Hello,

Please may someone provide a little more explanation with regards to how the value of g affects an accelerometer's performance?

My knowledge is limited in this field and I would like to know the following for generic accelerometer. For arguments sake I have chosen an Analog Devices ADXL103. It is available in 1.7g, 5g and 18g variants. Datasheet can be found here

Does an increase of g (1.7g, 5g, or 18g as given in the ADXL103 datasheet) of an accelerometer have any affect on:

1. Range of measurement
2. Accuracy
3. Resolution
4. Speed / response of measurement / output

I have read the datasheet and some supporting documents but I can't quite connect all of the pieces in my head.

Many thanks in advance for your wisdom and also your patience.

Ben

 

1. Range of measurement, Yes

Most of the time the range change is from a sensor signal attenuator being laser trimmed/fused (as an alternative to several sets of manufacturing masks). The reset of the circuit remains unchanged so 2 3 4 internally won't change much. Look at datasheet page 3 for specifications. Selecting a device in a range similar to the expected force measurements usually is the best way to go if you don't have a range switchable device.

 

Welcome to AAC!
From the ADXL103 datasheet we have

For those ICs the rated typical sensitivities are 1000mV/g, 312mV/g and 100mV/g, so if you are interested in the +-1.7g range only then the 1000mV/g gives the best resolution.
Regarding accuracy, all those ICs have the same rated maximum non-linearity of +-1.25% and the same temperature sensitivity error of +-0.3%

 

Thank you for the welcome and quick response.

Additionally, is it possible to define a range of angular measurement from the g value?

 

Gravitational acceleration is 1g acting vertically at a given location on the Earth's surface. If the IC is static and oriented so that the output voltage is half the supply voltage, that orientation indicates '0g', i.e the local horizontal acceleration. An angular deviation of x° from that will produce a corresponding output change proportional to sin(x).

 

Thank you for the welcome and quick response.

Additionally, is it possible to define a range of angular measurement from the g value?

Yes
https://www.digikey.com/en/articles/using-an-accelerometer-for-inclination-sensing

but it's usually better to use a IMU with integrated tilt data if you need accurate position data like this:
SCL3300 3-axis (XYZ) inclinometer
https://www.murata.com/en-us/products/sensor/inclinometer/overview/lineup/scl3300

https://forum.allaboutcircuits.com/...nd-sensor-node-for-canbus.189388/post-1814078