Thank you for the feedback,
Finally I found a solution that seems to work for me.
The register of my sensor is based on a FLOAT type variable (32 bits) coded in AB-CD or CD-AB.
I chose CD-AB.
- I read this one with a 4 x 8 bit array variable.
- Then I transfer the results with: 'Copy array part' by making my bit shifts:
- bit 0 to 2
- bit 1 to 3
- bit 2 to 0
- bit 3 to 1
- then I copy the new variable of type ABCD in a temporary 'buffer' then in a variable 'Real'.
It works for me.