DRV5013: Hall Latches and Hall Switches

Hello Sweshwaran, 

Thanks you for posting to the sensing forum! Just as a preface we have also released a new device the TMAG5115, the device was with BLDC motor applications. It offers a higher bandwidth, less delay, and jitter than the DRV5013 for more precise switching.

 To answer your questions:

• Why magnetic switches cannot be used instead of the Latches for this Motor control application ? Can you please send any reference design/ application note explaining the use of magnetic switch in Motor control application instead of Latch, if there is any ?
  • Unfortunately we do not have the document you are requesting but I can explain why you would not want to do this. The hysteresis is set up differently for Hall switches vs latches. Hall switches will change output as soon as the field drops below the release point, while latches will change output until the polarity of the magnet changes. You may be able to use unipolar switches (switches that are only sensitive to one polarity) but it would be very sensitive to placement. Placements would need to guarantee a 50% duty cycle of hall effect sensor transitions during a rotation. A Hall latch essentially guarantees a natural 50% duty cycle between poles and is not as sensitive to placement. It is not possible to use an omnipolar switch since they are sensitive to both north and south fields. Please look at the output images below: 

                     First image is the output of a latch, while the second image is the output of an  omnipolar switch for the same rotor.

• In DRV5013, apart from the pull up resistor, is it necessary to connect the Capacitor (C2 as per datasheet) if we are already having a single pole first order RC filter deployed on the line ? What purpose do the capacitor C2 serve apart from filtering ?
  • The C2 capacitor is not required. This is an optional component as noted in the datasheet most applications do not require the C2 filtering capacitor. There is no other purpose for the capacitor other than the filter provided.
• What should be the cut off frequency of that RC filter (question-2),as recommended by TI if the required bandwidth of the sensor is 4500Hz ? On What basis the filter cut off frequency is decided ?
  • If a filter is required, it is recommended to provide at least two times the bandwidth required for your system. In your case this would mean that the filter cutoff should be max 9000Hz but ideally you would want to give your system 3 times the bandwidth required, so the cutoff should be around 13.5kHz. Keep in mind most systems do not require the filter so you may not need to limit the sensors bandwidth.
• Why this DRV5013, is rated with higher voltage, up to 38V ? For what application it is rated with this high voltage ? For those application, is it recommended to directly bias this with, say 30V ? What are the benefits seen here?
  • There are a couple of reasons why the DRV5013 is rated up to 38V mainly done for designers flexibility, but changing the voltage does not offer any benefits compared to 3.3V. In a BLDC application the main reason for such a high voltage is due some systems have the motor and the device connected to the same bus. When the motor spins to a stop it generates a voltage on the bus that exceeds the normal bus voltage. Adding bulk capacitors here helps or having the system break instead of spin to a halt limits this voltage but if no protection is added here the device is able to protect itself. In some other instances some users will have a 12V bus or 24V bus but don't want to get an extra regulator, that is not an issue since the DRV5013 can work on that voltage as well. 

Here is some other literature that might help: https://www.ti.com/lit/an/slvaeg3b/slvaeg3b.pdf

The TMAG5115EVM is compatible with NEMA17 BLDC motors so it can be used as a module to place for trap commutation: https://www.ti.com/tool/TMAG5115EVM

I hope this helps!

Best,

Isaac