 Good morning everyone, I am Roshad Mistry, assistant professor in mechanical engineering department and today we will be discussing sensor characteristics. The learning outcome of the sessioners, the student will be able to define static characteristics sensors and dynamic characteristics of sensors. Now, what are static characteristics? Because sensor characteristics affect their measurement, capabilities and define the suitability of a sensor for a particular application. So, when you choose a sensor for a particular application, obviously you have to go through the list of characteristics which helps basically build the specification and then make a choice. Obviously there is a range of static and dynamic characteristics that you must consider and depending upon the application, certain weightages will have to be given of one particular characteristic over the other. So, let us review actually these characteristics which we already studied as a part of the measurement measurements and systems course. So, first obviously is accuracy. Now, I have written accuracy and error together because they are sort of complementary terms. So, accuracy is defined as the term in terms of closeness to the true value and error is defined as the difference between the measured value and the true value. So, obviously a smaller error that means the instrument is more accurate that is why we tend to speak in terms of accuracy and error alongside one another. So, accuracy is defined as the terms of closeness to the true value and error obviously being defined as the difference between the measured value and the true value. Coming back to errors, we can classify them basically as precision errors or random errors and second is bias or systematic errors. Now, bias and systematic errors typically are present in all instruments. They are difficult to detect and correct and cannot be very easily compensated for statistical methods. And these typically are due to calibrations, loading errors that is effect of sensor and the measurement wrong choice of signal conditioning and so on so forth and errors due to the effect of environment such as temperature. Most the properties of most substances which are used as transistors are temperature dependent they are not fixed. So, they change as the temperature changes and if the way and the change in temperature over the course of measurement is significant then obviously errors will be introduced. So, you have to be careful regarding the errors which are corrected. So, again it is necessary to understand what the types of errors are these are precision errors or random errors and then bias or systematic errors. Calibration errors to a certain extent can be managed, but the other errors are very difficult to predict and compensate. Then another important characteristic is often called as reproducibility. Some textbooks and authors tend to term it as repeatability and in other books the term use this precision. They are all often interchangeably used and it is typically defined as the ability of the measuring system to give the same or in the back obviously almost the same output for repeated application of an input with the with the value being kept constant. So, input value is been kept constant, but it is repeatedly given and then we see what the output is for that particular value. So, the more closeness to this particular value gives the idea of a repeatable reproducibility or repeatability. Random errors reduce the repeatability and also do hysteresis and noise. So, typically if there is a poor repeatability of an instrument you might want to check for hysteresis effect and any noise which is associated noise to a certain extent can be compensated by using proper signal conditioning. Another property is obviously the range and span. Now range or span is the maximum will and minimum value of the measurement that the sensor can that the sensor can sense and obviously output. For example, a sensor may have a range of 1000 degree Celsius that is minus 100 to 900. So, the range is 1000 degrees, but it can measure typically from minus 100 to 100. Now there is actually a bit of an issue when it comes to the nomenclature here. Some authors define the range and span separately. So, they in one case range has been defined as the maximum and the minimum value that can be measured like for example, in the above case the maximum value can be that is measured is 900 degree Celsius and the minimum value is 100 degree Celsius. And span is then defined as the absolute value of the difference between the maximum minimum. So, the span of this instrument you can say is 1000 degrees whereas, the range is from minus 100 to 900. So, this is one issue with the nomenclature some books do not tend to differentiate this particular definition. So, that is bear in mind the differences in nomenclature which occur between different authors and different books. Then another important property is resolution. Resolution is the smallest increment that the sensor or measurement system can accurately and reliably detect. In case of analog sensors the quality of the primary transducer often decides the resolution. In case of digital sensors the onboard signal conditioning circuitry is as important as the transducer itself. This is absolutely essential in all digital sensors. The most important obviously being the analog digital. Poor resolution of the ADC will result in poor resolution of the transducer. No matter how good the original analog primary transducer is. So, resolution is another important property which will determine how the smallest increment sensor can detect. Another important property again is stability and drift. Here also I have spoken of stability and drift alongside because stability refers to the ability of a sensor to give the same output when the applied input is held constant over a period of time. A deviation of this value is termed as drift. So, let us say if the sensor temperature sensor is measuring a certain temperature which is supposed to be held constant in a bath let us say a salt bath and you have to measure it over a period of time. Now obviously this temperature is being held constant but let us say the sensor measures a slight deviation over a period of time then that is obviously a lack of stability and the deviation from this value is termed as drift. This is very critical in process monitoring, medical equipment and surveillance equipment. So, this is one property which you have to bear in mind when you choose sensors in such applications. When there is a one-shot use this is typically not important but where prolonged measurement of a value is expected then obviously stability has to be considered. Now like I said this is a review now see if you can recall some of the remaining characteristics that we discussed as a part of this sensor characteristics both the static and dynamic characteristics and then we will see what are the other characteristics as well. Now let us continue with the remaining sensor characteristics. One of these characteristics is saturation. Now all transducers reach a point after which there will be no output no matter what the input. This is called as a point of saturation. So here for example in this case you notice this is a linear range of measurement after which there will be no output no matter what will be the change in input. So this saturation point also you have to bear in mind when you choose because the range typically will be decided on the basis of this. Another important characteristics is hysteresis. Now measurement typically is bi-directional as a quantity moves as the quantity we measure along the rising as the quantity rises in value and as also when it decreases in value. Now obviously this measure is never constant so do you typically get a band and this is called as the hysteresis loop. So no linear error resulting due to the difference in output when measuring from the lower end of the range to the higher end and then back to the lower end this is called as hysteresis and remember this is one of the major sources of errors in bias errors in sensors. Another characteristic again is dead space or Z zone. This is a small region near the lower extreme of the input range or often near the zero starting where there is no change output remains zero in spite of any change in input. Now this can often be a desired feature as well if you do not want any shunting effect to occur. So once the input causes this Z zone then the sensor can display the output. This is also a non-linear characteristic. The property is sensitivity. Sensitivity of the sensor is defined in terms of the change in output with respect to the change in input on a per unit basis. Therefore for analog sensors it is the slope of the transport and input output characteristics. So if it changes with over a period of time typically we tend to choose as per the choice where the measurement range could be. So this is a region of high sensitivity, medium sensitivity and again a region of low sensitivity. Now linearity, most sensors manufacturers specify accuracy in terms of the sensor range of measurement. This is often the case. For example accuracy would be let's say 1% of range. This is because to account for errors due to any non-linearity in the sensor input output characteristics a least square line of fit between the sensors input and the output is used to determine the accuracy. So this least square line is fit and then this band will define what is actually the accuracy of the instrument. So that is why often it is measured in terms of the range of measurement. Now let's look into dynamic characteristics. There aren't too many but these are again important when and where we come across applications involving dynamic measurements. Now see this is the overview of all dynamic characteristics. You have the peak time. This is called as the settling time. This is the rise time. This is the overshoot. Now certain characteristics are not mentioned here which we will study it at a later stage. But remember at least these four rise time, peak time, settling time and the overshoot. And these are actually defined as follows. This is the peak time over here as measured. This is the time taken by the instrument to reach a peak value. Now this is actually the value what it is supposed to measure. It often overshoots that and the maximum point it reaches that is termed as the overshoot. And the time taken to reach this particular value is the peak time. Then another time is settling time. Now often especially for this is actually a stepped input, the sensor will settle to a certain value near the expected value. Now when it comes within a certain band, let's say within 1% of the true value and it settles within that 1% of the true value we call it as a settling time. So this is marked typically as a time required for the output characteristics to be within let's say 1% of the expected value. Then one more time is what is called as rise time. Now that is measured as the time taken by the instrument to move from a specific range let's say from 10% output to reach 90% output. So how much time does the instrument requires to go from 10% of the value sorry to let's say 90% of the value that is given as the rise time. And the amount of overshoot that is the measurement from the expected value to the maximum value is from this overshoot. Time constants is another property which is not shown, but we will discuss this at a later test and so and also our response time. Okay, so this will discuss at a later stage. Now again I refer, recommend that you refer mechatronics handbook, Bishop chapter 18 which covers sensors and actuator characteristics in detail and also you can refer mechatronics by Bolton or any other mechatronics books for any supplementary information. Thank you.