Dale, Thank you again for the instant helpful response.
Regarding the 3 possibilities. 1) I actually used ShimmerConnect to show the streamed Magnetometers on the fly, instead of plotting the data in Matlab after data acquisition. 2) I will chose another location to run the test again. 3) I will confirm this possibility by a test. Thank you again for pointing out these likely causes of the problem. I'll share my investigation once I have a conclusion.
Best, Weiguang On 21/01/2015 3:51 PM, Dale Roberts wrote:
... Sorry, I meant to go back and correct #1 below. The RAW values are in fact signed. So, if you were expecting them to be unsigned, then maybe you are subtracting an offset when you should not be. dale ________________________________________ From: [email protected] <[email protected]> on behalf of Dale Roberts <[email protected]> Sent: Wednesday, January 21, 2015 3:47 PM To: Weiguang guan; [email protected] Subject: Re: [Shimmer-users] Magnetometer puzzle You are correct in your expectation that if you rotate the Shimmer 360 degrees, you should generally see both positive and negative numbers on the X and Y axes due to the Earth's magnetic field. There are a few possibilities I can think of. 1) Maybe you are making an error in handling the data in Matlab. Run ShimmerConnect, enable the Magnetometers, and see what numbers you get there on the screen. I just did that, and Z stays relatively stable, and X and Y RAW values both go above and below zero. The raw sensors may be "unsigned", in which case you may have to subtract an offset to get properly "signed" numbers. 2) If you do indeed have a large magnet very nearby (like w/in a few cm of a neodymium magnet, or you are near an MRI machine!), the sensors can saturate and the sign may appear to change erroneously. 3) Your calibration is messed up. Collect a CSV file with ShimmerConnect. Open in Excel or some other spreadsheet. Look at both the RAW and CAL mag data. If the RAW X and Y signals are going positive and negative, and the CAL is not, then your calibration is no good. dale ________________________________________ From: [email protected] <[email protected]> on behalf of Weiguang guan <[email protected]> Sent: Wednesday, January 21, 2015 2:47 PM To: [email protected] Subject: [Shimmer-users] Magnetometer puzzle Hi Shimmer users, This morning I played with a single Shimmer3 sensor and watched the plots of magnetometer x, y, z (all CAL) on the fly. I placed the sensor on a table, and slowly rotated it while keeping one of its face in full contact with the table surface. I found one of the axes (which is Cal Z) remained roughly constant so I believe the Z axis points up. What confused me is that the other two components are mostly negative during the rotation, especially one of them never had a positive value no matter how much I rotated it. How to explain this? Even if some electrical device in the room had impact on the local magnetic field, the combined field (with earth magnetic field) should point to certain direction. I would very much appreciate it if someone could shed a light on the confusion. Thank you very much. Best, Weiguang _______________________________________________ Shimmer-users mailing list [email protected] https://lists.eecs.harvard.edu/mailman/listinfo/shimmer-users _______________________________________________ Shimmer-users mailing list [email protected] https://lists.eecs.harvard.edu/mailman/listinfo/shimmer-users
_______________________________________________ Shimmer-users mailing list [email protected] https://lists.eecs.harvard.edu/mailman/listinfo/shimmer-users
