Actions
Forum Thread Details
  • State Not Answered
  • Replies 2 replies
  • Subscribers 287 subscribers
  • Views 271 views
  • Users 0 members are here
Related

Current Sensing; need a small solution

Catwell
I have to sense current through 3 coils of a 3 phase DC motor. There are many solutions: shunt resistor, current sense ICs, etc. But what if you are building a motor driver circuit on a 1 inch diameter PCB? Almost the entire PCB is used for the motor driver circuitry. The bosses proposed a "management solution" for me;  "the smallest option."
 
After I tried on my own to no avail, I turn to you, my comrades in engineering, for the ultimate resolution to this tiny problem. I am willing to entertain any option, even the obscure and conceptual.

Catwell
  • 0

    What magnitude of current are you expecting?

    Do you really have to measure all three phases? One usually can rely on current measurements on two phases, providing the modulation and the motor is symmetric.

     

    Two solutions pop my mind.

    Have you considered the Allegro ACS7xx series of integrated current sensors? SO8 isn't that small, but it's still a good reference of what you are able to squeeze in on your board.

    Have you considered external ferrite beads on the motor cables/connectors, with a gap for a linear hall sensor? This solution is found on many medium power motor drivers.

  • 0 in reply to Former Member

    I do have to sense each coil, unfortunately. I am expecting up to 2A through any one coil, max.

     

    My backup plan is the Allegro, actually. But the chip is a little large for what is on the wafer already.

     

    I like the idea of moving the sensing outside of the drive PCB. However, this have to be a closed and protected system. In other words, no external wires or components. Another company makes the motor/electronics housing. So getting it changed would be a stretch, but I'm not beyond demanding it.

     

     

     

     

     

    I was also thinking of using the BEMF (Back ElectroMotive Force) to inversely calculate the current. Has anyone tried this?