【Guide】This article shows how the performance of high-precision MEMS accelerometers can be degraded without adequate consideration of environmental and mechanical influences. Through holistic design practices and system-level attention, discerning engineers can achieve superior performance for their sensor systems. Since many of us experience unprecedented levels of stress in our lives, it’s useful to realize that, like an accelerometer, it’s never the stress that kills us – it’s our reaction to it!
In Part 1 of this series, we reviewed the internals of a 3-axis high-precision MEMS accelerometer. In the second part, we review how to obtain a good starting dataset to establish baseline performance and verify the type of noise levels to expect in subsequent data analysis. In this first installment of our series, we explore other factors that affect stability, then provide mechanical system design recommendations to improve the overall performance of a 3-axis high-precision MEMS accelerometer.
Another important aspect of inertial sensors is their long-term stability, or repeatability, once the thermal stresses in the design are well understood. Repeatability is defined as the accuracy of continuous measurements under the same conditions over a long period of time. For example, take two measurements of the gravitational field at the same temperature and in the same direction over a longer period of time and see how well they match. Repeatability of offset and sensitivity is critical when evaluating long-term sensor stability in applications where regular maintenance calibration is not possible. Many sensor manufacturers do not describe or specify long-term stability in their data sheets. Using the ADXL355 data sheet from Analog Devices as an example, predicting the repeatability of the 10-year lifetime, including the measured temperature cycling (? 55°C to +125°C and 1000 cycles), velocity random walk, broadband noise, and temperature hysteresis. The repeatability shown in the datasheet is ±2 m g and ±3 mg for the X/Y and Z sensors, respectively. These measurements are important for evaluating long-term performance.
Repeatability under stable mechanical, environmental and inertial conditions follows the square root law related to the measurement time. For example, to obtain an offset repeatability of 2.5 years on the x-axis (the mission profile may be shorter for the final product), use the following equation: ±2 mg × √(2.5 years/10 years) = ±1 mg. Figure 1 shows Example HTOL test results for 0 g offset drift for 32 devices over 23 days. The square root law can be clearly observed in this figure. It should also be emphasized that due to process differences in the fabrication of MEMS sensors, each part behaves differently—some perform better than others.
Figure 1. 500-hour long-term stability of ADXL355. (: Analog Devices)
Mechanical System Design Recommendations
Based on the knowledge discussed earlier, it is clear that the mechanical mounting interface and housing design will contribute to the overall performance of a 3-axis high-precision MEMS accelerometer sensor, since they affect the physical stresses propagated to the sensor. Typically, the mechanical mount, housing, and sensor constitute a second-order (or higher-order) system; therefore, its response varies between resonant or overdamped.
Mechanical support systems have modes (defined by resonant frequencies and quality factors) that represent these second-order systems. In most cases, the goal is to understand these factors and minimize their impact on the sensing system. Therefore, the geometry of any housing in which the sensor will be packaged, as well as all interfaces and materials, should be chosen to avoid mechanical attenuation (due to overdamping) or amplification (due to resonance) within the accelerometer’s application bandwidth. The details of such design considerations are beyond the scope of this article; however, a few useful items are briefly listed:
Printed Circuit Boards, Mounting and Enclosures
Attach the PCB firmly to the rigid substrate. Use multiple mounting screws and adhesive on the back of the PCB to provide support.
Place the sensor near the mounting screw or fastener. If the PCB geometry is large (several inches), use multiple mounting screws in the middle of the board to avoid low frequency vibrations of the PCB from coupling to the accelerometer and taking measurements.
If the PCB is only mechanically supported by the groove/tongue structure, use a thicker PCB (thicker than 2 mm recommended). For PCBs with larger geometries, increase the thickness to maintain the stiffness of the system. Use finite element analysis, such as ANSYS or similar software, to optimize the PCB geometry and thickness for a specific design.
For applications such as structural health monitoring where sensors are required to measure for long periods of time, the long-term stability of the sensor is critical. Packaging, PCB, and adhesive materials should be chosen to minimize degradation or changes in mechanical properties over time, which could cause additional stress on the sensor, resulting in offset.
Avoid making assumptions about the natural frequency of the enclosure. Calculations of natural modes of vibration for simple enclosures and finite element analysis for more complex enclosure designs would be useful.
The stress buildup from soldering the accelerometer to the board has been shown to cause offset offsets of up to a few milligrams. To mitigate this effect, it is recommended that the PCB landing pattern, thermal pad, and conduction paths through copper traces on the PCB be symmetrical. Follow closely the soldering guidelines provided in the accelerometer datasheet. It has also been observed that, in some cases, performing solder annealing or thermal cycling prior to any calibration can help alleviate stress buildup and manage long-term stability issues.
Potting compound
Potting compounds are widely used to secure Electronic devices within enclosures. If the sensor package is an overmolded plastic, such as a land grid array (LGA), potting compounds are strongly discouraged as their temperature coefficient (TC) does not match the housing material, causing stress to be applied directly to the sensor , then offset. A 3-axis high-precision MEMS accelerometer in a hermetically sealed ceramic package significantly protects the sensor from TC effects. However, as the material degrades over time, the potting compound can still cause stress buildup on the PCB, potentially straining the sensor through small warpage of the silicon chip. It is generally recommended to avoid potting the sensor in applications requiring high stability. 8
Airflow, heat transfer and heat balance
To achieve sensor performance, it is important to design, position and use the sensing system in an environment where temperature stability is optimized. As this article demonstrates, even small temperature changes can show unexpected results due to differences in thermal stress on the sensor die. Here are some tips:
The sensor should be placed on the PCB so that there is no thermal gradient across the sensor. For example, linear regulators generate a lot of heat; therefore, their proximity to the sensor causes a temperature gradient across the MEMS that can vary over time with the current output in the regulator.
If possible, sensor modules should be deployed in areas away from drafts (such as HVAC) to avoid frequent temperature fluctuations. If not possible, thermal isolation outside or inside the package is helpful and can be achieved with thermal insulation. Note that conduction and convection heat paths need to be considered.
It is recommended that the thermal mass of the enclosure be selected so as to dampen ambient thermal fluctuations in applications where ambient thermal variations are unavoidable.
in conclusion
This article shows how the performance of high-precision MEMS accelerometers can be degraded without adequate consideration of environmental and mechanical influences. Through holistic design practices and system-level attention, discerning engineers can achieve superior performance for their sensor systems. Since many of us experience unprecedented levels of stress in our lives, it’s useful to realize that, like an accelerometer, it’s never the stress that kills us – it’s our reaction to it!
【Guide】This article shows how the performance of high-precision MEMS accelerometers can be degraded without adequate consideration of environmental and mechanical influences. Through holistic design practices and system-level attention, discerning engineers can achieve superior performance for their sensor systems. Since many of us experience unprecedented levels of stress in our lives, it’s useful to realize that, like an accelerometer, it’s never the stress that kills us – it’s our reaction to it!
In Part 1 of this series, we reviewed the internals of a 3-axis high-precision MEMS accelerometer. In the second part, we review how to obtain a good starting dataset to establish baseline performance and verify the type of noise levels to expect in subsequent data analysis. In this first installment of our series, we explore other factors that affect stability, then provide mechanical system design recommendations to improve the overall performance of a 3-axis high-precision MEMS accelerometer.
Another important aspect of inertial sensors is their long-term stability, or repeatability, once the thermal stresses in the design are well understood. Repeatability is defined as the accuracy of continuous measurements under the same conditions over a long period of time. For example, take two measurements of the gravitational field at the same temperature and in the same direction over a longer period of time and see how well they match. Repeatability of offset and sensitivity is critical when evaluating long-term sensor stability in applications where regular maintenance calibration is not possible. Many sensor manufacturers do not describe or specify long-term stability in their data sheets. Using the ADXL355 data sheet from Analog Devices as an example, predicting the repeatability of the 10-year lifetime, including the measured temperature cycling (? 55°C to +125°C and 1000 cycles), velocity random walk, broadband noise, and temperature hysteresis. The repeatability shown in the datasheet is ±2 m g and ±3 mg for the X/Y and Z sensors, respectively. These measurements are important for evaluating long-term performance.
Repeatability under stable mechanical, environmental and inertial conditions follows the square root law related to the measurement time. For example, to obtain an offset repeatability of 2.5 years on the x-axis (the mission profile may be shorter for the final product), use the following equation: ±2 mg × √(2.5 years/10 years) = ±1 mg. Figure 1 shows Example HTOL test results for 0 g offset drift for 32 devices over 23 days. The square root law can be clearly observed in this figure. It should also be emphasized that due to process differences in the fabrication of MEMS sensors, each part behaves differently—some perform better than others.
Figure 1. 500-hour long-term stability of ADXL355. (: Analog Devices)
Mechanical System Design Recommendations
Based on the knowledge discussed earlier, it is clear that the mechanical mounting interface and housing design will contribute to the overall performance of a 3-axis high-precision MEMS accelerometer sensor, since they affect the physical stresses propagated to the sensor. Typically, the mechanical mount, housing, and sensor constitute a second-order (or higher-order) system; therefore, its response varies between resonant or overdamped.
Mechanical support systems have modes (defined by resonant frequencies and quality factors) that represent these second-order systems. In most cases, the goal is to understand these factors and minimize their impact on the sensing system. Therefore, the geometry of any housing in which the sensor will be packaged, as well as all interfaces and materials, should be chosen to avoid mechanical attenuation (due to overdamping) or amplification (due to resonance) within the accelerometer’s application bandwidth. The details of such design considerations are beyond the scope of this article; however, a few useful items are briefly listed:
Printed Circuit Boards, Mounting and Enclosures
Attach the PCB firmly to the rigid substrate. Use multiple mounting screws and adhesive on the back of the PCB to provide support.
Place the sensor near the mounting screw or fastener. If the PCB geometry is large (several inches), use multiple mounting screws in the middle of the board to avoid low frequency vibrations of the PCB from coupling to the accelerometer and taking measurements.
If the PCB is only mechanically supported by the groove/tongue structure, use a thicker PCB (thicker than 2 mm recommended). For PCBs with larger geometries, increase the thickness to maintain the stiffness of the system. Use finite element analysis, such as ANSYS or similar software, to optimize the PCB geometry and thickness for a specific design.
For applications such as structural health monitoring where sensors are required to measure for long periods of time, the long-term stability of the sensor is critical. Packaging, PCB, and adhesive materials should be chosen to minimize degradation or changes in mechanical properties over time, which could cause additional stress on the sensor, resulting in offset.
Avoid making assumptions about the natural frequency of the enclosure. Calculations of natural modes of vibration for simple enclosures and finite element analysis for more complex enclosure designs would be useful.
The stress buildup from soldering the accelerometer to the board has been shown to cause offset offsets of up to a few milligrams. To mitigate this effect, it is recommended that the PCB landing pattern, thermal pad, and conduction paths through copper traces on the PCB be symmetrical. Follow closely the soldering guidelines provided in the accelerometer datasheet. It has also been observed that, in some cases, performing solder annealing or thermal cycling prior to any calibration can help alleviate stress buildup and manage long-term stability issues.
Potting compound
Potting compounds are widely used to secure Electronic devices within enclosures. If the sensor package is an overmolded plastic, such as a land grid array (LGA), potting compounds are strongly discouraged as their temperature coefficient (TC) does not match the housing material, causing stress to be applied directly to the sensor , then offset. A 3-axis high-precision MEMS accelerometer in a hermetically sealed ceramic package significantly protects the sensor from TC effects. However, as the material degrades over time, the potting compound can still cause stress buildup on the PCB, potentially straining the sensor through small warpage of the silicon chip. It is generally recommended to avoid potting the sensor in applications requiring high stability. 8
Airflow, heat transfer and heat balance
To achieve sensor performance, it is important to design, position and use the sensing system in an environment where temperature stability is optimized. As this article demonstrates, even small temperature changes can show unexpected results due to differences in thermal stress on the sensor die. Here are some tips:
The sensor should be placed on the PCB so that there is no thermal gradient across the sensor. For example, linear regulators generate a lot of heat; therefore, their proximity to the sensor causes a temperature gradient across the MEMS that can vary over time with the current output in the regulator.
If possible, sensor modules should be deployed in areas away from drafts (such as HVAC) to avoid frequent temperature fluctuations. If not possible, thermal isolation outside or inside the package is helpful and can be achieved with thermal insulation. Note that conduction and convection heat paths need to be considered.
It is recommended that the thermal mass of the enclosure be selected so as to dampen ambient thermal fluctuations in applications where ambient thermal variations are unavoidable.
in conclusion
This article shows how the performance of high-precision MEMS accelerometers can be degraded without adequate consideration of environmental and mechanical influences. Through holistic design practices and system-level attention, discerning engineers can achieve superior performance for their sensor systems. Since many of us experience unprecedented levels of stress in our lives, it’s useful to realize that, like an accelerometer, it’s never the stress that kills us – it’s our reaction to it!
The Links: 7MBR25SA120-50 HD4890
