2021-03-22 06:23 PM
Hi ST
we are seeing wrong response for WHO AM I command,
SPI Configuration:
sent 0x8F00 on MOSI line and seen 0x0040 and it was random some times we are seeing 0x007F as a response. can you please help us on this
Solved! Go to Solution.
2021-03-24 01:27 AM
Hello @Eleon BORLINI
As we see it is working fine with the I2C lines for reading the WHO AM I register, but as mentioned above it is not working as expected over SPI lines.
can we take this post to high priority please,
2021-03-24 07:17 AM
No clue from my side, unfortunately :( it should work from chip side...
You could try reducing the Vdd / VddIO values.
And you could also try writing the I2C_disable bit in the CTRL4_C (13h) register, that is a further insurance in disabling the I2C interface (could be useful in case ofin case of huge cross-talks).
You can try asking to the online support for more help.
-Eleon
2021-03-25 02:48 AM
Hello @Eleon BORLINI
found the root cause for it, MISO line is with high impedance , just need to place the resistance to avoid reflections on the line. In my case we have added 20ohms resistor is added.
2021-03-25 02:54 AM
Glad to hear you solved your issue @SChin.2 !
-Eleon
2021-03-26 01:21 AM
Hi @Eleon BORLINI
Is their any calibration process that we need to perform on this sensor ?.
As here we are seeing gyro reading even though it is stationary, When it is at stationary we are suppose to see zero values but we are seeing non zero values any clue ?
reading (FS: +/- 500dps, ODR:104hz):
X: 350mdps
Y: -5285mdps
Z: -4550mdps
2021-03-26 01:48 AM
Hi @SChin.2 ,
please note that the G_TyOff parameter, i.e. the angular rate typical zero-rate level after soldering (datasheet p.20) has typical values in the range of ±10dps.
Your results are actually in this range, so no issue.
-Eleon
2021-03-26 02:42 AM
Can you please comment on the calibration part too !!
2021-03-26 03:39 AM
It's always a good step calibrating a gyroscope before using it.
The LSM6DS3 has no internal registers to store the ZRL compensation value (datasheet p.28), so you should subtract it to your data-out value after having characterized in a stationary condition.
Alternatively, if you were using an STM32 as application processor, you could evaluate the use of dedicated libraries, such as the MotionGC of the X-CUBE-MEMS1 function package.
-Eleon