2.3.7 45° Alignment
The TIDA-060040 also shows an additional alignment option for
Figure 2-23. TMAG5170 45° Alignment
This second sensor location option has the device physically rotated 45°. Instead of the normal orthogonal
alignment, the X and Y axes of the sensor each observe the same input magnitude. The effective input for each
axis is the average of the two axes. The benefit of installing the device with this rotation is that it allows the
sensor to be placed in locations where one input might be excessively large, one input might not have sufficient
SNR, or both. This alignment offers a potentially more compact solution, but at the cost of phase alignment. By
rotating the sensor in this fashion, a phase error is introduced which is a function of the total mismatch between
the two channels. This phase error is depicted as β in
. To properly use inputs with a phase error, the
following trigonometric adjustment to the arctangent calculation is helpful:
Acos θ − β = Acos θ cos β + Asin θ sin β
(5)
In2
In1
=
Acos β cos θ + Asin θ sin β
Asin θ
(6)
In2
In1
=
cos β
tan θ
+ sin β
(7)
tan θ =
cos β
In2
In1 − sin β
(8)
θ = tan
−1
cos β
In2
In1 − sin β
(9)
, it is important to verify the actual phase shift of the vector components.
Manufacturing alignment errors and the combination of unequal amplitude inputs may result in a value of β
which differs from the physical rotation of the sensor.
System Overview
16
Absolute Angle Encoder Reference Design With Hall-Effect Sensors for
Precise Motor Position Control
Copyright © 2022 Texas Instruments Incorporated
