presented by:
Dr.B.Ramamoorthy

---

[attachment=9916]
SENSORS FOR INTELLIGENT MANUFACTURING
Introduction
Between1920 and 1960 expansion of large scale processes created problems in measurement and control.
 1960’s development of Semiconductor technology and opportunities for electronic signal processing and control.
 lack of sensors for converting non electrical parameters in to electronic compatible signals.
 sensors – precise, reliable, capable of integration, acceptable dimension and price.
 More than 10000 publications 1000s of companies yet many variables need solution
 Wilhelm von Siemens ,built first sensor in 1860 – temp. dependence of resistor wire of Copper for measurements.
Definitions and Standards
 output is an electrical quantity, measurand is a physical quantity, property, or condition to be measured.
 output could be even optical signals (future).
 No limit on size, cost and complexity is defined. .For ex. MR System weighing several tons costing several million dollars and a thermometer are sensors as per definition.
 sensors, transducers, detectors, meters, pick ups, gauges, cells all synonyms have different historical origin.
 No rules governing mechanical and Electrical interfaces of sensors. Neither standardized dimensions for sensor housing, signals (except temperature sensors).
 Lack of standards hampers Dissemination of sensors.
 Distinction between Sensor and Sensor Element is not easy.
 A device that provides a usable output to a specified measurand.
Physical and Chemical Transduction Principles
Mechanical - length, area, volume, pressure etc.
Thermal - temp., entropy etc.
Electrical - Voltage, current, resistance etc.
Magnetic - Field intensity , flux density, permeability etc.
Radiant - intensity, phase, wave length, refractive index etc.
Chemical - composition, concentration etc.
 Usable Output in response to a specified Measurand
 Specified measurand is inherently different from the desired output signal. At least one transduction principle is needed to make a sensor work.
 Transduction principles are better known as Physical or Chemical effects. More than 350 effects in dictionary. Six groups as follows.
Problems in Classification
 1) Some of the transduction principles are reciprocal (table) whereas others are not. eg. Piezoelectric effect (mechanical stress generates electrical charge and vice versa).
 2) More than one output signal. eg. A diaphragm( mechanical / mechanical) converts pressure in to deflection and stress. Either of these could be used for conversion in to electrical signal. Capacitive or inductive effects for deflection, piezoelectric or piezoresistive for stress.
 Transduction principle characterized by one input and one output?
 3) Each box could be sub divided. More basic studies are required.
 4) Most sensors operate in a certain range ( temp. sensors).
 5) Definition of signal domain has ambiguities Eg. Radiant signals could be considered as thermal, high frequency electrical and magnetic signals. Thermal energy is microscopic mechanical ( Brownian) motion.
 So, Sensors can not be classified based on Physical / chemical effects
Classification of Sensors
- Physical and Chemical principles ; for students
- Measurand (primary input variable) ; for application engineers, But sensors for a measurand will look very different in different fields of application.
- Technology and material ; for development engineers, may render the possibility of producing inexpensive sensors. Rapidly emerging chemical and bio technologies are otherwise difficult to cover.
- Application ; application engineers; pressure sensors in automobile are cheaper, biomedical are smaller and so on.
- Cost
- Accuracy
Trends
System oriented rather than the supply of Single Sensors.
New Principles, Technologies, materials produced a wealth of possible new sensors in 1980’s.
Reliable, Robust and render the possibility of economic / mass production. Eg. Silicon sensors principle was known in 1960’s but only after a real breakthrough in mass production it became popular.
Miniaturization ( especially in medical, automotive and aerospace applications).
Mtl. Properties/ behavior of components are different for macroscopic / microscopic / nano scopic components.
Sensor signal processing supported by integrated / modular designed electronics. Higher speed of signal processing.
Provision of suitable and standard interfaces .Adaptable multi sensor Systems. multiple simple sensors and parallel processing of signals. Theoretical conception, modeling, CAD will play a key role in future.
Need for Object Detection/ Detectors;
1) Conveyor assisted applns. Jam/ empty line detection etc.
2) Safety and accident avoidance; loading of machines.
3) Inspection of products; containers filled/ not filled , incorrect closures etc.
4) Counting; detecting missing parts
5) Sorting; by size, color etc.
6) Hopper level detectors, feed cut off controllers;
7) Host of miscellaneous applns.; Including edge guidance, over aging roof detection etc.
Fundamental detector methodologies:
1985 survey indicated inductive and photoelectric detectors will dominate. Others are capacitive, Hall effect, Wiegand effect, magnetostrictive approaches etc