Why are SD Converters Used in Motor Drives?

Discussion in 'General Electronics Chat' started by NLightNMe, Aug 15, 2011.

  1. NLightNMe

    Thread Starter New Member

    Apr 20, 2011
    Hi Everyone,

    I have a question I hope someone can shed light on for me. I have a pretty good high-level understanding of motor control. One thing I've wondered for a long time, though, is why do they typically use sigma-delta converters for the current sense portion. This would be from phase motors such as BLDC and AC Induction.

    Consider Avago Tech's solution from their website: http://www.avagotech.com/pages/en/industrial/motor_control/ . The "Phase Current Sense" block suggest several amplifiers build from sigma-deltas and then a stand alone sigma-delta block without any digital filter stage. What is the advantage of using that over, for instance, just a straight SAR converter. The SAR converter would give the true value without another filter block.

    I've seen this basic block diagram from several suppliers, but I've never figured out the driving benefit.

    Does anyone know why sigma-deltas are preferred for motor drive current sense?
  2. GetDeviceInfo

    Senior Member

    Jun 7, 2009
    I suspect it's in the inherent noise rejetion.
  3. praondevou

    AAC Fanatic!

    Jul 9, 2011
  4. NLightNMe

    Thread Starter New Member

    Apr 20, 2011
    Thank you for the information!

    I understand the differences between SAR ADC's and DS ADC's. However, motor control seems like an area where the capabilities overlap. So I have difficulty seeing why delta-sigmas are consistently chosen over SAR's for this applications. It leads me to believe that one of two things is going on. Either DS ADC's are used because of their legacy benefits over SAR (lower noise, inherently differential inputs, etc...) . These benefits may not be true anymore at this performance node. Or DS ADC's are used because of something in the motor control algorithm that I don't understand. Is there some reason why a constant bitstream is preferable to a sampled conversion?