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.

DRV8313: Alternative?

Part Number: DRV8313
Other Parts Discussed in Thread: DRV8323

We currently use the DRV8313 on hundreds of robot drives with small BLDCs using 3 phase PWN channels and while its great I would love to lower the pin count for our micro controller.  

IS there an alternative IC that would provide good control of a BLDC(robotic drive sensorless) without the 3 phase signal inputs?

With the current 3 phase signal we can operate the BLDCs with pretty good precision like a stepper motor so Id still be looking to keep that aspect of the design just reduce pin count.  Perhaps i2c control?

Any suggestions welcome though I would like to avoid adding additional micro controllers to increase pin count 

  • Adam,

    Let me ask a colleague who covers the devices that may work for you.

    Regards,

    -Adam

  • Adam That would be great

    Best

    Adam

  • Hey Adam D.,

    1) What is the voltage supply and max current needed for your application? It has to be sensorless and any specific commutation type like sinusoidal or trapezoidal?

    2) I want to clarify the problem.

    a) Is the problem that you need too many pins from your mcu to control the DRV8313? If so, are the pins required 3 pwms for the 3 INx pins, and another 3 pwms for the 3 ENx pins?

    b) Or is the problem that you dont want to use an MCU at all?

  • Adam,

    Do you need integrated FETs? Our newer gate drivers have external FETs.

    Are you currently providing three PWM signals? You mentioned three phases but I'm not sure if you are including the negative complements (that would make six PWM signals in total).

    Regards,

    -Adam

  • We are currently only using 3 PWM channels generated as table arrays for the DRV8313, this works well enough to control the small motors.  The motors we use are quite small using around 300ma peak but I could add in fets as well.

  • 1) we have 3 motors running quite low power around 300mA each max and as low as 100ma (max and min power factor for acceleration tuning).  The motors run on 8.4 volts and id like to avoid going any higher than this as the robots are battery powered and we step up from as low as 5.8 volts when the the batteries are low:  We have some very fine rotational accuracy while at the same time the option for quite fast movement(gimbal motors).  I wouldn't mind using feedback if the IC can manage that without too much effort but it has to be able to be operated in position mode(we cycle through a table array) but I dont really want to do sensor calculations on the micro as there is a lot going on the robots already(3 lidar distance sensors an IMU with sensor fusion wifi and IR led 10bit code blinking for trackingg)

    A) Yes the problem is we use most of the pins of our micro controller and 9 in total for PWM and 1 to enable 3 DRV8313 drivers in total.

    B)  The table array we use to drive the motors is modified sinusoidal you can see the following link on how we generate the table array http://www.berryjam.eu/2015/04/driving-bldc-gimbals-at-super-slow-speeds-with-arduino/

    Im trying to reduce complexity and pin count so I dont have to either use another larger more expensive micro or add in an additional micro.

    You can see the research here http://hochschuh-donovan.com/portfolio/empathy-swarm/

    I hope that helps shed some more light on what Im trying to improve

  • Adam,

    I wonder if you could use our DRV8323 device. Here's the datasheet: http://www.ti.com/lit/ds/symlink/drv8323.pdf

    You could use this device in 1x PWM mode if the motor has hall sensors which would really reduce your pin count. The device accepts hall input to automatically commutate the motor.

    It accepts down to 6V input so maybe this works for you?

    As mentioned this uses external FET but with dual packages and small FET requirements you should be okay.

    Regards,

    -Adam

  • We dont really have the option for hall sensors for the motors as they are small brushless motors typically found on camera gimbals.   We have almost a thousand motors and wouldn't really be cost effective to change motors at this stage.  I had tried previously the DRV0983 which I failed to understand wasnt for the type of control I wanted though it did have some very good performance being able to set it very slowly with nice smooth movement, in the end I think something like this but with variable speed and position would be fantastic.  The rest of our hardware is not likely to change as its all pretty optimized just the pin count is high.  If the drv8313 was to exist to have i2c control and allow the same motion movement as we use with the table array I think you would have a pretty unique product.

  • Adam,

    Please checkout the DRV8323, you could likely still save a few pins even if using the 3x PWM mode without hall sensors.

    Regards,

    -Adam

  • If it cant do 1 x PWM without hall sensors I dont see how thats possible.  The DRV8313 is only using 3 PWM channels and 1 pin for enabling all 3 motor drivers.  total pin count of 10. Saving one pin for enable or disable wouldn't really make much of a difference.

  • Adam,

    Unfortunately we may not have a better sensorless option at this time.

    Regards,

    -Adam