Accelerometer Basics

If you use a smartphone, you benefit from accelerometers each time you pick up the device. These sensors monitor a variety of parameters, ranging from acceleration and rotation to orientation. But no one type of sensor can measure all of these factors. As a result, each smartphone contains a number of motion sensors. Combined, they enable the phone to determine location and context, elements essential to the services provided by apps.

The complexity of these sensors increases from one day to the next, as new technologies, materials and production processes are tapped to expand their functionality. Even so, the fundamentals don’t change—so it’s important to go back to the basics every now and then and review the principles that define these sensors.

Measuring Acceleration

One of the most commonly used motion sensors is the accelerometer. This electromechanical device measures non-gravitational acceleration forces in a given axis, quantifying changes in velocity. The forces can be static, like the constant force of gravity pulling on an object, or they can be dynamic, which is caused by movement or vibration. Design engineers often combine several of these devices to create multi-axis accelerometers.

Accelerometers typically provide measurements in terms of g force, which represents acceleration in increments of Earth’s standard acceleration. For example, a stationary object experiences 1g of acceleration. The g forces involved in a car crash can measure as high as 100g.

Accelerometers for Different Applications

Accelerometers come in a number of forms. The most commonly deployed sensors use the piezoelectric effect. These devices contain microscopic crystals that become stressed when the accelerometer begins to move. The change in state generates a voltage. Piezoelectric accelerometers support a wide measurement frequency range and come in a variety of sensitivities, weights and sizes. This type of accelerometer is well suited for shock and vibration measurements.

Next, you have the piezoresistive accelerometer. This type of sensor generally has low sensitivity, making it suited for shock measurements, but it is less appropriate for vibration measurements. Piezoresistive accelerometers support a wide bandwidth and their frequency response can be as low as 0Hz, which means that they can measure long-duration transients.

Among the newest types of accelerometers, variable capacitance sensors, consist of two microstructures positioned next to each other. In a stationary state, a defined capacitance exists between the two structures. If an accelerative force moves one of the structures, the capacitance changes. The device converts the change to a voltage. This design offers a frequency response as low as 0 and has high sensitivities, narrow bandwidth, and exceptional temperature stability. Automakers often use variable capacitance accelerometers in crash tests.

Features to Consider

When selecting an accelerometer for an application, you have to consider several factors. The trick here is to match the sensor’s operating parameters with the application’s requirements. One of your decisions will be whether you need analog or digital outputs. To a large extent, the hardware that you are interfacing the accelerometer with will provide the answer to this question. For example, if your application involves a microcontroller with digital inputs, you will need to use a digital output accelerometer.

Analog accelerometers output a continuous voltage proportional to acceleration. Digital accelerometers use pulse width modulation (PWM) for output. With PWM, you get a square wave signal. The amount of time the voltage falls within the high range is proportional to the acceleration.

Next, decide whether you require single or multiple axes. Usually, two will meet your needs, but if your application involves 3D positioning, you’ll need a 3-axis accelerometer.

Finally, be sure the accelerometer’s bandwidth meets the needs of the application. In this case, bandwidth represents the number of times per second you can take a reliable acceleration reading. For slow moving applications, a bandwidth of 60 to 75Hz will be adequate. If you intend to measure faster motion, look for a bandwidth of several hundred Hz.