• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 82 subscribers
  • Views 1443 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Open Source BLDC Motion Control project w/optical feedback (Advice Needed)

Mike De
Prodigy 30 points

Other Parts Discussed in Thread: DRV8302, DRV8301

Need help selecting a baseline MCU, DRV83XX and KIT to use as a starting point in the following machine motion control project.

BLDC Controller/Driver capable of driving two motors on a single linear stage. With two motors driving the same stage additional processing power in the PID will be needed to keep them in sync.

Power: 24-48V 50-100W
Feedback: InstaSPIN + Encoder
Rotary Encoder: Optical Incremental or Absolute with Index (500 - 2000) CPR Quadrature
Linear Encoder: Absolute, used for power up positioning and optionally integrated into the PID loop
Hardware Safety: Limit Switch
Software Safety: Motor or Position Sync Failure
Communication: Ethernet IEEE 1588 *Optional based on price and reliability concerns

Any advice is greatly appreciated, since we are planning to open source this project development cost is of some concern however the final goal is to provide open, flexible and affordable BLDC motion control solution with variable power and resolution based on motor, encoder and power supply selection.

  • 0
    TI__Guru** 119680 points

    What do you mean two motors on a single linear stage?  I assume that's an x-y type table but you need two inverters?

    Hmmm, this is tricky.  The HW that best meets the requirements right now is the TMDS2MTRPFCKIT. It has two inverters in the power range. The problem is that the driver device (DRV8402) is NRND. 

    We are replacing this kit in '13 with new inverter stages that use DRV8302 + TI NexFETs.  It will come withn encoder feedback capability and an option DYNO add-on that includes two Teknic servo motors (with incremental encoder and halls built in), coupling, and frame mount so you can mate the two motors and do active testing.  The problem for you here is that we are only building this hardware for up to 24V.  The DRV8302 and the motors can handle more, but we specifically are keeping this to a lower voltage for training/demo purposes. 

    Are you looking for turn key HW or will you build yourself?

     

  • 0 in reply to ChrisClearman
    Prodigy 30 points

    RE: What do you mean two motors on a single linear stage?

    Driving a single axis with two BLDC motors located at opposite ends of the axis to balance the mechanical forces on the axis and reduce motor size. Because high precision and tight synchronization between the two motors is required each motor will have it's own encoder and both motors will be driven by a single controller.

    I realize this will require a more complex PID loop to keep both motors in sync and also fault when they go out of sync to prevent mechanical damage. The main concern is selecting a MCU with enough processing power to handle this type of PID with high CPR optical encoders and still have the option of adding an additional absolute linear encoder.

    One option might be to use a couple of NexFET eval kits combined with other kit(s) however i'm new to TI solutions so I'm looking for some guidance.

    Regarding the Dyno: We are also building a Dyno but we have ruled out a second motor for load testing and will be going the foil-SG route with a frictional brake and speed sensor. This will also be controlled by a TI MCU. The purpose of the Dyno is for low cost BLDC motor evaluation and verification of BLDC controller designs.

    RE: Turn key: The plan is to leverage the eval kits and light breadboarding for the initial design if possible, the main concern is processing power and I/O, for example we could select a high power MCU and scale back once the design is completed to the lowest cost capable MCU, again just looking for some guidance in initial product selection.  




  • 0 in reply to Mike De
    TI__Guru** 119680 points

    ok, at the processor side, the lowes performance I would go is a 90 MHz F28069.  I'd also look at the 150 MHz F28335 or the 180 MHz Hercules RM48. 

    I think for kits that can get you furthest right now you would want to look at something based on DRV8301 (DRV8301-HC-C2-KIT, but you would need a new 28069 or 28335 controlCARD; DRV8301-RM48-KIT).  Both are single inverters with encoder inputs, but all have pins brought out to headers so you could feasibly drive two inverters from the same controlCARD.