 This is a lecture on vibration transducers as you know in condition based maintenance about 70 percent of the cases of machinery health monitoring is done usually through vibration. So, in this class we thought that you should look into the merits and demerits of the available vibration transducers and what are the different types of vibration transducers used in industrial health monitoring and in the subsequent classes we will be talking in details regarding how actually vibration monitoring is done. So, as you know what is vibration you all know vibration can be represented either as a displacement velocity or acceleration well if displacement is denoted the amplitude is denoted as a you know velocity is omega a and acceleration is omega square a. So, you see there is always a strong relationship between the amplitude of the quantities measured and the frequencies. So, in one way at high frequencies you will see because of this omega square a term the acceleration is going to have a high magnitude compared to displacement. That is actually a frequency independent phenomena and it is actually a low frequency phenomena I always I am asked this question what is the best parameter to measure you will see as we go through this class sometimes at low frequencies it is actually the displacement for example, if you think of a shaft which is there in a casing this is the casing of the shaft or a housing what could happen is this shaft while rotating if it is not on the rotating at its center there may be a subtle slight eccentricity and because of this at any of the points you will see the displacement at one location is a 1 and another location is a 2 for the shaft to be uniformly or concentric to the center all these a's should be equal if you measure in the four quadrants or four directions at 90 degrees to each other. But if you will see a 1 is not equal to a 2 is not equal to a 3 etcetera then the what happens is we know there is some problem because of eccentricity and how can this be detected if I have some sort of a sensor here displacement sensor or some sensor which senses this gap between the rotating shaft and the casing it can tell there is something wrong with the machine. So, this is where actually displacement sensors are used if you could be slowly turning the shaft it may not be powered by a motor you could be slowly rotating by some mechanical device think of a large cement plant where we have the large very long rotary kiln could be of the diameter of 2 to 3 meters the diameter of the kiln and this could be about to know about 2 to 3 meters in diameter. So, what happens in such a case if there is a slight eccentricity we are going to have a problems in the forces unequal forces etcetera. So, this actually you know you would have seen when they go to install and commission such equipment they actually put lot of dial indicators this is a nothing but mechanical string loaded they have a plunger and stylus. So, the plunger is going to move depending on this a 1, a 2 etcetera then we will have a corresponding reading and this dial indicators are very good in the sense you know if you want to measure the surface roughness in a bearing to out of roundness in a bearing the small variations here could be measured by a dial indicator and this could be in the order of microns. So, to measure displacements we have a mechanical device a dial indicator looks something like this and this is spring loaded this is the plunger. So, as you intuitively see that if the motion is very slow almost close to static I can use a dial indicator to move the see the displacements. But imagine if the bearing or the rotating sharp was moving at a very very high speed because of the inertia in the plunger this plunger is not able to respond to quickly changing stimulus. So, the need for high frequency measurements arises and that is where our dial indicators or L V D T is another linearly variable displacement transducers have a limit in the frequency range as to how much they can measure. I mean for low speeds they are good in fact L V D T these are good for less than 50 hertz. But imagine I want to monitor the vibrations of a gas turbine gas turbine could be having the about close to 30,000 rpm. So, 30,000 rpm will become 30,000 by 60 there is about 500 hertz and that is will be the fundamental frequency. So, such a high speed measurements are not possible through mechanical devices like an L V D T or dial indicator and so on. So, we have to see what are the instruments or transducers available to measure high frequency vibrations. Before I go into that let me just give you a general overview of what are the different types of vibration transducers available for us. One is the piezoelectric accelerometers and we will be discussing in details regarding the piezoelectric accelerometers because 99 percent or 98 percent or 95 percent of the vibration transducers are actually piezoelectric accelerometers. Nowadays because the convenience of measuring omega square a which is acceleration is very easy and then once I have this accelerations I can divide it by omega to get velocity and divide it by omega square to get displacement. So, this covers the entire frequency range from you know 0 all as close to 0 to about 5000, 10000 hertz which is not possible with the L V D T or the dial indicator we just discussed because another one is the L V D T and then one thing you would notice that be it an L V D T or a dial gauge or for that matter an accelerometer they have to be always in contact with the surface at which we are measuring the vibration. But there are instances where we cannot touch this surface for example, the surface is so soft, so surface is so hot temperature is very high surface is not accessible very minute location in a very intricate position we cannot use an accelerometer or an L V D T. So, that is where we have to use certain laser based techniques be it to measure the normal vibration or the rotational vibration of course, there are some probes like the eddy current proximity probe which can be used to measure the gaps or the measure the it is like a non-contacting type of transducer it we can measure the displacement or acceleration in such or velocity in such cases. The eddy current proximity probe and moving coil velocity pickup they essentially measure the velocities and then they are non-contacting in nature, but one again disadvantages of the eddy current proximity probe or the moving coil velocity pickup is that they cannot be measured on surfaces which are non ferritic surfaces has to be ferritic to measure the eddy current or the to use the eddy current or the moving coil velocity pickup. So, let us come to the brief principles of operation of m piezoelectric accelerometer essentially what the piezoelectric accelerometer has is basically if I put these crystals piezoelectric crystals either in this form either I compress them or I shear them they are going to develop a generator force and this force is responsible for generating the charge q and this charge could be converted to a voltage with a suitable charge to voltage converter amplifier and basically I will get the voltage signal. These are known as the charge type of piezoelectric accelerometer what happens here suppose I make this accelerometer here this is mounted on a surface wherein it is having a certain motion x. So, essentially in this piezoelectric type accelerometer all I have is set of piezoelectric crystals and mount within certain disc and they are bolted and then I will get a charge q which is proportional to the displacement of this seismic mass this q is proportional to y minus x and with mathematical equations you will see eventually this q is proportional to the base motion could be in the form of acceleration. So, once I get this charge as you know charge cannot be stored charge has to charge will decay with time. So, immediately we have to have a circuit wherein this charges has to be converted to voltage. So, this is where a piezoelectric charge type accelerometer would require a charge to voltage amplifier because this amplifier is is outside the accelerometer. The advantage of thus the charge type accelerometer is and this is a sealed unit this is a sealed unit. Advantage of such accelerometer is this can be subjected to very very high temperatures the charge type of accelerometers, but the disadvantage is that you have to carry a charge to voltage preamplifier with this accelerometer. So, with the miniaturization of electronics what has happened is people have put the amplifier inside the accelerometer in a small I C chip, but then to power the accelerometer I have to give it a power source it requires a power source. So, once such accelerometers are there then what we can do is. So, this is the same accelerometer, but the I C chip is inside install inside the accelerometer and all you have to do is just take a cable and this is a voltage output because you have to the same cable could be used another conductor could be used to give the power supply to the I C which is used to convert the charge to the voltage, but the disadvantage of this is because the I C chips are integral and kept inside the accelerometer housing and usually this housing is you know stainless steel very hard and you can never break it open they are hermetically sealed. So, this accelerometer cannot be subjected to high temperatures high temperatures we cannot subject them to high temperatures because this I C chips will be get damaged, but for simple viruses measurements in the room where the ambient temperature is you know 30 to 40 degree or around 30 to 35 40 degree Celsius is good. They are not good beyond 100 degree Celsius the I C P type you know there are many trade names for this you know like you will have the I C P types the delta tron delta tron piezotron isotron you know this tron comes from the electronics and delta piezo iso I C P these are different trade names the companies used to sell their charge amplifiers which have the I C inside it. So, all you will require is just put that accelerometer give it a power supply and you will straight away get the voltage. So, this is very convenient from a vibration monitoring point of view because nowadays many of the signal analyzers have the built in power source which could be used to power such a I C P type accelerometers and they are very convenient, but the strongest disadvantage is that they cannot be subjected to high temperatures. In fact, while doing experiments on measurements on very high speed gear boxes I have lost couple of accelerometers because the technician mistakenly put the I C P type accelerometers on a bearing which are very very hot and we are monitoring for about 30 40 minutes. Another day I get no signal and see that the I C chip has got fried because the high temperatures. So, we have to be careful when using such I C P type accelerometers. We will go to the details of the accelerometers later, but let me just tell you about the other important accelerometers or vibration transducers and this is how actually in the inside of the accelerometer this is how it looks like and this is the mass which is loading it and this is the piezoelectric crystal and this is a tightening bolt which goes and this whole unit is sealed. This is a ceramic unit as there is a steel unit and this is sealed and because of the relative displacement here and the piezoelectric mass we will get a charge which is proportional to this displacement or acceleration and then we will see a voltage and this voltage could be measured as the other. So, the charge sensitivity could be usually picoculums per meter per second square or in the I C P type it will be the voltage sensitivity milli volts usually meter per second square. I will move on to the and this is how the typical piezoelectric accelerometers look like. If you will see this is the size of a fifty paisa coin and this is where you connect the cable to connect the charge or to measure the charge which has been generated and similarly this is the top connector this is the side connector and this is what is known as the triaxial accelerometer triaxial accelerometer because of the fact that you know acceleration is highly directional. So, at any one point you will see in a machine that we have to measure in the x direction in the y direction in the z direction. So, rather than putting three uniaxial transducers in one location the three piezoelectric crystals are put at one location and then we can see you know this means I am measuring in the x direction this is perpendicular to x in the y direction and this z is out of the plane of the projection and so then I get the 90 degree three directions which are 90 degree perpendicular to each other mutually perpendicular. And if you see this hole this is just a hole to mount the accelerometer at location just by tightening a screw or you could have you could put studs like this on the accelerometer and then tap a hole on the surface where you are measuring the acceleration and then we can measure the acceleration. Another very very important note you have to make while using this charge accelerometer is the cable. Cable plays a very very important role because you know this conductor which is there and there is a signal wire in the ground wire and if there is a relative motion between them they are going to generate a charge. The cable is violently moving having motion and this cable is laying on a surface which is having excessive vibration because then you have your accelerometer here is having this accelerometer here and then there is excessive vibrations. Now because of this this cable is going to have lot of motion and because this motion they will be charge generated which is not actually your charge due to the acceleration but charge due to stray noise. So, lot of stray noise will occur if I do not take good care of the cables. So, these cables are very very special cables and they reduce what is known as the tribo electric noise and a good way to do that to reduce tribo electric noise is to ensure that the cables do not whip around or slash around or. So, what you have to do is you put this accelerometer here usually you tie the cable some sort of a tape. So, that the cable does not move because if this is a charge type of accelerometer lot of charges are going to get generated if the cables are going to move and they are going to interfere with the measurement charges and the best idea is to best ways to have this amplifier the charge to voltage converter or amplifier very close to the accelerometer. And usually this cables are not very very long about 1 meter to know 3 meter maximum because beyond that the cable noise will be so much that you will contaminate your vibration signal. So, this is another precaution you have to take while using the charge type accelerometers for measurements, vibration measurements. But this kind of issues are not there when you take when you have the ICP type of accelerometers wherein we straight away get a voltage signal in the cable rather than charge. Here we have the charge and charge will decay and charge you are generating. So, you have to have the charge to voltage amplifier. In fact as soon as the charge is generated you have to convert it and then transmit over long distances because sometimes if you have to measure at a one particular location and record at may be 100 meters away you cannot have 100 meter tribe electric cable which is slashing or moving around and then generating lot of noise. So, your vibration measurements will be no good and that is where you have to be careful. And you can get a feel of how large or how small this accelerometers are. Obviously, to increase the sensitivity of the accelerometers we have to have more piezoelectric crystals. So, more piezoelectric crystal means heavier it will be. So, heavier accelerometers will have high sensitivity. Usually, typically this kind of accelerometers which you see in the palm of somebody's hand it is about 10 picaculum meter per second square. And if you have the right type of voltage converter this could be about 10 milli volts per meter per second square. There are accelerometers wherein we have to measure very very low levels of acceleration. For example, a floor vibration wherein you have put a tunneling electron microscope or a scanning electron microscope. The vibrations level could be about 10 to the power minus 6 g micro g's you know micro gravity levels. For such measurements we have to have may be transducer having 1000 milli volts or 1 volts per meter per second square. So, this calls for very very massive amounts of piezoelectric crystals. So, whenever you have high sensitivity requirements the transducer will be big and heavy, but you have to pay a price for the transducer being big and heavy. The frequency response will be low. So, you obviously cannot measure very very high frequencies with high sensitivity accelerometers. You know this could be the frequency of such heavy accelerometers could be up till may be 10,000 hertz. Suppose I want to measure very very high frequency vibrations high frequency vibrations. To measure high frequency vibrations my frequency response of the accelerometer has to be very very high may be about 7200 kilo hertz. So, they have to be very very lightweight. So, the sensitivities are very low may be 1 milli volts per meter per second square. We want to measure accelerations very very high accelerations for example, during a shock generation something fell on the ground. So, in such a case and shocks are usually very very high frequencies. So, we have to use light accelerometers. So, if you go to the catalogue of any manufacturer you will see different types of accelerometers made available to you whether. So, if I was to just specify the generic specifications of an accelerometer. By accelerometer I mean piezoelectric accelerometers and that is what we are going to discuss. One of course, it is sensitivity whether and of course, before that could be the type of accelerometer by type I mean whether it is charge or I C P type where the electronics inside by sensitivity it could be whether it is in picoculum per meter per second square. If it is the charge type or milli volts per meter per second square next important is the frequency response. The frequency response I mean that every transducer is also a mechanical system. So, this is also going to have its natural frequency may be. So, measurements up to this range about you know 0.7 times if this is its natural frequency 0.7 times f n. So, this is the useful range wherein or this is the useful range wherein you will measure the. So, this accelerometer is good only up to 0.7 of f n. Then of course, the other maximum temperature it can be subjected to can withstand. This is very very important because when you talk about industrial health monitoring there will be gear boxes and bearings which are subjected to very very high temperature 300 400 degree Celsius. May be the you should know from the manufacture whether the accelerometer can be send or subjected to such a high level of temperature. The directional sensitivity whether the accelerometer is sensitive to which direction. So, we have to mount the transducer in that direction to which it is highly sensitive. It could be uniaxial or triaxial and then of course, you know subjected to which it should not get affected by a nuclear radiation by EMI by moisture. Does it require external power supply or not? It is mounting methods and so if you and of course, the very important is the mass of the accelerometer. Because as you know, vibration is a very very important dynamic property of a system and the dynamics of the system is going to change if the mass of the accelerometer is larger than the mass of the system which whose vibration you are measuring it is going to influence it. Imagine we have a very very thin desk which is rotating or which is having a motion some sort of transverse motion and this desk may be weighs you know 100 gram. I obviously cannot put an accelerometer on this which is weighing 500 grams because the entire dynamics of the system is going to change. So, one rule of thumb which we have to always keep in mind that mass of the accelerometer has to be less than one tenth of the mass of the system on which you are placing the accelerometer. Otherwise, this what is known as the mass loading effect will occur and then you are not going to get the true reading of the acceleration. Now, before I continue more on the accelerometers, I just want to mention you about couple of two important probes which you also use and this is essentially an eddy current proximity probe which is a non contacting probe basically what happens we have a probe here on to which we generate we apply a high frequency signal very very high frequency signal about few megahertz and that is there in the primary of this there is a coil here and primary and because of this high frequency waves which are close to the shaft which is iron shaft eddy currents are going to generate get generated and this eddy current will be generated in this gap and so the secondary voltage which is sensed by this proximity probe will be proportional or will be a function of this eddy current and this eddy current is a function of the gap. So, we can measure such eccentricities or gaps which are created between the shaft and its journal of the casing because of a radial run out eccentricity by putting such eddy current probe. They are non contacting type of course, they have a very very low frequency response compared to the accelerometers, but the advantage is I obviously cannot put an accelerometer here because accelerometer is only going to measure the vibration of the casing here I can no way measure the vibration of this shaft because I cannot put the accelerometer in contact with the shaft surface. So, relative displacement can be easily measured by such eddy current proximity probe. In fact, many of the steam turbines etcetera they come up with this pre installed with 2 or 3 eddy current probes and if this is your x and this is your y if you plot them x and y and if they are identical and of the same frequencies you will get a perfect square. So, sorry perfect circle and these are good for what is known as orbit analysis and if there is just looking at a phasor diagram of the x and y on a 2 plane system you will see if it is a perfect circle you know this system is a concentric to the casing and it is rotating perfectly and with time this may change like this this may change like this and then you know that something is wrong with the eccentricity of the shaft and this is a very good way to report it or record it or observe such eccentricity by having an eddy current proximity probe. So, usually in steam turbines large compressors axial we can put such eddy current proximity probe to measure the relative motion between the shaft and the casing. Another non-contacting type transducer is the velocity pickup. Essentially there is a permanent magnet and a moving coil and because of the motion of this system here I am going to get an EMF and this motion is because of this is contact this is in touch with the surface the velocity type pickup and this will get we will get a velocity proportional to this EMF which has been generated. But a modification to this is if I have something moving here and this is the coil is not there and this is ferritic we can have a key phasor kind of signal it will give a signal like this. So, this is these are also used to measure rotational speed because with virus and health monitoring you know measuring the rotational speed is also very very important. I cannot obviously put a tachometer always to a rotating shaft when end of a rotating shaft, but if I have an arrangement like this wherein I have a shaft on which there is a key way or a key and then if I put such a pickup. So, at every time the key way comes below the pickup this voltage this gap is going to change. So, there will be a voltage generated with time. So, just measuring the time period I will know the frequency of rotation as the in us of the time period just by having a frequency counter I can measure the rotational speed and in the industry all such rotational speed measurements is done through an inductive pickup and all you and then you can transmit this signal over many many meters or kilometers for that matter and on real real time always you will have this rpm trace. Now, earlier days when the piezoelectric accelerometers were not developed people were using such moving coil kind of velocity pickups and dedicated pickups, but nowadays with the piezoelectric accelerometers available we have only two important parameters which you need to measure one is acceleration and other is the rotational speed. So, these are the two most important vibration though rotational speed is strictly not vibration, but I am vibration transducers which we have to use for machinery health monitoring. I will now focus on some applications as to how this accelerometer is used and mounted on systems and so on. So, now that we know about the three types of transducers you see this is the eddy current probe velocity transducer and piezoelectric accelerometer. If you see their frequency range and the relative amplitude which they can measure the eddy current probe can measure low frequency and the limit is about 2 kilohertz and the range is not much velocity transducer has a low frequency limit of about you know 10 hertz to about you know 1000 hertz, but the piezoelectric accelerometers can measure all the way from 0 to very very high you know 70 80 kilohertz and the range is also very very high the dynamic range is very high. So, that is what I am telling to measure the acceleration we now will be only talking about piezoelectric accelerometer. Before the piezoelectric sensing elements were used people used to also use strain gauges instead of the piezoelectric crystals to measure the seismic accelerations and this is how a typical chart of an accelerometer will look like this is the frequency response and this is the resonant frequencies. And then what are the methods to mount the accelerometer and what are the voltage and charge sensitivity as a function of temperature etcetera and this is typically which we will find in any of the accelerometers which we purchase. The question is we have to use this vibration measurements and question always you would be asking well I have got a certain value is that good is that bad I do not know. So, there is a standard ISO 2372 standard and recent it has been updated and that is the new standard I will tell in the next class that tells the machines power as to the maximum permissible level of vibration in a frequency range of 10 to 1000 hertz and this is usually in millimeters per second. So, if you know 100 half power machine you know it should not be more than 3 millimeters per second and so on. So, that gives you feel like when you are doing vibration measurements on machinery is for the health monitoring you know the power of the machine. So, the maximum acceleration or maximum RMS level in this 10 to 1000 hertz band should be not beyond a certain level which is given in that standard. And this is how a typically this is a essentially a charge to voltage amplifier and this is the accelerometer here and this is the special cable I was mentioning about which reduces the triboelectric noise and sometimes you will see the cables are costlier than the accelerometer itself because otherwise you know if this cables are beating around if they move also they are manufactured in such a way that there is so much of reinforcing mechanism in the cable itself the actually the copper conductor which is there inside hardly has any motion. It is actually there are multiple sheets of protection in this cable there will be a layer of plastic insulation steel reinforcement and on top of the copper wire and then there will be a plastic coating sometimes silicon coating. So, that it can be used under water etcetera. So, this cables are very very expensive you know when you are talking about underwater accelerometers or hydrophones this we have accelerometers underwater accelerometer wherein you can measure the variation of underwater structures, but the cables there are very very costly they will not allow water to creep into the or seep into the conductor and contaminate the readings etcetera. And usually when we have this is an accelerometer any actual accelerometer of the top connector and this is actually a gold plated connector here to make sure that there is no corrosion etcetera and is that cable here and which goes to a vibration meter wherein you can select the knob to see the readings of other acceleration velocity or displacement. Now, the next question you would be asking yourself is how do we actually physically fix this accelerometer on to the surface on which we are measuring the vibration. Surface could be as soft as the flapping wings of a bee or as delicate as a buzzing bee wings or as robust and heavy and hot as a gear box casing. Everywhere there is vibration how do you how do you measure that. Obviously, on a gear box casing I can mount my accelerometer many ways. One is I could attach if this is a usually the gear box etcetera these are cast iron. So, I can very easily tap and put a stud and screw the accelerometer on to the stud. So, one is the stud other is I can attach a soft magnet to it, but again out of my experience I am telling you this magnets the soft magnets are very soft. This magnets cannot be subjected to high temperatures which time suddenly you will see that the magnet has cracked. So, usually this magnets cannot withstand high temperatures. So, usually permanently if you have to mount you make a hole and put a stud and then you sometimes tap the accelerometer on to it. Sometimes the surfaces are so hard or some machine the OEM would not allow you to do any tapping which would permanently damage the structure weaken the structure. How do you do that? In such a case what you do is on to the surface you glue a block which could be glued permanently by a cement like the cyanide. Like you know we talk about those very quick type of glues which are available in the market etcetera. You can use them to glue the block on to the surface whose vibration you are going to measure. People will not ask you to let you do a drill a hole and do a tapping that is not allowed. So, usually you will use such a device a block on to this block you can screw your accelerometer with a tap. And this actually the stud mounting is the best and then comes the glues and magnetic base if it is soft and they have pretty high frequency limits. But sometimes even just you know a layer of bees wax on a surface the honey beehive bee wax you can just put it here and then stick the accelerometer on to it. Just press it and this is the bees wax they are very strong. But the problem is we cannot use them beyond 40 degree. In room temperature these are good, but beyond that they will melt and they cannot bear. Another type is if you are quickly serving a surface you can put a hand hold probe and then move around accelerometer. Now, what I am going to do is I am going to show you some of the mounting methods through pictures. This is how the accelerometer if you buy it looks like in a kit. This is the accelerometer and this is that special cable I was talking about this is the magnet. This is the bees wax and there are lot of studs washers etcetera. And this is the tap given to make the tap and this is a typical kit which you will see of an accelerometer. And then accelerometers always need to be calibrated. So, this is the hand held calibrator we were in you can give a known levels of 10 meters per second square at 1000 radians per second. And then you will see the acceleration values. Whatever voltage it comes it gives you know that it gives 10 meters per second square of acceleration. And this is typically we are measuring using a magnet here the vibration in a test rig. And this is where another test rig wherein we have put an any axial accelerometer and that special cable. And this is an industrial type of accelerometers which can be subjected to very very high temperature this cable is special. And then we have an meter, vibration meter which you will actually see the you can see the spectrum. These are all there in our lab you can go to this website and see this system as well. And this is one of the reluctance type velocity pickup which we have used on the shaft. We can also use an optical encoder to measure the rotational speed. And we can use single point lasers also as an accelerometer to measure the vibration. Because this is the refrigerator compressor wherein we are measuring can see this spot here where we have put a laser beam. And once we put a known laser beam because of the reflected wave there will be a Doppler shift. And the Doppler shift will be proportional to the velocity of the surface. And we can measure the for surfaces where when we cannot approach for example, here in another case we have a transformer which is there in our lab. You can see these points which are shining actually these are the points we have put reflected tapes wherein we can measure the velocity which is with which this fins are vibrating. And then we can shoot the same laser vibrometer to measure the vibrations of the surface. Because here we are measuring the sound intensity here. Another application of this laser based vibration measurement system is using a rotational laser vibrometer wherein we shoot two beams if you can see there are two dots. If you have two laser beams separated by a particular distance if you know v 1 and v 2 if you know delta r you can find out omega. And that is what we are using this laser vibrometer to measure the rotational vibrations. Another way of the rotational vibrometer and this comes with the signal conditioning system wherein we will straight away get an omega as a function of frequency. This is another view of the same rotational laser vibrometer this is a laser head which shoots in shoots two beams. Thank you.