This article was updated March 10, 2023. It was originally published Aug. 16, 2018.
READ MORE: Surviving Shock and Vibration
Industries and R&D sites depend on sensors to measure, monitor and analyze vibrations. Unfortunately, the best ways to get accurate and repeatable vibration data are not clear to many engineers; they are unfamiliar with the nuances of test equipment and vibration sensors.
Here’s a quick refresher course on the sensors themselves.
The Basics
Vibrations are oscillating motions of a device or object about a position of equilibrium. They are often detected and measured by accelerometers and both contact and non-contact sensors.
Their outputs from these sensors are translated into usable vibration data and used to diagnose problems with motors and assemblies for maintenance purposes to assess the stability of structures subject to periodic stresses (such as monitoring vibrations caused by vehicles crossing a bridge).
The sensors are usually connected to a data network that then delivers 24-bit vibration data at high sample rates. Modern data acquisition equipment can plot, log, filter and analyze vibration data in real time and can handle wirelessly streaming vibration data over Wi-Fi.
Accelerometers
Accelerometers are the most commonly used sensors for detecting and measuring vibrations. When mounted on or in a vibrating structure, accelerometers proportionally convert mechanical energy into electrical energy. There are at least five types of accelerometers used for this:
Capacitive MEMS accelerometers are the smallest and least expensive accelerometer option, and are often used in smartphones. They can be mounted on printed circuit boards, which has made them a favorite with electrical engineers.
Although their low cost and small size have made MEMS accelerometers popular, they provide lower quality data compared to other kinds of accelerometers, especially for higher frequency and amplitude vibrations. They are considered unsuitable for industrial applications, but DC-coupled versions have been widely used in consumer appliance and electronics. Their low cost and power consumption also make them a good choice for monitoring patients’ health.
Piezoresistive (PR) accelerometers use strain gauges, giving them a wide bandwidth, which lets them measure short-duration (high-frequency) shocks, such as in crash testing. They can measure down to 0 Hz, so they’re useful for calculating velocity or displacement.
However, PR accelerometers typically have low sensitivity, which makes them less able to measure vibrations accurately. They are also sensitive to temperature changes, so they require temperature compensation.
Because PR accelerometers are usually more expensive than MEMS versions, they’re rarely called on for low frequency and amplitude testing. PR accelerometers, however, are the best choice for measuring impulses and impacts where frequency range and amplitude are typically high.
Piezoelectric (PE) accelerometers are well-suited for industrial applications. Their lead-zirconate-titante sensing elements generate electric charges (outputs) under acceleration. These accelerometers have low noise and offer better performance than MEMS and PR accelerometers in all vibration and most shock applications.
There are several versions of PE devices, including triaxial and single axis, and high sensitivity types for seismic applications as well as low sensitivity devices for shock testing. Some PE sensors can handle extreme environments—even high-levels of radiation. Their most significant downside is that they are AC coupled, so they can’t measure gravity or sustained accelerations.
PE accelerometers are well-suited for shock testing, but they also do well at simply measuring vibrations. They rely on strain gauges, so they need amplifiers and temperature compensation. Their bandwidth ranges from 0 to several thousand Hz and they generate little noise. PE sensors can use gas or fluid damping for protection and to prevent them from reaching their internal resonant frequency. They are DC coupled, so their output can be integrated to calculate velocity and displacement during shock events. Charge mode piezoelectric accelerometers are extremely durable and can survive hostile conditions, including extreme temperatures: −320°F to 752°F (−200°C to 400°C). Some can even operate in radioactive environments. However, their high resistance calls for special cabling to shield against noise. They also need a charge amplifier, which adds cost and complexity.
READ MORE: Is Vibration Detection the Canary in the Mines for Manufacturing?
Voltage mode internal electronic piezoelectric accelerometers are the most common type of accelerometer. They are like charge mode accelerometers but have a built-in charge amplifier. This eliminates the need for special cabling and simplifies the task of adding the accelerometers to a data acquisition network. This addition requires a constant DC power source, which many data acquisition networks now include.
The electronic circuit built into these accelerometers prevents them from matching a charge mode accelerometer’s ability to tolerate hostile environments. But their typical temperature range of −40°F to +257°F (−40°C to 25°C) is good enough for most applications.
Getting the right accelerometer almost always depends on the application. For example, in shock testing that requires integrating acceleration data for velocity or displacement, design teams should choose a capacitive MEMS or piezoresistive accelerometer. And although piezoelectric accelerometers are ideal for measuring vibrations, it takes high sensitivity accelerometers to measure low-frequency vibrations applications.
Other Sensors
Strain gauges detect and measure vibrations by electrically monitoring a minute electrically conductive grid etched on a thin foil that expands and contracts with the surface it is attached to. As the grid is stretched or compressed, its electrical resistance increases or decreases proportionally.
These gauges are versatile and accurate. They can be put on virtually any surface and shape, and they measure static and dynamic loads. They also measure strain directly, simplifying the task of calculating stress in the material.
READ MORE: Building a MEMS-Based Vibration Detector
Strain gauges are also small and lightweight and don’t significantly alter the stiffness of the test specimen. Relative to other types of vibration sensors they are inexpensive, ranging from just a few dollars to a few hundred.
They must be properly attached to the surface they are measuring, and that can be difficult; they must be perfectly bonded to the material. They also require costly and cumbersome electronics to amplify the signals and acquire data.
Microphone or acoustic pressure sensors are cost-effective tools for measuring high-frequency vibrations. This is particularly useful when determining how a piece of equipment’s vibrational profile changes over time. They also provide clear and accurate representations of a vibration’s frequency content. And from an engineer’s perspective, they are simple and inexpensive.
Laser displacement sensors bounce a laser off the equipment or product being tested, then recover a portion of that light that comes back to the sensor, much like radar. As the target vibrates, the angle of the reflected light changes as the target gets nearer or father from the sensor. Sensor data is processed to detect and measure vibrations.
Displacement sensors are good for applications that cannot employ accelerometers, such as rotating components or when the accelerometer’s mass would influence the equipment’s or product’s vibrations. Another advantage is that nothing needs to touch the object being tested.
Eddy current or capacitive displacement sensors use magnetic field to measure vibrations. An AC powered probe generates a magnetic field. This field induces a filed in test object, so it must be conductive. As the distance between the fields change as the object vibrates, so does the sensor’s output voltage. The output is converted into vibration measurements. They can only measure relative motion and need to be fixed, but no contact is called for between sensor and object. They are best used for labs given the fragility of the overall sensor.
Stephen Hanly was vice president of sales and marketing at Midé Technology Corp. when this article was first published.