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.

Use of Instaspin with BLDC for zero speed, hold position application

Hi,

I have a BLDC motor control application in which there will be a inner rate loop/outer position loop holding the motor at its current position. I have hall sensors available and am currently using standard trapazoidal commutation methods. I am interested in moving to a more sophisticated drive scheme to help minimize noise from PWM switching. The motor is low inductance (around 90uH) and thus the PWM frequency is between 100kHz and 150kHz.

I would like to use the launch XL-F28027F with BoostXL drive EVM kit to experiment with.

Can you tell me if the instaspin solution will:

1. Be a less noisy EMI solution vs trapazoidal drive.

2. Will work holding torque at zero speed?

Thanks,

Mitch

  • 2. No, InstaSPIN-FOC is sensorless and will not hold torque at 0 speed. It needs velocity to track the rotor flux angle. Hall sensors will not give enough resolution to do true position control so the InstaSPIN-MOTION labs use an encoder.

    1. yes, in general moving to a space vector modulation can help some with noise. those PWM rates are much higher than typically required in my experience. you can do sinusoidal modulation with a BLDC technique (hall sensor based commutation), but we don't offer an example. For our solutions we go the extra step and do FOC with our sinusoidal/space vector modulation.

    For your application I don't think our solutions are the best fit...
  • Chris,

    Is this still the case? We are pursuing a similar application (traction drive), and are evaluating our options. We will also have a hall-sensor with an optional encoder. It's been almost a year since this post, so I'm wondering what improvements have been made since then that might support zero-speed torque control.

    Thanks,
    Elliott
  • zero speed holding torque control requires the use of sensors.
    most traction drives use a resolver as the feedback sensor.
  • Thanks for the quick reply, Chris.

    We will have an encoder input, as well as hall sensor feedback. Will this be enough sensor feedback for InstaSPIN to control the drive properly at low speeds?

    I also wonder whether the HFI feature comes into play at low speeds.
  • all you need is an encoder. you can use InstaSPIN-MOTION.
    Proj_lab12 introduces encoder for velocity control. In a traction application you are normally just controlling torque (commanded from a throttle) so you don't really need all of the motion components, but it makes it easy to get started.

    once you have proj_lab12 working you can disable the speed controller and enter an IqRef_A torque setting directly while still using the SpinTAC converter features.

    be sure to note that the encoder is producing a mechanical angle and FOC requires an electrical, so you need to do the offset alignment (done in the example using RsRecal flag) and use the converter feature.

    since you are using an encoder you don't need to waste time on HFI. That is a sensorless technique for zero and low speed that is very difficult to get working well unless the motor is designed very specifically and you limit the load at start-up.
  • I have an application where my customer wants to use a BLDC for full position control, like a stepper. It can have hall sensors, but they do not want an encoder.

    They want to use a lookup table to create the sine wave controls for the 3-phases, and they believe they can use that to control position, with the hall sensors there for use at high speed and to verify position.

    So there are two questions here:
    #1 - Will the whole concept work?
    #2 - Can InstaSPIN-MOTION add any value to this? They are concerned InstaSPIN-MOTION only works in non-stalling situations.
  • Stepper motor control is different with InstaSPIN-FOC, it's a open loop control without rotor angle to do position control, so the control mode is ok.
  • So ignore my comparison to a stepper motor, maybe not a good comparison, let me state it a differnt way.

    They want a BLDC motor with servo control. If there was no encoder, could InstaSPIN-MOTION still work and provide position control? Would it be able to maintain a known position during a stall?
  • This is very difficult to implement the servo performance without position sensor, it's better to use sensor InstaSPIN-Motion.
  • When you say position sensor, are you talking about the three hall sensors that are typically used on a BLDC, or a full-blown encoder, like with 500-1000 steps per rev? We do have the hall sensors, and could generate a "virtual" encoder position in software.