Sensing and Control
|
Sensing and Control |
Design
|
ApplicationsMost first-time applications probably will include existing appliance designs augmented with sensors and a new control algorithm. Retrofitting into a system may be difficult in some cases, depending on the physical amount of room to add sensors within the system. And, while retrofitting helps reduce the tooling expense for the OEM, it places constraints on the sensor designer. Slight modifications to some system components to accept the turbidity sensor or to help get optimal output signal may be required. Most of these modifications should be simple and major retooling should not be necessary. The location of the sensor inside an appliance will depend on the type of sensor chosen and the physical room available within the machine. Transmissive sensors may be placed near or around a flow-tube like an upper shower hose or drain. Scattering or ratio sensors will more likely be placed near the sump area of the machine. This is a critical factor in the design of the system. In dishwashers, consideration must also be given to such things as food settling, filming, drying of the sensor, and circulation within the system. Bubbles create an optical scattering effect similar to a particle in suspension. Most dishwasher designers will quickly tell you that it is virtually impossible to avoid bubbles inside the machine . While bubbles are certainly an issue, there are many things that can be done to isolate the sensor from such effects. In fact, the sensor can be made to operate in this environment and still give a clean output signal. Detergents add another variable into the system. For a system to be effective it must be compatible with all detergents and cleaning agents. Some detergents cause more sudsing than others, and some may have dyes that color the wash solution. Testing is required to observe the sensor's output in the presence of these variables. Determining the turbidity may require the development of an average turbidity signal. Dishwashers for example, have rapidly varying conditions such as soils changing from solids to a slurry. Instantaneous readings may vary considerably during these periods. At these times, a sliding window average of the past several readings can help to stabilize the signal. Information such as the rate-of-change of turbidity also can be obtained from the sensor. This information can tell the controller such things as how fast the soil is coming off. |
SensorsThe major challenge for the sensor engineer is to insure that the sensor will provide useful and reliable information. Turbidity sensors require special attention in this regard. Being an optical-based sensor, the turbidity sensor needs to be in a location where it will give a good representation of the events taking place during the wash cycle but where large amounts of soil or foreign objects are not allowed to block the optical path. Turbidity sensors must also be protected from the effects of air bubbles because bubbles can create false turbidity readings. In any event, it is very important that the data supplied by the sensor is understood by the appliance designer. The designer must be able to relate the sensor output to events taking place inside the appliance. Understanding the value of this information is the key to designing a SMART system. The sensors must be able to survive the harsh environment inside the appliance. Care must be taken to select the proper packaging materials to survive the chemical elements while maintaining good optical and mechanical characteristics. Filming and buildup also must be considered if the sensor is to remain functional throughout the appliance's life. The harsh nature of wet appliance applications - especially in the presence of chemical additives - makes for a rough life for the turbidity sensor. In the dishwasher, high ph solutions are extremely aggressive and have devastating effects on some materials. Of special interest is the effect of these conditions on the optical clarity of the materials the sensor will use for seeing into the system. Cracking, crazing and clouding must be avoided if sensor performance is to remain stable over the life of the machine. Optical clarity must also consider the effects of residue and scale buildup in the machine and on the sensor. Such conditions could render the sensor inoperable. The material the sensor package is made from can aid in this area if it creates a smooth, non-stick surface that deposits don't adhere to. Ratio sensors can tolerate limited amounts of build-up due to these conditions. Scattered and transmissive sensors, however, cannot tolerate such build-up nearly as well. |
Control algorithmAdding a Turbidity sensor to a wash system will require some development time to achieve control algorithms and sensor operation best suited to the application. Control algorithms that utilize the sensor's output will need to be developed. This type of control moves away from the traditional time-based designs toward the area of process control. This will be new ground for some designers. Selection of a sensor supplier knowledgeable in both sensors and control systems will make this job easier and help the OEM avoid many pitfalls. For the appliance to act smart or to make intelligent decisions, it must have the ability to adapt to the specific loads it will encounter. It is these adaptive algorithms that take advantage of the turbidity sensor's visibility into the machine and the washing process and use that information to provide shorter cycles, less water consumption and shorter washing time. Fuzzy Logic control is one way to accomplish this. Fuzzy control has recently gained popularity for its ability to be used in situations where it is difficult to model the system. An advantage of fuzzy control is its ability to accommodate a large set of process variables like loading, user preference input, detergent variables, water conditions load conditions, etc. An adaptive algorithm may be able to benefit from the use of fuzzy control as it offers a way to combine multiple sensor inputs and assign relative and appropriate importance to each of the various sensors and their influence on the controlled output.
|
| Terms & Conditions | Privacy Statement | Webmaster |
