SBAA510 October 2021 DRV5032 , TMAG5170 , TMAG5231 , TMAG5273
The first test compares the performance of the Reed switch and the DRV5032 hall effect switch. For this test, several distances from each side of the X-axis are chosen for detection sensitivity. The Z-axis value is also varied to show sensitivity over multiple measurement nodes in the {Z,Y} plane. This test set also includes an identical testing process with variation nodes on the {X, Z} plane to show sensitivity dependency with respect to device positioning. Figure 5-1 shows the magnet approaches used for both tests.
![]() | ![]() |
The cylindrical magnet used in this test is magnetized through the length of the body. This creates a very low flux value at the center point of the magnet. For this reason, the center-point position along the y-axis provides no useful testing results for the DRV5032, but is added for accurate modeling purposes.
For each respective nodal coordinate (Y-axis, Z-axis), the magnet is slowly moved towards the device and the distance recorded. A standard tape measure is used for measuring the detection distance at an accuracy of 1/16th (0.0625) in. In the case of the DRV5032, the device is powered from a stable 3.3V DC power supply and populated on a small breadboard. An LED is also leveraged for visual indication of detection as well. The tape measure is only used to measure distance to the DRV5032 after the fact from a mark on a piece of masking tape and is removed during testing to negate any influence from the metal. This is shown in Figure 5-2.
![]() | ![]() |