs, since with long springs each spring a relatively large wire length is consumed, which must be discarded if it turns out that the finished spring is outside the tolerances.

It has also been proposed to check the diameter, the length and the pitch of the spring by means of suitable measuring means during production and to change manufacturing parameters in the case of deviations outside tolerance limits in such a way that the spring geometry remains within the tolerances.

The DE 103 45 445 B4 shows a spring coiling machine having an integrated measuring system with a video camera, which is directed to that area of the spring coiling machine in which the formation of the spring begins. A connected to the video camera image processing system with appropriate evaluation algorithms should allow to check the diameter, length and pitch of the spring during manufacture and it should be possible to change these spring geometry parameters by feedback to the motorized machining tools during manufacture. An evaluation algorithm for determining the current spring diameter is described in detail.

TASK AND SOLUTION

It is an object of the invention to optimize a method and a device of the generic type so that, in particular in the production of relatively long coil springs with great reliability from wire materials of very different quality coil springs can be manufactured within narrow geometrical tolerances. In particular, long coil springs with low dispersion of the total length and low rejection rate should be produced.

These objects are achieved by a method for producing coil springs by spring winds with the features of claim 1 and by a spring coiling machine with the features of claim 12. Advantageous developments are specified in the dependent claims. The wording of all claims is incorporated herein by reference.

In the method, first a desired desired geometry of the helical spring to be produced and a corresponding NC control program suitable for generating this desired geometry are defined. Thus, the sequence of coordinated working movements of the machine axes of the spring coiling machine is set, which are to go through in the manufacture of a spring.

During the manufacture of a coil spring, a measurement of an actual position of a selected structural element of the coil spring relative to a reference element is performed. By measuring a distance between the selected structural element and the reference element can be determined. The measurement takes place at a measuring time which is after the beginning and before the end of the production of the helical spring, ie during the course of the working movements of the spring coiling machine intended for spring production. So it is made at the time of measurement only part of the spring. The selected structural element lies in a measuring range which has a finite distance from the shaping device in the longitudinal direction of the helical spring. This distance is smaller than the total length of the finished coil spring, ie smaller than the total length resulting from the target geometry. By comparing the Actual position of the structural element with a target position of the structural element for the measurement time, a current position difference is determined, which represents the difference of the actual position to the target position at the measurement time. Depending on the position difference, the position of at least one tool influencing the pitch of the helical spring is then controlled in order to achieve an approximation of the actual position to the desired position.

A control intervention remains off when the actual value corresponds to the setpoint. If, on the other hand, a significant deviation (position difference) is detected, then the pitch of the spring produced at the moment of deformation is changed by changing the position of the pitch tool and / or another tool influencing the pitch (eg a controllably rotatable and / or tiltable wind pen), that a reduction of the position difference in the next measurement can be expected. The currently generated slope is thus regulated on the basis of the measurement. Preferably, only the position of a pitch tool is controlled or regulated for this purpose.

Due to the fact that the measuring range lies at a finite distance from the location of the forming process on the forming device, a cumulative length error in the spring section lying between the forming device and the measuring range can be determined by the measurement. Furthermore, since the distance of the measuring range from the forming device is smaller than the overall length of the finished coil spring, the measuring time with respect to the total duration of the production of a helical spring can be so early that a possibly performed due to the measurement control intervention in the forming process for correction any misadjustments can be used to obtain a total length of the coil spring within the tolerances after completion of the production.

The distance of the measuring range from the forming device is preferably adapted to the total length of the finished coil spring such that this distance is between 5% and 70% of the total length, in particular between 10% and 50% of the total length. If these preferred minimum values for the distance are adhered to, a longitudinal fault can build up in the case of non-perfect forming conditions via the spring section, which is sufficiently large in comparison to the measuring accuracy of the measuring system to allow significant measurement results. If the preferred upper limits for the distance are met, there is usually enough time left to obtain a coil spring with the desired overall length by means of one or more control interventions at the end of production.

Within the distance there are preferably one or more spring coils so that the measuring range is e.g. two, three, four, five, six or more spring coils can be located away from the place of the transformation or from the forming device. Depending on the slope, meaningful results can often be achieved even from two to three turns apart.

In preferred embodiments of the method, the measurement of the actual position relative to a machine-fixed reference element is performed. A machine-fixed reference element is an element whose coordinates are known or determinable with respect to a machine-fixed coordinate system. Since the reference element in this case has defined coordinates with respect to the machine coordinate system of the error winding machine, this measurement is an absolute measurement. As a result, a particularly high measurement accuracy is possible.

Alternatively, the reference element may also be a structural element of the helical spring, in particular one closer to the shaping device lying Windungsabschnitt or the contour of a Windungsabschnittes. In this case, a relative measurement is performed. In order to ensure that any accumulated length error between the structural element selected for the measurement and the reference element is large enough to be reliably measured, several windings should be present between the structural element and the reference element, for example two, three, four, five or two more turns.

The measurement is preferably carried out without contact, in particular with optical measuring means. In this case, for example, a laser measuring system could be used. Preferably, a camera with a two-dimensional image field (field of view, detection area) is used for the measurement, and the measurement area is placed in the image field of the camera. Camera-based measurement systems with powerful image processing hardware and software are commercially available and can be used for this purpose. The camera should be as low vibration mounted on a support which is firmly connected in operation with the frame of the spring coiling machine. Preferably, the camera is seated on or on a longitudinal guide, which allows a fixation of the camera at different distances from the forming device in order to set the optimum distance for different spring geometries. The position of the carrier may be vertically adjustable, e.g. to allow adaptation to springs of different diameters. If necessary, an adjusting device should also allow an arrangement of the carrier inclined at an angle to the spring axis.

In some method variants, the reference point for the measurement lies at the edge of the eg rectangular image field of the camera, which has known coordinates with respect to the machine coordinate system. In this case, a virtual reference element is formed by the image field edge, preferably by that side edge of the image field, which faces the forming device. The measurement of the actual position of the structural elements can then be reduced to a simple distance measurement within the image field.

In another, alternative or additionally usable variant of the method, a machine-resistant reference body is provided, which is positioned in the image field of the camera at a distance from the measuring area, wherein a structural element of the reference body, e.g. a straight edge, is used as a reference of the measurement. Any vibrations of the camera during the measurement can not affect the measurement accuracy of the measurement in this process variant, since these vibrations have no effect on the visible in the image field of the camera distance between the measurement element underlying the measurement of the coil spring and a reference point on the reference body.

When measuring with the aid of a 2D camera, it has proven to be particularly advantageous if the selected structural element of the helical spring used for the measurement is a contour section of a spring coil appearing more or less rectilinear in the image field, which is transverse to the longitudinal direction of the spring, in particular in one Angle between about 45 ° and about 90 ° to the longitudinal direction of the coil spring runs. This makes it possible with the help of simple contour detection algorithms of the image processing system to determine the lstposition of the structural element in the longitudinal direction of the spring very accurately. Alternatively, it would also be possible, for example, to place the measuring range at the outer edge of a spring coil, to determine the location of the maximum distance (maximum location) of this winding section to the longitudinal axis of the coil spring and to determine the distance of this maximum location to the reference element.

The target position of the structural element at the time of measurement should be known as accurately as possible in order to enable targeted regulation of the manufacturing process. Preferably, the desired position of the structural element for each time point during production is known, so that from a corresponding program time function, the target position can be derived directly to the measurement time. In the manufacture of coil springs, which have a more or less constant section (constant pitch section), the measurement preferably begins only when a possibly existing spring section with variable pitch has passed the measuring range. For measurements in the constant section, it is possible to exploit the fact that the set position of a selected structural element remains constant over a longer period of time, resulting in relatively simple measured value detection and evaluation. In principle, it is also possible to measure in spring sections with changes in pitch. Here, as a rule, temporally variable, i. wandering setpoint positions, which are then used as a basis for the comparison step with the setpoint value valid for the measurement time point.

In general, the coordinates of the nominal position of the structural element at the time of measurement are derived from a program time function, determined before the measurement, for the coordinates of the nominal position of the structural element. The correct setpoint can then be determined unambiguously for each measurement time. The program time function for the coordinates of the target position can be determined on the basis of a simulation on a computational basis. In general, however, an experimental determination within a relatively short time is possible and useful. In some method variants, the program time function for the coordinates of the nominal position of the structural element is determined on the basis of a reference production process of at least one reference helical spring, ie experimentally.

The term "program time function" here refers to a function that refers to specific locations within the NC control program. The achievement of a specific NC block corresponds to a specific program time or a point in time within the program sequence. In that regard, a program time corresponds to a sequence position in the sequential sequence of program steps during program execution. If, for example, a trigger signal (trigger) is required to control image acquisition by a camera in a specific phase of the program execution, then this trigger signal can be triggered by a program line present at the appropriate location. Such signals are directly linked in the program to certain positions of the machine axes, e.g. with the machine axis of the wire feed and / or with the machine axis for the position of the pitch tool. A time in a program time function thus corresponds to a location in the movement curve of one or more machine axes. The program time function results in times (program times) within an NC program that are synchronous with the progress of the spring production. In that regard, the program time function is also a path function with respect to the movements of machine axes. In particular, a program time function also corresponds to a path function of the wire feed.

In some manufacturing processes, such as relatively short coil springs, a single measurement and a single thereafter possibly performed control intervention may be sufficient to produce a coil spring with sufficiently small length error. In particular with relatively long coil springs, a plurality of measurements are carried out at intervals with successive measurement times during production of the coil spring, so that a temporal change in the spring geometry can be observed during the manufacturing process and optionally several control interventions can be carried out.

The number of measurements per unit time is theoretically limited by the recording and evaluation capacity of the measuring system. However, it has been found that a high measuring frequency is usually neither necessary nor useful. In preferred variants of the method, the time interval between directly successive measurement times is adapted to the feed rate of the wire such that at least one turn is generated in a time interval between two immediately consecutive measurements, preferably between one and two turns being generated in the time interval. In this way it can be ensured that any accumulated length errors are then large enough to be reliably detected within the measuring accuracy of the measuring system. The significance of the measurement results is thereby improved and the control works more stable.

Preferably, a plurality of measurements are made during the generation of a constant portion of the coil spring. Under these conditions, an observed structural element should not change its position over a period of time. The setpoint used for the comparison step remains constant during this time. If the structural element migrates in the direction of the forming device during the production of a constant section, this indicates a too low pitch during forming and can be corrected accordingly. Conversely, migration of the structural element away from the forming device can be compensated by reducing the slope.

In some method variants, a running average value for the actual values is determined from the actual values of a plurality of successive measurements after a predefined number of measurements, in particular after each measurement. From this running average can be meaningful information about the effectiveness der der Regeleingriffs derived. Preferably, a temporal development of the running average value is displayed on a display unit of the spring coiling machine. This allows an operator to immediately recognize whether the settings made on the operating device for effective control sufficient to obtain a coil spring of desired overall length at the end of a manufacturing step.

Different control concepts or control algorithms can be implemented. In some variants, a weighted difference value is determined for each determined position difference, and the position of the tool is changed based on the weighted difference value. In particular, a weighted difference value proportional to the position difference can be determined, wherein preferably a proportionality factor can be set by the operator and changed if necessary. Any deviation from the setpoint determined during a measurement can lead to a control intervention in this variant, so that it is possible to respond quickly to deviations. It is also possible to correct the position of the tool only when the position difference or a weighted difference value derived therefrom exceeds a specific threshold value.

In order to avoid a permanent control deviation, a temporal integration of the control deviations is preferably carried out in the manner of an I-controller, so that overall the control characteristic of a PIL controller can be realized.

As part of the method, a measurement of the position of a selected structural element can be triggered in different ways. For example, a trigger signal (trigger) for triggering a measurement can be triggered by a program line present at the corresponding position of the NC control program.

This ensures automatic synchronization with the program time function. Typical accuracies in the determination of the measuring time then lie in the order of magnitude of the cycle time of the control, which may be on the order of one or a few milliseconds, for example. In particular, in measurements during the generation of a constant section such accuracies are quite sufficient, since the structural section to be measured practically does not move. In other variants of the method, a timer independent of the NC control program is used to determine the measuring time, which is synchronized with the NC control program at a reference time. Such a timer can be realized for example by an additional card in the control unit. As a result, regardless of the cycle time of the controller, high accuracies can be achieved when determining the measuring time. In some variants, the measurement time relative to a reference time of the program time function is determined with an accuracy of 100 microseconds or less. As a result, sufficiently precise measurements are possible even when measured in the range of spring sections with slope change. In such cases, a time varying, i. a wandering target position that is to be used as a basis for the comparison step. Therefore, it is important to be able to recognize the measurement time as accurately as possible in order to be able to determine with sufficient accuracy the setpoint position of the observed structural element associated with the measurement time.

In particular, in the measurement of spring sections with changes in slope, it may also be advantageous if multiple measurements are made in different measuring ranges and the measurement results are taken into account in the scheme. In some embodiments, at a first measurement time, a first measurement is performed in a first measurement range that is at a first distance from the forming device is located, and in a chronologically subsequent second measuring point, a second measurement is performed in a second measuring range offset from the first measuring range, which lies at a second distance from the forming device, which is greater than the first distance. If the results of two or more staggered measurements are processed together in spatially offset measuring ranges, precise statements can be made about the temporal development of the spring coil process and tendencies in the event of possible deviations in the target. As a result, an even more precise control of the spring coil process can be ensured.

The invention also relates to a numerically controlled spring coiling machine, which is particularly configured for carrying out the method. It has a feed device for feeding wire to a forming device and a forming device with at least one wind tool, which essentially determines the diameter of the coil spring at a predeterminable position, and at least one pitch tool whose engagement with the developing helical spring determines the local pitch of the helical spring ,

The spring winding machine preferably has a first camera which is arranged such that a measuring area in the field of view of the camera detects a part of a spring section with a finite distance from the tools of the shaping device. The distance of the measuring range from the forming device is preferably adapted to the total length of the finished coil spring such that the distance between 5% and 70%, in particular between 10% and 50% of the total length and / or that within the distance one or more spring coils , For example, at least two or three spring coils. Furthermore, a second camera may be provided which is positioned at a distance from the first camera such that a free spring end section in an end phase of the production of the coil spring in the Detection area of the second camera runs into it. When using a camera with a sufficiently large detection range, a single camera may be sufficient to cover the measuring range at finite distance from the tooling tools and the measuring range for detecting the end portion.

In some modern CNC spring-winding machines, which already have a suitable measuring system with camera, the invention can be implemented with existing structural requirements. The ability to carry out embodiments of the invention may be implemented in the form of additional program parts or program modules, or in the form of a program change in the control software of computerized control devices.

Therefore, another aspect of the present invention relates to a computer program product stored on a computer readable medium or implemented as a signal, wherein the computer program product, when loaded into the memory of a suitable computer and executed by a computer, causes the computer to perform a method according to the invention or a preferred embodiment thereof.

These and other features will become apparent from the claims but also from the description and drawings, wherein the individual features each alone or more in the form of sub-combinations in an embodiment of the invention and in other fields be realized and advantageous and protectable Can represent versions.

BRIEF DESCRIPTION OF THE DRAWINGS

• Fig. 1 shows a schematic overview of an embodiment of a spring coiling machine with parts of the feeder and the forming device,
• Fig. 2 shows in perspective view of assembly groups for in Fig. 1 shown spring coiling machine, including two cameras of a camera-based optical measuring system for contactless real-time data acquisition on the geometry of a spring currently produced, and a spring guide device;
• Fig. 3 shows a spring section of the spring currently generated by the forming device from a viewing direction parallel to the direction of the wire feed or parallel to the optical axis of the camera optics of the first camera, wherein a winding portion of the spring lies in a lying within the image field of the camera measuring range;
• Fig. 4 FIG. 4 shows plots of the time evolution of the running average value for the actual values determined during a series of individual measurements during the manufacture of a spring, FIG. 4A showing the time development without regulation and FIG. 4B the time development with active regulation;
• Fig. 5 shows histograms and diagrams of the dispersion of actual values in a series of individual measurements during the manufacture of a spring, the actual values without regulation being shown in FIG. 4A and the actual values obtained with active regulation in FIG. 4B;
• Fig. 6 shows a rectangular image field of the first camera, wherein in the image field, a portion of a spring to be measured and the image of a machine-mounted reference element can be seen.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The schematic overview in FIG Fig. 1 shows essential elements of a CNC coil winding machine 100 according to a per se known construction. The spring coiling machine 100 has a feeding device 110 equipped with feeding rollers 112, which feeds successive wire sections of a wire feed wire 115, which is guided by a straightening unit, into the area of a forming device 120 with a numerically controlled feed rate profile. The wire is converted into a helical spring by means of numerically controlled tools of the forming device. The tools include two angularly offset wind pins 122, 124 that are radially aligned with the central axis 118 (corresponding to the location of the desired spring axis) and are designed to determine the diameter of the coil spring. The position of the wind pins can be changed to the basic setting for the spring diameter when setting along the lines shown in phantom and in the horizontal direction (parallel to the feed direction of the feeder 112) to set up the machine for different spring diameters. These movements can also be carried out with the aid of suitable electric drives under the control of the numerical control.

A pitch tool 130 has a tip oriented substantially perpendicular to the spring axis which engages the turns of the developing spring. The pitch tool is moved by means of a numerically controlled adjustment of the corresponding machine axis parallel to the axis 118 of the developing spring (ie perpendicular to the plane of the drawing). The advanced in the manufacture of spring wire is from the pitch tool according to the Displaced position of the pitch tool in the direction parallel to the spring axis, wherein the position of the pitch tool, the local slope of the spring is determined in the corresponding section. Gradient changes are effected by axis-parallel process of the pitch tool during spring production.

The forming device has another, vertically downwardly deliverable incline tool 140 with a wedge-shaped tool tip, which is introduced when using this pitch tool between adjacent turns. The adjustment movements of this pitch tool are perpendicular to the axis 118. This pitch tool is not engaged in the manufacturing process shown.

Above the spring axis, a numerically controllable cutting tool 150 is mounted, which separates the manufactured coil spring with a vertical working movement of the supplied wire supply after completion of the forming operations. In Fig. 1 the supplied wire is shown in a situation immediately after separation of the previously completed coil spring. In this position, the wire has already formed half a turn and the wire end forming the spring beginning is located 0.3 turns before the position of the pitch tool 130.

The machine axes of the CNC machine belonging to the tools are controlled by a computer numerical control device 180, which has memory devices in which the control software resides, to which i.a. an NC control program for the working movements of the machine axes heard.

To produce a helical spring, the wire is advanced from the "spring-finished position" shown by means of the feeder 110 in the direction of the wind pins 122, 124 and through the wind pins deflected to the desired diameter to form a circular arc-shaped curvature until the free wire end reaches the pitch tool 130. Upon further wire feed, the axial position of the pitch tool determines the current local pitch of the developing coil spring. The pitch tool is axially displaced under control of the NC control program when the slope is to be changed during spring development. The adjusting movements of the pitch tool essentially determine the gradient along the helical spring.

When setting up the spring coiling machine, the forming tools are brought into their respective basic positions. In addition, the NC control program is created or loaded, which controls the positioning movements of the tools during the manufacturing process. The geometry input is made in the spring coiling machine by an operator on the display and control unit 170, which is connected to the control device 180.

Based Fig. 2 Now some mounting assemblies for the in Fig. 1 shown spring coiling machine, which are advantageous for the implementation of the method. From Fig. 1 already known elements are denoted by the same reference numerals as in FIG Fig. 1 characterized. Fig. 2 shows the spring coiling machine during the production of a relatively long, cylindrical coil spring 200, of which about twenty turns are already made to the time shown in the figure. It is a long spring with a ratio UD between total length L of the finished spring and diameter D of the spring of more than ten. To ensure that the increasing with increasing wire feed spring remains straight and does not bend with its free end down, a spring guide device 210 is provided. The spring guide device has a fastened with a horizontal longitudinal axis on the frame of the spring coiling machine Angle plate 212 with V-shaped profile. The downwardly converging, planar inclined surfaces of the angle plate support the spring downwardly and laterally so that the longitudinal axis (central axis) of the developing spring is coaxial with the central axis 118 of the developing spring. The angle plate is fastened by means of a holding device, not shown on the machine position and adjustable in height and in the lateral direction to allow for springs of different diameters the desired, to the central axis 118 of the spring coaxial guide. The angle plate can be pivoted after completion of the production of a spring automatically by means of a hydraulic rotary actuator down, so that the finished spring can slide into a sump.

The forming device facing the end of the angle plate is located at a distance of a few inches from the forming, so that between the tools of the forming device and the machine-side beginning of the angle plate a free-floating spring portion 202 remains. The length of the angle plate is adapted to the total length of the finished coil spring, that the first manufactured Federendabschnitt protrudes freely beyond the machine-distal end of the angle plate in the final phase of production. The machine-near, free-floating spring section 202 and the machine-distal spring end section 204 are thereby accessible for an optical measurement with an observation direction perpendicular to the longitudinal axis of the helical spring.

The spring coiling machine is equipped with a camera-based optical measuring system for the non-contact, real-time acquisition of data on the geometry of a currently manufactured spring. The measuring system has two identical CCD video cameras 250, 260, which in the exemplary case at a resolution of 1024 x 768 pixels (pixels) up to 100 frames per second via an interface to can deliver a connected image processing system. The image acquisition of the individual images is triggered in each case via trigger signals (trigger) of the controller. This determines the measurement times. The software for image processing is housed in a program module, which cooperates with the control device 180 of the spring coiling machine or is integrated in this.

Both cameras are mounted on a torsion-resistant support rail 255, which is laterally mounted next to the spring guide device in the region of the guide rollers of the feeder on the machine frame of the spring coiling machine so that the longitudinal axis of the support rail is parallel to the machine axis 118. The measuring cameras are longitudinally displaceable on the carrier rail and can be fixed to arbitrarily selectable longitudinal positions.

The machine-proximate first camera 250 is mounted so that its rectangular image field 252 (image capture area) captures a portion of the free-floating spring portion 202 away from the transforming tools (see FIG Fig. 3 ). The optical axis of the camera optics is in the example, approximately at the height of the central axis of the coil spring (ie at the height of the axis 118) arranged and perpendicular to this axis. Within the rectangular image field 252, a smaller rectangular measuring range 254 can be seen, through which a winding section of the spring facing the camera extends obliquely from top left to bottom right. The image of this (in the spring production in the longitudinal direction of the wire moving) Windungsabschnitts or its off-machine contour serves as a structural element for the length measurement.

The second camera 260 is intended for the detection of the free spring end 204 and therefore positioned on the support rail that the free Spring end runs in the final phase of the production of the coil spring into the detection range of the second camera.

Diametrically opposite the cameras, a lighting device is mounted at the level of the axis 118, which flashes at the measurement times predetermined by the control in response to trigger signals (trigger) of the control and enables measurement in transmitted light. On the side of the cameras, a reflected-light illumination device may be provided in order to improve the visibility of interesting details of the spring for the measurement.

Fig. 3 shows the in Fig. 2 dargstellte situation from a line of sight parallel to the direction of the wire feed (C-axis of the spring coiling machine) or parallel to the optical axis of the camera optics of the first camera. On the left is a section through the wire 115 recognizable, which is advanced in the feed direction (perpendicular to the plane of the drawing) on a curved inclined surface of the lower wind tool 124. By the wind tool, the wire is urged upward on a circular curved path in the direction of the upper wind tool and thereby reshaping permanently. Above the wind tool, the tip of the pitch tool 130 can be seen, which bears against the developing winding with a side working surface. The pitch tool can be displaced parallel to the spring axis 118 by means of the associated machine axis (in the direction of the arrow) so that the local pitch of the spring at the location of the forming is determined by the position of the pitch tool.

In Fig. 3 Fig. 12 is a situation in the initial stage of manufacturing a cylindrical coil spring 200, which includes an already-formed end-side abutment portion 206 of continuously increasing pitch, a constant-pitch constant portion 208 thereafter, and an opposite unillustrated one at the time shown Anlageabschnitt with decreasing slope has. At the time shown, the production has already progressed so far that the free spring end with the contact section passes the measuring range 254 and has already reached the angle plate of the spring guide device and thus the free-floating spring portion 202 with a constant pitch is stable coaxial with the axis 118.

The first camera 250 is oriented so that the measuring area 254 viewed in the longitudinal direction of the coil spring has a relatively large distance 210 from the tools 122, 130 of the forming device. In the example, about four turns of the coil spring are at this distance. The distance in the example case is between about 10% and about 20% of the total length of the finished spring, in particular for short springs it can e.g. up to 30% or 40% or 50% of the total length.

In the series production of coil springs using this spring coiling machine can proceed as follows. First, the desired desired geometry of the coil spring is input to the display and control unit 170 or, for example, by entering an identification number, corresponding existing geometric data is loaded from a memory of the spring coiling machine. Based on the geometry data, a so-called NC generator calculates an NC control program whose individual NC blocks and their sequence during the subsequent production control the coordinated working movements of the devices and tools of the spring coiling machine.

After the tools of the forming device have been filed, a first helical spring is manufactured in a first reference manufacturing process, without the control built up with the measuring system being activated. The first camera 250 captures with its measuring range 254 a selected structural element of the spring, in the example, the obliquely from top left to bottom right through the measuring range extending winding section. This appears dark in the camera image and clearly stands out from the bright background, forming a straightforward light / dark contour. To improve the visibility of the contours, the coil spring can be illuminated in the region of the measuring range from the side of the camera and / or inside. The machine-remote boundary appearing in the image field or the edge of this winding section is used to determine the actual position of the structural element. In this case, for example, the coordinates of the upper intersection point 256-1 and the lower intersection point 256-2 of the light / dark transition with the upper or lower boundary of the measuring area can be determined by the image processing system and the coordinates of the intermediate linear area can be determined by interpolation. For a measuring point 270 located centrally between the upper and the lower intersection point, the axis-parallel distance to a machine-fixed reference point is then determined with the aid of a "distance tool" of the image processing software in order to obtain a first actual value for the position of the structural element. In the example of Fig. 3 the machine-near (left) rectilinear boundary of the image field 252 serves as a virtual reference element or as a "fixed stop" for the measurement. The distance measured parallel to the axis (relative to the axis 118) between the measuring point 270 on the selected structural element and the reference element is then adopted by the control as the first setpoint value for the further production.

Thereafter, the total length of the finished spring is measured independently. If this total length is within the specified tolerance, it is assumed that the measured first setpoint can be taken as the starting value for the following series production. On the other hand, if the total length is outside the tolerance, then settings for the production process are changed to a subsequent one Spring to be able to perform a corresponding further reference measurement. These individual reference measurements are repeated stepwise until a manufactured spring is well within the manufacturing tolerance for the overall length of the coil spring. The target value for the structural element determined in the production of this "good" spring is then adopted for series production.

It should be noted in the example case that the determination of the desired value takes place at a time when the constant portion 208 of the spring is already in the measuring range 254. Under these conditions, the absolute value of the nominal dimension is then constant over a longer time interval, so that ideally nothing changes in the appearance of the projection of the developing spring detected by the camera, as long as turns of the constant portion are moved through the detection range of the camera.

Thereafter, the control can be adjusted and activated for the production of subsequent springs of a series. In this case, a measurement expediently begins only when an optionally existing contact area with varying pitch has traveled through the measuring area and the measuring area is located in the constant part of the spring. Thereafter, the control cycle begins with a first measurement of the distance of the selected structural section to the defined reference element (edge of the image field). The determined actual position or the determined distance is then compared by an evaluation software with the previously determined desired position or the nominal distance of the structural element for the measurement time. This computational comparison generates a value for a current position difference, which represents the difference between the actual position and the setpoint position at the time of measurement. In the following example, the figures are given for reasons of clarity without dimension, the dimension is eg millimeters.

For example, if the setpoint is 10.5 and the actual value is 10.7, the result is a position difference of -0.2. From this position difference, a weighted difference value is determined. For this purpose, in the example, an operator-adjustable weighting parameter called "control step" is used, which is defined in percent and which is applied to the determined position difference. If, for example, a control step of 50% is set, the result for a position difference of -0.2 is a weighted difference value of -0.1. This value remaining after the weighting is then added to a correction value in order to obtain a new (modified) correction value. The correction value may initially be set to the value 0 (zero), for example, and is then changed stepwise during the regulation. In the example case (correction value first 0), a new correction value is calculated according to the calculation relation 0 + (-0,1) = (-0,1), which is then sent to control the spring coiling machine as a correction.

The NC control program is prepared for control at specified points in such a way that an immediate change of an NC block in accordance with the received correction value can be made by the programmable logic controller (PLC) in the NC program. This change has an immediate (real-time) effect on the position of the pitch tool 130 in terms of reducing the position difference.

In the immediately following second measurement, for example, an actual position is determined with the actual dimension 10.6. At the still valid nominal value of 10.5, a position difference of -0.1 results. If the weighting factor is unchanged (control step 50%), the result is a weighted difference value of -0.05 and thus a correction value according to: (-0.1) + (-0.05) = -0.15. Here it can be seen that the renewed correction does not affect the original correction value (= ₀ ), but the due to the previous measurement changed correction value (-0,1). After the second measurement, therefore, a correction value of -0.15 is sent to the controller as a correction and processed in the manner previously described for immediate changes of the NC control program.

This exemplary processing of the measured data corresponds to a PI controller with an adjustable proportional component and the integrating effect of an integral component.

These steps are now carried out during the manufacture of the constant section of the coil spring at a plurality of successive measuring time with a time interval, so that a plurality of control actions takes place or take place. The wire is continuously advanced during the measurements, no stopping is necessary. The time interval between the successive measurement times is adapted in this process variant to the feed rate of the wire so that between two directly successive measurement times about 1.4 turns are generated. As a result of this measurement sequence, which is relatively slow in comparison with the possible frame rate of the camera, it can be achieved that an error of sufficient magnitude can build up between the individual measurements in the case of a non-optimal process sequence in the spring, which becomes a significant measured value within the measurement accuracy of the optical measuring system leads, so that a correction of the right size in the right direction is initiated.

The precision-enhancing effect of this scheme can be assessed on the basis of Fig. 4A, 4B and 5A, 5B be demonstrated. These figures show measurement results obtained in the manufacture of clutch springs with 47 turns of spring wire of 3.8 mm diameter. The feathers had a diameter of about 27 mm and a Total length of about 350 mm. The diagrams in the Fig. 4A, 4B each show the temporal evolution of the current average value for the actual values determined during the individual measurements during the production of a spring. On the abscissa each dimensionless counter for equidistant measuring times are given, so that the abscissa is a time axis. The ordinate shows in each case the values for the running average of the actual value in comparison to the nominal value of 10.55 mm shown with a bold line. FIG. 4A shows a typical measurement diagram for a conventional production without control. At counter time 351 begins the production of a new coil spring. To the left of this is shown the final stage of the previous fabrication, which ends with a too small average value (about 10.48 mm), so that the overall length of this spring made is too short. In the case of the new coil spring, the actual values are initially too high; the running average initially approaches the nominal value and then falls below this value as the distance increases, so that this coil spring too is clearly too short after completion.

Fig. 4B shows the corresponding representation for a production with the control switched on. At counter time 405, the fabrication of the previous spring ends at an average value that is very close to the set point so that the total length of the spring is very close to the setpoint for the total length. When manufacturing the following coil spring, the actual values are initially well below the setpoint. However, the intervention of the control leads to an approximation of the running average from the third measurement to the target value (10.55 mm), which approximates the running average towards the end of production asymptotically, whereby at the end of production the running average value again almost exactly Setpoint is.

The Fig. 5A, 5B show in another illustration the effect of the control, with the results without control in FIG. 5A and in FIG. 5B Results with closed-loop control are shown. The diagrams shown on the right show in their abscissa again the measuring times in arbitrary units of a counter and the ordinate the respectively measured position difference between actual value and setpoint. The bold lines running parallel to the zero line above and below represent the limits of the tolerance range for the production. The measurement results in the form of histograms are shown in the left-hand sub-figures. At the in Fig. 5A In the case of production without regulation, the actual values scatter strongly around the setpoint in both directions, with all values still within the tolerances. Is the control activated ( Fig. 5B ), this results in significantly lower variations around the setpoint, so that it is ensured that all coil springs manufactured with the aid of the control have an overall length very close to the setpoint value for the total length.

The first camera 250 is arranged relatively close to the forming tools on the carrier rail 255, so that any vibrations at the location of the first camera can only have small amplitudes, which hardly affect the measuring accuracy. Nevertheless, it may be that the measurement result is impaired by movements of the camera. Based on Fig. 6 is explained a way to make the measurement independent of any camera vibrations and thereby increase the accuracy of measurement. Shown is a rectangular image field 652 of the first camera. A smaller rectangular measuring area 654 includes an almost vertically from top to bottom extending contour of a lying in the focus range of the camera, the camera facing winding section. Between the intersections of the light / dark contour with the upper and lower edges of the measuring range, the coordinates of the actual position of the observed structural element of the spring are determined by interpolation. Furthermore, in the image field, the image of a reference element 680 can be seen, which is formed by a vertically oriented bolt, which with the help of a stable support is attached to the machine frame. The bolt protrudes from below into the image field and forms in the sharpness zone of the camera a sharp, vertical contour with a light / dark transition. During the measurement, the distance between the structural element and the edge of the reference element 680 facing the structural element is determined and the evaluation is taken as the basis. This measured distance is independent of any vibrations of the camera and any associated shifts of the image field relative to the observed spring. Any movements of the camera are therefore not included in the measurement errors.

The measurements of the distance between the structural element of the helical spring (e.g., a winding section contour) and a virtual or physical reference element may be performed as described in the direction parallel to the axis 118 or in any other suitable directions obliquely thereto.

The embodiments described in detail were explained with reference to the production of a long spring with over 30 turns. In non-illustrated experiments, an approximately 65 mm long coil spring was produced with only 7 turns. During production, only two measurement times were measured and corrected if necessary. The scattering of the total length could be reduced from approx. 0.3 mm in unregulated operation to approx. 0.15 mm in controlled operation.

As an alternative or in addition to the described absolute measurement relative to a machine-fixed reference element, in some cases a relative measurement with respect to a reference element which is formed by a part of the spring is also possible. If, for example, that in Fig. 3 shown image field 252 is sufficiently large to detect more windings in the longitudinal direction of the spring, the longitudinal distance between the measuring point 270 at the lying in the measuring range 254 winding contour and a closer to the forming tools, 3 or 4 turns remote, corresponding Windungskontur measured and based on the scheme. For example, the first complete turn 214 or its machine-distant contour could serve as a reference element.

Claims (15)

A method for producing coil springs by spring-winding by means of a numerically controlled spring coiling machine, wherein a wire is fed under the control of an NC control program by a feeder of a forming device of the spring coiling machine and converted by means of tools of the forming device into a coil spring, comprising the following steps: Defining a desired setpoint geometry of the coil spring and an NC control program suitable for generating the desired geometry; Measuring a lstposition of a selected structural element of the coil spring relative to a reference element at least one after the beginning and before the end of the preparation of the coil spring measuring time in a measuring range, which has a finite distance from the forming device in the longitudinal direction of the coil spring, wherein the distance is less than the total length of the finished coil spring is; Comparing the actual position with a target position of the structural element for the measurement time to determine a current position difference representing the difference of the actual position to the target position at the measurement time; Controlling the position of at least one of the pitch of the coil spring-determining tool of the forming device in dependence on the position difference. The method of claim 1, wherein the distance of the measuring range from the forming device to the total length of the finished coil spring is adjusted such that the distance between 5% and 70%, in particular between 10% and 50% of the total length and / or that within the distance one or more spring coils lie. Method according to Claim 1 or 2, wherein a camera with a two-dimensional image field is used for the measurement and the measuring region lies in the image field of the camera. Method according to one of the preceding claims, wherein the measurement of the actual position is carried out relative to a machine-fixed reference element. Method according to claim 4, wherein a virtual reference element is used which is formed by an edge of the image field of a camera, preferably by the side edge of the image field facing the forming device, or in which a machine-fixed reference body is provided which is in the image field of the camera is positioned at a distance from the measuring range, wherein an element of the reference body, in particular a straight edge, is used as a reference element of the measurement. Method according to one of the preceding claims, wherein the selected structural element of the helical spring used for the measurement is a contour of a winding section which appears rectilinear in the image field and which is transverse to the longitudinal direction of the helical spring, in particular at an angle between approximately 45 ° and approximately 90 ° to this Longitudinal direction, runs. Method according to one of the preceding claims, wherein the coordinates of the target position of the structural element at the time of measurement are derived from a pre-measurement program time function for the coordinates of the target position of the structural element, preferably the program time function for the coordinates of the target position of the structural element based on at least one reference manufacturing process a reference coil spring is determined experimentally. Method according to one of the preceding claims, wherein during the manufacture of the helical spring several measurements are carried out at successive time intervals, wherein the time interval is preferably adapted to a feed rate of the wire such that in a time interval between two consecutive measurements at least one turn is generated, preferably in the time interval between one and two turns are generated. Method according to one of the preceding claims, wherein a plurality of measurements are carried out during the generation of a constant section of the helical spring, and / or wherein a running average value for the actual values is determined from the actual values of a plurality of successive measurements after a predefined number of measurements, in particular after each measurement In particular, a temporal evolution of the running average value is displayed on a display unit of the spring coiling machine. Method according to one of the preceding claims, wherein a weighted difference value, in particular a weighted difference value proportional to the position difference, is determined for each determined position difference, and the position of the tool is changed on the basis of the weighted difference value. Method according to one of the preceding claims, wherein, in particular in the measurement of spring sections with change in pitch, at a first measuring time a first measurement is carried out in a first measuring range, which lies at a first distance from the forming device, to a temporally thereafter lying second Measuring point is performed a second measurement in a first measuring range offset second measuring range, which is at a second distance from the forming device, which is greater than the first distance, and wherein results of the first measurement and the second measurement are processed together. Spring winding machine (100) for producing coil springs (200) by spring windings under the control of an NC control program, with a feed device (110) for feeding wire (115) to a forming device (120), wherein the forming device at least one wind tool (122 , 124), which essentially determines the diameter of the coil spring at a predeterminable position, and at least one pitch tool (130) whose engagement on a developing coil spring determines the local pitch of the helical spring, characterized in that the spring coiling machine for carrying out the method is configured according to one of the preceding claims. A spring coiling machine according to claim 12, wherein the spring coiling machine comprises a first camera (259) arranged such that a measuring area (254) in the image field (252) of the first camera forms part of a finite pitch spring portion (210) of the tools of the forming device (120), wherein the distance (210) is preferably adapted to the total length of the finished coil spring such that the distance is between 5% and 70%, in particular between 10% and 50% of the total length and / or within the distance or more spring coils lie. A spring coiling machine according to claim 12 or 13, wherein the spring coiling machine comprises a second camera (260) positioned at a distance from the first camera (250) such that a free spring end portion (204) in an end phase of the production of the coil spring into the detection range of the second camera runs. A computer program product, stored in particular or implemented as a signal on a computer-readable medium, wherein the computer program product, when loaded into the memory of a suitable computer and executed by a computer, causes the computer to perform a method according to any one of claims 1 to 11.

Images