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.

DRV8412 mode 3 or 4 - implementing a brake

Other Parts Discussed in Thread: DRV8412, LM3S3748

Hello all,

how is the most secure way to implement a brake for mode 3 or 4 ?  Is it possible to switch to mode 0, put PWM-A and PWM-B inputs to the same value and to change back to mode 3 or 4,  or it must be done by another (hardware) way?  Suppose inputs M1 and M2 are connected to a processor.

My second, hope the last, asking. If I remember information in the datasheet well, cycle by cycle over curent  doesn't work with PWM 0% or 100%. Am I right ?

 

With best regards

 

Igor

 

 

 

 

 

 

 

  • Igor,

    For mode 3 (M2 = 1 and M1 = 0), that is the parallel full bridge mode which means the PWM-A is controlling one side of the motor and PWM-B is controlling the other.  If you drive these both to GND, then you will have braking across the motor with no need to change the mode.

    For mode 4 (M2 = M1 = 1), this is commonly used for a case when you have limited GPIOs from the microprocessor.  I am not sure how your motor is connected in this case as this mode is reserved for 2x brushed DC motors....one across A&B and the other across C&D.  But, you are correct that you would need to switch to another mode to break and drive the inputs to each half-bridge to be equal.  

    To answer, your question.....you can switch modes and the device will respond to mode changes "on-the-fly".  If you have the GPIOs, I would just use mode 1 instead of mode 4 and fix the mode pins.  If you are limited on PWM channels, I can understand the desire to switch modes. 

    For you other question, the answer is yes.  If you are in cycle-by-cycle mode and the output trips into protection, a rising edge on the channel that tripped will be required to recover from CBC.  Otherwise, you will be stuck in CBC.  You can still use nearly 100 or 0 percent duty cycle and just insert a minimum 50ns pulse to avoid getting stuck in CBC.

     

  • Hello Ryan,

    Thank you much for your very clear explanation. You were guessing well I am short of GPIOs - I am trying to develop a driver for four stepper motors as well as for brushed motors controlled by a microprocessor like LM3S3748 having eight PWM outputs only. Yeah, there is no problem to use even Z80 to control DRV8412, but I would prefer to use 100 pins for controlling display, keyboard, DROs and drivers for my CNC project..

    If I can have another  humble question to member of TI stuff - I am very eager to try some of the last produced Cortex M3 or M4 Stellaris processors but all my asking about samples to European distributors went to nowhere ...  Could you send me an E-mail of some helping person?

    With best regards

    Igor

     

  • Igor,

     

    Send me your email address and a list of Cortex devices you would like to sample and I will see if I can help.  It is possible that we are out of stock, but let me see what I can do.

     

     

  • Hello Igor/Ryan

    This post is useful for me also.

    Hello Ryan,

    How can we implement the braking using DRV8412 when there is power fails or no power @ MCU side(Controlling the DRV device)?

    With Regards,

    Harshal

  • Harshal,

    Assuming the DRV8412 still has a 12V gate drive supply and motor supply is active, then simple pull-down resistors on the input will ensure the outputs go to a LOW-LOW state across the motor and cause a slow decay braking.