EP0286950A1 - Feed chutes for fibre-processing machines - Google Patents

Abstract

The actual height of a fibre column (62) in a feed chute (40) is sensed, and a corresponding signal is compared with a desired level in an evaluation unit (56). A controller (58) reacts to any deviations, in order to control the supply of material by means of feed rollers and a feed-roller drive (34A). <IMAGE>

Description

The invention relates to a hopper for a fiber processing machine, e.g. a card or a cleaning machine in the blow room of a short batch spinning mill.

Such filling shafts have been state of the art in short-staple spinning for some time, with newer variants of these devices in our European patent application no. 175851 and Swiss patent application No. 2751/86 have been described. The mode of operation of such a shaft is explained in the article "The new card feed Aerofeed-U" by Messrs R. Waeber and U. Stähli in the February 86 edition of "mittex". As shown in this article, the height of a flake or fiber column should be monitored and controlled by two light barriers arranged one above the other, whereby the target production of the system must be set by the operator. The two light barriers work together with a feed roller drive, which operates at high speed when the lower light barrier is actuated and when actuated the upper light barrier should run in slow speed to convey flakes from a feed shaft into the feed shaft.

In principle, this arrangement should enable an uninterrupted fiber flow from the feed shaft into the feed shaft. In practice, however, it often happens that the production set by the operator is set too high, so that the slow speed is nevertheless too fast and the delivery of the flakes in the feed chute has to be stopped by an overfill protection system, i.e. the conveyance of the flakes in the feed chute is discontinuous, which is known to be undesirable and leads to strong impacts on the cotton-forming material column (upsets).

An arrangement for the continuous delivery of fibers or flakes in a filling shaft is known from the German patent no. 2834586 and its American equivalent No. 4321732 known. According to this arrangement, however, it is not the height of the fiber column that is regulated, but rather the pressure in the feed shaft, as also in DE-PS 2658044 (equivalent - US 4404710) and in DE-OS 1510302 (FIG. 4). Since the mode of operation of the controller has not been completely described in DE-PS 2834586, it cannot be derived from this how the system as a whole should work. From recent publications by the patent owner in the textile press, however, it seems necessary to feed an additional control variable (a setpoint set by hand or a display of the basic speed obtained from the card feed) into the control loop in order to achieve the desired constant pressure in the feed chute.

Together with the order of the German patent No. 2834586 has an arrangement according to this invention the features listed in the preamble of claim 1. Compared to this prior art, the new arrangement is characterized in that means are available to define or determine a target height for the fiber or flake column in the filling shaft and to determine the size of any deviation of the current column height from this target height. The controller is then arranged in such a way that it reacts to any deviation and controls the subsidy in order to eliminate the deviation.

In a second aspect, the invention relates to a device for determining the level of a flowable material. In this aspect, the invention is not limited to the use in connection with a filling shaft for fiber material, but in its preferred form, the invention is intended in this second aspect to provide a device which is used to determine the desired height and to determine any deviation in the column height can be used in a filling shaft according to the first aspect of the invention.

According to this second aspect of the invention, the device is characterized by a series of sensors, each of which is able to determine the presence of the flowable material in a certain working range of the sensor. In addition, the device contains an interrogation device to interrogate the sensors as to whether material is present in their work areas or not. A specific sensor in this row can then represent a setpoint for the level in use. Furthermore, by determining which sensors are there for the presence react to material in their work areas and which do not, the current material level is determined. Any instantaneous deviation of the current level from the target height is thus given by the distance between the current level and the sensor representing the target height.

The distance between each two adjacent sensors can be the same along the row, but in a preferred variant this distance is chosen to be smaller in the vicinity of the sensor representing the setpoint and larger in the areas further away from the setpoint. This last arrangement gives a more precise level detection and possibly also a smaller "level hysteresis" in the vicinity of the setpoint, while the whole row of sensors still covers a sufficiently large level range without a large number of sensors (with corresponding additional devices, as will be described below) desire. The number of sensors in the row will depend on usage conditions, but for a hopper on a fiber processing machine, 6 to 10 sensors will normally meet the requirements.

The sensors can be light barriers. Each light barrier can be designed as either a one-way barrier (without reflector) or a two-way barrier (with reflector). The sensors can be switched individually or in groups in order to sense the presence of material in their respective work areas, so that cross talk between the sensors is avoided. The sequential switching of the sensors can be controlled by the interrogation device.

The interrogation device can be arranged in such a way that it regularly repeats a specific interrogation cycle, with each sensor being interrogated within each cycle. The results for each cycle are then sent to an evaluation, where a possible deviation of the current level from the target level is determined. The size of this deviation is made available to the controller as a suitable signal.

The evaluation can be designed in such a way that it can react differently to a change in level indicated by the sensors in one direction than to a change in level indicated by the sensors in the other direction. For example, the evaluation can be arranged in such a way that, under predetermined circumstances, it immediately and fully passes on an "apparent" level change in one direction (eg downward), while only delaying an "apparent" level change in the other direction (eg upward) and / or to a reduced extent in a predetermined manner. In this way, "deceptive effects" can be reduced.

If the row of sensors is viewed from the bottom up, the first sensor that shows no material in its working area can be accepted as determining for the apparent current level.

In combination with a filling shaft according to the first aspect of the invention, a signal representing the deviation can be processed by the controller according to a predetermined control function in order to generate an output signal supplied by the controller. This variable output signal can be used as a reference variable (setpoint) for serve another controller, which regulates the funding directly. As in the case of DE-PS 2834586, a feed roller (or a pair of feed rollers) can serve as funding for flakes stored in a feed shaft. The controller can then regulate the speed of the feed roller.

The said control function can preferably be changed as a function of the size of a detected deviation between the desired and actual heights of the fiber column. The entire measuring range defined by the height measuring device can e.g. be divided into two or more (preferably four) zones, a respective control function being determined for each zone. However, each control function is preferably a so-called PI function.

• Fig. 1 Eine Seitenansicht eines Füllschachtes gemäss dem Stand der Technik,
• Fig. 2 eine Modifikation des Schachtes von Fig. 1, so dass der umgebaute Schacht gemäss dieser Erfindung arbeiten kann,
• Fig. 3 ein Blockdiagramm des Regelkreises, welcher gemäss der in Fig. 2 dargestellten Modifikation gebildet wird,
• Fig. 4A eine Seitenansicht von einem Teil des Schachtes der Fig. 2 mit einer darin gebildeten Faser­ säule und einem Höhenmessorgan zur Ermittlung der Höhe dieser Säule,
• Fig. 4B einen Querschnitt des gleichen Schachtes,
• Fig. 5 eine Anordnung von Sensoren eines Höhenmessor­gans,
• Fig. 6 ein Diagramm zur Erklärung von einem vorge­schlagenen Verfahren zur Ermittlung der Säulen­höhe mit einer Anordnung gemäss Fig. 5,
• Fig. 7 ein Blockschaltbild eines Höhenmessorgans,
• Fig. 8 eine Auswertungseinheit zur Auswertung von Ausgangssignalen des Höhenmessorgans,
• Fig. 9 ein Diagramm zur Erklärung einer gedachten Ein­teilung des vom Höhenmessorgan abgedeckten Messbereiches,
• Fig. 10 ein Ablaufdiagramm für den in Fig. 3 gezeigten Regler,
• Fig. 11 ein Diagramm einer möglichen Variante des Regelkreises,
• Fig. 12 ein Diagramm zur Erklärung der bevorzugten Anordnung der Sensoren des Höhenmessorgans.
• Fig. 13 zeigt eine Variante der Anordnung von Fig. 8

As examples, some embodiments according to the invention will now be explained in more detail with reference to the drawings. All figures are schematic. It shows:

• 1 is a side view of a filling shaft according to the prior art,
• FIG. 2 shows a modification of the shaft of FIG. 1, so that the modified shaft can work according to this invention,
• 3 is a block diagram of the control loop, which is formed according to the modification shown in FIG. 2,
• 4A is a side view of part of the shaft of FIG. 2 with a fiber formed therein column and a height measuring device to determine the height of this column,
• 4B shows a cross section of the same shaft,
• 5 shows an arrangement of sensors of an altimeter,
• 6 shows a diagram for explaining a proposed method for determining the column height with an arrangement according to FIG. 5,
• 7 is a block diagram of an altimeter,
• 8 shows an evaluation unit for evaluating output signals of the height measuring device,
• 9 is a diagram for explaining an imaginary division of the measuring range covered by the height measuring element,
• 10 is a flowchart for the controller shown in FIG. 3,
• 11 is a diagram of a possible variant of the control loop,
• Fig. 12 is a diagram for explaining the preferred arrangement of the sensors of the height measuring element.
• FIG. 13 shows a variant of the arrangement from FIG. 8

The filling shaft 20 shown in FIG. 1 will only be described briefly, since it was explained in detail in the "mittex" publication mentioned above is. In operation, the shaft 20 is supplied with fiber flakes from machines upstream in the blowroom line via a pneumatic feed channel (not shown). Flakes are excreted from this channel in a separating head 22, so that they flow downward together with transport air into a so-called feed shaft 24. A wall 26 of the feed shaft is formed by a perforated plate, so that the transport air can flow through this wall 26 into a calming space 28 and from there an exhaust air housing 30. The flakes themselves cannot pass through the holes in the perforated plate and form a wadding (not shown) in the feed shaft 24 above a pair of feed rollers 32.

The feed rollers 32 can be driven by a motor 34 in order to deliver material from the wadding formed in the feed shaft 24 to an opening roller 38 driven by the motor 36. The feed shaft therefore serves as a reserve-forming shaft. The construction of the roller 38 and the guide plates surrounding it have been dealt with in the patent applications mentioned at the outset and will therefore not be described further here.

The material conveyed by the feed rollers 32 and opening roller 38 falls as smaller flakes or as individual fibers into a so-called feed chute 40, where it forms a fiber or flake column (not shown) above a pair of draw-off rollers 42. Material from the lower end of this column, not shown, can be supplied through the take-off unit 44 comprising the rollers 42 to the feed cylinder 46 of a card, not shown. Even though The filling shaft 20 shown in FIG. 1 is designed in particular for card feeding, an essentially identical shaft can be used for feeding fiber material to other machines in a blowroom line. For this purpose, the shaft parts are mounted in a housing 48 in the arrangement already described.

In this known arrangement, a lower light barrier 50 and an upper light barrier 52 are provided to regulate the column height in the feed shaft 40. The intended mode of operation of these light barriers in cooperation with the motor 34 has already been indicated in the introduction to this patent specification and has been completely described in the article from the specialist press mentioned at the outset, so that a further description can be dispensed with here. According to an embodiment of this invention, the known height control system of FIG. 1 is to be replaced by the new system of FIG. 2.

The feed chute 24, feed chute 40, feed roller pair 32 and opening roller 38 shown in FIG. 2 remain unchanged and are therefore provided with the same reference numerals as in FIG. 1. The drive motor 34A for the feed roller pair 32 must enable a constant or continuous speed regulation instead of simply switching between slow and fast speed in the arrangement according to FIG. 1. Instead of the light barriers 50, 52, a height measuring element 54, which is described in more detail below, is now provided, which outputs signals delivers to an evaluation 56. A controller 58 receives output signals from evaluation 56 and responds to control the speed of motor 34A. A control loop is thus formed, which can have the form shown in FIG. 3.

The control path of the control loop includes the feed shaft 24 with the lower end of the wadding 60 (FIG. 2) and the feed shaft 40 with the fiber or flake column 62 (FIG. 2), the height, i.e. the position of the uppermost surface 64 of the column 62 in the feed chute 40 should be regulated. A material outflow MFa takes place from the lower end of the shaft 40 and cannot be influenced by the control loop itself. Between the feed shaft 24 and the feed shaft 40, a material flow MFe takes place, which takes place within the control loop and can be controlled by the control loop via the control device (ie via the actuator in the form of the feed roller pair 32 and the actuator in the form of the motor 34A) in order to keep the column height in the feed chute 40 as constant as possible.

The measuring device of the control loop is formed by the height measuring element 54 attached directly to the feed shaft 40 together with the associated signal evaluation 56. The evaluation 56 contains a first unit 56A, which emits a signal h representing the current current column height (the column level) and a second unit 56B, which carries out a comparison between this actual height and a predetermined target height, by a possible height deviation generating signal e.

In the example of FIG. 3, controller 58 in FIG. 2 is formed by two elements, namely by a microprocessor 58A and a motor controller 58B. The latter must be adapted to the motor 34A and can be of an ordinary construction, so that another Description can be omitted here. The microprocessor 58A processes the signal e in accordance with a predetermined control algorithm in order to output a signal n representing the setpoint for the motor controller 58B.

It will be clear to the person skilled in the art that the division of this control loop into various elements, in particular with regard to the evaluation 56 and microprocessor 58A, has been carried out in part for the purpose of a complete description. In practice, thanks to modern electronics, many operations can be carried out in a single component (chip).

Height or level measurement

Examples of a height measuring device for use in a shaft arrangement according to FIG. 2 will now be described with reference to FIGS. 4 to 7. As a first step, the position of the height measuring device in relation to the feed chute 40 will be explained with reference to FIG. 4, which at the same time will clarify certain requirements for the device itself.

4 includes a main diagram 4A and a sub-diagram 4B. In the side diagram, FIG. 4B, the feed chute 40 is viewed in cross section from above in order to show the position of the device 54 approximately in the middle of a long side of the rectangular cross section. The device 54 contains a number of sensors, each of which is formed as a light barrier in this example. In the example shown, each light barrier is designed as a two-way device, with a transmitter-receiver unit in a housing 54A on one side of the feeder shaft 40 and a reflector 54B on the opposite side Manhole side. The light barriers of the device 54 could, however, be formed as one-way devices so that each contains a transmitter element on one side of the shaft and a receiver element on the opposite side of the shaft, the reflector 54B then being omitted.

The housing 54A in FIG. 4A is mounted on a transparent plate 66, and the latter is mounted in the side wall 68 of the shaft so that the light beams of the barriers can be transmitted across the width of the shaft to and from the reflector 54B. As will be explained in more detail later in connection with FIGS. 5 to 7, the individual sensors (not shown in FIG. 4) are arranged in a vertical row. A predetermined position within this row (preferably approximately halfway between the upper and lower ends thereof) represents a target level SN (a target height). This target level SN is preferably as far as possible from the take-off rollers 42 (FIG. 1) (column height as large as possible) ) without taking the risk of overfilling the shaft 40 to the opening roller 38. A suitable distance between the desired level SN and the outer surface of the opening roller 38 is in the range 200 to 300 mm, the said outer surface containing all rotating parts of the opening roller 38 (including the clothing).

The target level SN can be clearly defined or determined. Determining the current level (represented in simplified form by surface 64 in FIG. 2) is a relatively complicated operation, which requires a determination process. Certain problems with this determination can be seen from the representation of the fiber column 62 in FIG. 4A and the schematic representation of the newly arriving flakes 70.

As shown schematically in FIG. 4A, the uppermost surface of column 62 will never form a horizontal plane, and new flakes 70 will normally be present above this column surface and at the same time within the total measuring range of the height measuring device 54. It will therefore be clear, firstly that the device 54 cannot sense a "more precise" or "absolute" current level (because there is no such level) and secondly that any "deceptive effects" must be taken into account, which come from newcomers nearby existing flakes can be caused on the column surface.

A suitable method for overcoming these difficulties will now be explained using the diagrams of FIGS. 5 and 6.

5 shows a vertical row of n sensors which have been numbered successively from bottom to top. For the sake of clarity, each sensor is shown as a one-way light barrier, with respective transmitter elements S1 to Sn and corresponding receiver elements E1 to En. FIG. 5 shows the two lowest light barriers 1 and 2 and the three uppermost barriers n-2 to n The target height or target level SN (FIG. 4A) lies somewhere between the barriers 2 and n-2. The fiber column should therefore normally cover the receivers E1 and E2 with respect to the correspondingly assigned transmitters S1 and S2, while the receivers EN-2 to En should be left blank with the correspondingly assigned transmitters Sn-2 to Sn. However, as already indicated in connection with FIG. 4A, one or the other or each light beam from these upper light barriers can be replaced by new ones Fiber flakes 70 (Fig. 4A) are interrupted.

An imaginary "work area" is defined for each sensor by the path of the light beam from the transmitter to the receiver (of the same sensor). If there are flakes in this "work area", the light beam is interrupted.

The sensors can be queried continuously or at any intervals for their respective "states", i.e. whether their respective light beams have been interrupted (state - covered) or not (state - free). With such a detection, the column height must be lower than the working area of the first free barrier from the lower end of the row. The position of the adjacent, deeper, covered sensor can be referred to as the "apparent" current level.

Assuming that the light barriers are not queried for their states sequentially, but sequentially (according to a repeatable polling cycle), a certain "apparent" level can be determined for each polling cycle by evaluating the output signals from the light barriers. Possible results of such an evaluation are shown for 13 successive query cycles in the bar diagram of FIG. 6. In this diagram, each bar represents the result of a query cycle, and the bar height indicates the number of light barriers that are covered by fiber material, as seen from the bottom end of the row, ie between the bottom end of the row and the first free light barrier.

The vertical axis of the bar chart is divided according to the sensor numbering, whereby the same distances between the sensors are assumed.

As will be explained in more detail below, the column height determined by the evaluation is not necessarily the same as the apparent column height represented by the bar height. The processing of the height signals within the evaluation depends on the direction of any changes in height. The resulting column heights are indicated in Fig. 6 by the dashed line.

For the sake of simplicity, it is assumed that in the first interrogation cycle in FIG. 6, the determined value is set equal to the apparent value. In the given example, both values are given with four "units" (sensors covered from the bottom end of the row). In the second polling cycle, the apparent value drops to two units. From a consideration of FIGS. 4A and 5, it will be clear that the current level can in no case be above the apparent level. The determined value also corresponds to the apparent value in the second query cycle.

In the third polling cycle, the apparent value increases to five units. It becomes clear from the consideration of FIGS. 4A and 5 that the current level can very well be below the apparent level, so that an increase in the apparent level between two successive query cycles cannot be accepted as "valid" without further ado. The evaluation therefore passes on an increase, but not fully, but to a reduced extent according to a predetermined function.

6, the bar diagram and the dashed line diverge accordingly during the third polling cycle. According to the dashed line, a column height of two units is determined.

In the fourth polling cycle, the apparent level goes up further, which causes a further increase in the dashed line (and the correspondingly determined value for the column height). In the fifth cycle, the apparent level comes down slightly again without bringing the bar chart and the dashed line together again. The latter happens in the sixth polling cycle when the apparent value falls below the determined value again. The implementation of this step will be described below in connection with FIG. 8.

The query cycles 7 to 10 show that periodic, smaller fluctuations in the column height are practically not passed on by the evaluation, since they are "smoothed" by the delay emanating from the determination function. As shown for the last three polling cycles in FIG. 6, the apparent value is obtained from the determined value after a certain time delay if it remains at the top.

FIG. 6 shows that a step-like or step-like increase in the signal at the input of the evaluation causes a step-like increase in the level (represented by the signal h). As will be described in more detail below, this result can be brought about by a digital low-pass filter, the filter characteristic being set such that a stu fen-shaped increase in the input signal only after a predetermined number of polling cycles (time delay) is fully reflected in the determined level (in FIG. 6, after 4 polling cycles - see cycles 11, 12 and 13). Thus, while the signal x only accepts discrete values (corresponding to a number of sensors), the signal h can assume any values caused by the filtering.

FIG. 7 shows further details of a height measuring device which can work according to the system described in connection with FIGS. 5 and 6. The device contains eight one-way light barriers with transmitters S1 to S8 and respective corresponding receivers E1 to E8. The light barriers are operated in multiplex mode so that they are switched on in sequence, with each light barrier being switched on once within a single polling cycle. The period of the polling cycle is so short that movements of falling flakes within a polling interval can be neglected. The light barriers are switched on in order to avoid mutual interference. If such mutual interference can be avoided in other ways (e.g. by suitable modulation of the light signals or by color filtering), the light barriers can be switched on at least in groups at the same time or even continuously. In the given example, the polling interval is determined by the multiplexer 72. In the case of simultaneous activation, a polling interval could be determined differently, e.g. by a suitable clock.

During a specific polling interval, the status signals from the receivers E1 to E8 are transmitted via a Data line 74 read into a shift register 76, the read operation being controlled by signals on a trigger line 78 from multiplexer 72. At the end of the query interval, there are data relating to all eight receiver states on the outputs 80 of the shift register 76. The state data can be read out by an output gate 84 by means of a suitable strobe signal on line 82, so that the corresponding information is still available to them Descriptive evaluation unit 56A (Fig. 3) can be forwarded via a connector 86 (Fig. 7).

evaluation

The evaluation unit 56A is shown again schematically in FIG. 8, the scheme being chosen less to represent reality than to explain the operations carried out. Unit 56A includes an input stage 88 which receives the data from shift register 76 (Fig. 7). From this, the unit 56A derives the apparent level by determining the number of light barriers between the lower end of the row and the first free light barrier and supplies a corresponding signal x to an output stage 90.

The signal x can therefore represent discrete values from 0 to 8. For the input stage 88, the lower end of the row, ie sensor no. 1 (and not the desired height), serves as the reference level, everything below this reference level being regarded as "0". The signal x thus corresponds to the distance between this reference level and the lowest point within the measuring range, where no material is detected, for example if sensor 1 itself does not measure determines the signal, the signal x remains at "0" - In general, the signal x corresponds to the number (L-1), where L is the number of the first free light barrier from the lower end of the row.

The output stage 90 processes the signal x into an output signal h, the type of processing depending on the development of the signal x over time (on its "history").

• -1. die direkte Weiterleitung des Signals x (unverändert) als Signal h,
• -2. Tiefpassfiltrieren des Signals x und Weiterleitung des filtrierten Signals als das Signal h.

In the example represented by FIG. 8, two "processing methods" are provided:

• -1. the direct forwarding of the signal x (unchanged) as signal h,
• -2. Low pass filter the signal x and pass the filtered signal as the signal h.

In practice, all operations performed in the evaluation unit 56A are carried out by the software of a microprocessor. A corresponding “hardware solution” is shown in FIG. 8 for the purpose of illustration and has been described below.

The direct forwarding is represented by the signal path (bypass) 91, while a second signal path comprises a low-pass filter 89. A controllable switch 87 sends the signal x to either the bypass 91 or the filter 89.

The switch 87 is controlled by a comparison element 93, which compares the instantaneous signal x with the signal h just given previously (a memory - not shown - for the signal h can be between the filter 89 and the comparison element 93 are switched on).

If element 93 determines that signal x is below or equal to previously output signal h, switch 87 is controlled so that signal x is provided to bypass 91. Conversely, if element 93 determines that signal x is higher than previously output signal h, switch 87 is controlled so that signal x is provided to filter 89.

The evaluation unit 56A thus receives a signal from the device 54 which corresponds to the apparent level determined by the device 54. The unit 56A supplies a signal h which corresponds to the determined level. The output signal h of the unit 56A changes as a function of the signal supplied by the device 54, the function itself being changeable depending on the "behavior" of the input signal. In the given example, this function depends on how the instantaneous “signal level” of the input signal (in discrete values 0 to 8 of the signal x) behaves in relation to a previously determined output signal h. The function can be adjusted between two forms (low pass filtering and unchanged forwarding) according to the input signal, i.e. in one case the function corresponds to a 1: 1 reproduction of the input signal.

However, the invention is not restricted to the variant listed as an example. For example, it could prove advantageous to filter both a decreasing and an increasing input signal, but with different cutoff frequencies. Where the determination of the actual height requires a change in the signal x representing the apparent height, this change can be carried out by an operation other than filtration. For example, empirical values for the "real" meaning of "level jumps" can be determined by tests and entered in the programming of the evaluation. As a further variant, "average values" of some combined query cycles could be passed on.

Changes in the processing function could e.g. caused by changes in the input signal alone, rather than by changes in the input signal compared to the output signal (in which case the filtering must be changed accordingly). The main difference from the example given would be seen in the behavior of the system when (slowly) descending from a peak of the apparent level, e.g. in cycles 4, 5 and 6 in FIG. 6, in that the descending tendency alone is sufficient to bring the unchanged forwarding of the input signal back into effect. The function can take more than two forms, and additional forms would have to bring significant benefits to justify the complications involved.

In general, the optimal evaluation can be determined empirically (by considering the system's response to various level fluctuations). In any case, the evaluation should recognize a rising tendency in its input signal and only pass it on modified.

In the comparison element 56B (FIG. 3), the signal h the determined height shown is compared with a predetermined target height and an output signal e is supplied in order to represent any deviations. This output signal comprises two components, namely a directional component ± and a variable. The signal is processed in accordance with the control algorithm of the microprocessor 58A, as will be described below in connection with FIGS. 9 and 10. For the time being, however, certain possible operating states in the controlled system will be dealt with.

Zones within the total measuring range

The control according to this invention works preferably at least during normal production operations without operator intervention. This means that the regulation is not provided with any information about the target production of the card (or other machines to be supplied). The regulation must therefore also function if the supply shaft 40 (FIG. 1) is completely empty at the start of its work (start of work). Furthermore, the control not only absorbs general fluctuations during normal operation, but also compensates for the effects of a new adjustment of card production carried out by the operator.

Of course, these different operating states must be taken into account when designing the overall system. For this purpose, it would be possible to include various "setting conditions" as reference variables in the control system. However, the system is preferably designed in such a way that it works "self-regulating" during normal production operation, that is to say that it finds its own level, no matter (within certain limits) how the upstream and downstream machines are set.

To make this possible, the control algorithm (the control function) itself is determined as a variable function of the deviation from the target level. The change in the control algorithm (the control function) can be carried out in stages, so that the measuring range defined by the height measuring device is divided into several zones, each zone being assigned to a predetermined control algorithm. The control algorithms of neighboring zones are always different, but zones that are not adjacent to one another can be assigned to the same algorithm.

A corresponding division of the measuring range is shown schematically in FIG. 9. The vertical line corresponds to the entire measuring range covered by the height measuring device. The cross lines NAZ max and NAZ min correspond to the upper and lower limits of a "normal working zone", ie during normal operation, level fluctuations within this zone must be expected. A first ("normal") control algorithm is assigned to this zone. The section above the normal working zone (NAZ) is referred to as the upper working zone (OAZ) and is assigned to a second control algorithm in order to break down the accumulated flakes more quickly. An overfill protection device (not shown) is provided above or at the upper end of the work zone OAZ in order to switch off the supply of flakes if the control system can no longer counteract a further increase in the column height.

The cross line NP represents a "relative zero point", so that a lower working zone UAZ is defined between point NP and point NAZ min. This working zone UAZ is also assigned to a control algorithm which differs from the algorithm of the zone NAZ. The algorithm of the lower work zone can be equated with that of the upper work zone or can differ from the latter.

Below the relative zero point NP there is an "idle zone" (LZ) which is assigned to a further algorithm which differs from that of the lower working zone UAZ. Normally, the current column height should only be in (or below) the idle zone LZ when the shaft is to be filled (new start of work) or to be emptied. The corresponding control algorithm can be determined in such a way that the shaft is filled up particularly quickly if the flake feed is regulated according to this algorithm. In order to let the shaft run empty, the flake feed can be switched off, so that the control loop is no longer able to compensate for the material flow from the lower end of the shaft by feeding from above.

The cross line SN represents the target level, which of course lies within the normal working zone NAZ. By evaluating the two components (size and direction) of the deviation signal e, it is possible to determine in which zone the level (represented by the signal h) is currently located.

As already indicated by the word "algorithm", Nowadays the processing of the deviation signal to a target speed signal can be carried out by a microprocessor. A flowchart for a corresponding loop (routine) in the programming of this microprocessor is shown in FIG. 10.

Box 100 in FIG. 10 represents an initialization step which must be carried out when the shaft or the control system is switched on again in order to determine the initial states. This step automatically sets a predetermined speed of the motor 34A (e.g. 30 to 50% of the maximum speed of the motor).

For the time being, a description of the steps shown with dashed lines is omitted, i.e. it is provisionally assumed that the system immediately proceeds to determine the actual height (the actual level) represented by box 102. According to the step shown in box 104, it is then determined whether the actual height is greater than the maximum height (NAZ max) of the normal working zone, i.e. whether the actual height is in or above the upper working zone OAZ. In the latter case, the deviation signal is processed according to a first control algorithm A1, as indicated in box 106 in FIG. 10. This step determines a new base speed (target speed) of engine 34A and the corresponding data is stored, the final step being represented by box 108.

The stored value is now output in the form of a target speed signal N (box 110). After the total touch time (the duty cycle), as by the Box 112 is indicated, the loop leads back to the new determination of the actual height (box 102), or to the steps indicated by the dashed lines, which will be dealt with later in this description.

If step 104 determines that the actual height is not in or above the upper zone OAZ, step 114 first determines whether the actual height is within the normal working zone (between the normal maximum and minimum heights NAZ max or NAZ min ) lies. In this case, the deviation signal is processed according to a second control algorithm A2 (box 116) in order to determine the new basic speed, after which the loop continues through steps 108, 110 and 112 described above.

If it is determined by step 114 that the actual height is below the normal minimum height NAZ min, the last branching step 118 determines whether the actual height is within the lower working zone UAZ, i.e. lies between the normal minimum height NAZ min and the zero point NP. In this case, the deviation signal is processed according to a third control algorithm A3 (box 120), after which the processor proceeds to steps 108, 110, 112. If it is determined by step 118 that the actual height is below the zero point (i.e. inside or below the idle zone LZ), the deviation signal is processed according to a fourth control algorithm A4 (box 122) in order to determine the new basic speed.

• A1: N = S0 + Fo.e - Exo
• A2: N = S0 + FN.E
• A3: N = S0 + Fu.E + Exu
• A4: N = Fl(t),

wobei die verschiedenen Symbole die folgenden Bedeutungen haben:
S0 - aktuelle Grunddrehzahl
N - Solldrehzahl des Speisewalzenmotors 34A
e - Höhendifferenz ("Abweichung" - Sollhöhe minus Isthöhe)
F0 - Regelparameter für die obere Arbeitszone OAZ
FN - Regelparameter für die normale Arbeitszone NAZ
Fu - Regelparameter für die untere Arbeitszone UAZ
Fl(t)- eine Regelkennlinie für die Leerlaufzone LZ
Exo - ein Zusatzabschlag bei deutlich zu hoher Dreh­zahl
Exu - ein Zusatzzuschlag bei deutlich zu kleiner Drehzahl.The following control algorithms are suggested as examples:

• A1: N = S0 + Fo.e - Exo
• A2: N = S0 + FN.E
• A3: N = S0 + Fu.E + Exu
• A4: N = Fl (t),

where the different symbols have the following meanings:
S0 - current base speed
N - target speed of the feed roller motor 34A
e - height difference ("deviation" - target height minus actual height)
F0 - control parameters for the upper working zone OAZ
FN - control parameters for the normal working zone NAZ
Fu - control parameters for the lower working zone UAZ
Fl (t) - a control characteristic for the idle zone LZ
Exo - an additional discount if the speed is clearly too high
Exu - an additional surcharge if the speed is significantly too low.

) dargestellt werden, wobei der Regelalgorithmus durch Anpassung der Komponenten K und Tn an verschiedene Arbeitsbedingungen angepasst werden kann.The controller therefore normally works according to a control algorithm of the general form: N = eF + S0, the control parameter F being different from zone to zone. The controller preferably works as a PI controller, and thus the control parameter F can be determined by the equation F = K (1 + $\frac{\text{T0}}{\text{Tn}}$

) are shown, whereby the control algorithm can be adapted to different working conditions by adapting the components K and Tn.

The control parameter FN within the normal working zone NAZ is preferably independent of the size of the deviation e, so that the components K and Tn are determined as constants for this zone. In the upper and lower work zones OAZ and UAZ, however, each can because control parameters F0, FU are set in proportion to the deviation e by appropriate adaptation of the components K and Tn, ie in these zones K and / or Tn is a function of e.

If the actual height is below the zero point NP, the target speed N is no longer determined by processing the deviation signal e, but directly from a characteristic curve Fl (t). The symbol t indicates that the target speed N is a function of time, so that the longer the actual height remains below the zero point NP, the higher the target speed N is set. The characteristic curve itself can be made dependent on the basic speed and / or the sink rate, e.g. the slope of the characteristic can be changed as a function of one or both of these parameters.

The microprocessor switches on a time measurement process. when the actual height drops from the lower working zone UAZ to the idling zone LZ, and the determination of the target speed N is made dependent on the subsequent measured time according to the characteristic curve Fl (t). If this time measurement method is not stopped within a predetermined interval by the return of the actual height to the lower working zone UAZ, the microprocessor issues an error message "shaft empty", after which the card can be switched off. A corresponding time measurement method can be used to determine the sink rate mentioned by determining the expiry time for lowering the actual height through predetermined intervals within the lower working zone, after which, for example, the slope of the characteristic curve can be adjusted accordingly.

Additional measures

The additional step indicated by dashed lines in Fig. 10 will now be described. Branch step 124 determines whether the carding machine is operating at a normal speed S0 = Sn or a "creeping speed" S0 = SK. If the card works at a normal (production) speed, the control runs as already described. However, the card is sometimes switched to a low (creep) speed (e.g. to start a belt break) to make operation easier. If this state is set, the normal base speed Sn can be replaced by a "creep speed" SK via the branch 124. The current basic speed S0 used to determine the target value N is then suddenly increased to a value corresponding to the creeping speed of the card, wherein a large fluctuation in the column height over a certain period after the changeover to creeping speed can be avoided.

If the card is set to a normal production speed again, the creep speed SK can be replaced by the last stored value of the normal basic speed Sn. This measure enables a malfunction to be eliminated, which could otherwise be caused by foreseeable operating conditions outside the normal machine sequence.

The arrangement shown in FIG. 11, which provides a so-called disturbance variable feed-in, serves a similar purpose. The control loop shown in Fig. 3 influences only the speed of the feed rollers 32, which should lead to a change in the material flow MFe. However, this material flow is dependent on other sizes, for example on the density of the material stored in the shaft 24. In order to eliminate disturbances caused by density fluctuations, the lift of the feed rollers 24 can be measured by suitable means (not shown), a corresponding signal S (FIG. 11) can be generated and combined with the signal generated by the microprocessor 58A in order to give a corrected nominal value N. . The signal i representing the lift (FIG. 11) can be processed, for example, by forming the reciprocal value (box 126 in FIG. 11), and the output signal of the device 126 can be adapted by a proportional factor in the device 128 to form the signal S. Signal S can then be multiplied by the output of microprocessor 58A. If a linearization function can be built into the formation of the signal S, the multiplication point 130 (FIG. 11) can be replaced by an addition point.

12 shows the preferred arrangement of eight light barriers in order to optimally cover an entire measuring range. This total area is again represented by a vertical scale, as in FIG. 9. The length of this scale has been given as 150mm as an example, but the arrangement is not limited to this example. The cross lines on this scale represent the positions of the individual light barriers, these barriers being identified from number one to one after the other as numbers 1 to 8. In this arrangement, light barrier No. 4 represents the target height or the target level, which for example is 95 mm from the lower end of the scale (0mm) can be set. This figure shows that the light barriers above and in the immediate vicinity of the desired height are separated from one another by a relatively small distance A (in this example of 10 mm) and that the corresponding distance increases further up and down. It is emphasized that these precise distances have only been given as examples to illustrate the principle of the increasing distance.

The sensor arrangement according to FIG. 12 can be divided into zones according to FIG. 9 as follows:
Zone LZ - below sensor 1
Zone UAZ - sensor 1 to below sensor 2
Zone NZ sensor 2 to below sensor 7
Zone OAZ - sensor 7 to below sensor 8
(Overfill protection - sensor 8)

The "sensor density" is therefore highest in the normal working zone. There is also a relatively high "density" just above the target height. The concentration of the sensors in the NAZ zone is intended to maintain the actual level within this zone. The higher concentration above the target height is advisable because of two circumstances:
- firstly, because the regulation cannot influence the removal of fiber material from the shaft, so that a tendency to overfill is more delicate for the regulation than a tendency to run empty,
- secondly, because the target height is in any case set as high as is practically possible and a tendency to overcrowding must therefore be counteracted, which is due to the additional information (finer division of the measuring range above the target height) is facilitated.

The proposed arrangement aims for optimal use of a limited number of sensors. The goals mentioned could of course also be achieved by increasing the number of sensors, but this would considerably increase the overall effort (not only for the sensors themselves, but also for the corresponding downstream elements for signal processing).

Modifications

The invention is not restricted to the details of the illustrated embodiments. In particular, it is not dependent on the use of light barriers. Other sensors, e.g. Ultrasonic transmitter / receiver units can be used. Where the actual height is to be determined by light radiation (which in this context includes the infrared and UV ranges), it is not essential to the invention to use "discrete" signals (generated by individual sensors). The actual height could e.g. can be determined by a so-called image analysis of the entire measuring range. Where individual sensors are needed, the arrangement (array) can be more complex than the simple series of the example described. The complexity must of course have a corresponding advantage, e.g. bring with it greater accuracy by determining an average value.

The sensors of the height measuring device should have the same sensitivity as far as possible. In connection with a shaft depth (distance transmitter - The receiver or transmitter reflector) of approx. 190mm can be arranged in such a way that a thirty percent gray filter does not interrupt the beam path (minimum range of 350mm). Each receiver unit can be equipped with its own amplifier and threshold switch.

The determined height values of the described versions are digital values, i.e. the system can only take into account predetermined, coded height values. However, they are also discontinuous values because they are determined according to a periodically repeated query cycle. If the sensors are switched on continuously, which requires an adapted evaluation, continuous digital values can be determined accordingly and taken into account by the controller.

The use of analog signals is not out of the question, but it will usually be easier to obtain digital values. E.g. in the case of an image analysis, the actual height could be determined using a raster device (scanner).

Pneumatic sensors could be used as a further possibility, e.g. in the shaft wall. The air flow through small nozzles could be used to determine the actual height.

Compared to the prior art (pressure regulation, eg according to DPS 2658044), a level regulation according to this invention has the following advantages:
- no compressor (fan) is necessary to achieve the required operating pressure in the supply shaft (in Fig. 1, manhole part 40) to build; there is therefore no pre-compression of the cotton pad in the shaft (such pre-compression has to be relaxed again until it is drawn into the card, which requires large, uncontrolled distortions; in addition, the pre-compressed material column is very sensitive to small level fluctuations, so that a higher control accuracy is required in order to avoid spinning disadvantages)
- The level regulation is based on a direct recording of the decisive size (height of the wadding column) instead of an indirect recording on the operating pressure (or the air volume); the level control therefore avoids deceptive effects such as the dependence of the operating pressure on the material type
- The regulation does not require any air-guiding elements that can impair operational safety due to the risk of clogging or contamination
- the scheme does not require any essential information from the card; the shaft can therefore produce perfect cotton wool regardless of the type of card attached.

According to a further variant, shown schematically in FIG. 12, the filtering (FIG. 8) can be replaced by a signal delay means. FIG. 12 again represents a hardware solution, whereas in today's practice a software solution in the programming of a microprocessor would be preferable.

Fig. 12 shows time delay elements V1 to Vn between the inputs E1 to En (only three inputs shown) and an evaluation stage 88A, which performs the same operation as the input stage 88 of the variant of Fig. 8, namely the determination of the number of light barriers between the lower one Row end and the lowest free light barrier in the row.

Each element V1 to Vn is assigned a respective redirection U with two switches S, with each switch S reacting to a signal change in the sense of "exemption" of the corresponding light barrier in such a way that the signal comes from the respective output of the shift register 76 via the respective redirection U. the evaluation stage 88A is passed on, ie without delay. In the event of a signal change in the sense of "covering" the corresponding light barrier (by flocculation), the switches S react in such a way that the signal is passed from the respective output of the shift register 76 to the respective delay element V1 to Vn, and then only after the predetermined delay time has elapsed the evaluation stage 88A is forwarded.

The effect of this arrangement is that the "exemption" is immediately communicated to the evaluation stage, while the "coverage" arrives only after a delay, which means that an "increase" in the apparent level is not noticeable until after a certain delay at stage 88A can make a "decrease" in the apparent level immediately.

Accordingly, a temporary increase in the apparent level with a duration shorter than the predetermined delay has no influence on the value determined by the stage 88A, since the (subsequent) "decrease" coincides with or even before the "increase" in the stage 88A reports, and the latter outputs an output signal h1, which only takes into account the lowest free light barrier in the row.

Furthermore, if a signal change in the sense of "coverage" on a particular input, for example E1, is overtaken by a signal change in the sense of a "release" on the same input before the time delay expires, then the coverage at stage 88A cannot change at all Make it noticeable, since the "release" and the switch as well as the time delay element are reset, thereby completely suppressing the short coverage. Each element V1 to Vn can be formed as a counter, the counter being prompted by a signal change in the sense of a cover for counting clock pulses and only emitting an output signal when a predetermined number of counting pulses is registered have been.

The elements V1 to Vn may have different time delays, shorter delays (approx. 1 second) close to the target level and longer delays (2-3 seconds) towards the end of the row being advantageous.

An "overfill protection" can be provided in that a "shaft vol" signal (all light barriers covered) triggers a stop signal after a time delay predetermined by an element US. If the "shaft vol" signal disappears from the output of stage 88A within this delay, the element US is reset and no stop signal is output.

In general, it has proven advantageous to respond to a falling level with a different intensity than to an increasing level. A zone of the measuring range (Fig. 9/10) can therefore be assigned two (or more) different control algorithms, one taking effect when the level falls and the other when the level rises. The decrease or increase in the level can be determined by comparing the currently determined value with a previously stored value and the corresponding control algorithm can be selected. Above the target level, it is possible to react relatively strongly to an increasing level (reducing the speed), but to reacting to a decreasing level relatively weakly (reducing the speed). Below the target level, a rising level can be relatively weak, a falling level relatively strong, in both cases in the sense of an increase in speed be reacted to.

From these remarks it will be clear that the response should be relatively weak near the target level and relatively strong further away. A base speed for the feed roller (s) No can accordingly be defined for the target level. Above the target level, the current target speed N for the controller is given by N = No - AN, below the target level by N = No + AN, where AN is a function of the deviation from the target level and this function is different for a decreasing tendency than for an increasing one Tendency is chosen.

Claims (27)

1. A filling shaft for a fiber processing machine with a shaft part, in which a fiber or flake column can form during operation, conveying means for conveying fibers into this shaft part and a regulator for regulating the conveying means depending on the amount of fibers present in the shaft,
characterized,
that means (54, 56) are present to define or determine a target height for the fiber or flake column (62) in the filling shaft and to determine the size of any deviation of the current column height from this target height, the controller (58) is arranged in such a way that it reacts to a possible deviation and controls the funding (32, 38) in order to compensate for the deviation. 2. A shaft according to claim 1,
characterized,
that the conveying means (32, 38) for conveying fibers from a reserve-forming shaft part (24) is arranged. 3. A shaft according to claim 1 or claim 2, wherein the conveying means (32, 38) is arranged above the column-forming shaft part (40),
characterized in that the target height is between 100 and 400 mm (preferably 200 and 300 mm) from the conveyor (38). 4. A shaft according to one of claims 1 to 3,
characterized,
that the funding comprises a pair of feed rollers. 5. A shaft according to one of claims 1 to 4,
characterized,
that the means (54, 56) for determining a possible deviation comprises a measuring device (54) which reacts to the instantaneous column height along a certain vertical line. 6. A shaft according to claim 5,
characterized,
that the device contains a number of sensors distributed along the line, each sensor reacting to fiber material in a working area assigned to the sensor. 7. A shaft according to one of claims 1 to 6,
characterized,
that the column-forming manhole part has an elongated cross-section and the means (54, 56) for determining any deviation contains a device (54) which reacts to the interruption of a beam directed transversely to the manhole part caused by fiber material. 8. A method for feeding a uniform cotton wool to a fiber processing machine, after which a column of fibers or flakes is formed in a shaft,
characterized,
that the column height is sensed and compared with a target height, with a possible fixed posed deviation is used to regulate the fiber feed. 9. A method according to claim 8,
characterized,
that the height of the halls is scanned along a vertical line. 10. A method according to claim 9,
characterized,
that the column has an elongated cross section and the column height is determined by scanning in the transverse direction. 11. A device for determining the level of a flowable material,
marked by
a plurality of sensors (S, E), each of which is able to determine the presence of the flowable material in a specific working area of the sensor, and an interrogation device (72, 76) to query the sensors whether material is present in their working areas or Not. 12. A device according to claim 11,
characterized,
that the interrogation device (72, 76) is arranged for the sequential activation of the sensors (S, E) according to a specific interrogation cycle. 13. A device according to claim 11 or 12,
characterized,
that the sensors (S, E) are light barriers or light sensors. 14. A device according to one of claims 11 to 13,
characterized,
that the sensors (S, E) are distributed in a row along a line. 15. A device according to claim 14,
characterized,
that the distances between the sensors (S, E) in a region between the row ends are relatively small and increase towards at least one row end (and preferably against both row ends). 16. A device for generating a level-dependent signal with a device (54) which reacts to the level and which delivers a signal corresponding to the level determined by the device,
characterized,
that there is an evaluation means (56A) which receives the latter signal as an input signal and supplies the level-dependent signal as a function of its input signal, the evaluation means (56A) being arranged such that the function itself depends on how the input signal changes over time. 17. A device according to claim 16,
characterized,
that the function depends on how a current input signal behaves in relation to a previously output level-dependent signal (output signal of the evaluation means). 18. A device according to claim 16 or 17,
characterized,
that the function takes a first form when the input signal shows a rising trend and a second form when the input signal shows a decreasing or constant trend. 19. A device according to claim 16, 17 or 18,
characterized,
that the function corresponds to a low-pass filtering if the input signal shows a rising trend. 20. A method for generating a level-dependent signal as a function of a signal supplied by a device which responds to the level,
characterized,
that the function itself depends on how the delivered signal changes over time. 21. An apparatus for determining a material level within a measuring range, which is determined by a device (54) that reacts to the material,
marked by
Means (88) for supplying a signal which corresponds to the distance between a reference level and the lowest point within the measuring range, where no material is detected by the device (54). 22. A method for determining a material level within a measuring range,
characterized,
that the lowest point free of the material is scanned within the measuring range and the distance between this point and a reference level is determined. 23. A unit for level regulation with a device (54) which reacts to the level and a device (58) which defines a control function,
characterized,
that the device (54) defines a measuring range that can be divided into zones (OAZ, NAZ, UAZ, LZ) and the device (58) is arranged so that the control function depends on the zone (OAZ, NAZ, UAZ, LZ) the current level is determined. 24. An aggregate according to claim 23,
characterized,
that a normal work zone is defined with neighboring upper and lower zones. 25. An aggregate according to claim 24,
characterized,
that a further zone is defined below the lower zone mentioned. 26. An aggregate according to claim 24 or 25,
characterized,
that the level control switches off the supply of material when the level rises to the upper end of the upper zone. 27. A procedure for level regulation,
characterized,
that the control function depends on the current deviation between the target and actual level.

Images