Mp6543hb - faults

Good Morning,

I have some questions concerning the faults that may occur during the operations of the driver MP6543HB (Hall sensors inputs).
We have purchased your evaluation board, and we’ve also mounted the driver IC on an electronic board prototype of the device that my company is developing.

We are currently using a series of evaluation boards for performing a stress test on some samples of the motor that we’re planning to use.
We are driving the motors using the PWM at 20kHz and a variable sinusoidal duty cycle at 5Hz frequency, with sine amplitude that linearly increases from 0 to 80% in 10 seconds and then stays constant. In this condition, the current consumption is 1A RMS with peaks of ca. 3A. The motors are driven for 90 seconds, then they are stopped and the drivers are disabled; after 30 seconds, the drivers are re-enabled and the cycle starts again.

Sometimes (and this happens randomly on the samples of the motor) it may occur that in the first few seconds of increment of the sinusoidal duty cycle amplitude, a motor stops spinning: the fault output of the driver is kept low; in the next cycle (since the driver is disabled and then re-enabled), the motor works normally.

  • For my understanding of the operations of your driver, this fault cannot be related to an overcurrent protection event: if such condition occurs, the outputs should be re-enabled automatically after 4ms, but this does not happen, as the motor does not restart; there cannot be a continuous overcurrent condition, as in this case I should see the nFault output toggling every 4ms, but it stays low. Moreover, the current consumption of the motor that I measure (peaks included) is lower than the minimum overcurrent protection threshold. This is true especially in the first linear increase of the duty cycle amplitude, but strangely this is where the problem may occur. It never occurs when the sine duty cycle amplitude reaches its maximum of 80% and the current consumptions are the highest. I’ve also tried forcing an overcurrent condition by applying a square wave duty cycle at 5Hz with 100% amplitude and I can indeed see the nFault output toggling.
  • It cannot be a thermal shutdown also; I measured the PCB temperature using a thermocouple and it is far less than the safe limits; the problem occurs in the first seconds of driving, when the system has rested for 30 seconds; it also occured at the beginning of the test, when the system was basically at room temperature; moreover, when the problem occurs, if I manually and immediately disable and re-enable the driver (using the manual switch on the evaluation board) the motor starts spinning and the problem does not occur anymore for the rest of the cycle.

What could be the cause of the problem? What is the fault condition that makes the nFault output stays low and the only way to recover is to reset the driver? Do you have any suggestions?

Thank you! Kind regards.

Any update?

Thank you. Kind regards.

I am also interested in an answer.
Bump.

We have the same problems, there has to be a solution. Like this the driver is not useable.

Hello,

the issue might be caused by noise on HA/HB/HC signals, which can cause an invalid combination of these inputs and generate a fault in the driver. It is suggested to add some small caps (around 100 pF) on each input to mitigate the effect of the noise.