Different question about serious drift

[mmarkell]

Hello!

First of all, thank you for the code you've written for the LSM9DS0. It's taught me a lot and I am very grateful!

However, sadly, something is going very wrong with my LSM9DS0 measurement drift.

Basically, let's say I hold the device at rest. The yaw, pitch, and roll start at some value (arbitrary seeming, definitely not zeroed for a flat and upward facing orientation, for instance.)

Next, they drift, slowly at first, and then in very large jumps. Within a few minutes, the yaw pitch and roll I'm calculating might go in complete circles, even though the device is stationary.

It's not totally inaccurate - it measures acceleration quite well from point A to point B, and knows in general which way I am turning the device. However, the drift over time is pretty insane.

Steps I have taken: I am using a mahony filter, and recently implemented the calLSM9DS0 method from your code.

I would appreciate any advice you may have - do you think my device is faulty?

Thank you for your help!

[kriswiner]

Likely you need to caliubrate the magnetometer. https://github.com/kriswiner/MPU-6050/wiki/Simple-and-Effective-Magnetometer-Calibration

…

On Sun, Apr 9, 2017 at 7:12 PM, Michael Markell ***@***.***> wrote: Hello! First of all, thank you for the code you've written for the LSM9DS0. It's taught me a lot and I am very grateful! However, sadly, something is going very wrong with my LSM9DS0 measurement drift. Basically, let's say I hold the device at rest. The yaw, pitch, and roll start at some value (arbitrary seeming, definitely not zeroed for a flat and upward facing orientation, for instance.) Next, they drift, slowly at first, and then in very large jumps. Within a few minutes, the yaw pitch and roll I'm calculating might go in complete circles, even though the device is stationary. It's not totally inaccurate - it measures acceleration quite well from point A to point B, and knows in general which way I am turning the device. However, the drift over time is pretty insane. Steps I have taken: I am using a mahony filter, and recently implemented the calLSM9DS0 method from your code. I would appreciate any advice you may have - do you think my device is faulty? Thank you for your help! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#12>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AGY1qhDx6aWtduFfw58XHOOQ8S9Cm7Wgks5ruY_7gaJpZM4M4PzV> .

[mmarkell]

Taking a look now. I was looking at https://github.com/kriswiner/MPU-9250/blob/master/MPU9250_BMP280_BasicAHRS_t3.ino and I noticed that you have float magCalibration[3] = {0, 0, 0}; // Factory mag calibration and mag bias Then, you use this as a multiplicative factor in numerous equations. However, you never set it to anything other than 0,0,0. Then, wouldn't any operation being multiplied simply equal 0? Maybe I'm missing something [image: Brown logo] <http://brown.edu/> *Michael Markell ‘18* B.S. Computer Science Candidate 302 229 0419 On Sun, Apr 9, 2017 at 10:19 PM, Kris Winer <notifications@github.com> wrote:

…

Likely you need to caliubrate the magnetometer. https://github.com/kriswiner/MPU-6050/wiki/Simple-and- Effective-Magnetometer-Calibration On Sun, Apr 9, 2017 at 7:12 PM, Michael Markell ***@***.***> wrote: > Hello! > > First of all, thank you for the code you've written for the LSM9DS0. It's > taught me a lot and I am very grateful! > > However, sadly, something is going very wrong with my LSM9DS0 measurement > drift. > > Basically, let's say I hold the device at rest. The yaw, pitch, and roll > start at some value (arbitrary seeming, definitely not zeroed for a flat > and upward facing orientation, for instance.) > > Next, they drift, slowly at first, and then in very large jumps. Within a > few minutes, the yaw pitch and roll I'm calculating might go in complete > circles, even though the device is stationary. > > It's not totally inaccurate - it measures acceleration quite well from > point A to point B, and knows in general which way I am turning the device. > However, the drift over time is pretty insane. > > Steps I have taken: I am using a mahony filter, and recently implemented > the calLSM9DS0 method from your code. > > I would appreciate any advice you may have - do you think my device is > faulty? > > Thank you for your help! > > — > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub > <#12>, or mute the thread > <https://github.com/notifications/unsubscribe-auth/ AGY1qhDx6aWtduFfw58XHOOQ8S9Cm7Wgks5ruY_7gaJpZM4M4PzV> > . > — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#12 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AKYxm2npZxBGA6_FTuZufws0bTtmp-wNks5ruZGbgaJpZM4M4PzV> .

[kriswiner]

Yeah, your missing the mag calibration functions. One is in initAK8963 and one is in magcalibration. But try this instead anyway: https://github.com/kriswiner/LSM9DS0/blob/master/Teensy3.1/LSM9DS0-MS5637/LSM9DS0_MS5637_Mini_Add_On.ino On Sun, Apr 9, 2017 at 7:41 PM, Michael Markell <notifications@github.com> wrote:

…

Taking a look now. I was looking at https://github.com/kriswiner/MPU-9250/blob/master/MPU9250_ BMP280_BasicAHRS_t3.ino and I noticed that you have float magCalibration[3] = {0, 0, 0}; // Factory mag calibration and mag bias Then, you use this as a multiplicative factor in numerous equations. However, you never set it to anything other than 0,0,0. Then, wouldn't any operation being multiplied simply equal 0? Maybe I'm missing something [image: Brown logo] <http://brown.edu/> *Michael Markell ‘18* B.S. Computer Science Candidate 302 229 0419 <(302)%20229-0419> On Sun, Apr 9, 2017 at 10:19 PM, Kris Winer ***@***.***> wrote: > Likely you need to caliubrate the magnetometer. > > https://github.com/kriswiner/MPU-6050/wiki/Simple-and- > Effective-Magnetometer-Calibration > > On Sun, Apr 9, 2017 at 7:12 PM, Michael Markell < ***@***.***> > wrote: > > > Hello! > > > > First of all, thank you for the code you've written for the LSM9DS0. It's > > taught me a lot and I am very grateful! > > > > However, sadly, something is going very wrong with my LSM9DS0 measurement > > drift. > > > > Basically, let's say I hold the device at rest. The yaw, pitch, and roll > > start at some value (arbitrary seeming, definitely not zeroed for a flat > > and upward facing orientation, for instance.) > > > > Next, they drift, slowly at first, and then in very large jumps. Within a > > few minutes, the yaw pitch and roll I'm calculating might go in complete > > circles, even though the device is stationary. > > > > It's not totally inaccurate - it measures acceleration quite well from > > point A to point B, and knows in general which way I am turning the > device. > > However, the drift over time is pretty insane. > > > > Steps I have taken: I am using a mahony filter, and recently implemented > > the calLSM9DS0 method from your code. > > > > I would appreciate any advice you may have - do you think my device is > > faulty? > > > > Thank you for your help! > > > > — > > You are receiving this because you are subscribed to this thread. > > Reply to this email directly, view it on GitHub > > <#12>, or mute the thread > > <https://github.com/notifications/unsubscribe-auth/ > AGY1qhDx6aWtduFfw58XHOOQ8S9Cm7Wgks5ruY_7gaJpZM4M4PzV> > > . > > > > — > You are receiving this because you authored the thread. > Reply to this email directly, view it on GitHub > <#12 (comment)>, > or mute the thread > <https://github.com/notifications/unsubscribe-auth/AKYxm2npZxBGA6_ FTuZufws0bTtmp-wNks5ruZGbgaJpZM4M4PzV> > . > — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#12 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AGY1ql3JDOc7s1n7XP14lIdu5EyPGsEdks5ruZbMgaJpZM4M4PzV> .