WO2008026942A9

WO2008026942A9 - Modifying washing machine operation based on wash load characteristics

Info

Publication number: WO2008026942A9
Application number: PCT/NZ2007/000233
Authority: WO
Inventors: Adam John Darby
Original assignee: Fisher & Paykel Appliances Ltd; Adam John Darby
Priority date: 2006-08-31
Filing date: 2007-08-28
Publication date: 2009-07-09
Also published as: WO2008026942A1

Abstract

A washing machine executes a method of modifying a wash process that uses reduced water volume during agitation. According to the method a controller obtains at least one value of a load parameter indicative of one or more wash load characteristics. The machine agitates the wash load in a spin basket with a low profile wash-plate, the wash-plate being operated by a motor. The controller obtains at least one value of a feedback parameter indicative of a motor load characteristic for a period during agitation of the wash load. The controller determines a relationship of the feedback parameter value and the load parameter value to predetermined data, and modifies a wash process of the washing machine based on the relationship of the feedback parameter value and load parameter value to the predetermined data.

Description

MODIFYING WASHING MACHINE OPERATION BASED ON WASH LOAD

CHARACTERISTICS

FIELD OF THE INVENTION The present invention relates to washing machines, and in particular washing machines adapted to modify operation based on characteristics of the wash load.

BACKGROUND TO THE INVENTION

Washing machines typically have an agitate cycle for cleaning the wash load loaded into the spin basket. An agitate cycle comprises numerous agitation strokes, each one comprising at clockwise and counter-clockwise rotation of the spin basket or agitator. This moves the items and the water in the spin basket in order to clean the items in the wash load. The items in the wash load are typically clothing, linen and the like.

The composition or make up of a wash load can vary considerably. A wash load might comprise a substantially homogenous load of similar items, or a mixed load comprising a number of different item types. In each case, the individual characteristics of the items in the load, and the load itself might vary considerably. As a result, the particular manner in which the spin basket or agitator is agitated during an agitate cycle might be suitable for a first wash load type, but not another. For instance, a particular agitate stroke might be suitable for a wash load of heavy items, but might be too vigorous for a load of lighter items. This could result in unnecessary wear of the items. Similarly, a less vigorous agitate stroke could be used for lighter items, but such a stroke would not be suitable for effectively cleaning a different load type, such as one with a mixture of heavy items.

This is particularly relevant in the case of a washing machine with a low profile wash plate that washes the clothes in a reduced volume of water by setting up a toroidal wash movement.

Washing machines using this wash pattern are described in US patent 6,212,722 and in US patent application 11/470,658. The whole contents of each document is hereby incorporated by reference. This is in contrast to a washing machine with a central agitator post. A low profile wash-plate does not extend above the level of a typical wash load. Certain load types couple strongly to the wash plate which results in good toroidal movement and therefore washing. Other wash types might not couple so well to the wash plate resulting in less toroidal movement and less washing. In this case a different wash process might be necessary to ensure sufficient cleaning action.

It would be desirable to have a washing machine which could alter the nature of its wash process based on the nature of the wash load in the spin basket. SUMMARY OF INVENTION

It is an object of the present invention to provide a method and washing machine for modifying operation of a washing machine in response to the nature of the wash load.

In one aspect the present invention may be said to consist in a method of modifying a wash process in a washing machine that uses reduced water volume during agitation, the method comprising: obtaining at least one value of a load parameter indicative of one or more wash load characteristics, agitating a wash load in a spin basket with a low profile wash-plate, the wash-plate being operated by a motor, obtaining at least one value of a feedback parameter indicative of a motor load characteristic for a period during agitation of the wash load, determining a relationship of the feedback parameter value and the load parameter value to predetermined data, and modifying a wash process of the washing machine based on the relationship of the feedback parameter value and load parameter value to the predetermined data.

Preferably, the feedback parameter is indicative of the coupling between the wash load and the low profile wash-plate for the period during agitation of the wash load. Preferably, the load parameter is indicative of a wash load mass or size.

Preferably, the load parameter is the water level at which the spin basket floats.

Preferably, the load parameter is the recirculation water level shift.

Preferably, the load parameter is one of: load absorption, rotational inertia, motor torque, motor current, motor voltage, or motor speed. Preferably, the feedback parameter is indicative of motor loading and is one or more of: a motor current, a motor voltage.

Preferably, modifying a wash process of the washing machine comprises modifying an agitate velocity vs. time relationship of an agitate stroke in the agitate cycle.

Preferably, modifying a wash process comprises introducing more water into the washing machine.

Preferably, modifying a wash process comprises modifying a rinse cycle.

Preferably, modifying a wash process comprises modifying the time of an agitate cycle.

Preferably, the predetermined data relates one or more values of the feedback parameter to one or more corresponding values of the load parameter for one or more wash load types. Preferably, the predetermined data are associated with respective wash process parameters, and wherein the determined relationship of the feedback parameter value and the load parameter value to predetermined data is used to select respective wash process parameters for use in modifying the wash process.

Preferably, the period begins at least 25 seconds after agitation commences. Preferably, the period covers the range of 25 seconds after the agitation commences until 60 seconds after the wash cycle commences.

Preferably, the feedback parameter is indicative of average motor current over the period. Preferably, the predetermined data is empirical data obtained through experimentation. Preferably, determining a relationship of the feedback parameter value and the load parameter value to predetermined data comprises: determining where the feedback parameter and load parameter lie in relation to a threshold line, the threshold line relating to the predetermined data, wherein if the feedback parameter and load parameter lie to a first side of the threshold line, then the wash process is modified to operate in a first manner, and if the feedback parameter and load parameter lie to a second side of the threshold line, then the wash process is modified to operate in a second manner.

Preferably, determining a relationship of the feedback parameter value and the load parameter value to predetermined data comprises: determining where the feedback parameter and load parameter lie in relation to two threshold lines, the threshold lines relating to the predetermined data and dividing the predetermined data into a plurality of sets of data, wherein if the feedback parameter and load parameter lie to a one side of a first threshold line, then the wash process is modified to operate in a first manner, if the feedback parameter and load parameter lie to one side of a second threshold line, then the wash process is modified to operate in a second manner, and if the feedback parameter and load parameter lie between the threshold lines, the wash process is operated in another manner dependent on the position of the feedback parameter and load parameter between the threshold lines.

Preferably, each threshold line is a regression line relating to at least a portion of the predetermined data.

In another aspect the present invention may be said to consist in a washing machine adapted to modify a wash process in a washing machine that uses reduced water volume during agitation, the washing machine comprising comprising: a spin basket adapted to hold a wash load, a motor for operating a low profile wash-plate to agitate the wash load in the spin basket, a sensor for obtaining at least one value of the load parameter indicative of one or more wash load characteristics, and a controller coupled to the sensor and the motor, wherein the controller is programmed to: obtain at least one value of a load parameter indicative of one or more wash load characteristics, agitate a wash load in a spin basket with the low profile wash-plate, obtain at least one value of a feedback parameter indicative of a motor load characteristic for a period during agitation of the wash load, determine a relationship of the feedback parameter value and the load parameter value to predetermined data, and modify a wash process of the washing machine based on the relationship of the feedback parameter value and load parameter value to the predetermined data.

Preferably, the load parameter is the recirculation water level shift.

Preferably, the load parameter is one of: load absorption, rotational inertia, motor torque, motor current, motor voltage, motor speed. Preferably, the feedback parameter is indicative of motor loading and is one or more of: a motor current, a motor voltage.

Preferably, modifying a wash process comprises modifying a rinse cycle.

Preferably, the period begins at least 25 seconds after agitation commences. Preferably, the period covers the range of 25 seconds after the agitation commences until

60 seconds after the wash cycle commences.

Preferably, the feedback parameter is indicative of average motor current over the period.

Preferably, the predetermined data is empirical data obtained through experimentation.

Preferably, to determine a relationship of the feedback parameter value and the load parameter value to predetermined data, the controller is programmed to: determine where the feedback parameter and load parameter lie in relation to a threshold line, the threshold line relating to the predetermined data, modify the wash process in a first manner, if the feedback parameter and load parameter lie to a first side of the threshold line, and modify the wash process in a second manner, if the feedback parameter and load parameter lie to a second side of the threshold line. Preferably, to determine a relationship of the feedback parameter value and the load parameter value to predetermined data the controller is programmed to: determine where the feedback parameter and load parameter lie in relation to two threshold lines, the threshold lines relating to the predetermined data and dividing the predetermined data into a plurality of sets of data, modify the wash process in a first manner, if the feedback parameter and load parameter lie to a one side of a first threshold line, modify the wash process to operate in a second manner, if the feedback parameter and load parameter lie to one side of a second threshold line, and modify the wash process in another manner dependent on the position of the feedback parameter and load parameter between the threshold lines, if the feedback parameter and load parameter lie between the threshold lines.

Preferably, each threshold line is a regression line relating to at least a portion of die predetermined data.

In this specification where reference has been made to patent specifications, other external documents, or other sources of information, this is generally for the purpose of providing a context for discussing the features of the invention. Unless specifically stated otherwise, reference to such external documents is not to be construed as an admission that such documents, or such sources of information, in any jurisdiction, are prior art, or form part of the common general knowledge in the art.

The term "comprising" as used in this specification means "consisting at least in part of. Related terms such as "comprise" and "comprised" are to be interpreted in the same manner. This invention may also be said broadly to consist in the parts, elements and features referred to or indicated in the specification of the application, individually or collectively, and any or all combinations of any two or more of said parts, elements or features, and where specific integers are mentioned herein which have known equivalents in the art to which this invention relates, such known equivalents are deemed to be incorporated herein as if individually set forth.

BRIEF DESCRIPTION OF THE DRAWINGS

Preferred embodiments of the invention will be described with reference to the following drawings, of which: Figure 1 shows in schematic form a washing machine according to one embodiment of the invention,

Figure 2 shows a method of altering operation of the washing machine in accordance with one embodiment of the invention,

Figure 3 shows a graph of a feedback parameter vs. load parameter in schematic form, Figure 4 shows a graph of average motor current vs. bowl float including regression lines showing data points for a first load type and a second load type measured during experimentation for a washing machine of a wash load capacity of a first size,

Figure 5 shows an example of wash profiles for an agitate stroke for a first load type and a second load type,

Figure 6 is a flow diagram showing a method of operating a washing machine in accordance with a first embodiment of the invention, and

Figure 7 shows a graph of average motor current vs. bowl float including regression Lines showing data points for a first load type and a second load type measured during experimentation for a washing machine with a wash load capacity of a smaller size.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

Overview

Figure 1 shows in schematic form a washing machine 10 adapted to modify its operation based on the type or nature of a wash load 14 in the spin basket 13. The washing machine 10 is of the top loading, vertical axis type. The washing machine 10 comprises an outer housing or wrapper 11, an internal tub 12 and a spin basket 13 within the inner tub 12. The spin basket 13 is adapted to hold and wash items comprising the wash load 14. The washing machine 10 includes a motor 15 for operating a low profile wash-plate 19 in order to wash the items in a wash load. In a washing machine where the spin basket decouples from the motor, the wash-plate 19 is agitated. A washing machine with a low profile wash-plate 19 can wash the wash load in a reduced volume of water by setting up a reverse toroidal wash movement through agitation of the wash-plate 19. This moves the wash load in a reverse toroidal manner, thus circulating the items in the spin basket, resulting in cleaning. A low profile wash-plate 19 does not extend above the level of a typical wash load. This is in contrast to a washing machine with a central agitator post, which generally extends beyond the level of a typical wash load. The machine also includes a controller 16 which is adapted to control various aspects of the function of the washing machine 10, including the motor 15. The controller can also operate various other functions of the washing machine. These functions will be known to those skilled in the art and will not be described here. The controller 16 operates the agitate cycle by controlling the motor 15 to oscillate the low profile wash-plate 19 for each agitate stroke. An agitate stroke comprises a clockwise and then counter-clockwise rotation or oscillation of the or low profile wash-plate 19. The controller 16 might be programmed to operate the agitate stroke based on a wash profile, which will be described in further detail later. It might alternatively operate the agitate stroke based on another agitation regime. The controller can also operate or modify other aspects of the wash process implemented by the washing machine.

The washing machine 10 also includes a sensor 17 to determine or obtain at least one value of a load parameter indicative of one or more characteristics of the wash load. In a preferred embodiment, the spin basket 13 can detach from the motor 15 and float, although other arrangements are possible. In the preferred embodiment, the load parameter is bowl float, that is, the water level at which the spin basket 13 floats in the inner tub 12 after water 18 has been introduced and the spin basket 13 decouples from the motor. This bowl float can be measured by way of the sensor 17. Bowl float is indicative of the characteristics of the wash load. Alternative load parameters indicative of one or more of the wash load characteristics could be obtained using suitable sensors. This might for example be load weight, water flow rate, an agitate inertia parameter or the like. A range of other parameters could be used and those listed are exemplary only. Where a low profile wash-plate is used, the parameter could be indicative of the degree of coupling between the wash load and the wash-plate. In turn, this degree of coupling can be used to gain an indication of the wash load type and/or the nature of the wash process required in order to effectively clean the wash load. The load parameter in combination with a further feedback parameter, such as motor current, can be used as a furdαer indication of the degree of coupling. For example, a higher motor current might indicate a greater degree of coupling of the wash load to the low profile wash-plate, as the load will provide greater resistance to oscillation. In contrast, a lower motor current might indicate a lower degree of coupling as the load will provide less resistance to oscillation. It should be noted that the load and feedback parameters provide a good indication of the coupling between the wash load and low profile wash plate. In a washing machine with a central post agitator that uses full immersion in the agitate cycle, the same parameters do not provide as good an indication of coupling between the wash load and central post agitator.

The nature or type of a wash load varies considerably. A wash load 14 might, for example, contain the same or similar items, or alternatively a mixture of different items. The items themselves might have different characteristics, such as being heavy or light, delicate or sturdy. The individual items might also have characteristics such as being hydrophilic or hydrophobic, and they might have different densities and different bending stiffness. A large range of different individual characteristics of the items is possible. The combination of all these factors, such as load weight, load composition and the individual characteristics of the items in the load combine to produce different load characteristics of the wash load overall. The differing overall wash load characteristics determine how well a wash load will respond to a wash process, including the agitate cycle. Loads with certain characteristics may couple strongly to a low profile agitator/wash plate where such is used. The degree of coupling between the wash load and the low profile wash plate/agitator will affect the effectiveness of any washing. Therefore, determining the degree of coupling enables the wash process to be modified accordingly.

In general terms, a washing machine 10 according to the invention carries out the following method with reference to Figure 2. The controller 16 is programmed or otherwise adapted to operate the washing machine in accordance with the method. After the washing machine has been loaded with a wash load and turned on, the controller 16 operates, step 20, the washing machine 10 to fill the inner tub and carry out the usual functions. The controller 16 then monitors, samples or otherwise receives output from the sensor 17 to determine at least one load parameter indicative of one or more characteristics of the load, step 21. The load parameter could be any suitable parameter indicative of the load characteristics mentioned above. Preferably, in the case of a washing machine with a floating bowl, the load parameter is indicative of the floatation level of the spin basket. This provides an indication of the nature of the wash load in the spin basket, such as its mass. Alternatively the load parameter could be some other suitable parameter such as inertia (possibly measured from a motor acceleration parameter ), a measurement from a strain gauge indicating mass, or the like. The controller 16 then initiates an agitate cycle of the washing machine comprising a number of agitate strokes of the low profile wash-plate as described above, step 22. The agitate cycle preferably cleans the items in the wash load, although a "test" agitate cycle could be used in order to obtain die parameters required to determine the nature of the load and hence the required washing machine operation.

The controller 16 also obtains or otherwise receives at least one value of a feedback parameter indicative of a motor load or operation characteristic during a period of the agitatation, step 23. The feedback parameter is any suitable parameter from die motor indicating a motor characteristic or loading. This could be preferably average motor current over the period, although could be another parameter such as motor voltage. The average motor current would be measured when bowl float has occurred, that is the bowl has decoupled from the motor. The motor feedback parameter in combination with the load parameter provide an indication of the coupling between the wash load and the low profile wash-plate. From this, the degree of cleaning likely to be provided by the agitation cycle could be inferred and die wash process altered accordingly if the degree of cleaning would not be otherwise satisfactory. The parameter is measured, and appropriately averaged, over a period of the agitate stroke. This period is not necessarily the entire agitate stroke. Preferably, the period extends from 25 seconds after initiation of the agitate stroke until 60 seconds after the agitate stroke commences. More preferably, the period extends from 30 seconds to 60 seconds after the commencement of the agitate stroke. The period could extend beyond 60 seconds. The period could also commence before 25 seconds, although the accuracy of the invention diminishes prior to this point.

Once the controller 16 has received the one or more values of load parameters indicative of the one or more load characteristics, and also received the value(s) of the feedback parameter, it then determines the relationship of this to predetermined data, step 24. This relationship could be where the measured data falls compared to a threshold, data set, look-up table or the like. Based on the relationship of the obtained load and feedback parameters to the predetermined data, the controller 16 can obtain an indication of, or otherwise determine how, the agitate stroke should be operated in order to improve the efficiency of the washing of the particular wash load. For example, the predetermined data might be associated with respective wash process parameters, such as various parameters of an agitate or oscillation velocity vs. time profile, spin cycle, or water introduction phase. The relationship of the load and feeback parameters to the predetermined data can dien be used to determine a suitable wash process based on their respective wash process parameters associated with the predetermined data. For example, where the load and feedback parameters match particular predetermined data, then the wash process parameters associated with that predetermined data can be used. Similarly where the load and feedback parameters fall within a certain subset of predetermined data that has associated wash process parameters, these wash process parameters can be used. The relationship above also might provide an indication of the nature or type of the wash load, or at least provide a general indication even if it does not provide a specific indication of the nature. However, it is not essential for the controller to actually determine the load type.

For example, the relationship may determine that the wash load is a mixed cotton wash load, or a load with similar properties thereto, or a load that could be washed with a similar agitate stroke to that required for a mixed cotton load. Alternatively, the relationship of the load parameter and feedback parameters to the predetermined data might indicate that the wash load is a mixed composite load, or some load similar to it, or a load that could be cleaned by a similar agitate cycle mat would be suitable for a mixed composite load. Alternatively, the relationship does not indicate the type of load, but rather is used as a basis for modifying a wash process direcdy.

Once the controller 16 has determined the relationship, step 24, it can then obtain the wash process parameters based on the relationship, step 25. Modifying the wash process can comprise modifying any aspect of the wash process. For example, this might be altering the wash profile, which is the agitate velocity vs. time profile used to control the agitate stroke. Alternatively, it could comprise modifying the spin/rinse cycle, the water introduction phase, the agitate cycle time or any other part of the wash process or combination thereof. The wash profile itself specifies to the controller 16 the manner in which the agitate stroke should be operated or controlled. The wash profile might be a profile such as that in Figure 4 and as will be described later. In general terms, the wash profile is the agitate velocity vs. time profile of the agitator during the agitate stroke. Alternatively, it might be some other type of profile that specifies how the agitate stroke operates. Yet further, the controller 16 might generate some other type of regime or provide some other type of control parameters to determine or specify how to control the agitate stroke based on the relationship of the load parameter and the feedback parameter to predetermined data. The controller then modifies the wash process based on the wash process parameters determined from the comparison of the feedback and load parameters to the predetermined data, step 26.

Preferably, the predetermined data is obtained through experimentation. For example, the load parameter and feedback parameter might be measured for a particular "test" load for various volumes and weights of such a load. This could result in a range of load parameters and corresponding feedback parameters for that test load type for a range of different weights and volumes. A similar process could be undertaken for one or more other test loads of different types. This would result in a range of test data correlating different load parameters to corresponding feedback parameters for a range of load test types, each with different weights or volumes. This could be tabulated or graphed or put into a look-up table. The predetermined data can be the data measured during experimentation, or data based on the experimental data.

The controller 16 can then use the actual load parameter and feedback parameter that it determines through a period of the agitate stroke to determine the relationship of these parameters to predetermined data.

In one possible embodiment, the predetermined data comprises average motor current over a specific period of the agitate stroke corresponding to water level of the spin basket float point. The predetermined data of this embodiment is schematically shown in Figure 3. The bowl float level for a spin basket and motor current for a range of different wash loads of a first load type could be measured 30. This data (shown as 'x' on the graph) could then be recorded, plotted and a group regression line 31 indicating the average of these data points determined. Similarly, average motor current and bowl float for a range of different sized wash loads of a second load type could be recorded 32 and a regression line 33 found. An average of these two regression lines 34 might be also determined. Above the average regression line 34 provides one set 35 correlating average motor current to different bowl floats and below the line defines another set 36.

The actual measured bowl float and average current (i.e. measured parameters) over a period for a particular load could then be compared to the first 35 and second 36 set. Depending on which set the measured parameters fall within, a different wash profile could be selected or generated. Clearly, other load and feedback parameters could be measured and graphed or otherwise arranged in order to determine a first set and second set. Note, the first set and second set 35, 36 may contain more points of average current vs. bowl float than actually determined experimentally 30, 32. Rather, the first and second set can contain points that fall generally in the same region as that which the experimental points fall within. The experimental data 30, 32 is used to determine a threshold 31, 33 or in some other manner determine different sets of average currents vs. bowl float that can be used to determine a suitable wash process in response to measuring an actual average current and bowl float for an actual load used. A diird set could also be defined (e.g. 37) between the thresholds 31, 33. This could correspond to another wash process, or a range of custom wash processes based on the first and second set wash process. This might be for example when the wash profile is modified. Alternatively to data sets, or thresholds, a look-up table could be generated, at least partially on the basis of empirical data. Upon measuring an actual load parameter and feedback parameter in an actual load, this can be correlated to the look-up table and then used to determine suitable wash process parameters. Yet other arrangements are also possible.

In a preferred embodiment, the test load types will have a wash process generated for them that provides a suitable level of cleaning. This wash process will be determined experimentally. For example, it could be a standard wash profile. Actual loads that produce load parameters and corresponding feedback parameters diat fall within the first set 35 will then be assigned a wash process the same or similar to that which was determined as suitable for the first load type during testing. It should be noted that any actual load that produces a load parameter and feedback parameter that falls in this first set may not be the same as the actual test load type. However, the wash process determined for the test load type will be a wash process that is suitable for the actual wash load that produces load and feedback parameters that fall within the first set. Similarly, actual loads that produce load and feedback parameters that fall within the second set 36 will be assigned a wash process the same or similar to that which was determined as suitable for the second load type during testing. In this way, by comparing measured load and feedback parameters to predetermined data (that data being determined based on experimental data) a suitable wash process can be generated. It may also be used to identify or classify the wash load itself also, although this is not essential to the invention.

In summary, the predetermined data can be anything that is used for comparison to the actual load and feedback parameters that are measured for gaining an indication of or generating/ modifying the wash process. The modification of the wash process can comprise modifying the wash profile, spin cycle, water introduction phase or any other aspect of the wash process. The predetermined data could be actual data measured during experimentation, a set of data points determined from that information, a threshold or other regression line, a look-up table or any combination of the above or any other information which is determined and used by the controller 16 in order to classify the information and determine a wash process, or parameters for a wash process.

Preferred Embodiment A preferred embodiment of the invention will be described with reference to Figure 4. This embodiment relates to a machine with a low profile wash-plate 19 that has a mode that uses a reduced volume of water in the wash process. Rather than the wash load being fully or predominandy immersed in water, a smaller amount of water is used during agitation. Reverse toroidal water movement is set up by the agitating low profile wash-plate 19. This circulates the items in the wash load through the water, effecting cleaning. This is in contrast to a full volume immersion wash processes/mode, whereby the wash load is totally or predominandy immersed in water. Figure 4 shows a graph in which average current for a particular bowl float for two test load types are plotted. This is data determined through experimentation. The data plotted in Figure 4 relates to two different test load types. The first test load is a mixed cotton load, which comprises predominandy or entirely relatively lightweight cotton items. The second test load type is a mixed composite load comprising a range of different item types. It will be appreciated that while these test load types were used for obtaining the experimental data shown in Figure 4, other test load types could be used where appropriate.

For each test load type, a number of experimental agitation cycles were performed. Further, the actual weight and/or volume of items making up the test load type was varied from experiment to experiment. Therefore, for each experimental mixed composite load, multiple agitate cycles were carried out for different mixed composite loads, each having a different volume and/or weight. Similarly for each experimental mixed cotton load, multiple agitate cycles were carried out for separate mixed composite loads, each having a different volume and/or weight. For each load, the floatation level of the spin basket was measured after water was added into me washing machine. The floatation level of the spin basket is associated with a number of characteristics of the load, including its weight, volume, density of the items and the load in total. As each test load was agitated, die average motor current during a period of agitation was measured. Preferably the agitate period was 25 seconds to 60 seconds after commencement of the agitate stroke, and more preferably 30 - 60 seconds. It will be appreciated that odier time periods could be defined, and also periods that commence before 25 seconds. However, it should be noted that the accuracy of the invention diminishes when experimental data on the average current is taken prior to 25 seconds after commencement of the agitate stroke. Then, for each load, the bowl float and average current in the period is measured and recorded. This resulted in a range of data points 42 correlating average current to bowl float over the period for the test load type. The same was carried out for the second test load type to obtain data points 45.

Referring to the graph in Figure 4, the average current for a particular bowl float of the first load type 42 generally differs from the average current for that same bowl float for the second load type 45. The regression lines 40, 41 correlating each data shown in the graph identify the different average current characteristics for a particular bowl float for a test load type. This data can then be used to determine predetermined data 43, 44 for comparison to actual values obtained during washing in order to generate a wash profile. The regression line 40 relates to data for the first load type. The line 41 relates to data for the second load type. The upper regression 40 line has an equation y=12.123x — 1736 where x is the bowl float and y is the average current. Regression line 41 has the equation y=9.5534x — 1441. These apply for bowl floats between 175-270 mm. If the bowl float is less than 175 mm or above 270 mm then the regression line 40 applies. The regression lines define first 43 and second sets 44 to which actual measured load and feedback parameters can be compared. These sets are the predetermined data. Average current vs. bowl float points that lie above the first threshold or regression line 40 relate to a first set 41 and average current vs. bowl float points that lie below the second threshold or regression line 41 relate to a second set 44. A third set 46 relates to average current vs. bowl float data points that fall between the first and second threshold or regression lines 41 , 42. Note, the data points that fall within the first, second and third sets do not necessary relate to actual data points that are measured during testing. Rather the test data that is obtained can be used to determine the first, second and third sets of data points 43, 44, 46, which comprise the predetermined data.

The controller 16 is then programmed with sets of data points 43, 44, 46 and/or threshold or reference lines 40, 41 that can be used for comparison with the actual average current and bowl float measured for an actual wash load during an agitate stroke. By making this comparison, a relationship to the predetermined data is determined. The relationship is used to modify the wash process based on wash process parameters associated with the predetermined data. More particularly, the relationship in this case is used to generate a wash profile. If the measured parameters are in the first set 43, a first wash profile is generated by selecting a first standard profile. If the measured parameters are in the second set 44, the wash profile is generated by selecting a different second standard profile. If the parameters are in the third set 46, a profile is generated by modifying a one or both of the standard profiles.

In the preferred embodiment a wash profile is used by the controller 16 to determine how to operate the motor and thus the low profile agitator in order to clean the wash load within the spin basket. Figure 5 shows two possible standard wash profiles. Each wash profile 50, 51 specifies the speed of rotation of the wash-plate 19 at various stages throughout one half of the agitate stroke, for example the clockwise stroke. Each wash profile has a ramp portion A, which specifies the time taken B to reach a rotational speed C from a stationery point O in the agitate cycle. The stationary point is where the spin basket is between the clockwise and counter-clockwise rotations. The next section of the wash profile specifies the plateau time D, namely the length of time the clockwise phase of the agitate stroke remains at the rotational speed C. The final stage is the ramp time E for slowing the clockwise agitate stroke from the plateau speed C. Once the spin basket is stationary, the direction of rotation reverses to counter-clockwise. The counter-clockwise stroke wash profile is the same as that for the clockwise stroke. By altering the various parameters A-E of the wash profile curve the nature of the agitate stroke can be altered.

In this case there are two wash profiles 50, 51. The first 51 relates to the standard wash profile, which during experimentation, was determined as suitable for washing mixed cotton loads. The other standard wash profile 50 was determined during experimentation a being suitable for washing mixed composite wash loads. Any actual wash load that produces load and feedback parameters that fall within the first set 43 might have similar properties to the first (mixed cotton) load type and therefore the wash profile suitable for the mixed cotton load might also be suitable for that load and can be used. Similarly any actual wash load that produces load and feedback parameters in the second set 45 might have similar properties to the second (mixed composite) load type. However this is not necessarily the case. The actual loads might differ in nature from the test loads. The measured parameters might also indicate the degree of coupling between the wash-load and low profile agitator as mentioned earlier.

Where the actual measured average motor current for a bowl float sits in the third set 46, that is, between the regression lines 40, 41 , another custom wash profile might be suitable. This can be generated by taking either the first 51 or second 50 standard wash profiles as a starting point and altering them accordingly.

An actual method according to the preferred embodiment that the controller 16 is programmed to carry out will be described with reference to the Figure 6 flowchart. The controller 16 as described above is programmed with a regression lines 40, 41, data sets 43, 44, 46 or some other predetermined data as set out above which can be used by the controller 16 to compare with actual parameters measured. First of all the controller 16 controls the machine introduces water into the inner tub, step 60. Once this has been done, the controller measures the spin basket floatation level through the sensor, step 61. The level of the spin basket float will be different based on the load type, the mass, the volume and/or the nature of the items in the load. The controller 16 then controls the washing machine 10 to begin an agitate stroke, step 62, or number of agitate strokes. The controller 16 starts off with a default agitate stroke. This agitate cycle could be a test cycle solely for determining the actual wash profile, or the agitate cycle could be the actual cycle used to wash the clothes. Where a test agitate cycle is used, the wash profile can then be determined and used for the real agitate cycle. Where the agitate cycle is used for determining the wash profile, the wash profile can at various points throughout the agitate cycle be updated and modified as required depending on load parameters measured. For a period during the agitate cycle, the controller 16 measures the average motor current throughout that period, step 63. Preferably the period is between 25 seconds and 60 seconds from commencement of the agitate stroke and more preferably 30-60 seconds.

Once the controller 16 has measured the average current over the period and the corresponding floatation level in the spin basket, it uses the sets 43, 44, 46 and/or threshold 40, 41 to determine or otherwise generate the wash profile for the agitate stroke by determining the relationship of the measured average motor current and bowl float to predetermined data, step 64. It does this by determining a load sense number. Where the controller determines that the average current vs. the bowl float sits above the first threshold line 40, which correlates to the first set 43, the controller determines a 0% load sense number and selects the first standard wash profile 51 as the agitate stroke wash profile, step 65. In contrast, where the controller determines that the average current vs. bowl float sits below die second threshold line 41, which correlates to the second set 44, the controller determines a 100% load sense number and selects die second standard wash profile 50 as the agitate stroke wash profile, step 67. For example, referring to Figure 4, if the measured current is 1250 mA and the bowl float is 200 mm the controller will determine that the measurements fall within the first set 43 and therefore the first standard wash profile 51 should be used step 65. Conversely, if the measured average current is 750 mA and bowl float 240mm die controller uses die second standard wash profile 50, step 67, as these parameters fall in the second set 44.

Alternatively rather than determining whether the measured parameters fall within the first set or second set, a comparison to the threshold 40, 41 could be made. Each could be defined as a mathematical function as noted earlier and the measured load parameters compared to function determine their relationship to the threshold and therefore the first set 43 or second set 44. The load sense number and wash profile could be determined based on the relationship to the diresholds.

Where the measured load and feedback parameters sit within the third set 46 then neither die first 51 or second profiles 50 are selected. Rather they are used as a starting point to generate a custom wash profile. This is determined as follows. The load sense number is determined, step 65, which is the position at which the measured load and feedback parameters sit between the two threshold lines 40, 41. For a particular measured bowl float this corresponds to where the measured current sits between the two lines 40, 41. The motor controller solves the two algebraic equations for y as noted earlier. The first equation being y = 12.123x — 1736 and the second equation being y = 9.5534x — 1441. This gives two limit points on the motor current vs. time graph shown in Figure 4. As noted above if the measured value y is above the uppermost threshold function then a first wash profile is selected and the load sense number is 0%. If the measured y value is below the lower threshold 41, then the load sense number is 100% and the second wash profile is selected. Where y falls between the two thresholds 40, 41 the percentage between these is calculated as follows:

yl — y measured / yl — y2 ( 1 )

Where yl is the average measured current on the first regression line 41 , y measured is the measured average current for that bowl float, and y2 is the average current on the lower regression line 41 for that particular bowl float. This sense number is then turned into a percentage.

For example, with bowl float at 240mm, the first regression or threshold line 41 current is approximately 1150 mA and the second threshold or regression line 40 current is approximately 850 mm. If the measured current were 1000 mA then this is halfway between the two and the fabric sense parameter is 50%. This load sense parameter is then used to generate a custom profile, step 68, based on a modification the standard profiles 50, 51 as follows.

Each part of the profile (ramp time, plateau speed, and plateau time) is modified independently. Referring back to Figure 5, the standard wash profile for the first load type 51 and second load type 50 are shown. The wash profiles include ramp times, plateau times and plateau speeds respectively. Referring first to the ramp time, as shown in Figure 5, the ramp time for the mixed cotton wash profile for the mixed cotton load is 108 milliseconds, while the ramp time for the mixed composite wash profile is 107 milliseconds. The ramp time for the custom wash profile would therefore be halfway (50%) between these two values. As this is rounded up, the ramp time for the custom profile is 108 milliseconds. Similarly, the plateau speed for the mixed cotton wash profile is 89 revolutions per minute, while the plateau speed for the mixed composite wash profile is 110 revolutions per minute. Halfway (50%) of the way between those two valves is (approximately) 100, and therefore the plateau speed for the custom wash profile is 100 revolutions per minute. Similarly for the plateau time, the plateau time for the mixed cotton wash profile is 29 milliseconds while the plateau time for the composite mixed washed profile is 38 milliseconds. Halfway (50%) between these two points provides the custom or modified wash profile plateau time of 34 milliseconds. Therefore the modified wash profile generated for these particular load parameters would be 100 rpm/108 ms/34 ms. Clearly, where the actual measured current parameter falls at a different point between the two thresholds 40, 41, then the fabric sense parameter would be a different percentage and the three aspects of the custom wash profile would also be altered in accordance with the different fabric sense percentage.

After determining or generating the wash profile, the controller uses it to modify the agitate stroke, step 69. The controller optionally then remeasures current, step 63, and the process continues again. However, in a preferred embodiment, the modified wash process is determined and applied just once. In a preferred embodiment, there might be more than just a single standard wash profile corresponding to the first 43 and second sets 44. For example, several standard wash profiles might be specified for each test load type depending on particular bowl floats. More particularly, for the first test load type, a first standard wash profile might be set for all bowl floats for that load type below 200mm; a second standard wash profile defined for that type where the bowl float is between 200-270 mm, and a third standard wash profile for where the bowl float is 270 mm and above. Similarly several standard wash profiles could be defined for the second load type. In this case this would produce six different standard wash profiles, which then might be amended accordingly if the actual measured load and feedback parameters sit within the third set. It should be noted that even more initial wash profiles could be provided where required for different bowl floats. Further, the actual thresholds of bowl floats for which standard wash profiles are defined might also change.

In a further alternative of this embodiment, where a bowl float measured is below 200 mm, just one standard wash profile might be defined for all motor currents. Similarly, where the bowl float is above 270 mm one standard wash profile for the second load type could be defined irrespective of the average motor current.

The above method and apparatus for modifying the wash profiles used by a washing machine have been explained in relation to a set of experimental data that was obtained and determined as providing acceptable results. The data relates to a washing machine with a "large size" (8kg) capacity for wash loads. It will be clear to those skilled in the art that other experimental data could be obtained that might result in different "models" (meaning data sets, regression lines), which a washing machine could use in order to modify wash profiles. In such cases, the general method and hardware will be the same as described above.

For example, referring to Figure 7, a different set of experimental data was obtained, and corresponding regression lines found 70, 71, for a "medium size" (7kg) capacity for wash loads. A pre-spin step was added to improve the bowl float consistency. In this case, the different experimental data resulted in the thresholds y = 8.052x - 750.35 (70) and y = 3.2286x + 1.0572 (71). These thresholds separate the three data sets 74, 75, 76. The model in Figure 7 obtained from the experimental data can be used to determine how a wash process should be modified, based on where measured motor current and bowl float lie in relation to the threshold lines 74, 75, 76. The same process occurs, whereby if the measured motor/bowl float parameters lie in data sets 74 or 75, then a standard wash profile is used. If the measured motor/bowl float parameters lie in data set 76 (between the regression lines 70, 71), then a custom wash profile is derived based on a percentage value, which in turn is based on the position where the parameters fall between the thresholds 70, 71 as described in relation to Figures 4 and 5. Clearly, where a washing machine has different configurations or parameters or capacity, different experimental data might be obtained that will be suitable for use with that machine. This will result in a different "model" or thresholds for use. However the method and hardware for carrying out the general load detection and wash process modification will remain the same.

In effect, the preferred embodiment is measuring parameters indicative of the coupling between the wash load and the low profile wash-plate 19. The degree of coupling effects the efficiency of cleaning. Therefore, by determining directly or indirectly the level of coupling between the low profile wash-plate and the wash load, the controller can determine how to improve the wash process to improve cleaning. In the case of a low profile agitator which is designed to cause torodial flow of water and wash load in order to clean, a lower degree of coupling reduces torodial movement. Improving the torodial movement will improve the washing. This can therefore be done by altering the wash process. Preferably, this is achieved by altering the agitate cycle.

In an alternative embodiment, the predetermined data might be separated into just two sets which are delineated by a single threshold or regression line. In this case the single threshold or regression line will be the average between the first and second regression lines 40, 41. In this case, the controller will only determine whether the actual measured load and feedback parameters fall into a first set and therefore should adopt the first standard wash profile, or fall within the second set and should adopt the second standard wash profile. There would be no modification of these profiles. And yet another alternative embodiment, rather than having particular first, second and third sets and/or threshold or regression lines, there would be a continuous change in the wash profile depending on the actual value of the measured average current and bowl float compared to experimental or predetermined data. Starting from a first wash profile and a second wash profile at two extremes, the actual measured parameters will be calculated as some percentage between these and the wash profile parameters altered accordingly in accordance with the proportions. In yet a further alternative, a look-up table might be used instead of thresholds and/or first, second and third sets. In this case, for each combination of current and bowl float a particular wash profile might be specified, or else some modification required to a particular wash profile. This predetermined data again could be based on the experimental data as discussed above. In all the alternative embodiments, more than just two initial wash profiles could be stipulated, and for example six might be stipulated depending on the initial bowl float as discussed above.

It should also be noted mat while preferably the predetermined data is based on experimental data, such data could be obtained in other ways, for example through calculation. It should be appreciated in all the embodiments die actual load parameter and feedback parameter need not necessarily be average current and bowl float. Other suitable parameters could be measured. Further, the actual experimental data obtained is exemplary only. Other experimental data could be obtained and predetermined data and suitable wash profiles generated based on it. Other types of test wash loads could also be used. It should be appreciated that by comparing the obtained load and feedback parameters and comparing them to determine the relationship between predetermined data does not necessarily mean that die identification of the nature or characteristic of the load is actually determined. Rather, the relationship information is used to generate a suitable wash profile. However, it might also be possible to use this to determine the actual nature of the wash load, or at least some characteristics of this. If suitable, this could be used to operate the washing machine accordingly.

Odier experimental data and wash profile parameters could be obtained using different wash loads in accordance widi different machine requirements. The above example is illustrative and shoud not be considered the only possible solution. Also other parts of the wash process could be modified, as mentioned earlier. It will be appreciated that die data and values in Figures 4, 5, and 7 exemplary only, in order to demonstrate the general concept of die invention. Odier data and values could be used, which could implement the same general concept.

In another embodiment, the load parameter that is used to gain an indication of one or more wash load characteristics is die recirculation water level shift. This gives an indication of the wash load size or mass.

An agitation phase of die wash cycle used by the washing machine utilises low water consumption facilitated by wash liquid recirculation. This embodiment utilises the discovery that during recirculation of wash liquid, the saturated wash load holds a volume of free water diat is closely related to the load size. By load size we mean weight. By free water we mean water that will drain quickly from the wash load after recirculation ends. During recirculation, diis free water is draining through the laundry load and through the various openings in the wash basket to return to the sump. Some free water will also be airborne. Some water may also return to the sump from the recirculation conduits. Furthermore it has been discovered that the rate of flow of diis free water back to the sump after the recirculation ends is closely related to the load size. Accordingly, in this embodiment, the washing machine measures the water level at the sump during steady state recirculation where mere has been sufficient water to fully saturate the load. The control also measures the water level in die sump with the same total volume of water in the machine and with the load fully saturated, but where the recirculation is inactive. The difference in water levels (that is, the recirculation water level shift) is related to the load size. In the most preferred control, die water level is measured first during recirculation, then recirculation is halted and after a predetermined time the water level is measured again. The predetermined time is sufficiendy short that the free water has not substantially completely drained from laundry load for any of the load sizes. This technique is described in US patent application 60/938,840 filed on 18 May 2007 and is incorporated herein by reference. In this embodiment, the same method is used for modifying the washing machine operation as is used for the first embodiment described above in relation to bowl float being the load parameter. That is, in this embodiment, experimental data is obtained for a range of test loads, in which recirculation water level shift is used to determine load size, and this is mapped against average motor current. Thresholds are obtained, and these thresholds are used to determine which wash profile should be used for actual measured values of recirculation water level shift vs motor current for a real load.

In yet further embodiments, odier load parameters indicative of load characteristics (such as size) could be used as an alternative to bowl float or recirculation water level. These alternative load parameters could be one or more of the following. Load absorption (determined using a water level and/or flow meter).

Rotational inertia (using motor speed and/or power feedback). Motor torque. Motor current. Motor voltage. Motor speed.

Again, where one of these load parameters is used, the same method is used for modifying the washing machine operation as described above.

Claims

1. A method of modifying a wash process in a washing machine that uses reduced water volume during agitation, the method comprising: obtaining at least one value of a load parameter indicative of one or more wash load characteristics, agitating a wash load in a spin basket with a low profile wash-plate, the wash-plate being operated by a motor, obtaining at least one value of a feedback parameter indicative of a motor load characteristic for a period during agitation of the wash load, determining a relationship of the feedback parameter value and the load parameter value to predetermined data, and modifying a wash process of the washing machine based on the relationship of the feedback parameter value and load parameter value to the predetermined data.

2. A method according to claim 1 wherein the feedback parameter is indicative of the coupling between the wash load and the low profile wash-plate for the period during agitation of the wash load.

3. A method according to claim 1 or 2 wherein the load parameter is indicative of a wash load mass or size.

4. A method according to any preceding claim wherein the load parameter is the water level at which the spin basket floats.

5. A method according to any one of claims 1 to 3 wherein the load parameter is the recirculation water level shift.

6. A method according to any one of claims 1 to 3 wherein the load parameter is one of: load absorption, rotational inertia, motor torque, motor current, motor voltage, or motor speed.

7. A method according to any preceding claim wherein the feedback parameter is indicative of motor loading and is one or more of: a motor current, a motor voltage.

8. A method according to any preceding claim wherein modifying a wash process of the washing machine comprises modifying an agitate velocity vs. time relationship of an agitate stroke in the agitate cycle.

9. A method according to any preceding claim wherein modifying a wash process comprises introducing more water into the washing machine.

10. A method according to any preceding claim wherein modifying a wash process comprises modifying a rinse cycle.

11. A method according to any preceding claim wherein modifying a wash process comprises modifying the time of an agitate cycle.

12. A method according to any preceding claim wherein the predetermined data relates one or more values of the feedback parameter to one or more corresponding values of the load parameter for one or more wash load types.

13. A method according to any preceding claim wherein the predetermined data are associated with respective wash process parameters, and wherein the determined relationship of the feedback parameter value and the load parameter value to predetermined data is used to select respective wash process parameters for use in modifying the wash process.

14. A method according to any preceding claim wherein the period begins at least 25 seconds after agitation commences.

15. A method according to any preceding claim wherein the period covers the range of 25 seconds after the agitation commences until 60 seconds after the wash cycle commences.

16. A method according to any preceding claim wherein the feedback parameter is indicative of average motor current over the period.

17. A method according to any preceding claim wherein the predetermined data is empirical data obtained through experimentation.

18. A method according to any preceding claim wherein determining a relationship of the feedback parameter value and the load parameter value to predetermined data comprises: determining where the feedback parameter and load parameter lie in relation to a threshold line, the threshold line relating to the predetermined data, wherein if the feedback parameter and load parameter lie to a first side of the threshold line, then the wash process is modified to operate in a first manner, and if the feedback parameter and load parameter lie to a second side of the threshold line, then the wash process is modified to operate in a second manner.

19. A method according to any preceding claim wherein determining a relationship of the feedback parameter value and the load parameter value to predetermined data comprises: determining where the feedback parameter and load parameter lie in relation to two threshold lines, the threshold lines relating to the predetermined data and dividing the predetermined data into a plurality of sets of data, wherein if the feedback parameter and load parameter lie to a one side of a first threshold line, then the wash process is modified to operate in a first manner, if the feedback parameter and load parameter lie to one side of a second threshold line, then the wash process is modified to operate in a second manner, and if the feedback parameter and load parameter lie between die threshold lines, the wash process is operated in another manner dependent on the position of the feedback parameter and load parameter between die tiireshold lines.

20. A method according to claim 18 or 19 wherein each threshold line is a regression line relating to at least a portion of the predetermined data.

21. A washing machine adapted to modify a wash process in a washing machine that uses reduced water volume during agitation, the washing machine comprising comprising: a spin basket adapted to hold a wash load, a motor for operating a low profile wash-plate to agitate the wash load in the spin basket, a sensor for obtaining at least one value of the load parameter indicative of one or more wash load characteristics, and a controller coupled to the sensor and the motor, wherein the controller is programmed to: obtain at least one value of a load parameter indicative of one or more wash load characteristics, agitate a wash load in a spin basket with the low profile wash-plate, obtain at least one value of a feedback parameter indicative of a motor load characteristic for a period during agitation of the wash load, determine a relationship of the feedback parameter value and the load parameter value to predetermined data, and modify a wash process of the washing machine based on the relationship of the feedback parameter value and load parameter value to the predetermined data.

22. A washing machine according to claim 21 wherein the feedback parameter is indicative of the coupling between the wash load and the low profile wash-plate for the period during agitation of Λe wash load.

23. A washing machine according to claim 21 or 22 wherein the load parameter is indicative of a wash load mass or size.

24. A washing machine according to any one of claims 21 to 23 wherein the load parameter is the water level at which the spin basket floats.

25. A method according to any one of claims 21 to 24 wherein the load parameter is the recirculation water level shift.

26. A method according to any one of claims 21 to 24 wherein the load parameter is one of: load absorption, rotational inertia, motor torque, motor current, motor voltage, motor speed.

27. A washing machine according to any one of claims 21 to 26 wherein the feedback parameter is indicative of motor loading and is one or more of: a motor current, a motor voltage.

28. A washing machine according to any one of claims 21 to 27 wherein modifying a wash process of the washing machine comprises modifying an agitate velocity vs. time relationship of an agitate stroke in the agitate cycle.

29. A washing machine according to any one of claims 21 to 28 wherein modifying a wash process comprises introducing more water into the washing machine.

30. A washing machine according to any one of claims 21 to 29 wherein modifying a wash process comprises modifying a rinse cycle.

31. A washing machine according to any one of claims 21 to 30 wherein modifying a wash process comprises modifying the time of an agitate cycle.

32. A washing machine according to any one of claims 21 to 31 wherein the predetermined data relates one or more values of the feedback parameter to one or more corresponding values of the load parameter for one or more wash load types.

33. A washing machine according to any one of claims 21 to 32 wherein the predetermined data are associated with respective wash process parameters, and wherein the determined relationship of the feedback parameter value and the load parameter value to predetermined data is used to select respective wash process parameters for use in modifying the wash process.

34. A washing machine according to any one of claims 21 to 33 wherein the period begins at least 25 seconds after agitation commences.

35. A washing machine according to any one of claims 21 to 34 wherein the period covers the range of 25 seconds after the agitation commences until 60 seconds after the wash cycle commences.

36. A washing machine according to any one of claims 21 to 35 wherein the feedback parameter is indicative of average motor current over the period.

37. A washing machine according to any one of claims 21 to 36 wherein the predetermined data is empirical data obtained through experimentation.

38. A washing machine according to any one of claims 21 to 37 wherein to determine a relationship of the feedback parameter value and the load parameter value to predetermined data, the controller is programmed to: determine where the feedback parameter and load parameter lie in relation to a threshold line, the threshold line relating to the predetermined data, modify the wash process in a first manner, if the feedback parameter and load parameter lie to a first side of the threshold line, and modify the wash process in a second manner, if the feedback parameter and load parameter lie to a second side of the threshold line. .

39. A washing machine according to any one of claims 21 to 38 wherein to determine a relationship of the feedback parameter value and the load parameter value to predetermined data the controller is programmed to: determine where the feedback parameter and load parameter lie in relation to two threshold lines, the threshold lines relating to the predetermined data and dividing the predetermined data into a plurality of sets of data, modify the wash process in a first manner, if the feedback parameter and load parameter lie to a one side of a first threshold line, modify the wash process to operate in a second manner, if the feedback parameter and load parameter lie to one side of a second threshold line, and modify the wash process in another manner dependent on the position of the feedback parameter and load parameter between the threshold lines, if the feedback parameter and load parameter He between the threshold lines.

40. A washing machine according to claim 21 to 39 wherein each threshold line is a regression line relating to at least a portion of the predetermined data.