• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 30 subscribers
  • Views 659 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.

DRV8842 - Torque Loss

thiago cruz
Prodigy 30 points
Other Parts Discussed in Thread: DRV8842

Hi Everybody.

I just started drive an DC motor with DRV8842 and' i'm seeing strange, strange things. I already tried many things until two boards burned up.

Well, this unhappy event occurred when i was trying fix the lost torque problem.....

If I hold INx low and the other INx high, everything is just perfect (current, torque and etc..) but when i put pwm in one of the inputs.. for example 95% Ton and 05% Toff and hold the other input low, i get a torque about 10% of the one in the first case.

Any ideas??

Ps.: sorry about my english..

Thanks.

  • 0
    Prodigy 80 points

    Hello,


    Please give more details about the frequency of the pwm output, motor you are driving and if possible connections used from the MCU.

    Too low frequency and your motor will have time to break (since it's terminals are shorted for 5% of the time).


    Thank you.

  • 0 in reply to Valerian Sarbu
    Prodigy 30 points

    I'm using a "poor" motor...  24V - irms_no_load 0.7A - i_stall_torque - 10A.

    My board is based in MSP430F16611, i use Timer B port and i've tried 250Hz and 2KHz pwm frequencies.

    I have Vref = 1.6V, Rsense = 0.1R and full range current ( [i4..i0] = 0x1F).

    Thanks again. 

  • 0 in reply to thiago cruz
    Prodigy 80 points

    No problem and sorry for the slow reply.

    The frequency can go up to 100khz.

    Do you use fast decay, slow decay or mixed? I would recommend fast decay and a bit higher frequency.

    With that setup you have 3A current limit so it's within working current of the IC

    Good luck

  • 0 in reply to Valerian Sarbu
    TI__Expert 7455 points

    Hi Thiago,

    Sorry I have not had a chance to answer this post. When you say you have burned two boards, is this a board of your own design or one of our EVM's? If it is your board, how is the device being soldered in place? The Power Pad connection is important so I want to make sure this is not a source for problem.

    Your ITRIP is set to 3.2A which means that you will start the current regulation engine when this motor runs with this much current. This may be a problem if the motor can go up to 10A as this pretty much tells me at some point in time you will get the current regulation internal PWM at the same time you are inputting your own PWM. Ideally you will want to have one PWM or the other, but not both. The problem is the internal PWM is asynchronous to your INx PWM so the outputs will be all over the place and control will be lost.

    Since this motor is so big, an option would be to have your algorithm monitor current (by reading the analog voltage on the SENSE pin) and then adjust PWM according to how close you are to the maximum current you want to see (3.2A in this case). Once you are about to hit the current regulation point, you drive your PWM to 100% duty cycle and let the current regulation take over.

    Another option is to only use the internal current regulation and have your MSP430 modulate the VREF pin by using a DAC output function. In this case, the device will regulate torque with the internal current regulation and you can decide where the Torque set point is by moving VREF accordingly.

    A third option I can think of is realize the motor will most likely use large amounts of current for starting up and during stalling. It is possible that during run time you will never see more than the 3.2A. In this case, you can start the motor by using 100% duty cycle and then start your own PWM once the current has gone down in magnitude. It will require some small tuning but I think it is the easiest solution.

    Hope the info helps. Best regards,

    Jose Quinones