• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 6682 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

How to decide the angle at which hall sensors has to be placed in a bldc motor

annet rose
Prodigy 40 points
Other Parts Discussed in Thread: DRV5013, DRV8308

Hi all,

We are trying to control a bldc motor using two hall sensors. But before proceeding further i wish to be clear about the calculation behind the hall sensor placement angle (electrical angle). 

In three hall sensor configuration why 120/60 angle is most commonly used? What is the calculation behind it. How did we reach at this angle configuration. 

We have two hall sensor assemblies available with us(These are two hall sensor configuration i.e only two sensors are placed for control.Third signal is generated using the combination of these two signals). One is with hall sensors placed 90 degrees apart and other with 67 degrees apart. What might be the reason behind this angle? Why not 180 degrees?

Also one thing observed is that with 67 degrees, speed achieved is more. What might be the reason behind that? Is it because the change in states are detected faster in 67 degree configuration? 

Regards,

Ann

  • 0
    TI__Guru 71526 points

    Hello Ann,

    What Hall-Effect Sensors are you using now? 

    Also, when you say at 67-degree angle "speed achieved is more" - does this mean the motor is spinning faster?  Or that it achieves the desired speed in a faster time?

    Thank you for using E2E.

    ~Leonard

  • 0
    TI__Mastermind 20985 points

    Ann, please see what I wrote on this topic in the DRV5012 datasheet on Page 12.

    If you're going to use 2 sensors instead of 3, they will no longer directly map to the phase voltages; instead, the 2 sensors will describe the electrical period.  They need to have a linear output.  The best spacing when using 2 sensors is 2 / [number of poles] * 90°.  That produces a sine and cosine period for each electrical cycle.

    Best regards,
    RE

  • 0 in reply to LeonardEllis
    Prodigy 40 points

    Hi Leonard,

    Thank you for replying.

    At present hall sensor used is A3290 by Allegro . 

    We are doing trapezoidal control. Our motor has 12 slots and 8 poles.

    LeonardEllis said:
    Also, when you say at 67-degree angle "speed achieved is more" - does this mean the motor is spinning faster?  Or that it achieves the desired speed in a faster time?

    "Speed achieved is more" means motor could achieve desired speed . But with 90 degree placement motor is not able to achieve desired speed. There is a difference of around 2000 rpm.

  • 0 in reply to RE
    Prodigy 40 points

    Hi RE,

    Thank you for the datasheet. It was helpful. But still i have few doubts.

    "The best spacing when using 2 sensors is 2 / [number of poles] * 90°"

    Why is that in the equation it is 90 degree and not 180 degree. I couldn't understand the logic behind it. Is it because 2 hall sensor combination generates four 2-bit states (360/4=90)?
    But again 00,11 are invalid states or not?

    Our motor has 8 poles and we are using trapezoidal control.
    Out of the two hall sensors, one signal will be used to generate the third signal using an integrator and a ZCD ( I am not sure how much feasible that is).

  • 0 in reply to annet rose
    TI__Mastermind 20985 points
    Generally, if you're using 2 Linear Hall sensors to measure angle, it is optimal for them to have a quadrature (90°) phase offset. Then, when one signal reaches its peak and has little change per degree, the other signal has its maximum slope, allowing you to accurately measure the angle. If you used a 180° phase offset like you propose, then both signals would reach their peaks (one high, one low) at the same time, and you'd have lower accuracy since it's harder to distinguish the small change per degree.

    However, in your last message it sounds like you indicated you want the normal 3 sensors and 6 Hall states, while only using 2 actual sensors and then estimating the 3rd. In that case I guess you want the normal 120° electrical cycle offset using DRV5013 Hall Latches.

    Best regards,
    RE
  • 0 in reply to RE
    Prodigy 40 points

    Hi RE,

    RE said:
    However, in your last message it sounds like you indicated you want the normal 3 sensors and 6 Hall states, while only using 2 actual sensors and then estimating the 3rd. In that case I guess you want the normal 120° electrical cycle offset using DRV5013 Hall Latches.

    It is that , with two hall sensors I couldn't think of any other logic other than generating the third signal using external circuitry. But if that is the case, the angle will not be 67 nor 90. It would be 30 ( 2/8*120). But the problem is that the hall sensor assembly available with us uses these two combinations.

    Can you help me with the 2 hall sensor logic? It would be of great help if you could share any document related to this as i am unable find any.

    Regards,

    Annet

  • 0 in reply to annet rose
    TI__Mastermind 20985 points
    For Hall sensor logic, you can check out the DRV8308 datasheet on the bottom of Page 17. However, if you only have 2 digital-output sensors, it's not possible to generate the 3rd in software, because there's ambiguity.

    Best regards,
    RE