FIELD OF THE INVENTION

This invention relates to devices in which strain is measured to determine the value of a force, particularly an applied force, typically weight. The invention is especially suited to the measurement of loads on structures not specially designed for weight measurement.

BACKGROUND ART

In conventional weight measurement, a carefully designed and machined device such as a load cell is employed, in conjunction with a load receptor such as a platform, bin, tank, hopper, silo, pan, etc.

The load cell is a costly device of high grade material such as heat-treatable high tensile strength tool steel or aluminium, machined with precision to a geometry designed to minimise the sensitivity of the cell to extraneous strain and to minimise variations in the strain field orientation at the strain sensor location or locations.

An example is what is commonly known as a single point loadcell. This type of loadcell relies on complex and costly machining to produce a parallelogram structure and general geometry intended mechanically to prevent unwanted bending and twisting forces from acting on the strain sensing devices of the loadcell.

Another example is what is commonly known as a shearbea loadcell. Again, the structure of this type of loadcell is intended to provide sensitivity only to

SUBSTITUTE SHEET

shear forces at the strain sensing location. Such a loadcell is, however, still sensitive to secondary forces at the strain sensing location despite most careful and accurate mechanical design, and complex machining. This sensitivity results in stringent requirements for accurate installation and load line maintenance. Costly components are necessary to meet the requirements.

Strain in materials can arise from a variety of stress patterns due to a variety of forces such as shear forces, bending forces, torsional forces, tension forces, compression forces, etc. Where apparatus is supported on more than one weight sensing device, or where two or more weight sensing devices are mechanically linked by a weight-receiving structure, for example through a bin, pallet, bag or baled materials, and particularly where it is required to measure weight by strain measurement on a pre-existing structure which has not or cannot be designed for the minimisation of extraneous strain, such problems are less controllable. As result, little success has been achieved in obtaining reliable and accurate load measurement in such circumstances.

Many instances exist, however, where weighing of general purpose accuracy is required, but load cells are too costly or too difficult to use. In many of these cases weighing would desirably be implemented by fitting strain sensors to existing structures or structural components, or by incorporating components that would be simple and economic to produce, and convenient and non-critical to install.

The problems of implementing such an approach arise from the multitude of sources of extraneous strain affecting the proper operation of the weight reading

arrangement, sources such as distortion of structures due to variation in loading, variation in temperature with resultant stress changes, sagging, tilting, unintended torque, bending, side loading, end loading and so on. Such influences generate secondary strain and variations in the orientation of the stress field at the location at which principal strain is measured for weight determination, such problems being of course the very reason for the existence of the various types of load cells.

A further source of inaccuracy lies in the fact that electrical strain gauges exhibit a degree of transverse sensitivity, that is to say, they respond not only to strain in the direction of the primary axis of the gauge, but also to strains which are perpendicular to that axis. In environments of relatively uncontrolled secondary strain, such transverse sensitivity will lead to unpredictable error.

Throughout this specification the force which is to be measured (typically weight) will be referred to as the principal force and the strain which is measured as a measure of the principal force will be called the principal strain. Strain other than the principal strain will be called the secondary strain. Forces other than the principal force acting on the structure or otherwise giving rise to secondary strain will be referred to as secondary forces.

SUMMARY OF THE INVENTION

We have found that in weighing apparatus in which the primary sensing axis of a principal strain gauge is orientated for response to variation in a principal force, the effect of secondary forces or secondary

strain on the principal gauge can be corrected by means of a secondary strain gauge located and orientated so that its response represents the influence of the secondary forces on the response of the principal gauge.

According to one aspect, therefore, the invention resides in apparatus in which a principal force which causes principal strain in a structure is measured by the response of a principal strain sensing means, characterised in that the apparatus further comprises means responsive to secondary strain caused by force other than said principal force, said secondary strain representing the influence of secondary force on the response of said principal strain sensing means.

According to another aspect, the invention resides in apparatus for the measurement of a load in ^~ which shear strain is measured by principal strain sensing means having primary strain sensing axes orientated for response to shear strain arising from said load, characterised in that said apparatus includes secondary strain sensing means responsive to secondary strain being strain other than said shear strain, said secondary strain representing the influence on said principal strain sensing means of secondary force being force other than said load.

In another aspect the invention resides in apparatus for the measurement of a load in which compression strain is measured by principal strain sensing means having a primary strain sensing axis orientated for response to compression strain arising from said load, characterised in that said apparatus includes secondary sensing strain means responsive to secondary strain being strain other than said compression strain, said secondary strain representing the

influence on said principal strain sensing means of secondary force being force other than said load.

In another aspect the invention resides in apparatus for the measurement of a load in which tension strain is measured by principal strain sensing means having a primary strain sensing axis orientated for response to tension strain arising from said load, characterised in that said apparatus includes secondary sensing strain means responsive to secondary strain being strain other than said tension strain, said secondary strain representing the influence on said principal strain sensing means of secondary force being force other than said load.

In another aspect the invention resides in load measurement apparatus including a load receiving body, first strain gauge means located on said body and orientated for response to shear in said body due to said load, second strain gauge means located adjacent said first strain gauge means and orientated for response to secondary strain being strain other than that due to said shear.

In another aspect the invention resides in load measurement apparatus including a load receiving body supported at spaced support regions, first and second strain gauge means located on said body and orientated for response to shear in said body due to said load, third and fourth strain gauge means respectively located adjacent said first and second strain gauge means and orientated for response to secondary strain being strain other than that due to said shear.

In another aspect the invention resides in apparatus for the measurement of a principal force on a structure, in which principal strain arising from said

force is measured at a strain measurement location on the structure, characterised in that the apparatus further comprises means responsive to secondary strain at said location arising from secondary forces, said secondary strain representing the influence of said secondary forces on the measurement of said principal strain.

In another aspect the invention resides in apparatus for the measurement of a principal force on a structure, in which the primary sensing axis of principal strain gauge means is orientated for response to variation in said principal force, characterised in that the apparatus further comprises secondary strain gauge means the primary sensing axis of which is orientated for response to variation in secondary forces, the response of said secondary strain gauge means representing the influence of said secondary forces on the measurement of said principal strain.

It is to be observed that the principles of the present invention are quite opposed to those suggested in the prior art to counter the specific effect of load position sensitivity in traditional load cells, such as are described in Australian patent 594619 of Toledo Scale Corporation and Australian patent 579135 of Reliance Electric Company. In the first of these the position of the load on a load cell is determined by the use of additional load-position sensing strain gauges, calibrated by the positioning of a known load at selected locations. In such a device there is no measurement of secondary strain, only principal strain being measured, and correction is for load position only.

A somewhat similar approach is to be found in

Australian patent 579135 where strain gauges are mounted off the centre line of a parallelogram load cell to detect off-axis displacement of the load, and correction for this is made by means of a resistance network. Again, only strain arising from the principal force is measured and compensation is only for load position.

Such prior art devices suffer from the disadvantage that the only error which is correctable is that arising from load position variation, and most importantly, secondary strain arising from secondary forces in such devices will be interpreted as a load position error, so that a spurious correction will be applied to the load indication, resulting in an increased inaccuracy of measurement. As a consequence, such devices are quite unsuitable for application outside the field of single loadcells operating in environments of insignificant secondary force.

In accordance with the present invention, the origin of secondary strain is of no importance, as the correction of the principal force measurement proceeds without regard to the nature or mix of secondary strain. It is of course necessary to apply forces of known types to a device according to the invention during its calibration, when the influence of secondary strain on the principal strain sensor is observed, but once the apparatus has been calibrated it will operate transparently.

Several embodiments of the invention will now be described by way of example.

BRIEF DESCRIPTION OF THE DRAWINGS

In the accompanying drawings,

Fig. 1 shows a load-supporting member embodying the present invention; Fig. 2 is an end elevation of the member of Fig. 1; Fig. 3 is a fragmentary sectional plan view taken on the line 3-3 of Fig. 1; Fig. 4 is an elevation view of the gauging area from the direction 4 indicated in Fig. 3; Fig. 5 is an elevation view of the gauging area from the direction 5 indicated in Fig. 3; Fig. 6 is a circuit diagram illustrating the manner of interconnection of strain gauges in the embodiment with reference to Figs. 1 to 5; Fig. 7 is a further circuit diagram illustrating the interconnection of strain gauges; Fig. 8 is a side elevation of apparatus for calibrating the device illustrated in Figs. 1 to 5; Fig. 9 is a further side elevation of the apparatus of Fig. 8; Fig. 10 is a side elevation of a modified embodiment of the invention; Fig. 11 is a side elevation of a further modified embodiment of the invention;

Fig. 12 is an isometric view of a further embodiment of the invention; Fig. 13 is a side elevation view of a further embodiment of the invention; Fig. 14 is a side elevation of a silo to which apparatus according to the invention has been applied; Fig. 15 is a fragmentary detailed elevation of a support leg of the silo illustrated in Fig. 14; Fig. 16 is a side elevation of a further embodiment of the invention; Fig. 17 illustrates a further embodiment of the

invention; Fig. 18 shows in fragmentary side elevation a further embodiment of the invention; and Fig. 19 shows the device of Fig. 18 in end elevation.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

In the embodiment illustrated in Figs. 1 to 5 a load measuring device is provided in which shear strain is measured as the principal strain. This device may be used as a substitute for a shear beam loadcell in applications where cost cannot justify the use of such a device, or where the problems of the influence of secondary force and stress described above cannot practically be overcome.

The illustrated device 10 is in the form of an I-beam comprising a web 11 consisting of a flat steel plate referred to herein as a shearplate, and a pair of flanges 12. End plates 13 are provided for attaching the device to a support and to a load-bearing structure the load on which is to be measured. The I-beam may be fabricated for the purpose, or may simply be a commercially available beam of suitable dimensions for the purpose.

As seen in Figs. 4 and 5, the web 11 carries a pair of strain gauges 15a and 15b each of which comprises a pair of sensing grids orientated to respond to shear strain, with their primary axes at 45° to the axis of the beam. Normally one would locate such gauges at the neutral axis of the beam with respect to bending, shown at 14. In theory this ensures that cancellation of bending stresses will occur when gauges are connected in a Wheatstone bridge configuration.

However in practice, it is very difficult to know precisely the location of the neutral axis of a given

structure, and furthermore the neutral axis may shift with different loading and support conditions. In accordance with the invention therefore, a first pair of secondary strain gauges 16a, 16b are located near the gauges 15a and 15b. In this case the gauges 16a and 16b are Poisson ratio gauges, and comprise sensing grids orientated to respond to bending strain in that region in the plane of the shearplate 11 and compression or tension strain. The gauging region is, moreover, located away from the nominal neutral axis 14 to obtain a larger bending strain measurement than would be obtained at the neutral axis.

Optionally, a second pair of secondary strain gauges 17a, 17b may be mounted adjacent the gauges 15, with grids orientated to respond to bending of the shearplate 11 out of its plane in response to side loading of the device. In a structure such as that illustrated, the gauges 16 and 17 are preferably mounted on a common line with the principal gauges 15 parallel to the neutral axis 14, but in more complex structures the optimum position of the secondary gauges may be ascertained by experiment, in order to obtain a response from the gauges which mirrors as closely as possible the influence of the secondary strain on the principal gauge. Photoelastic examination of a model of the structure may be used to assist in the choice of location, particularly where secondary forces produce changes in the direction of the strain field in the region of the principal gauge.

Fig. 6 shows the manner in which the strain gauges 15 and 16 may be interconnected to provide for correction of secondary stress, where only a single pair of secondary gauges 16 is used. As is conventional, the sensing grids of the gauge pairs 15a and 15b, and 16a and 16b, are connected as Wheatstone bridges 18 and 19

between excitation voltage supply rails 20, 21. The bridge 19 is connected to the source of excitation voltage through dropping resistors 22 which serve to scale its output relative to that of the principal bridge 18, and the bridges are connected together with opposite polarity as shown.

In this way, changes in the output of the secondary strain gauges 16 will oppose related changes in the output of the principal gauges 15. In a device of the kind illustrated in Figs. 1 to 5, it is found that the output of the shear gauges 15 is influenced by bending strain and compression strain in the plane of the shearplate 11, so the gauges 16 will tend to correct for this. Providing the gauges 16 are located so that their output changes with this secondary strain in the same manner as does the output of the principal gauges 15, then by suitably choosing the value of the resistors 22, the response of the device will be made essentially independent of the secondary strain.

The device may be calibrated by the application of bending to the I-beam by any suitable means, to determine the necessary value of the resistors 22.

Where two or more secondary gauges are used the simple analogue technique for obtaining the requisite correction may also be used, although the presence of three interconnected bridges will require an iterative process to arrive at the correct value for the attenuating resistors. Alternatively, however, the outputs of the bridges may be processed digitally, as shown in Fig. 7, where the individual outputs of the three bridges 18, 19 and 23 are amplified and applied to a multiplexing analogue-to-digital converter 24 for passing to a computer.

It is to be understood that it is not essential that the correcting strain measurement be a measurement of strain actually influencing the principal strain sensor, since the desired correction can be achieved as long as the correcting strain as measured is a function of that influence.

A device of the kind illustrated in Figs. 1 to 5 and 7 can be calibrated for the effects of secondary strain by means of the apparatus shown in Figs. 8 and 9. The device 10 is mounted with the shearplate 11 vertical, by bolting down the end plate 13. The calibrating rig consists of a mounting plate 25. which is bolted to the upper end plate 13, this mounting plate carrying a vertical member 26 coplanar with the shearplate 11. A horizontal extension 27, also in the plane of the shearplate 11, is located near the upper end of the member 26, and is provided with two load positions 28 and 29 along its length. This configuration ensures that the vertical member 26, in line with the main axis of the device 10, applies to the device 10 only secondary force and secondary strain - in this case arising from compression force and the bending moment produced by the application of force at 28 or 29.

The calibration is carried out by locating a weight (the value of which may not be known) at position 28 and recording the output of the principal and secondary strain gauges. The same weight is then applied at 29 and the gauge outputs again recorded. With the apparatus configured as shown in Fig. 8, the recorded data will show the influence on the principal strain gauge of secondary strain in the plane of the shearplate 11, enabling the response of the principal gauge to be corrected for that influence.

The test device is now rotated through 90° so that the arm 27 is now normal to the plane of the shearplate 11 as shown in Fig. 9, and weight (which need not be known) is again applied at the points 28 and 29, the outputs of the gauges again being recorded. Thus the principal gauge 15 can be corrected for the influence of secondary force acting at right angles to the plane of the shearplate 11, which in normal use of the device 10 will be the horizontal plane.

It is to be observed that these calibrations are carried out without the application of any force in the direction of load measurement of the device 10. While this is not essential, it does demonstrate the fact that the present invention provides correction for influences which are essentially unrelated to the load to be measured.

Calibration by means of the arrangement shown in Figs. 8 and 9 has been found to provide in a device of the kind illustrated in Figs. 1 to 5 a reduction of 99% in the influence of the secondary forces and strains involved, which is quite remarkable for a device with no special machining and without the use of expensive materials.

In prder to provide devices of different load capacities for a standardised I-beam and end plate dimension, the device of Fig. 1 can be modified as shown in Fig. 10 by providing slots 30 in the upper an lower regions of the shearplate 11, or as shown in Fig. 11 where a thin section 31 is provided by simple machining of a vertical central portion of the shearplate, the gauges being located on this section.

Various approaches may be taken to the manner in which devices of the general kind illustrated in Figs. 1 to

5 are provided with means for attachment to supporting and load-bearing structures, given that the invention reduces the effect of secondary strains arising from the connection of the device with other structures and from the manner of application of the load. For example, the end plates 13 can be slipped in to slots provided on the supporting and/or supported structure, eliminating the need for bolting. An alternative arrangement is shown in Fig. 12, where the endplates 13 are provided with a horizontal flange 32 serving as footing or load support - an arrangement unthinkable with prior art devices. Fig. 13 shows another alternative, in which the end plates 13 are fitted with vertical or horizontal bushes 33 to receive load transfer pins.

The invention will now be described in relation to two embodiments of a different kind, in which strain gauges are applied to existing structures. The retro-fitting of weighing devices to existing structures, for example by the installation of loadcells or by the fitting of "bolt on" or "weld on" strain sensing devices has in the past met with varying degrees of success. In many cases the relationship between measured strain and weight turned out to be quite unpredictable due for example to the effects of wind forces and bending distortion of the support structure under varying load conditions, differential temperature between the sensor and the support structure, and differing temperature coefficient as between the sensor and the structure.

By attaching strain sensors (for example strain gauges) directly to the support structure, secondary strain due to temperature differentials can be largely eliminated, and the influence of secondary forces can be dealt with in accordance with the present invention

by locating secondary strain sensors in a location where their output represents the influence of secondary strain on the principal strain sensors, and using the scaled outputs of the secondary sensors to counteract the influence of the secondary strain on the principal strain sensors.

Figs. 14 and 15 show such an arrangement applied to a leg 35 of a silo support structure 34. Pairs of electrical resistance strain gauges 36a, 36b, 37a and 37b are applied to the opposite sides of the leg 35 with their grids aligned as shown in Fig. 15. The grids 36a and 36b are connected in a bridge for response to compression strain in the region, while grids 37a and 37b are connected in a second bridge for correcting response to bending strain in the region. Calibration of the gauges may be carried out by applying side loading and adjusting the correction applied from the grids 37a and 37b.

In the example shown in Fig. 16, the invention is applied to conveying apparatus, in this case a track 38 suspended on hangers 39. In order to measure the weight of an article being carried along the track, the force acting on the track between the hangers 39 is measured by means of shear gauges at gauging locations 40. The force to be measure will of course result in principal strain (shear strain) at the gauging locations. However, the section of track between the hangers will bend downwardly as the article passes between the hangers 39, resulting in secondary forces and secondary strain at the two gauging locations 40. By locating secondary strain gauges responsive to bending at the gauging locations and the subsequent use of the their scaled output in the manner already described in relation to the other embodiments, the weight on the rail section can be

measured independently of the secondary forces induced by the application of the weight.

A generalised example of the application of the invention to devices in which, for instance, torque is of significance, either as the source of principal or a secondary strain, and indeed of the applicability of the invention to quite generalised structures, is provided by the idealised device shown in Fig. 17. Here four pairs of strain gauges are used, the second gauge of each pair being located on the opposite side of the structural member 41. Gauges 42 respond to torque, gauges 44 respond to bending in the vertical plane as well as tension and compression, gauges 43 respond to bending in the horizontal plane while their response to tension and compression will be cancelled when connected in Wheatstone bridge configuration, and gauges 45 to both shear and torque. Weight is assumed to be applied as shown at 46. The grids of each gauge pair are connected in bridges and the outputs of each bridge employed for compensation in the manner described above. In this idealised example the principal strain may be torque, where for example the torque is used as a measure of applied load. Equally, however, the principal strain may be bending, shear, tension or compression, with the calibration technique being chosen appropriately.

Finally there is illustrated in Figs. 18 and 19, an embodiment of the invention in which the gauging arrangement described in relation to the device illustrated in Figs. 1 to 5 is utilized on the forks of a vehicle for rubbish removal.

The illustrated member 47 in this example of the invention is one of a pair of forks used for lifting and manoeuvring a rubbish skip, and for this purpose

is mounted at its end 48 on a supporting structure not shown here. A gauging region 49 is located near the supported end of the member 47. A pair of flanges 50 are attached to the member 47 to increase the resistance of this section of the member 47 to side loading. Such forks are subjected at times to considerable side loading, often beyond the yield point of the unstiffened member, and the flanges 50 serve to reduce the risk that overload will result in permanent distortion of the gauged region of the fork. The gauges may be mounted in a recessed portion of the member to provide mechanical protection.

The gauging region is provided with principal and secondary gauges in the manner described in connection with Figs. 1 to 5.

It will be understood that the location of the strain sensors and the orientation of the primary sensing axes of the principal and secondary strain sensors must be determined by the nature of the structure to which they are applied and to the forces with which they most cope, and the above examples will have demonstrated this. With the assistance of those examples and a consideration of the structure in question, the reader skilled in the behaviour of structures and materials will be able to apply the invention to many other situations.