Pb with self-test

[jmcornil]

Hello Kris

First of all, thank you for your work and for sharing it.

I am a beginner with MPU6050 and I am reading your code in order to understand the lacks in the documentation of Invense. I have two questions

1. In your SelfTest() function code, there are lines such as :
   selfTest[0] = (rawData[0] >> 3) | (rawData[3] & 0x30) >> 4
   I wonder if (rawData[0] >> 3) would not be replaced with (rawData[0] >> 3) & ~0b11
   in order to kill bit-0 and bit-1 ?
   Without & ~0b11, it could generate a little error up to 3, no ?
2. I have also a question about bit-0 in accel bias register.
   Can we arbitrarily set them to 0 or 1, and what is the influence of each choice
   on the temperature compensation?

Thank you in advance for your reply

J.M. CORNIL

[kriswiner]

2) the temp compensation bit is programmed by Invensense, this has to be read and preserved for rewrite after the bias is determined. 1) Yes, probably a good idea. Maybe better would be (rawData[0] & 0xE0) >> 3.

…

On Sun, Sep 9, 2018 at 10:32 PM jmcornil ***@***.***> wrote: Hello Kris First of all, thank you for your work and for sharing it. I am a beginner with MPU6050 and I am reading your code in order to understand the lacks in the documentation of Invense. I have two questions 1. In your SelfTest() function code, there are lines such as : selfTest[0] = (rawData[0] >> 3) | (rawData[3] & 0x30) >> 4 I wonder if (rawData[0] >> 3) would not be replaced with (rawData[0] >> 3) & ~0b11 in order to kill bit-0 and bit-1 ? Without & ~0b11, it could generate a little error up to 3, no ? 2. I have also a question about bit-0 in accel bias register. Can we arbitrarily set them to 0 or 1, and what is the influence of each choice on the temperature compensation? Thank you in advance for your reply J.M. CORNIL — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#36>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AGY1qrGo8pV9qAvVZZpvxUGAZTRo8qSVks5uZfmDgaJpZM4WgvFm> .

[jmcornil]

About the (2.) question
Yes, I saw it on your code but at the beginning, I executed a script by jF ROWBERG
https://github.com/jrowberg/i2cdevlib/blob/master/Arduino/MPU6050/examples/IMU_Zero/IMU_Zero.ino
and it seems not to me that he takes account of these bit-0.

Is it possible to retrieve the initial value that Invensense wrote in the register ?

Where can I find info about these bit ?
For the only one I found is on the page 7 the MPU Hardware Offset Registers Application Note
It is this line
Bit 0 on the low byte of each axis is a reserved bit and needs to be preserved
which caught my attention and encourage me to search on Google until I reach your work.

[kriswiner]

That is the only source I know.

…

On Tue, Sep 11, 2018 at 10:27 PM jmcornil ***@***.***> wrote: About the (2.) question Yes, I saw it on your code but at the beginning, I executed a script by jF ROWBERG https://github.com/jrowberg/i2cdevlib/blob/master/Arduino/MPU6050/examples/IMU_Zero/IMU_Zero.ino and it seems not to me that he takes account of these bit-0. Is it possible to retrieve the initial value that Invensense wrote in the register ? Where can I find info about these bit ? For the only one I found is on the page 7 the MPU Hardware Offset Registers Application Note It is this line Bit 0 on the low byte of each axis is a reserved bit and needs to be preserved which caught my attention and encourage me to search on Google until I reach your work. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#36 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AGY1qlkOkEAp6xkUHeAM6qi0dxYr0uMMks5uaJsvgaJpZM4WgvFm> .