GB2280751A - Multi-DETF load cell - Google Patents

Abstract

A load cell comprises a plurality of double ended tuning forks (DETFs) lying axially parallel between two end members 18, each DETF having an exciter means and a frequency pickup means, a load to be measured being applied across the two end members whereby the change in resonant frequency upon application of said load is a measure thereof. A modified load cell (Fig. 8, not shown) has a double ended tuning fork (DETF) fixed across an open frame and means for applying a load to be measured across the frame, wherein the frame acts to attenuate the load and to apply only a reduced but proportionate load across the DETF. <IMAGE>

Description

Multi-DETF Load Cell This invention relates to load cells.

The accurate measurement of physical quantities such a mass, force, and strain is of prime importance for many industries. The force measurement devices are commonly known as load cells and normally utilize the straingauge technology. The performance of such a load cell is limited due to associated problems such as creep.

Furthermore, most strain-gauge based load cells have an undefined zero-load output when they are operated between tension and compression regions. These problems have increased the need to develop load cells based on other technologies. Good sensing devices for the measurement of force are mechanical resonators of various structures whose natural frequencies are a function of applied force.

Mechanical resonators have excellent stability and potentially low hysteresis. These factors together with the fact of outputs in frequency form have made them the subject of considerable practical interest. Frequency domain transducers have many advantages over conventional analogue sensors in that their output can be measured accurately with inexpensive frequency counters, they have high reliability, low error rates, low susceptibility to degradation of transmitted signals by electrical interference, and low dependence on change in electrical characteristics with time.

For this purpose, a single double-ended tuning fork (DETF) has been known. This has the property that if one tine is excited, the whole device will resonate at a frequency which is altered by the application of a force along the tines, either tension or compression.

Although a load cell has been developed using this system, it has major disadvantages. For larger loading, the tines must be made stronger and thicker until the device becomes heavy and unwieldy, and the excitation power becomes excessive. In addition, it does not perform accurately if the loading is not axially applied. The invention aims to avoid or reduce these problems.

Accordingly, the invention proposes a load cell comprising (P) a plurality of double ended tuning forks (DETFs) lying axially parallel between two end members, each DETF having an exciter means and a frequency pickup means, a load to be measured being applied across the two end members whereby the change in resonant frequency upon application of said load is a measure thereof.

Such a load cell may also have support members extending parallel to the DETFs and acting as a load shunt, reducing the load on the DETFs by a factor. This enormously increases the load range which can be catered for, without requiring the DETFs themselves to become massive.

Further, the use of a plurality of DETFs, preferably symmetrically around a circle, the disturbing effects of off-axis or non-axial loads is all but eliminated, since the loads on all the DETFs need only to be summed (even when unequal) to give the correct total load.

According to a further feature of the invention, there is provided a load cell comprising a double ended tuning fork (DETF) fixed across an open frame and means for applying a load to be measured across the frame, wherein the frame acts to attenuate the load and to apply only a reduced but proportionate load across the DETF.

In order that the invention shall be clearly understood, several exemplary embodiments thereof will now be described with reference to the accompanying drawings, in which: Fig.l shows a single double-ended tuning fork load sensor for explanatory purposes; Fig.2 shows a multi-DETF load cell according to the invention, in perspective; Fig.3 shows a transverse cross-section of Fig.2; Fig.4 shows another version of a load cell according to the invention; Fig.5 shows the frequency response of a single DETF load cell; Fig.6 shows the simple closed loop excitation and detection circuit; Fig.7 shows the curve of a DETF load cell under varying load; Fig.8 shows a load cell using a single DETF load sensor; and Fig.9 shows a device using a single DETF load cell for measuring fluid pressure.

The double-ended tuning fork (DETF) basic structure consists of twin beams (tines) 10 which are coupled at their roots 11 (Figure 1). When one of the tines in this symmetrical structure is excited at its fundamental frequency, the opposite tine resonates in sympathy. Due to symmetrical vibration of the DETF, the vibrational counter forces R1, R2 and bending moments M1, M2 cancel at the roots, so that at each end of the resonator beyond the roots no vibration occurs. This unique property of the DETFs allow them to be clamped beyond their vibrational roots without significant loss of vibrational energy or reduction in their mechanical quality factor Q.

The DETF resonator can be investigated in an open-loop configuration using a spectrum analyser. A piezoelectric chip 12 is used to energize one of the tines, and a second piezoelectric chip 13 attached to the opposite tine is used to detect the natural frequency of the DETF. To maintain the high Q DETF in continuous resonance an electronic circuit is used to provide feedback and phase control around a closed-loop.

Because of the high Q of the resonator the perturbing effects of cross-talk between the piezoelectric chips and electronic amplification circuit are minimal.

Therefore, the performance of the DETF resonator may be assumed to be entirely mechanical.

The fundamental frequency of a beam is a function of its parental material property, and dimensions. This frequency increases with increasing applied axial force, and the equation governing the vibration of a beam under tension T is given by the following expression;

where I = second moment of area = (w.H); :2 E = Young's modulus of the material p = density of the material S = cross-sectional area of the beam '.H) t = time, and W and H are thickness and width of the beam, respectively. Assuming a simple harmonic motion and applying the boundary conditions, the natural frequency f01 when T = 0 (axial force T = 0) is given by the expression below;

where a = value determined by mode number n of vibration z (1 +2n) r/2, and I = length of the beam (Ref. 3J.

However, when T Z 0, the natural frequency of the beam under axial force fT,is given by;

These equations define the motion of a DETF under stress and they may be used to design a DETF to operate at a particular frequency; it is also possible to determine the total frequency shift of a DETF under full-scale load. However, in practice other important parameters should be considered in order to realize a DETF-based load cell.

The load bearing capability of a DETF is determined by its material's yield strength and cross-sectional area.

In designing a high capacity DETF-based load cell 14 (Fig.2) a simple approach of integrating the DETFs and their support was chosen (i.e. DETFs 15 and supports 16 are fabricated from one piece of material). This integration approach reduces the potential hysteresis due to the mismatch of materials of the DETFs and the support. Four equai-spaced DETFs 15 were fabricated by machining in the walls of a hollow stainless steel cylinder 17 at 900 relative to each other as shown in Figures 2 and 3.

The cylinders four identical DETFs have tines with 1 mm x 4 mm cross-sectional area and a length of 20 mm.

The load cell is designed for operation either in tension or compression. The load bearing capability of this type of load cell may be easily varied by changing the dimensions of the supports 15. The supports 15 act as a load shunt across the DETFs, so that greater loads can be accepted than without them. Finite element analysis can be used to establish an optimum static loading model for the device. The 20 kN load cell is quite insensitive to off-axial loading which may occur in real applications. This insensitivity is achieved by addition of the frequency excursions of all four DETFs around the cylinder.

Figure 4 illustrates a multi DETF load cell without the load shunt supports 16, and includes loading caps 18.

The cylindrical DETF based load cell was tested under open-loop condition. The open-loop test involves the use of a spectrum analyser to obtain the frequency response of the DETFs. The natural frequency and mechanical quality factor of each DETF was measured under zero load conditions. The natural frequencies of all four DETFs were found to be around 6000 Hz (less than 5% variation between them). A typical frequency response curve of one of the DETFs is shown in Figure 5.

This DETF has a natural frequency of 6029 Hz and a Q greater than 1500.

The open-loop characterisation of the DETFs showed that the lower the drive voltage the higher the Q. The high level of drive voltage causes loss of energy from the system by means of sound and/or heat. Frequency response curves also showed that the value of the Q is almost independent of the applied load (i.e Qs of DETFs were > 1500 for various loads).

An electronic closed-loop circuit tracks the resonant frequency (output of each DETF) and incorporates a digitally generated 900 phase-shift to bring input and output in phase. This closed-loop arrangement is shown in Figure 4. This arrangement keeps each device in continuous oscillation, and as the structure is loaded each closed-loop circuit tracks the resonant frequency and maintains respective oscillation. The output frequency of each DETF is monitored using a frequency counter.

Figure 5 shows a typical continuous loading curve for the load cell monitored from one of the DETFs for a load cycle from 10 kN tension to 10 kN compression. The graph shows a total frequency excursion of approximately 500 Hz corresponding to 20 kN. The total frequency excursion or span is approximately an 8.5t shift in the natural frequency of the DETF. The characteristic loading curve also shows that the load cell output is linear. The region of transition from tension to compression shows no discontinuity.

Characteristics of the load cell are indicated in table 1. The load cell was tested under different conditions and the fundamental frequencies of all the DETFs were monitored. Because of the compensation method (described in section 3) no significant off-axial loading was noted and the results shown are typical of one of the DETFs. The load cell has a safety margin of 100%.

Table 1 Typical characteristics of the DEIF based load cell.

Parameter Value (Hz) % of span Nat. Freq. 6030 Freq. shift (20 k 500 100 Repeatability 0.2 0.04 *Stability (0 k) 0.2 0.04 " (2 kN) 0.5 0.10 " (5 kN) 0.5 0.10 Linearity 5.0 r1.00 Sensitivity -(1 Hz/40 N) Resolution (better than 1 part in 2000) Stiffness (less than 100m displacement for FS) * Stabilities are measured over 4 hour tests.

The performance of the new load cell based on the resonating DETF structure has shown that this technology has the potential to be utilized in fields of mass, pressure, and force measurement. The benefits of this technology may be summarised as follows; i) the DETF-based load cell has low power consumption (less than 2 mW), and a frequency output which makes it ideal for interfacing with digital systems; ii) the load cell characteristics indicate good repeatability, linearity, stability, and high resolution; iii)the load cell requires only simple electronics, thus reducing the manufacturing cost.

iv) Any plurality of DETFs can be fabricated at 900 (or any other angles, eg. three DETFs at 1200) round a hollow cylinder.

v) Insensitivity to positional off-axial loading (all DETFs should be monitored).

vi) Insensitivity to angular off-axial loading.

An alternative form of load cell (Fig.8) utilises any spring element 20 (eg. a proving-ring) as an elastic medium to increase the load bearing capability of a single DETF 21. By applying a force 22 to the ring, the ring elongates and the DETF is tensioned. The DETF may be pre-stressed in tension so that it can function for both tension and/or compression forces 22. The prestressing can be introduced by applying force 23 to the ring before the DETF is fixed in place. The working range of the load cell may be increased by simply increasing the thickness of the elastic element. The DETF is attached to the ring 20 by welding, but other mechanical methods are also possible (i.e. bonding or integral). The DETF may be excited piezoelectrically or electromagnetically, and the pick-up technique may be piezoelectrically, optically and/or magnetically. The load cell can have one or any plurality of DETFs.

A fluid pressure sensor (Figure 9) uses a pre-stressed DETF 30 (as in Fig.8) (or any number of DETFs may be used) housed in a two cage structure. One end 31 of the DETF is fixed and the other 32 passes through a hole in the centre of an encastre beam. The latter end is threaded and a nut 33 is used to tighten it to give required pre-tensioning. The thread on the end of the DETF is identical to a thread cut in a boss on a diaphragm 34. The diaphragm is screwed onto one end of the DETF and locked in place by the locking cap head screw. The inner cage is mounted within the outer cage with the diaphragm sitting in a recess in the outer cage. The pressure cap 35 is then screwed down onto the outer cage clamping the diaphragm and allowing the connection of pressure supply through a Legree type quick fit push-in 36 to a chamber 37 above the diaphragm.

Claims (10)

Claims

1. A load cell comprising a plurality of double ended tuning forks (DETFs) lying axially parallel between two end members, each DETF having an exciter means and a frequency pickup means, a load to be measured being applied across the two end members whereby the change in resonant frequency upon application of said load is a measure thereof.

2. A load cell as claimed in claim 1, wherein the DETFs are positioned at points symmetrically placed around the circumference of a circle.

3. A load cell as claimed in claim 1 or 2, wherein the DETFs and the two end members are integral and are formed by machining of a hollow cylindrical body.

4. A load cell as claimed in any preceding claim, wherein support members which act as load shunts extend parallel to the DETFs between the two end members.

5. A load cell as claimed in any preceding claim, comprising four DETFs.

6. A load cell comprising a double ended tuning fork (DETF) fixed across an open frame and means for applying a load to be measured across the frame, wherein the frame acts to attenuate the load and to apply only a reduced but proportionate load across the DETF.

7. A load cell as claimed in claim 6, wherein said DETF lies across the frame perpendicular to the direction of the load to be measured.

8. A load cell as in claim 4 or 6, wherein the DETF(s) is/are held pretensioned and both tension and compression loads to be measured produce loads on the DETF(s) which are not compressive.

9. A fluid pressure measuring device comprising a load cell according to claim 6 or 7, having a closed chamber to which fluid is applied with one diaphragm wall, the wall being physically linked to the load cell which can generate a measured value indicative of the fluid pressure.

10. A load cell substantially as herein described with reference to Figures 2 to 7 or Figures 8 and 9 of the accompanying drawings.