 So, we will talk about I mean when we see what are the sensor characteristics that are required for application and how to go about selecting a sensor and or actuator I mean for example, I mean both are similar having similar requirements, ok. Now, these are basic formulations like the charge generated is proportional to some force applied, ok and where this D is a piezoelectric coefficient and, ok, this is like a static relationship. I am not considering here dynamic relationship, ok. So, in dynamics like you will have more more complicated, there is a there is a effect of capacitance also coming into picture in the constitutive basic fundamental relations for the piezoelectric sensor. So, this is a static relationship and then in that we see that this voltage is proportional to the force, ok. Then basic materials that are therefore, piezoelectric city are quartz crystal, then PZT like lead zirconium titanite and PVDF, ok. Now, out of these maybe like some of the materials like which are compatible with the MEMS processes, I mean they are basically used for see these piezoelectric sensors they are actuators also, ok. So, sensing and actuation is combined in this piezoelectric effect is reversible, ok. It is not just sensing, but like now if you apply voltage there will be a force produced. So, you can make use of these unique feature. There is a trick maybe I can talk at a little later point. When I talk about actuators, I mean it will be better, I mean I should talk it about that time because basically you can use the actuation and sensing together, ok and have the capabilities of your sensor enhanced, ok. So, say for example, if you have a accelerometer which is sensing some range of acceleration, you can have additional voltage, I mean whatever principle based on which it is working, like you can have additional actuation inside the sensor itself, ok, which can enhance like if you give some bias, ok. Now, the range of acceleration say for example, 1g to 3g it was measuring previously, then with the bias you can change that range to measure with 1 to 3, I mean like 3g change only, but you can have that at 10g to 13g, ok. So, that is possible with if you have sensor and actuator together, I mean that philosophy, I mean not many people have exploited that so far. I mean we will talk about when we come to MEMS like where it is it is very easily exploitable, ok. So, that is also possibility, ok. So, if you want a sensor which can perform very, very good and also have higher range, ok. See there is a dilemma here, ok. When you demand more range, ok, your sensing accuracy typically will go down, ok. That is you take any sensor specification and you see there the any sensor data sheets and you see the range, if they have the enhance range of sensing, their accuracy has to go down. I mean that is a way like I mean all the things in the nature behave. So, its resolution will go down, ok. But if you want to have both at the same time, ok, you can use I mean there again you can have a sensor itself as a mechatronics system where you have actuation inside the sensor, ok. That is very, very nice way of doing like high performance thing using mechatronics, ok and in the same cost, ok. When you demand high resolution and high range both at the same time, you will typically find in the current data sheets in the literature like you will find that sensor will be very costly, ok. So, higher range and higher resolution both at the same time the sensor will be costly and that that cost you can bring down by using mechatronics philosophy in there, ok. If you do not understand point, I mean this is like a important point for futuristic sensors, ok. Now, this is one of the application of piezoelectric sensor for the sensing pressure in the mold cavity. Advantage basically is again as I was talking about both actuation and sensing is possible and they are typically high frequency kind of sensors, ok. So, very high dynamics they can sense like, ok. Now, let us come to the sensor selection, ok. So, based on like, ok, you have some conceptualize some idea about your mechatronics system and you have come now to the level, ok. I want to sense variable, ok. You know the variable that you want to sense and there are many possibilities like for displacement sensing you can have encoder or I mean there are tons of possibilities there. So, what is to be selected? So, basically you come up first with the sensor requirements, ok. What exactly will suffice your job, ok? You should not have something which is overkill and at the same time something which will not work, ok. There are these two extremes like, ok. Either you have something which is overkill like, ok. You say, ok, you do not do detail analysis and then, ok, you come up with something, some specification you know for sure it will work. But I mean you see into detail like, ok, whether it is overkill for my system, ok. So, it should not be that overkill and it should also work, ok. It should not be that it is not working, I mean, ok. So, that is for that you need to do some system analysis, system modeling, ok and then like come up with system requirements and from system requirements derive the sensor requirements, ok. That is a formal way of like getting to the sensor, ok. Of course, there will be as we do in the mechanical domain like your first iteration, then second iteration of design like that. There will be some iterations, ok. You may have to go through like, ok. You select something and then you go and analyze your mechatronics like I mean the interfacing and you say that, ok, this is not working for this particular, let me change this and so that kind of a thing always you need to do, ok. But that is a design is a iterative process, I mean that is, but your analysis and your knowledge of mechatronics will decrease the number of iterations that you have to go through, ok. So, analysis you basically do either, I mean depending upon your application frequency, ok. Do you understand what is application frequency like application bandwidth, ok, where which frequency your thing is like operating at, ok, your device is operating at, ok. If it is like, say for example, this liquid level sensor, ok, that we will take it as a example and if I am just interested in like monitoring the level of a liquid, ok, not these dynamic changes in the level, ok. Just like simple level, that is very slowly changing level in, say for example, in the overhead tanks, ok. That is very slowly changing. So, there I do not need to bother about like dynamic bandwidth, ok. It is very slowly like say for example, if the liquid level in my tank changes and I get to see that only after say whatever, because of whatever like the response time of a sensor, I get to see only after 2 seconds. It is ok, right. I am not interested in like, ok, monitoring the liquid changes in my over a tank, I mean second to second, ok. But the same thing if I am going for the slosh application, where I am interested in knowing like how the liquid level is changing because of the motion of the tank, then it is a different matter. I cannot use that, ok. So, you understand the difference, ok. So, you need to see what are the end goals that your system is supposed to work for, ok. And accordingly you come up with bandwidth and like amplitude range requirements for your sensor, ok. And then once you know the analysis and at least rough analysis and you are clear about some of the numbers, then like it is a matter of just going to the data sheets and selecting, ok. This is appropriate, this is not appropriate like that, ok. Then based on the analog and digital sensor, you need to worry about the noise in the system, ok. This is very important practical point, ok. I mean theoretically everything is sound, ok. You have a linear sensor and all those things you say. But when you practically connect the wires, I mean every wire is sitting there to act as an antenna, ok. I mean will you may. So, it is very important to think ahead of time like how, what is the way you are going to connect, how much is the length of wire that you are going to use and what is a magnetic field or electromagnetic sources around your application, ok. Say for example, typically this is a case like all the mechanics engineers they will go for some motor. They will be some motor in the and for motor you need a motor driver, ok. High power motor is there, there is a high power driver. And typically all these drivers they are based on the something called edge bridge configuration. Let us see I mean I have that in the slides, but I do not know whether I will get a time to talk about in detail. But in those edge bridge there is high frequency switching that is happening. See any high voltage switching will cause electromagnetic disturbance, ok. Say for example, here in the tubeless you have a choke, ok, where there is a high frequency thing going on. High frequency and high electromagnetic that choke coils are really heavy, ok. So, high voltage electromagnetics is there. Similarly, like if you go for motor controller I mean edge bridge controller I mean they will have this kind of a switching going on there. So, wherever that acts as a source of noise to your sensors and your sensor sitting in a electron or mechatonic kind of an application by the side of this kind of a source is really damaging, ok. And actually we have seen in our application of this Losh rig a lot of noise, ok. We are using the pressure I mean not pressure the force sensor 6 axis load cell, ok. And it used to give a lot of noise then we will have to think about what are the strategies will eliminate this noise. And if not eliminate completely like what we can do to at least software process that thing. So, that like we get the signal which is of interest, ok. So, signal processing consideration I do not know whether they are getting covered in this course, but they are again important in this sense, ok. Then the cost implications as I was saying you have to balance against like what is the cost that your company is permitting you, ok, to spend, ok. And you may be recommending that within whatever this much cost it is not at all possible or this is overkill I mean you do not need really that higher cost I mean you can save on the cost. All those kinds of things like if you do proper analysis what are the requirements then you can come up with the numbers, ok. This is I am talking about specifically for robotic applications there, ok. See for robotics like first you need to do kinematic analysis, ok. Then only you can go for dynamic analysis. Basically this kinematic I have mentioned for the robotics application. What you need to do is basically a dynamic analysis, ok. Ah, ok. That is what I was telling when I consider the example of slosh, ok. When you are interested in monitoring the liquid level in the overhead tank you do not have to consider the dynamics, ok, because it is changing very slowly over the day. But in my slosh application like I will show you video the liquid is changing its level very fast and very having various kinds of ways in which it is sloshing. And I want to study those ways of sloshing. If I if that is my end goal I need to have a sensor which can respond faster than that application. Whatever the changes are happening. Oh, how is the dynamic analysis done? You need to go to then this next step which is mathematical modeling of a system, ok. This you have to mathematically capture the dynamics, ok. Say for example this slosh application, ok. I can just consider it to be a pendulum kind of a system. Simple pendulum kind of a system and study the dynamics, ok. Though it is a complex application I mean where lot of fluid motion and fluid structure interaction is involved like one of the ways, ok, depending upon the frequency of course. But like whatever this this kind of a slosh I can study by using pendulum analogy, ok. So, so like that you need to think about even an application what is the way I can come up with the mathematical model, ok. Once you come up with the mathematical model then like you analyze that see it should be a reasonably representing your phenomena that is happening. Like say for example one of the students in our mechatronics course like he was studying I mean he was studying the dynamics of motor and he asked me a question sir why why we should not take a air damping when the motor is rotating, ok. So, so the thing is that see these are all the phenomena that are definitely happening, ok. But are they relevant for my process? Are they affecting my process, ok? See motor is rotating air which is surrounding the motor how fast it is damping. If you compare the damping in the bearings and damping of the air which is rotating I mean which is like moving around I mean because of the shaft rotation it is weak gap, it is hardly I mean it is negligible you can safely I mean very safely ignore that, ok. So, so you I mean that quality is there with all of us engineers, ok that seeing the phenomena observing the phenomena very carefully you need to think about, ok, whether what is important what is not important to model, ok. See you cannot say that there is a perfect mathematical model for any phenomena in the world there will be no perfect model. All models whether it is linear or non-linear there will be approximations, ok. Approximations that you are specifying, ok. In this range that application will application range that approximation will be valid, ok. So, so that is very very important point to take home that you need to be able to see, ok. And we have all of engineers have that quality, ok to some extent we need to be conscious about it and develop that, ok. But we have that quality that, ok, seeing the phenomena we know, ok what is going to be important in this phenomena, ok. Now to translate that in the mathematics is a skill that we need to like develop and master, ok. Once we develop that skill then whatever this like how to come up with the dynamic requirements of a sensor will be very easy, ok. So, modeling of the phenomena, ok under consideration with a simple mathematical model which is representing like most of the essence of the application that is that is a key, ok. Case study I will see what time for me is like I mean I am already running late, but let us see. Running on a very bad road. Yeah, you do not need. What is the variation in this level? So, you see the. Exactly. What? Yeah, that is very right. In volume I need at least 3 sensors to find out the first about the plane of the movement and the response type of that, how fast it is going to move you cannot visualize this even. When car is running on a very rough road it will move like a. See, yeah, yeah. So, you will not be able to judge what is the level because like in car when the liquid is moving or in the petrol tank your fuel is moving and if you have very high response time sensor placed in there it will always be fluctuating, ok. It will not show you what is actual level I mean. So, you have to do something to like have this response time damped or like. Exactly. Yeah, that is a good example, ok. So, once this mathematical model is there and you do the analysis little bit of analysis dynamic analysis do use MATLAB to carry out simulation and then you can come up with proper specifications and then like it can you can use them with the manufacturer's catalog to select a particular sensor for your application, ok. So, that is about that is all about sensors. Now, we will focus on actuators now, ok. Some of the basic principles and things are similar I mean say for example, selection of actuators I mean again there you need to worry about bandwidth and how fast you want to have the dynamics going and things like that, ok. So, I will not spend the time on that, but I will give you the flavor of typical actuators that are used in the in the mechatronic system, ok. AC and DC motors, stepper motors and hydraulic actuators are three major classes that are used for mechatronic system. And solenoids of course, I mean they are not I mean they are not kind of a servo actuators, ok. They are like a I mean 01 kind of actuators, ok. So, piezoelectric motors and other pneumatic devices we will not talk about, but they can also be used in mechatronic system. Pneumatic devices mainly you will find go well with the PLC kind of a controller, ok. How many of you are working with the you are working with PLC? So, there it is kind of a logic which is very easy to be programmed and then you have this kind of actuators, pneumatic actuators working there quite well, ok. So, DC motors the basic principle is wire carrying conductor placed in magnetic field will generate a force when so that force is basically used for like moving and generating like whatever like this rotary motion in the motor or whatever, ok. So, there are two primary classes one is brush type DC motor and other is brush less type. So, brush less DC motors because of obvious reasons like there is no friction in the brush to like reduce the speed I mean we can have these motors going at very high speeds, ok. And then you can have also AC similar kind of AC motors also. Wide variety of motors are available in the market, ok. So, I will not get into the details about these, but there are the main thing that you need to look for is whether the torque or whatever torque and speeds are compatible for your application and whatever actuator you are selecting, ok. That is again based on that actuator similar to the sensor selection you have to do this analysis and come up with the actuator specifications, ok. Solid states, solid state switching is basically you use the semiconductor, ok. There is no physical switch, ok, a transistor switch which is solid state, ok. So, stepper motors are basically basically used for high torque and low speed kind of application, ok. So, we will see exactly how the stepper motor works I mean the principle and other thing. There is a stepping action as the name suggests in the motor, ok. So, it is not like a smoothly running I mean, ok. Of course, this smoothness will depend upon what is required for the application, but typically the stepper motor if you are feeling like you rotate the shaft of the stepper motor, you will not be able to rotate it smoothly. It will go in and lock into some steps, ok. So, that is the way the stepper motor is designed to work actually. This is ideal for implementation of digital control. So, why it will become clear in the next slide. Also there is a holding torque in the stepper motor even if you have no power given to the stepper motor, it has some holding torque, ok, which may be important for some applications, ok. So, that you can make use of. Then there are various types of constructions and configurations possible and these are the two different types of stepper motors, but basic principle is like this. You have a permanent magnet which is here and these coils surrounding it can be like excited impulses, ok. So, say for example, when you excite one of the coils, ok, the magnet orientation will change in that direction, ok. So, that is a basic basic principle of stepper motor, ok. I am exaggerating it here like it is not really just 90 degree step at a time, it is 1.8 degree step usually for many of the stepper motors, ok. So, that is like it is very small steps it can move, ok. So, your job, I mean as a mechatronic engineer is to make sure that, ok, this excitation sequence is proper, ok. So, there are many different kinds of sequences also possible for the stepper motor. Say for example, in this particular sequence you can say this is a half step mode, ok. You are exciting two coils at a time and one coil in the next insert, I mean next step, ok. So, these logically can be displayed as in here, ok. So, this is a half step sequence. You can see these two coils are excited at a time and in the next step only one coil is excited, ok. Then the next coil is excited in the next step, ok. So, this logic sequence you need to develop in your microprocessor, ok. Did you see yesterday digital input outputs? No, ok, but ok. So, in microprocessor it is possible to configure or like to program the pins of your digital, there are pins that are provided like maybe 16 pins or 8 pins will be provided for digital input outputs, ok. These pins you can program to give either 0 or 1. So, you can program this, like say for example, in this case we will require these 4 pins, ok. One a, one b, two a and two b, there are 2, 4 coils to be excited. So, you need 4 pins that you will program there and on that pin 0 and 1 you give according to this sequence and your motor will start running, ok. So, these are the various applications of stepper motors in your dot matrix printer. So, see these are the sources like say if you are in the college and you want like for your students to gain knowledge in in mechatronics. These are the sources CD-ROM drive, stepper I mean this printer floppy drives are the sources from where you can like some gone floppy drive you can connect collect and then like you can show them, ok. This is a stepper motor you can I mean so that is just like directly available at no cost that electronics is available for you. So, you can make use of that and demonstrate students to like what are the like sophisticated mechatronics you can demonstrate at no cost at all, ok. So, CD-ROM drive stepper motor for example, we have developed one setup around the CD-ROM drive to teach principles of CD-ROM, ok. And it has all the necessary things we have the laser of CD-ROM drive actuated separately not as in the CD-ROM drive, ok. We just take a head and then like actuate the laser and you can see now all the focusing signals and the tracking signals and all those things, ok. So, that is a way I mean if you are college teacher you can or in your industry also I mean you can think about what can be used because this is a high quality electronics believe me like anything you would take out from the from the printer or floppy drive or whatever scanner or whatever the very high quality electronics that is available at no cost very low cost. So, if you are like requirements and that match you as well borrow it from existing high industry, high quality industrial product, ok. The question is between DC and stepper, ok which one is more controllable? Oh actually see that will be told you told to you by your application, ok. See more controllable is not really see I would say what are the specifications that you are you are looking for. So, if you so see now, ok. This stepper motor technology has like really developed in the years and people are using stepper motor also in the floppy drives, ok. Of course, they are not using it for hard disk drive or like, ok. So, that kind of a precision whatever is required for floppy drive is possible with the stepper motor, but it should be high end stepper motor. But typically stepper motors are not used for very high resolution kind of application typically, ok. So, this is just industrial hydraulic motor you can see this is a axial motor there on the top here you can see axial piston motor with a swash plate and other thing and then it is driving the this industrial product, ok. These are you this is a use of hydraulic devices I mean where anywhere you see earth moving machinery, ok. You need to have hydraulics there, ok. Because you need typically very high forces, high power, ok. And in the in the mobile application it can be generated very easily by using pumps oil pumps oil pressure basically and use that for hydraulic actuators, ok. Now, so basically how now the the way is, ok we have selected actuator and sensor. Now, what are the different interfaces possible for interfacing them with your microcontroller, ok. This cannot be done independent of microcontroller selection I mean I mean I am not talking right now of microcontroller selection or microprocessor selection, but I am talking about interfaces, ok. So, you need to make sure that, ok your microprocessor is having those interfaces or appropriate interfaces for your application, ok. And, ok there are if you are using microprocessor then these interfaces you will have to create, based on like some chips which are used for input output ports for example, A255 chip for example, it can be interfaced for digital input outputs. So, like that you need to create those, ok. Other solution if you are not really in the electronic engineer and you do not want to mess up with electronics like you can directly go for microcontroller, ok. Microcontroller are the chips that provide these interfaces to you directly, ok. If you are working in a research kind of environment like it is not a product which is but your the product is in development, ok. Then instead of going for microprocessors or microcontrollers you can go for data acquisition systems directly which can be sitting in your computer and controlling your application, ok. They will have their own microcontroller, microprocessor on board, but you would not bother about like programming details of that microprocessor, ok. So, there are various kinds of microprocessor interfaces or digital control implementation interfaces possible. I am not going into details of that I am just talking what are the sensors and actuator interfaces. And, typically you see this is a configuration that I have shown here that your control computation will take place. There will be clock which will be synchronizing all these events like you have analog to digital converter here which will be taking your sensor. I am just showing this analog to digital for convenience, ok. This can be a encoder interface also I mean completely digital interface also, but there will be some interface from the sensor which will interface your sensor with the controller and computer or microprocessor or whatever, ok. It may be simply if it is in analog domain like it may be simply the electronics at it, ok. So, you have many possibilities in all of these elements, ok. Then from control you do control computation and then like you need to give it to your actuator which is typically in the analog domain. Say for example, your servo motor, ok. Saper motor is a special case which is in the digital domain you can say, but typically all the other actuators will be in the servo mode and you will have them as analog actuators, ok. So, you it is required to convert that digital signal into analog domain and that is done by this digital analog converter, ok, which can be interfaced with your microprocessor. So, microprocessor, ok. Yesterday you saw the fundamentals of microprocessor, but just microprocessor is not sufficient. You need to have these like peripheral interfaces, ok. So, it is like just the computer chip 4866 or Pentium chip you are given that is not sufficient for you to work. I mean there should be some interface where you can you can be seeing what is going on in the chip on the monitor, ok. So, you need that kind of a monitor interface. So, this is a similar thing, ok. Just this microprocessor chip is not sufficient that is a brain, but you need hands and legs or eyes to see and operate, ok. So, those are the interfaces through which like microprocessor can talk with the high power mechanical application, ok. Ok, this is a question that I was I paused on like few minutes ago. Like can your computer be connected to any motor directly? What is your response? Yeah, exactly. Small motors. Small motors, very small tiny motors you can think of, ok. Even small motors if you want to drive through your computer like you will need some something similar to sound card, ok or you can drive that motor through the sound card. If you connect it directly to your parallel port or serial port you will destroy, ok. So, so, so never, never, never, ever attempt, ok. This was one of the questions we asked in the exam like if what if I like one of the DD students connects the motor directly to the computer what will happen? Like power requirements on both the sides are totally mismatch, ok. So, you need something some device called amplifier in between to do the job and provide the extra power that your application needs, yeah, ok. So, the simple, simplest interface, ok, you might be familiar with is a sound card, ok. So, through sound card maybe you can drive some speakers which are having low power. If you want to drive like high power speakers you need some more monitor or some device to take care of that, ok. So, it is exactly same like the way you are familiar actually with all these concepts I mean in your day to day life, but I mean you see it from different perspective and it will be useful for your mechatronics application, ok. This sound card is an interface that you can use I mean this is for people who are in the college and they want to make use of CD-ROM at the example that I gave for control. So, they can interface the CD-ROM drive with the sound card, I mean sound card output instead of going to speakers they can drive the coils in the CD head, ok. And then like you have with no cost at all the complete mechatronic application in place, ok. So, digital analog converter is one of the interface PWM is another interface, ok pulse width modulation, ok. We will see what is that in the brief and then digital signals, ok digital port you can have 0s and 1s programmable on the digital port and you can use that for driving stepper motor. Then you can have, ok. The more high-ended cards like detection systems that you get in the market as a I mean as a product, ok. Mainly for the research purposes you can use that and those maybe having some kind of a serial or parallel interface with your computer, ok. So, that RS232 or some kind of a CAN is another interface for automobile applications. So, there are people have developed lot of interfaces other than these, I mean these are most typical kind of interfaces directly for the hardware, but other interfaces are for mainly for communication, ok. So, say for example, if you want to drive a simple servo motor with maximum current limit 4 amperes, let us say. Then this MC33186 is a Motorola chip which you can use for that kind of a application. This is very simple chip which is having this edge bridge configuration and that can be used for DC servo motor control with PWM, ok. And when you want to like have, see stepper motor also you cannot directly connect stepper motor terminals to your digital input output. So, that interface or amplifier interface for the stepper motor is this ULN 2003 chip. A lot of interfaces actually are chips available, but these are the things which you can directly start with for building your application one of the examples, ok. There are all these issues we are talking about. The other more the very important issue is reversing the direction, ok. We will see how it is done, what are the solutions. I mean you just amplifying a signal is not sufficient, ok. So, this direction reversal needs a special treatment or special attention, ok, especially in the servo motors not in the stepper motors. Stepper motors you just change the sequence and it will reverse the direction, but step I mean servo motors is that is an issue, ok. So, this power amplification let us not dwell into, ok. See typical transistor characteristics is having like very low, see it is kind of a dead zone in the near 0, ok. This voltage versus collector current, amplifier current, ok. So, this kind of a dead zone characteristics is where the problem is, ok. So, you will not be able to drive your motor at very low speeds, so to say, because of this kind of a characteristics. And you do not have these characteristics on the reverse side, ok. On the reverse side you do not have these characteristics, ok. That is another problem. So, that is why you cannot reverse the direction, ok. So, what you do is use this pulse width modulation, ok. You use this pulse kind of a signal, where these are very high frequency pulses which are square in square type of pulses. And the voltage of this pulse is remaining constant, but the time period, time period is remaining constant, but this on time all or called pulse width is what is modulated or changed, ok. By changing this pulse width, you can change the average voltage that is seen by the motor, ok. So, here like these pulses are changing very fast, but motor sees only the average level, ok. Motor does not see, because motor cannot respond that fast. These pulses are typically like 10 k or 20 k kilohertz kind of a frequency, ok. So, you cannot see it changing that fast. It is like a tubelight. Tubelight also there is a fluctuations happening, but our eyes cannot detect those fluctuations. We see that it is as a continuous source of a light, ok. But those fluctuations which are at higher frequencies, we cannot see. Like that, these motors will not be able to see these fluctuations, ok. So, you need to have this point in mind when using PWM for any other applications rather than motor, ok. Where say if you are saying like piezoelectric application I am using PWM, piezoelectric will respond to this, ok. This high fluctuation, immediately it will respond. But because response time for piezoelectric is much, much smaller, ok. So, you need to be wise in taking decision while using this PWM philosophy for any other application. The basic idea here is that you are not changing the operating point on your transistor characteristics, ok. The operating point on your transistor characteristics is say for whatever like fixed voltage that you have in the PWM this, this fixed voltage, ok. You are not changing that operating point on your transistor characteristics, ok. So, you are always operating at this point and so corresponding to that you will get current, ok. So, you are not at all touching this zone. That is a basic, basics of like using PWM. So, you avoid that dead zone by having your operating point fixed. But now how to change the energy that is if the operating point is fixed, how to change the energy like amount of time for which it is on you reduce or increase. That way it will allow you to change the energy that is going into your motor, ok. Finally, what matters for motor to see is average energy that is going into the motor, ok. So, that is how you can beat application around this philosophy. And all the components for this kind of a way are easily available in the market. You can program your digital input output ports to generate this PWM and you can use this 3D 186. See, this is a thing which we give in our mechatronics course for students to like I mean interface and see themselves, ok. So, this is a most simple way of constructing your own mechatronics system, ok. Then you can have like more higher ended systems built by using more higher ended like drives, ok. So, 3 3 186 that chip is basically edge tree edge bridge driver. It is it is it works basically on the switching of the transistors, ok. So, this say for example, if you have this out of this four switch switches these are transistors, ok. Basically switching is done by the transistor, ok. So, these transistors S 1 and S 4 when they are switched on the motor will go in one direction and when they are open and S 2 and S 3 are switched on motor will reverse the direction, ok. So, it is basically the reversal is done by using transistor switching in the edge bridge and edge bridge chip has that I mean everything is there on the chip to take care of your direction reversal. You need to give appropriate signals to the pins of the chip and then it will there is a pin on which like if you give 0, if you give 1 it will change the direction, ok. So, that is how in edge bridge I mean that 3 3 186 chip it will happen, but if you go to internally what is happening is this is what is happening internally into the chip, ok. If you want to have specified accuracy and range how that specifications affect like selection of actuator, ok. So, say for example, for one micrometer you cannot use anything which gives you the friction, ok. So, basically friction is a very killing factor in very high precision applications, ok. So, you need to avoid friction and backlash, there are two things to be avoided if you are going for one micron kind of a precision, ok. So, in that case what is done is basically like if you want to have a motion you use some kind of a cantilever, cantilever to which your stage is suspended and you keep it in the magnetic field, ok. So, as you change the magnetic field this cantilever will bend and there is no friction happening anywhere, ok. You get this point, so it is a simple philosophy of like flexures, this is called flexure, ok. Flexure mechanism or flexure joint you can use and you can have reproducibly same thing, I mean exactly at the same position it will stop when your voltage is at the same point, ok. So, that is a precision that you will get because there is no friction. If there is a friction like next time you cannot predict that at that point the same amount of force, frictional force will be there, ok. Because of the wear the surface might have changed its characteristics and it would have changed the friction force. So, high precision positioning will not be possible with the friction, ok. So, that is the main point of like this example, ok. So, those you are applying magnetic field you need to have the magnetic coils or permanent magnet and magnetic coil and you can like that magnetic force you apply at the end and then like flex this thing will have to bend, right. When you apply force on the cantilever it will bend. So, that is how the force is generated, ok. Ok. The sensor interfaces are essentially the same like you in addition to the digital and AD I mean analog to digital converter you have encoder interface which you do not have for actuators, ok. So, that is a additional interface. Again that has some something to do with the way the like I showed you in encoder. So, whatever 0s and 1 coming out of that encoder need to be processed in proper way to say that, ok this is one count this is another count like that. So, that is there is a, ok. How is it happening is a little complicated thing, but how to use that is very simple, ok. See so, you need to have a balanced understanding, ok. You need not be knowing everything that, ok how is it happening? You may be in specialized in whatever your field, ok. So, you may not know details about how microprocessor is working, ok or how flip flops and other things are changing to make the microprocessor work. I mean that is not your field of specialization. You need not know that, but how to use microprocessor that is for sure you should know, ok. So, some of the things you may be knowing may not be knowing does not matter, but like it is important to know what are the things which are important to know, ok. So, that as long as you think about and like make the sense of, ok what is important to know and then you are fine, ok. So, it is not like your skill in a particular area that will help you, but it is a thing, ok. What is important to know for this application to work with a high performance with a lower cost? What is important to know? As long as you are making that kind of a decision, ok, this is not sufficient. I need to know more about this to get this thing. That is a kind of a field that if you are getting for the application that will work best, ok. It does not matter what expertise you have, ok. You just need to have that kind of a with that kind of a understanding. I mean, it will come through, it will come through. I mean, most of you have that actually and you just need to be aware and enhance that, ok. And it will come through as you go along and consciously be aware, ok. Am I taking right decision? I am not taking. All those things it will slowly start developing, ok. Just being aware about that kind of a faculty, it will get developed, ok. This is the rig that I was talking about. So, these are the actuators, ok. This is a motor. There is a encoder on this motor at the back and then there is a ball screw which is hidden under this paper which is used for like preventing dust to enter there which is moving this linear stage which is placed on the linear bearings, ok. And this is a mechanism which will be giving me the tank a rotary motion, ok. And here again there is this linear slide. You can see that there is a stage which is moving in the linear direction as this rotary excitation motor rotates which is again having this encoder, ok. This has a big amplifier attached to this motor and then from that amplifier we are taking the signals into actually computer, ok. So, are giving the signal from computer to this amplifier and then that is driving this motor, ok. So, I will show you this video. You can see I am not able to point out because it is showing in some different monitor on this monitor it is not seen. But you can see how the linear slide is moving and along with that also the linear the pitching actuation motor is also working on the at the same time. And if you see at a liquid level closely it is not changing at all, ok. So, this is what we are doing actually is we are giving one motion to the linear degree of freedom like I mean linear translation we are giving some motion and we are adjusting the other motion such that the slosh is compensated, ok. So, this it is possible to do that, ok. So, this is a demonstration actually of slosh cancellation, ok. I mean this is not really a feedback kind of a control, ok. We have developed couple of feedback control strategies also on these. Actually at a base of this tank if you notice carefully you will find that load cell, ok. No, no it is hidden here, ok. This is a wire that you see for the load cell, ok. So, we have the load cell which is at a base of the tank which is capturing the forces and movements in all six whatever like three translate I mean three forces and three movements we are capturing, ok. And because of the sloshing whatever these movements and forces are generated they are getting recorded and that is can be used now for like identifying pendulum parameters, identifying different proposing different models, ok. So, those that data can be processed further to like get the meaningful information out of the system. I mean this is a project from ISRO, you know in launch vehicles that go in the space there are like lot of fuel they carry, ok. The 70 percent of the weight that goes in the space is liquid. So, the liquid motion has lot of effect on your trajectory. I mean for this Saturn trajectory I mean Saturn failure was because of the slosh. I mean these forces were not characterized before and because of that like the controller that they had developed was not sufficient to like take care of these and because of that there was a Saturn failure. So, slosh is very important for space rockets, ok. So, ISRO has like given this project to us to develop this rig and like develop different models and model parameters schemes algorithms for the space application, ok. Now, this technology has been transferred to them and they are also now developing a on a bigger size. This is like 10 kg tank, but they are developing on a bigger size like 50 kg or 100 kg tank to carry out their experiments and they will be all using our control algorithms to really identify various parameters, ok. That is a practical example. So, in conclusion like the automation engineer needs to have good knowledge about these elements, ok. Sensors actuator and they are interfacing to really synthesize your mechanical system, ok. All right.