• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 3 answers
  • Subscribers 30 subscribers
  • Views 1859 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.

DRV11873: the speed issue: 9000rpm~10800rpm

Xianrong Shang
Intellectual 855 points
Part Number: DRV11873
Other Parts Discussed in Thread: DRV10987

H3ello.

Now I use the DRV11873 on the BLDC board, I found under the light or no load, when the FG output is 900~980Hz, the corresponding Motor speed is 9000~9800RPM. and the motor will be accidental suddenly stop drive (RD pin = H), eventually to grind to a halt (inertial slowly stop). Why?

The Motor itself could be up to 10800 RPM, according to the motor's specification.

  • 0 in reply to Xianrong Shang
    TI__Mastermind 18900 points

    Hello Xianrong Shang,

    It sounds like the DRV11873 is triggering a false lock, and this situation is well documented and presented in this app note: http://www.ti.com/lit/an/slva778/slva778.pdf

    Essentially, if the FG and RPM (by extension) are a multiple of the internal sampling clock (101.6Hz typ.) then the devie will trigger a false lock. If the motor is spinning in the 9000 and 9800 RPM range, then this would correspond to the false lock. Some customer will vary the output electrical frequency (lowering and raising the frequency periodically) to prevent this from happening if they have a specific RPM.

    Please read the document and let me know if you have any quetsions. I apologize for the inconvenice of this false lock.

    Best,

    -Cole

  • 0 in reply to Cole Macias
    Intellectual 855 points

    Hello Cole,

      Whether the DRV10987 has the similar problem ? 

    Could the DRV10987 reach 10800rpm ?

    Can you recommend a suited IC ? 

  • 0 in reply to Xianrong Shang
    TI__Mastermind 18900 points

    Hey Xianrong Shang,

    The DRV10987 does not have this false lock problem.

    However, I will say that the DRV10987 might have a hard time obtaining this RPM. According to the electrical frequency to RPM equation, this RPM seems to be outside the capabilities of our device:

    f_electrical = RPM*(n_poles)/120

    f_electrical = (10800)*(12)/120

    f_electrical = 1080Hz

    As shown by out table below, the max electrical frequency (operating closed loop speed) our device can output is 1000Hz. Using the equation, 1000Hz will put the motor at 10,000RPM.

    Unfortunately, TI does not have a suitable replacement to obtain this high RPM. However, I encourage you to check if this driver might be suitable for you.

    Best,

    -Cole

  • 0 in reply to Cole Macias
    Intellectual 855 points

    Hi Cole,

      So,  only the DRV11873 can reach 10800RPM. But I refer to the recommendation of slva778, the false lock still existed. Why?

    the PWM is 12KHz, the Vcc is 12V, the FG is 900Hz, the duty cycle is about 55%, the motor still grind to a halt at 9200RPM.

    Below is Today's debug, I can't make the motor 10800RPM: 

    1. The PWM frequency, tried 12 KHZ, 24 KHZ, 50 KHZ, adjustable and duty from 6.2% ~ 6.2%, increasing/decreasing the amount of 0.15%, increasing/decreasing the amount of 1%, increasing/decreasing the amount of 3%, to 60 ~ 66%, the speed is about 9000 ~ 9800 RPM, and increasing, decreasing duty, RD = 1 will appear intermittently, motor when the fast slowly, when duty gradually up to 90%, the issue has been occurred.
    2. The motor stops, VCC = 12.01 V constant; noise peak-peak 50 mV
    Motor turns, normal VCC = 11.96 V constant , noise peak-peak 50mV,  Will the noise affect the speed?

    In addition, the motor turns off and on, it's abnormally. and the noise peak-peak 130 mV, VCC is11.92 V.

  • 0 in reply to Xianrong Shang
    TI__Mastermind 18900 points

    Hello Xianrong Shang,

    Apologizes for the late reply. Unfortunately, it sounds like your test results lead to fine tuning of the system.

    To give context into what is happening, RD is a pin that indicates the motor is blocked or stopped with an external force when it is high, as you've previous mentioned. It monitors FG and see if it toggles or remains constant at a frequency of 101.6 Hz +/- 30%. As a result, if the device gets the motor to a certain RPM and the triggers RD when the motor is not blocked or stopped by an external force, then the device is triggering the false lock.

    I've restated this is information because the duty cycle and VCC levels have been adjusted and are still causing the RD to go high when the motor is not blocked. This indicates a false lock. The only way to fix a false lock is to continue to adjust the duty cycle or VCC. I apologize because I know this is not ideal and I apologize that we don't have any other drivers that meet your needs (yet).

    To answer your question, changing VCC levels will affect speed. This is because VCC is how the driver delivers current to the motor. Thinking about a R,L, and V_BEMF motor model, more applied voltage (VCC) will mean more current.

    However, millivolts are probably not enough to change the RPM of the motor significantly. This depends on your system's requirements. If you are curious, monitor FG and see how changing VCC from 12.01V to 11.96V affects the electrical frequency of the system. 

    Best,

    -Cole

  • 0 in reply to Cole Macias
    Intellectual 855 points

    Hi Cole,

       According to your reply, it's difficult to tune the speed from 9000RPM to 10800RPM. The suggest method is not feasible: the PWM frequency, the duty cycle, the Vcc, because the motor would be false lockeas long as more than 9000RPM. 

  • 0 in reply to Xianrong Shang
    TI__Mastermind 18900 points

    Hey Xianrong,

    I apologize that it is not possible to fix this problem. The false lock issue is a very well known problem that has causes our customers a lot of grief. I apologize for the inconvenience. Hopefully, this evaluation has shown that this part is not a good fit for your application. I hope you find a replacement part that fits your requirements, even if it is not with TI.

    Thanks,

    -Cole