WO1996018084A1

WO1996018084A1 - Measuring method and apparatus

Info

Publication number: WO1996018084A1
Application number: PCT/GB1995/002868
Authority: WO
Inventors: Robert Eric Bradwell Holland
Original assignee: Hydronix Limited
Priority date: 1994-12-08
Filing date: 1995-12-07
Publication date: 1996-06-13
Also published as: GB2299406B; GB9424831D0; GB2299406A; AU4121496A

Abstract

The present invention relates to the measurement of a property, for example moisture content in a material subject to substantially transient presence in the region of a sensor. In a concrete mixer (1) the rotation of the mixing blades (8) causes the mix to lose intimate contact with a moisture content measuring sensor (9). This introduces noise, in the form of spikes, into the sensor output. The spikes adversely affect the accuracy of the moisture content measurement. The present invention solves this problem by sampling the output of the sensor (9) in the regions between the spikes. The timing of the sampling is triggered by the occurrence of a spike in the sensor signal, either directly by comparing the time derivative of the sensor signal with a threshold or indirectly, for instance by detecting the position of a mixing blade. A variety of samples may be taken between each pair of spikes and averaged to produce an output signal representing the moisture content of the mix in the mixer (1). In another embodiment, the invention is applied to sensing the moisture content of loaves (40) on a conveyor belt (41). The gaps between loaves (40) are detected and used to trigger sampling of the output of a moisture sensor (42).

Description

Measuring Method and Apparatus

Field of the Invention

The present invention relates to a method and an apparatus for measuring a 5 property of a material subject to transient presence in a region.

Background to the Invention

In the concrete industry, concrete is manufacture by mixing the constituents of the concrete, that is cement, water and aggregate, in a vessel by means of to rotating blades. It is during this mixing process that the water is added. Typically a closed-loop control system is used to control that addition of water to the mix, the feedback signal of the control system being provided by a moisture sensor mounted through the wall of the mixing vessel.

a It has been found that the rotation of mixing blades in the vessel causes cavitation, that is the formation of voids or regions of reduced density, at the trailing faces of the blades. This cavitation introduces noise into the moisture sensor reading and is particularly acute if the cavitation causes the mix to lose intimate contact with the sensor.

20

The noise in the sensor output signal, caused by the cavitation, takes the form of sharp spikes on an otherwise slowly changing signal. Hitherto, low-pass filters have been used to remove these spikes from the sensor output signal. However, if sufficient accuracy is to be achieved, a filter having a long time a constant is required. The long time constant can lead to the mixing time being extended with a consequent reduction in throughput. In a commercial operation, it is undesirable that more time be spent on a process than is absolutely necessary.

30 A similar problem has been found in the food industry where it is desired to measure the moisture content of discrete amounts of material, for example loaves or dough pieces, on a conveyor belt. The gaps between items on the conveyor belt produce a signal similar to that produced by the cavitation discussed above.

Summary of the Invention According to a first aspect of the present invention, there is provided a method of measuring a property of material subject to transient proximity to a sensor, comprising generating a timing signal delayed with respect to material entering the region of the sensor and sampling the output of the sensor in dependence on the timing signal, wherein the timing signal is delayed relative to material entering the region of the sensor such that the output of the sensor is sampled to make a measurement when the material is in the region of the sensor.

According to a second aspect of the present invention, there is provided a measuring apparatus comprising a sensor for sensing a property of a material subject to substantially transient presence in the region of the sensor, timing means for generating a timing signal delayed with respect to the entry of material into said region, and sampling means for sampling the output of the sensor in dependence on the timing signal, wherein the timing signal is delayed relative to entry of material into said region such that the output of the sensor is sampled when the material is in said region.

It will be apparent that at the end of the cavitation, mentioned above, the material re-enters the region of the sensor. The region of the sensor can be considered to be the space for which the sensor output is meaningful. For instance, the space occupied by material during calibration of the measurement system.

The present invention was devised in response to a problem existing in the concrete industry. However, it will be appreciated that it may be applied with beneficial effect in analogous situations such as the conveyor belt problem discussed above. In a broad aspect, the present invention provides for one timing signal to be produced for each entry of material into the region of the sensor. However, timing means may be provided for producing a plurality of time spaced timing signals in respect of each entry of material into the region of the sensor. This may advantageously provide improved accuracy compared with the use of a single sample. The plurality of samples may be averaged, thereby reducing the effect of random noise in the sensor output signal. Advantageously, ten or more samples are taken for each entry of material into the region of the sensor. to

It is important that the time of entry of material into the region of the sensor be known. One way of achieving this to detect characteristic features, for example the spikes referred to above, in the sensor output signal. In the exemplary situations, where entry of material into the region of the sensor is n indicated by sharp transitions in the sensor output, entry of material into the region of the sensor may be detected by differentiating the sensor output and comparing the result of the differentiation with a threshold.

Nevertheless, it is not essential that the occurrence of the entry of material 20 into the region of the sensor be detected from the sensor signal. Instead it may be reliably inferred in certain situation from other events. For example, a proximity sensor responsive to the proximity of a blade or a variable reluctance sensor, including a number of teeth on a shaft driving a blade or blades, may be used to detect a predetermined configuration of a blade ! corresponding to cavitation in the region of the sensor. The presence of cavitation in the region of the sensor may also be inferred from the detection of a predetermined condition of the driving current of an electric motor driving a blade which is responsible for said cavitation. The predetermined condition may be conveniently a zero crossing for an ac energized motor. 0

In a preferred embodiment the sensor output is read, the time derivative of the read sensor output calculated, the time derivative compared with a threshold, the steps of:

(i) starting a timer,

(ϋ) determining whether a predetermined period has been timed by the timer, and (iϋ) reading the sensor output and accumulating the sensor output readings if the predetermined period has elapsed;

repeated a predetermined plurality of times, if the time derivative is greater than the threshold, and accumulated sensor output readings divided by the number of readings taken to provide a measure of the property.

Preferably, the sensor output signal samples obtained in response to the timing signals, or an average thereof, are plotted against time on a display device. The display device may comprise an electronic display unit, e.g. a liquid crystal display panel, or a chart recorder. The displayed trace provides a good indication of the homogeneity of the material being monitored. This information is substantially lost with prior art low-pass filtering techniques due to the long time-constants employed.

Brief Description of the Drawings

Embodiments of the present invention will now be described, by way of example, with reference to the accompanying drawings, in which:

Figure 1 shows a static pan mixer and controller to which the present invention is applicable; Figure 2 shows the output of the sensor shown in Figure 1 and the derivative thereof;

Figure 3 shows a computer based embodiment of the present invention;

Figure 4 shows a flow chart illustrating a set up procedure for the embodiment of Figure 3; Figure 5 shows a flow chart illustrating the measuring operation of the embodiment of Figure 3;

Figure 6 shows an implementation of the present embodiment in hardware; Figure 7 shows a second embodiment of the present invention; and

Figure 8 shows the output of the sensor shown in Figure 7 and the derivative thereof.

Description of Embodiments of the Invention

Referring to Figure 1, a concrete mixing plant comprises a static pan mixer 1 and a controller 2. The mixer 1 comprises a tub-shaped mixing vessel 3, a gearbox 4 supported centrally over the mixing vessel 3 by a spider 5, a motor 6 mounted above and to the gearbox 4, and three equidistant mixing arms 7 depending from the gearbox 6 and supporting blades 8. A microwave moisture sensor 9, such as a Hydronix^* HYDRO-MDC sensor manufactured by Hydronix Limited of Cranleigh, Surrey, is mounted eccentrically through the floor of the vessel 3. The output from the sensor 9 is coupled by a signal line 10 to the controller 2. The controller 2 comprises processing circuitry for processing the signal from the sensor 9 and producing control signals to control the supply of water to the vessel 3 by water supply means (not shown).

In another form of concrete mixer installation, the mixing vessel is provided with side-wall scraping blades and the sensor is mounted through the side-wall of the vessel. The spikes in the sensor output signal have been found to be more easily detected where a side-wall mounted sensor is used in a mixer having side-wall scraping blades.

In operation, the vessel 3 is loaded with cement and aggregate and the motor 6 caused to rotate the blades 8 via the gearbox 4. The controller 2 determines the moisture content of the mixture in the vessel 3 from the output of the sensor 9 and controls the addition of water to the mixture in dependence on the sensed moisture content. As the blades 8 rotate through the mixture, they generate turbulence in their wake. The turbulence causes the mixture to lose intimate contact with the sensor 9 as each blade 8 passes by the sensor 9.

Referring to Figure 2, the upper trace shows the output of the sensor 9. The positive-going spikes indicate the passage of turbulence past the sensor 9. During the spikes, the sensor output does not give a good indication of the moisture content of the mixture in the vessel 3. However, the substantially horizontal portions of the trace between spikes do usefully reflect the s moisture content of the mixture.

Referring to Figure 3, the controller 2 comprises a microprocessor 11, a RAM 12, a ROM 13, an analogue-to-digital convener 14, a digital-to-analogue converter 15, an I/O unit 16, a display unit 17, a keypad unit 18 and a bus 19. to The ROM 13 contains program instructions for the microprocessor 11. Variable data processed by the microprocessor 11 is stored in the RAM 12. The display unit 17 displays information relating to the operation of the controller 2 or the signal received from the sensor 9 (see Figure 1). The display unit 17 includes a graphical display such as is described in our UK is patent application no. 9320956.7 and sold as a separate unit under the name HYDRO-VIEW by Hydronix Limited. The keypad unit 18 is provided to enable a user to control the operation of the controller 2. The I/O unit 16 interfaces the analogue-to-digital converter 14 and the digital-to-analogue converter 15 to the bus 19 so that they can be accessed by the microprocessor

20 11.

The digital-to-analogue converter 15 may be omitted if circumstances require that the controller 2 output a digital signal.

25

A set up procedure for the controller 2 will now be described with reference to Figures 3 and 4.

On selection of a set up mode by a user manipulating the keys of the keypad 30 unit 18, a set up procedure is performed. The set up procedure comprises steps si to s!3. Step si is performed when the set up procedure is entered and sets the initial values of ten timers T(0) - T(9) and ten period timers P(0) to P(31) to zero. The timers T(0) - T(9) will be used to store values indicating the times at which measurements are to be taken and the period timers P(0) - P(31) will be used to store measured cavitation periods.

After initialization at step si, a sensor reading is taken at step s2. The microprocessor 11 interrogates the analogue-to-digital converter 14 to obtain a sensor output sample. Then in step s3, the time derivative of the sampled sensor output is calculated.

If the calculated time derivative is less than or equal to a threshold (step s4) the program flow moves to step s6 where the current period timer P(n) is incremented. At step s7, following step s6, the current sensor output sample S is stored if the current period timer is the 32nd period timer P(31). The samples stored at step s7 are used to provide representative signal for display at step sll. After step s7, the program loops back to step s2.

If the calculated time derivative is determined to be greater than the threshold at step s4, it is then determined whether the current period timer is the 32nd period timer P(31) at step s5. If the current period timer is not the 32nd period timer P(31) then the next period timer is activated (step s8) and program flow passes to step s7.

If the current period timer is determined to be the 32nd period timer P(31) at step s5, program flow passes to step s9 where the average of the times stored in the period timers P(0) - P(31) is calculated. This average provides a refe ence cavitation period. Alternatively, the shortest of the periods stored in the period timers P(0)-P(31) could be selected as a reference cavitation period. Once the reference cavitation period has been determined, the timers T(0) - T(9) are set to 1/lOth of the representative cavitation period at step slO. At step 11, the samples stored at step s7 are displayed as a trace on the display unit 17 and, at step sl2, markers are added to the displayed trace to indicate the positions represented by the values set for the timers T(0) - T(9).

The positions of the markers may be changed by a user operating keys of the keypad 18 (step sl3). In this way, the user may ensure that all of the sampling times represented by the timers T(0) - T(9) occur during the substantially flat portion of the sensor output's waveform. Once the user has indicated that the markers are in acceptable positions, the timers T(0) - T(9) are reset with values corresponding to the marker positions.

In another embodiment, the timers T(0)-T(2) are not preset automatically and a user is prompted to position markers in a display of the sensor signal for one cavitation cycle.

In yet another embodiment, the timers T(0)-T(9) are preset with different proportions of the representative cavitation cycle period. For example, the first timer T(0) could be set to a larger value that the others in order to ensure that the first measurement sample is taken after the spike has completely passed. The particular algorithm for presetting T(0)-T(9) will depend on the characteristics of the sensor output signal.

In a still further embodiment, the waveform of the sensor output signal is analyzed automatically to determine the region during which measurement sampling is appropriate.

The measuring operation of the controller 2 will now be described with reference to Figures 3 and 5.

During measurement operation, the procedure shown in Figure 5 is repeatedly performed. Each performance of the procedure produces a new value representing the moisture content of the material being monitored. The first step of the procedure, step s20, sets an accumulation variable SUM to zero. The second step, step s22, sets to zero a pointer n for pointing to one of the timers.

Once the accumulation variable SUM and the pointer n have been initialized, the microprocessor 11 interrogates the analogue-to-digital converter 14 for a sensor output signal sample S at step s22. The time derivative of the sensor output is then calculated at step s23.

At step s24, the calculated time derivative is compared with a threshold to determine whether a spike in the sensor output signal is occurring. If the result of the comparison indicates that a spike is not occurring, program flow returns to step s22. However, if it is determined that a spike is occurring in the sensor output signal, the timer T(n) indicated by the pointer n is started at step s25. At step s26, the current value of the current timer T(n) is repeatedly tested to determine whether it has expired. This step may comprise repeatedly comparing a current value of a clock circuit output with a clock circuit output value stored at step s25.

When it is determined that the current timer T(n) has expired, the program flow moves to step s27 where the analogue-to-digital converter 14 is interrogated for a sensor output signal sample. The new sample is added to the accumulation variable SUM at step s28 and then, at step s29, the pointer n is incremented.

At step s30, it is determined whether all the timers have been used by comparing n with the number of timers T(0) - T(9). If one or more timers T(0) - T(9) are as yet unused, program flow returns to step s25 where the next timer is started. However, if the timers T(0) - T(9) have all been exhausted, the program flow moves on to step s31 where the average of the sensor output signal samples is calculated by dividing the accumulation variable SUM by the number of timers. The resultant value is then made available to update the control process performed by the controller 2 at step s32.

In a modified form of the apparatus of Figure 3, the microprocessor 11 causes s each of the samples s to be plotted on the display unit 17. This produces a trace corresponding to that exemplified by the upper trace in Figure 2 but with the spikes removed. The resultant trace provides a good indication of the homogeneity of the mixture in the vessel 3.

to It will be appreciated that the present invention is not restricted to microprocessor based systems. Accordingly, a dedicated hardware embodiment will now be described with reference to Figure 6.

Referring to Figure 6, a measuring circuit comprises a differentiator 30 which is receives the output from a sensor, a comparator 31 which compares the output of the differentiator 30 with a threshold set by a potentiometer 32, an monostable multivibrator 33 which receives the output of the comparator 31 at its SET input, a clock oscillator 34, a shift register 35 having a plurality of taps T_t - T„, having the output of the monostable 33 applied to its data input 20 and the output of the clock oscillator 34 applied to its clock input, a multi- input OR gate 36, the inputs of which are coupled to respective one of the taps T, - T_o., of the shift register 35, a first sample and hold circuit 37 which has the sensor output signal and the output of the OR gate 36 as its inputs, a low-pass filter 38 which receives the output of the first sample and hold circuit 25 37 as its input, and a second sample and hold circuit 39 which has the output of the low-pass filter 38 and the output of the last tap T„ of the shift register 35 as its inputs.

The operation of the circuit of Figure 6 will now be described.

30

The sensor output signal is differentiated by the differentiator 30. This has the effect of converting the sensor output signal into series of spikes on a substantially zero dc voltage as shown in the lower trace of Figure 2. The occurrence of a spike in the output of the differentiator 30 is detected by the comparator 31 which produces a pulse while the voltage of a spike exceeds the threshold set by the potentiometer 32. The monostable 33 produces a further pulse, which has a predetermined duration, in response to the pulse output by the comparator 31. The duration of the output pulse of the monostable 33 is set to be between 50% and 100% of the period of the clock signal from the clock oscillator 34. This ensures that only a single logic "1" is fed into the shift register in response to each spike in the sensor output signal and that each spike is registered by the circuit. The output of the monostable 33 is applied to the data input of the shift register 35. The shift register 35 is clocked by the output of the clock oscillator 34. Following the detection of a spike, the output of the monostable 33 is shifted along the shift register 35 at a rate determined by the frequency of the clock oscillator 34. After predetermined periods, the pulse applied to the data input of the shift register 35 appears at the taps T, - T„.

When logic "1" appears at one of the taps T, - T_n.„ the OR gate 36 outputs a logic "1" which is applied to the first sample and hold circuit 37. This causes the first sample and hold circuit 37 to sample the sensor output signal. The samples taken by the first sample and hold circuit 37 are low-pass filtered by the low-pass filter 38. The output of the second sample and hold circuit 39 samples the output of the low-pass filter 38 in response to a logic "1" appearing at the last tap T_n of the shift register 35. In effect, the sample and hold circuits 37, 39 and the low-pass filter 38 act to average the samples taken of the sensor output signal during one period of cavitation.

The output of the second sample and hold circuit 39 may be supplied to a chart recorder (not shown) for plotting.

In the above-described example, the clock oscillator is free running. However, the frequency of the clock for the shift register 35 could be set by the period of the cavitation cycle, for example by phase locking it to the spikes in the sensor output signal or to a signal indicating the speed of a prime mover, driving an element responsible for the cavitation. Locking of the clock to the cavitation cycle means that optimum sampling can be maintained in the event of gradual changes in the cavitation cycle period. Such changes may be due, for example, to changes in the load on a prime mover, driving an element responsible for the cavitation.

In a modified form of the circuit of Figure 6 means, for example jumpers, are provided so that a user may select the taps T^T_^, to be coupled to the OR gate 36.

While the present invention has been described with reference to sensing the moisture content of a mixture in a concrete mixer, it will be appreciated that it may be applied to other situation where periodic or substantially periodic cavitation causes spikes in a sensor output.

Referring now to Figure 7, a series of loaves 40 are transported by a conveyor belt 41. A microwave moisture sensor 42 is mounted above the conveyor 41 such that the loaves 40 pass through its active region.

As the loaves 40 proceed along the conveyor belt 41, the sensor 42 outputs a signal substantially of the form of the upper trace of Figure 8. In Figure 8 the positive-going pulses represent the gaps between loaves 40 and the regions between the pulses the moisture content of the loaves 40.

The output of the sensor may be applied to circuits of the type shown in both Figures 3 and 6. However, the gaps between the loaves 40 are relatively long compared with the duration of the cavitation found in the concrete mixer 1 described above. Consequently, the end of the gap in particular is deteαed by comparing the derivative of the sensor signal with a negative threshold (see Figure 8). - 13 -

In an industrial process, the loaves 40 can be expected to pass along the conveyor belt 41 at regular intervals. As a result, the pulses in the output signal of the sensor 42 are for practical purposes periodic.

The apparatus shown in Figure 7 may also be employed to determine whether the correct amount of dough has been placed in baking tins. In such a system, tins loaded with dough are conveyed by the conveyor belt 41 passed the sensor 42. The leading and trailing edges of the tins cause negative-going spikes in the output signal of the sensor 42. The spikes caused by the leading edges of the tins can be detected and used to trigger a measuring process as described above.

The signal output by the sensor 42 during the measurement process will be dependent on the amount of dough in the current tin, that is the proximity of the surface of the dough to the sensor 42. Thus, the measured "moisture content" can be compared with thresholds to determine whether too many or too few dough pieces have been placed in the current tin.

Typically, an excess of dough in one tin will be accompanied by a corresponding deficit in another tin. The reUability of excess dough deteαion can be improved by relying on the deteαion of both an excess of dough in one tin and a deficit of dough in a preceding or succeeding tin to trigger an alarm signal.

It will be apparent that this system is applicable to the determination of the quantity of materials other than dough.

An advantage of the synchronous sensing of the present invention is that the moisture profile of the loaves 40 or other items, for example planks of wood, can be determined without interference from the pulses caused by the gaps between items.

Claims

1. A method of measuring a property of material subjeα to transient proximity to a sensor (9;42), comprising generating a timing signal delayed s with respeα to material (40) entering the region of the sensor and sampling the output of the sensor in dependence on the timing signal, wherein the timing signal is delayed relative to material entering the region of the sensor such that the output of the sensor is sampled to make a measurement when the material is in the region of the sensor. 0

2. A method according to claim 1, wherein a plurality of timing signals are generated for each occurrence of material entering the region of the sensor, each timing signal being delayed by a different amount, and the output of the sensor is sampled in dependence on each timing signal. 5

3. A method according to claim 2, wherein the sensor samples taken for each said occurrence are averaged.

4. A method according to claim 1, 2 or 3, comprising deteαing the 0 occurrence of material entering the region of the sensor from the sensor output to produce a presence signal indicative of the occurrence of material entering the region of the sensor and generating the or each timing signal in dependence on the presence signal.

5 5. A method according to claim 4, wherein the occurrence of material entering the region of the sensor is deteαed by comparing the time derivative of the sensor signal with a threshold value.

6. A method according to claim 1, 2 or 3, wherein the occurrence of the entry of material into the region of the sensor is inferred from an event related to said occurrence. 7. A mαhod according to claim 6, wherein said event comprises a predetermined configuration of a blade (8) responsible for said occurrence.

8. A mαhod according to claim 6, wherein said event comprises a s predetermined condition of the driving current of an eleαric motor (6) which drives a blade responsible for said occurrence.

9. A mαhod according to claim 1, comprising the steps of: (a) reading the sensor output; o (b) calculating the time derivative of the read sensor output;

(c) comparing the time derivative with a threshold;

(d) if the time derivative is greater than the threshold, repeating a predetermined plurality of times the steps of:

(i) starting a timer, (ϋ) dαermining whαher a predetermined period has been timed by the timer, and

(iii) reading the sensor output and accumulating the sensor output readings if the predetermined period has elapsed; and (g) dividing the accumulated sensor output readings by the number of said readings taken to provide a measure of said property.

10. A mαhod of measuring the moisture content of a material in a mixing vessel (3), the mixing vessel including rotating cavitation inducing blades (8), according to any preceding claim.

11. A mαhod of measuring the moisture content of discrαe quantities of material (40) on a conveyor (41) according to any one of claims 1 to 6 or 9, wherein the sensor comprises a moisture sensor mounted with respeα to a conveyor such that material on the conveyor passes through the region of the sensor.

12. A mαhod according to claim 10 or 11, wherein a microwave moisture sensor (9;42) is used to produce said sensor output.

13. A measuring apparatus comprising a sensor (9;42) for sensing a property of a material (40) subjeα to substantially transient presence in the s region of the sensor, timing means (2;30,31,32,33,34,35,36) for generating a timing signal delayed with respeα to the entry of material into said region, and sampling means (14;37) for sampling the output of the sensor in dependence on the timing signal, wherein the timing signal is delayed relative to entry of material into said region such that the output of the sensor is to sampled when the material is in said region.

14. An apparatus according to claim 13, including timing means (2;30,31,32,33,34,35,36) for producing a plurality of time spaced timing signals (T₁ ..

in respeα of each entry of material into said region. is

15. An apparatus according to claim 14, including averaging means (2;38) for averaging the sensor output samples taken in response to each entry of material into said region.

0 16. An apparatus according to claim 13, 14 or 15, including deteαion means (2;30,31) responsive to the sensor output to dαeα entry of material into said region.

17. An apparatus according to claim 16, wherein the dαeαion means

25 comprises differentiator means (2;30) for differentiating the sensor output and comparator means (2;31) for comparing the output of the differentiator means with a threshold.

18. An apparatus according to claim 13, 14 or 15, including inferring means JO (2) arranged to dαeα an event related to the entry of material into said region and generating the or each timing signal on the basis of the d eαion of said event. 19. An apparatus according to claim 19, wherein the inferring means is arranged to dαeα a predαermined configuration of a blade (8) responsible for entry of material into said region.

s 20. An apparatus according to claim 18, wherein the inferring means comprises means arranged to dαeα a predαermined condition of the driving current of an eleαric motor (6) driving a blade which is responsible for said entry of material into said region.

lo 21. An apparatus according to claim 13, comprising: a differentiator means (2;30) for differentiating the sensor output; a comparator means (2;31) for comparing the differentiator means output with a threshold; timing means (2;33,34,35) responsive to the output of the differentiator means is to produce a plurality of timing signals (T, .. T„.J; sampling means (14;37) for sampling the sensor output in response to each of the timing signals; and averaging means (2;38) for averaging the samples taken in response to the timing signals.

20

22. A moisture content measuring apparatus according to any one of claims 13 to 21, wherein the sensor is mounted to a mixing vessel (3) having a rotating mixing blade (8).

25 23. An apparatus for measuring the moisture content of material on a conveyor (41) according to any one of claims 13 to 18 or 21, wherein the sensor comprises a moisture sensor (42) mounted with respeα to a conveyor such that material on the conveyor passes through the region of the sensor.

JO 24. An apparatus according to claim 22, wherein the sensor comprises a microwave moisture sensor (42). 25. A method of dαecting the quantity of a material comprising the steps of: producing a moisture content signal by an mαhod according to claim 11 or 12; and producing a signal representative of the quantity of said material on the basis the moisture content signal.

26. A mαhod of dαeαing excess dough in a baking tin according to claim 25.

27. A apparatus for dαeαing the quantity of a material comprising an apparatus according to claim 23 or 24 and means for generating a quantity signal in dependence on the measured moisture content.