Changes in axes directions across device/firmware versions?

[vincentvanhees]

Hello, could you clarify whether the axes directions have changed across device/firmware versions? If yes, which device/firmware version should I use as rule to identify the orientation of the axes?

I am looking at gt3x files collected in 2016 with device serial numbers starting with CLE and firmware version 2.5.0, they were worn on the hip.

My impression from the raw data is that the x-axis represents the longitudinal axis because it has close to zero gravity throughout the night and more gravity during the day as opposed to the other axes which have time periods when they pick up a lot of gravity during the night.

This seems to contrast with the ActiGraph axes definitions I can find online where the y-axis is consistently shown as the longitudinal axis for an ActiGraph worn on the hip. For example in this post on your website.

I remember hearing that the axes definitions have changed at some point in time, but I cannot remember the details. If you could clarify that would be much appreciated.

UPDATE: I see that axis rotation specific to "wGT3X-BT (Serial Numbers "MOS") firmware version 1.6.0" is discussed in here but that would not explain my observations for CLE and firmware 2.5.0.