 Hello, everyone. I'm J.S. Fendiari, and thank you very much for joining us today. I'm joined by my colleagues, Tom Boquino, Mauro Escan Yuzo, and Chris Kim, to discuss the industrial applications of our latest high-precision inclinometers. This webinar should last about an hour, and we will try to reserve the last 15 minutes to tackle as many questions as possible. So with that, let's get started. Here is our agenda for the next hour. We will briefly talk about the CMAC electronics and give you a quick overview of our sensor technology. We will then dive into SD inclinometers and the industrial applications. At the end of this webinar, we will introduce to you our free-of-charge software libraries and also the available evaluation tools that you can use to evaluate these sensors. Let's start with a brief overview of who we are. SD microelectronics is one of the world's largest semiconductor companies. Our revenue reached more than $9.5 billion in 2019. SD microelectronics employs about 46,000 people worldwide, and about 17% of these employees are in research and development, who are working on new and innovative technologies and solutions. We have 11 manufacturing sites, and our product portfolio includes a wide range of semiconductor products that serve many market segments. Please visit SD.com to get more information and details on all products we offer. We have been in MEMS and sense of business for more than 20 years. During the last two decades, we have been working on some of the greatest and most innovative solutions to serve our customers. If you compare the performance of early MEMS-based products with the ones we offer today, you can see the huge improvements we have made in terms of performance while trying to bring down the development cost and making the price of these devices more affordable. The performance improvements and the cost reduction have enabled many new applications and have helped increase the accuracy of the existing applications. All these advances have helped us to maintain our leadership in the market and further expand our customer base for MEMS sensors and micro actuators. We offer a wide range of sensors. Our sensor portfolio includes inertial, magnetic, pressure, humidity, and temperature sensors. We also offer MEMS microphones. On the macro actuators side, we are the leader in microfluidic, macro mirrors, and piezo actuators. We provide customers with a wide range of custom products and solutions. So before jumping into the inclinometers, let's take a minute or so to talk about some of our latest sensors. Our sensor products are divided in three major categories to serve automotive, industrial, and consumer markets. The automotive category of products we offer are ADC Q100 qualified to primarily serve automotive applications. The industrial parts here in the column in the middle are offered with 10-year longevity commitment from the date of the product release. Some of these industrial parts come with ceramic packaging and a wider operating temperature range, which goes from minus 40 Celsius to plus 105 Celsius, making these products ideal for applications in harsher operating environment. These products are being used in the Internet of Things, smart cities, smart homes, smart industry, industry 5.0, and so on. The column on the right, on this slide, provides you with a list of products that we call consumer and can be used in any applications in consumer electronics products. These products provide the most cost-effective solutions. Now I'm moving to the next session of this webinar. In the following section, I would like to talk about inclinometers. We'll learn about how the tilt or slope of an object with respect to gravity's direction is measured using an inclinometer and how the angles of slope or tilt are calculated. We'll also learn about the latest inclinometers that can be used to pretty accurately measure the tilt. Let me first explain what we mean by high accuracy and high precision. If you take different measurements, and each data point differs widely from each other, and each data point is far from the true value, this will indicate that the device has very low accuracy and low precision. As you can see on the illustration on the very left on this slide. As the middle illustration on this slide shows, if all the data points are very close to each other, but away from the true value, this will mean that the device offers high precision but low accuracy. On the other hand, if all the data points are very close to each other and close to the true value, this means that the device offers both high accuracy and high precision, as you can see here the very right on this slide. You might be wondering what the differences between an accelerometer and an inclinometer are. So I will try to explain the commonalities and the differences between these two types of devices. Let's take a look at these comparison charts. The two devices on the left, the LIS2DH12 and LIS2DW12 are two of our latest accelerometers. The device on the right side, the column very right here, the IIS2ICLX, is our latest high precision inclinometer. All these three devices use the capacity principle based on MEMS technology. So MEMS processes are used to make the mechanical sensing element of these devices. There are differences, however, on the technical specification side. If you look at here in this chart, the LIS2DH12 and LIS2DW12, they are designed for applications in consumer electronics. The IIS2ICLX is primarily designed for tilt measurement. I have highlighted the differences in green here, you see in terms of sensitivity, in terms of offset, in terms of operating temperature range, noise level, these are the advantages of the inclinometer over the two accelerometers. The inclinometer offers better sensitivity, lower offset, high stability and repeatability of sensitivity and offset over time and temperature. As I mentioned, it has much lower noise density, is only 15, micro G over square roots of hertz, versus 90 and 220. And it has the temperature range from minus 42 plus 105 degrees Celsius. Because of these improvements, a much more accurate tilt angle can be achieved using the IIS2ICLX. Okay, let's now talk about the tilt measurement. We can use a meme-based accelerometer or inclinometer to measure the tilt of an object with respect to a horizontal plane. The accelerometer measures the projection of the acceleration of the gravity on the axis of the sensor. So once we have those projections, we can then use basic trigonometric equations to calculate the tilt angles. On this slide, we assume a one-axis accelerometer to explain the tilt calculation. As you can see here on the left-hand side on this slide, if you rotate the object, the amplitude of the sensed acceleration changes according to the sine function of the tilt angle alpha between the sensing axis and the horizontal plane. From this object orientation and the graph on the top right, you can see that the sensor is most sensitive to changes in tilt angle when the sensing axis is close to perpendicular to the force of gravity. In this case, the sensitivity is very good. If you take the sine of one degree, so if you tilt it for one degree and you calculate sine of one degree minus sine of zero, it gives you a sensitivity of approximately 17.45 milligie per degree. So because of this derivative function of the sine function, the sensor has lower sensitivity, in other words, less responsive to tilt angle changes when the sensing axis is close to its plus one G or minus one G position. In this case, the sensitivity is only 0.15 milligie per degree if you do a similar calculation as the previous case. So if you take the sine of 90 degrees minus the sine of 89 to one degree tilt, then it gives you a sensitivity of 0.15 milligie per degree. The top graph on this slide shows the sensitivity of different tilt angles. As you can see, the sine function has good linearity between 0 and 45 degrees, between 135 and 225, and again between 315 and 360 degrees. The graph at the bottom here shows the relationship between the tilt angle alpha and the acceleration gravity from minus one G to plus one G tilt. So now there are, as I mentioned, the drawback, main drawback of this one axis tilt measurement is the linearity issue. So the high resolution is only around 0 degrees, so high sensitivity, high resolution. That's the best case if your object is almost perpendicular to the force of gravity. Now the drawback of this one axis solution is that it is not possible to measure over the entire 360 degrees. We saw the limitations of using a single axis sensor on the previous slide. Let's now consider a two axis device. When we use a two axis tilt sensing method, it helps overcome the drawbacks of the single axis tilt sensing that we saw in the previous slide. But the user should be aware of two different situations. To explain this, let's look at the two different scenarios. Look at the graph here on the left side on this slide. The accelerometer is rotated counterclockwise at an angle of theta. When theta is less than 45 degrees, the X axis has higher sensitivity, while the Y axis has lower sensitivity. And when theta is greater than 45 degrees, the X axis has lower sensitivity, while the Y axis has increased sensitivity. So therefore, when the two axis approach is used, it is recommended to calculate the angle based on the orthogonal axis to a plus minus 1G condition. As I mentioned on the previous slide, the tilt sensitivity equation, millilty over degree, can be calculated by taking the difference of the acceleration output between 1 degree at that point. For example, the tilt sensitivity at 89 degrees is calculated by the sine function of 90 degrees minus the sine function of 89 degrees, which gives you a sensitivity of 0.15 milligie per degree. Again, to repeat what I mentioned on the previous slide, the sensor is most responsive to changes in tilt when the sensitive axis is perpendicular to the force of gravity. That means when perpendicular to the force of gravity, the accelerometer experiences approximately 17.45 milligie per degree tilt. It is least responsive when the sensitive axis is parallel to the force of gravity in the plus 1G or minus 1G orientation. Again, the responsiveness would be only 0.15 milligie per degree. This is clearly displayed here in this graph where the absolute value of the tilt sensitivity was taken. So whenever the x-axis is at its minimum tilt sensitivity, the y-axis is at its maximum tilt sensitivity. By combining the x- and y-axis, solving for the tilt angle and using the arc tangent, so a x acceleration or the gravity component on the x-axis over the gravity component on the y-axis, a constant tilt sensitivity of 17.45 milligie can be maintained through 360-degree rotation, as you can see here the green line in this graph. The second scenario would be when both x- and y-axis are perpendicular to the force of gravity. That means x and y are on the horizontal plane and in this case, both x and y have high sensitivity because they are perpendicular to the force of gravity. But without the help of a third axis, for example, the z-axis, it is not possible to distinguish the tilt angle of 30 degrees from the 150 degrees because the x-axis has the same output at these two tilt angles. So with a three-axis accelerometer, you can use the z-axis to combine with the x- and y-axis for tilt measurement. So a three-axis accelerometer gives you the complete set of gravity components to measure the tilt with respect to the plane and offers all the benefits, including almost constant tilt sensitivity and the possibility to measure over 360 degrees. To be able to define the angles of the accelerometer in three dimensions, the pitch, roll, and yaw are measured using all three components of the accelerometer. So all three axes of the accelerometer. In terms of definition of pitch, yaw, and roll, so pitch is defined as the angle of the x-axis relative to ground. Roll is defined as the angle of the y-axis relative to the ground. Yaw is the angle of the z-axis relative to gravity. We can use the trigonometric equations shown here on this slide to calculate pitch, roll, and yaw tilt angles. Keep in mind that now the acceleration due to the gravity on the x, y, and z-axis are combined. So the square roots of ax squared, ay squared, and az squared is equal to 1 when the accelerometer is static. Okay, now let's talk about the static and dynamic inclinometers. A static inclinometer is used to determine the orientation angle of an object with respect to the gravity force. So a two-axis or three-axis inclinometer is designed for static measurements. This is when there is no dynamic motion or vibration because the two-axis or three-axis inclinometer or accelerometer does not differentiate between the vibration and gravity. On the left side here, on this slide, we have two inclinometers that we offer. There are many applications where you can use an inclinometer including some of them here listed like precise instruments and platform leveling, industrial automation, industrial vehicles, forklift, and construction machines just to mention a few. In the bottom on this slide, you can see three-part numbers. The IIS-3DHHC is a three-axis digital inclinometer. The IIS-2ICLX is a two-axis inclinometer with machine learning core and with extended temperature range. Any of these two devices can be used for any of these applications. We will look at the specs of these two devices on the next slide. But if you want to have the option of measurement of tilt in the presence of dynamic motion, you need to add a three-axis gyroscope. You can use the three-axis gyroscope to determine the changes in orientation at an extremely fast rate. With the gyroscope, you can measure how fast an object is rotating in a three-dimensional space. If you decide or you need to use a module with a three-axis accelerometer and a three-axis gyroscope in a single small package, you can consider the part over here in the bottom right, the ISM-330DHCX, which has again the machine learning capability and also the extended temperature range. On the previous slide, we talked about some of the applications of the inclinometers. As far as the static inclinometers are concerned, there are a few key performance indicators that can be relevant for the applications listed here on this slide. The first thing you need to consider is the number of axes of the device. Depending on your requirements, you can decide if you need a two-axis or a three-axis device. The measurement range can also be important, but for static inclinometers, usually a full-scale range of 1G should be sufficient because you are measuring the gravity components. The other performance indicator is the resolution. This is important when you want to know what angle resolution is required and achievable. This of course depends on the noise, bandwidth, filtering, and the ADC resolution of the device. The third performance indicator is the accuracy. The accuracy is determined by the stability of the device over temperature. That means the offset and sensitivity drift over temperature. Also, the repeatability, offset, and sensitivity drift over time is important. The accuracy depends also on the vibration rectification. That means the rectification to DC component of broadband AC vibration can shift the offset of the inclinometer that leads them to significant errors. Operating temperature range and interface, if you need analog or digital, could also be important in some applications. There are other parameters such as offset sensitivity calibration error, post-soldering drift, cross-axis, non-linearity that are less relevant because they can be measured and taken care of with post-assembly calibration at ambient temperature, which is usually required for higher accuracy applications. Industrial inclinometers. We touched on these two devices earlier. On this slide, I have listed the major specifications of these two devices, the latest inclinometers that we offer. The IIS 3D HHC on the left-hand side here is a three-axis digital inclinometer that is used for pitch, roll, and yaw drift measurement and offers high stability over temperature and time. The offset change over temperature of these devices is less than 0.4 mG over degree Celsius. The noise level has been reduced down to 45 mG over square roots of hertz, which translates to an RMS noise considerably less than 1 mG at about 235 Hz bandwidth. The operating temperature range goes from minus 42 to plus 85 degrees Celsius, and indeed offers a full-scale range of plus minus 2.5 g. Like all other industrial sensors in our portfolio, this part comes with the 10-year longevity commitment. On the right-hand side, this is the very latest high precision inclinometer that we offer. On the left-hand side, I would look at the specs in more detail. Okay, so the IIS 2-ICLX is a two-axis digital inclinometer with a selectable full-scale range that goes from plus minus 0.5 g to plus minus 3 g. So you can choose any of these four full-scale. In terms of noise level, the noise is significantly lower than any other comparable inclinometer that you see on the market in terms of performance and also for devices compared to devices at the same price point. So the noise level is only 1515 mG over square roots of hertz. It offers a resolution of 16 bits. So this device is a 16-bit device, and it offers an accuracy better than 0.5 degrees over entire temperature range and also over time. It has a very high stability over temperature. It is less than 75 mG over degrees Celsius. The repeatability is very good because it has an embedded compensation for high stability over temperature. In terms of interface, you have both IIS 2-ICLX and SBI that you can use. In terms of current consumption, it is significantly lower compared to the comparable devices. It is only 0.42 mA. It is very suitable for those applications with limited power budget. It comes with the sensor hop feature, and that helps you to really efficiently collect the data from additional external sensors if they are required for your applications. It has a FIFO. We call that Smart Embedded FIFO that goes up to 3 kilobytes. It gives you also the option of programmable a high pass and low pass filter, digital filter inside the device. In terms of temperature range, I mentioned earlier, it goes up to 105 degrees Celsius, so from minus 40 degrees Celsius to 105. Of course, like all other MEMS devices that we offer, it has also embedded temperature sensor in it that you can use in temperature data. The supply voltage goes from 1.71 to 3.6 volts. All our devices, they use our MEMS processes, which are very shock-assistant, so these devices, like all other devices we have, provide very high shock survivability. In terms of embedded features, it has the programmable machine learning code, and that is actually designed to integrate artificial intelligence algorithm inside the device to reduce the power consumption at the system level. Also, it comes with the programmable finite state machine to process data from accelerometer and one external sensor. That will also help you in terms of speeding up the implementation of some of the applications. We will talk about the machine learning code briefly in one of the next slides. Let's take a look at one of the emerging applications for an inclinometer that has been significantly growing over the last few years because of high performance and cost-effective MEMS-based sensors that have been introduced in the market. This application is called structural health monitoring, which is about implementing a strategy to identify the potential damages to the structures before they take place. It is about predictive monitoring for safety and cost reduction reasons. So why sensors are so widely used for this application? The basic approach here is to collect data from a number of sensors over a long period of time, and these sensor data can then provide information on the overall condition of the structure. For example, data on vibration, inclination, temperature, pressure, strain, and force are considered the major components of a structural monitoring system. The sensor data are collected continuously, and that can be done at desired time interval to detect and monitor the structure. The sensor data and the statistical analysis of this data are then used to determine the system health at any given time. We talked earlier about some of the key performance indicators for the industrial applications of the sensors of inclinometers, so they apply here, too, for these applications, and some of them that we talked about were resolution, stability, accuracy. These are all also very relevant for this application here. Okay, we talked about static inclinometers, and now let's take a closer look at dynamic inclinometers. If you are considering a dynamic inclinometer, it is important to know what dynamic range your application requires. The MEMS-based inclinometers integrate a three-axis accelerometer and a three-axis gyroscope in a single package that can be used to measure the tilt in real time. The data from the accelerometer are usually impacted by external acceleration. A gyroscope, on the other hand, offers pretty much clean angular velocities or rate, angular rate, around all three axes, X, Y, Z. But the data from the gyroscope drifts over time. The solution here is to fuse the data from the gyroscope and from the accelerometer to have a reliable and accurate solution for dynamic tilt measurement. The dynamic inclinometers can be used in many applications in different markets. We talked about some of them listed on this slide. This includes industrial vehicles, mining cranes, and the reason this inclinometer is used so widely is because of the six degrees of freedom that it is offering because of the combination of the three-axis accelerometer and the three-axis gyroscope. Another widespread use of the dynamic inclinometers is in robotics to monitor multiple movements at the same time. In autonomous driving, for example, multiple dynamic inclinometers are used to provide lots of data at a very high rate, which is very important for self-driving cars. One of the latest modules that we offer for this kind of application, we have a number of these devices, but the one here mentioned is the ISM330DHCX. On the next slide, I would like to give you more details on the specs of this device. Okay, let's look at the specs of the dynamic inclinometer. The ISM330DHCX has a very good stability and robustness because the sensing elements of the accelerometer energizers are implemented on the same silicon die, so that is one of the major advantages of this device compared to the ones you see on the market. This device has a full-scale acceleration range from plus minus 2G all the way up to plus minus 16G and also for the gyroscope, it has a full-scale range of about 60 different options available to you from plus minus 125 degrees per second all the way to 4000 degrees per second. That enables the usage in a broad range of applications because in some of the applications you would need much higher full-scale range. The devices that you see on the market they usually goes up to 2000 degrees per second. This one we have increased to 4000 degrees per second to enable other applications that require higher full-scale range. All the design aspects and the calibration of the ISM-330DTX has been optimized to reach the best accuracy and stability. It comes with a very low noise and full data synchronization. If you look at the data of the device you will see that the offset versus temperature is only plus minus 0.005 degrees per second for degrees Celsius. The bi-assistability is 3 degrees per hour. The rate noise sensitivity has gone down to 5 millidegrees per second per square root of hertz and accelerometer noise density is only 60 micro-G over square root of hertz and that is compared to some of the earlier devices with 220 micro-G significant improvement. In terms of output data rate for the accelerometer like any other accelerometer that we have most of them the output data rate goes up to 6.6 kilohertz. In terms of embedded features we have in this device machine learning core. I will talk about that briefly on the next slide. It has a program of wall-fine state machines and also it comes a big cycle of 2-9 kilobytes. It has the sensor hop functionality, event decoding and also the interrupts. In terms of availability like any other industrial parts that we have this device comes with 10-year longevity commitment so you don't need to worry about the availability of the device for many years and I guess that if about the specs of the device I will talk about the machine learning core in a couple of minutes. So in the last few years we have been trying to provide a number of software libraries to our customers free of charge of course to help them reduce their development time for these inclinometers we have a couple of software libraries that are listed here on this slide but just keep in mind we have about 30 different software libraries at least that you can use with our device so if you need more information on this and other software libraries feel free to contact us and we would be more than happy to support you with those libraries. So now motion BL is a library for static tilt measurement that is developed for our inclinometers we can use this library with any of our inclinometers that we discussed during this webinar it also includes calibration algorithms to achieve much higher accuracy. The inclinometer calibration is an thing that is included here and the second software library that we have is the motion DI it includes a six point calibration for the accelerometer it comes also with gyroscope calibration and also the six axis sensor fusion that we talked about earlier to achieve much higher accurate orientation angles in the presence of vibration and motion we talked about the inclinometer and the sensor fusion on the previous slide. So as I mentioned earlier if you need more information on this software libraries please feel free to visit our website or contact one of our sales offices in your region to give you more information and support that you need. So we talked about the components we talked about the software libraries now I would like to talk about the evaluation tools that are available to evaluate our products. The fastest way to start with the evaluation of these parts is to use these boards that we have here listed on this slide. The one the board the bigger one on the left hand side that you see this is the motherboard it includes a high performance 32 micro control that functions as a bridge between the sensor evaluation board and a computer. Now for each of these components we have an evaluation board that goes with this motherboard for the IIS 2XDLX there is a board that has embedded this part in it and it is connected through a flat cable to a simple adapter board to make it compatible with that motherboard that you see on the left hand side. For the IIS 3DHTC we have also an adapter board that is designed to facilitate the evaluation of this part. The board offers a very good solution for fast system prototyping and the device evaluation directly in your application. At the bottom right here on this slide you can see the parameters of the evaluation board of the three parts we discussed about. So the two first ones are the static inclinometers and the last one the IIS M330DHCX is the dynamic inclinometer in the parameters of the adapter boards and evaluation boards are listed over here. So if you want to order them you do through our distribution partners or you contact our sales offices and we would be happy to provide you with these boards. These boards are compatible with the ST Unico which is the graphical user interface that is developed for all our sensors. Quick and modular prototyping we have created the STM32 open development environment to provide a fast and affordable way to develop and prototype. The combination of a broad range of expandable boards and modular software gives you the capability of fast prototyping of ideas that can smoothly be transferred into final designs. To start your design you choose the appropriate STM32 nuclear development board and expansion or ex-nuclear boards for the functionality that you need. Next step is to select your development environment and use the free STM32 cube tools and software. You can download from ST website all the software pieces that you need to run the functionalities on the selected STM32 nuclear expansion boards. You can then compile your design and upload it to the STM32 nuclear development board. Then you start developing and testing your applications once you are done. You can use the software and that software will be developed on the STM32 development environment in product design. The X-Cube MMS1 expansion board what you see here on the top right is a software package for the STM32 cube and runs on the STM32 and it includes drivers that recognize the sensors and collect a number of sensor data like temperature, humidity, pressure, motion, acceleration and in addition to the onboard sensors that are supported by the X-Cube MMS1 expansion software package for the STM32 cube the extended X-Cube MMS1 here the new version supports also devices that are connected via the DIL 24 socket. The graphical user interface for the X-Cube MMS1 and X-Cube MMS X-T1 software expansion board and STM32 nuclear expansion board is called Unuclio GUI. Okay the last topic I would like to touch on is the machine learning core that we have embedded in some of our latest devices including the devices that we discussed here so the two access inclinometer in the dynamic inclinometer. The machine learning core allows moving some algorithms from the application processor to the MMS sensor itself to reduce the overall system power consumption. The process is illustrated here on this slide. The idea behind is to use internal and external sensor data to compute a set of statistical parameters. These parameters can be then used as inputs for a configurable decision tree which can be say stored in the device itself, in the sensor itself. There are also configurable filters embedded in the sensor to define new inputs in addition to the sensor data input. The machine learning core is implemented using the decision tree logic. You can store the decision tree as a binary tree that is composed of a series of nodes. In each node of the decision tree a statistical parameter is evaluated against the threshold to establish the evolution in the next node. So when a leaf which is one of the last nodes of the tree is reached, the decision tree generates a result which is readable through a dedicated device output register. The device with the machine learning core can be also configured up to eight decision trees simultaneously and independently. Also, there is the possibility to generate an interrupt for every change in the result in the decision tree. So this was a very quick overview to make you aware of this new feature. If you would like to learn more about this, we have lots of publications and webinars on this topic and we would be happy to share them with you. For collateral materials and the devices we discussed today and other SD sensors, please feel free to visit sd.com. We have dedicated pages for each sensor in our portfolio. I have also listed on this slide the links to the three devices. We discussed today, but you can find on our website the latest publications, samples, ordering tools in software libraries, info on all of our sensors. Of course, you can contact our sales office in your respective region for any support you might need. Okay, we are at the end of the presentation now. I would like to thank you again for taking time to join us today. We really appreciate your time and interest.