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.

DRV8825 microstepping large errors

Other Parts Discussed in Thread: DRV8825, DRV8880, DRV8886, DRV8886AT

I have been doing some testing on a DRV8825 Pololu module from China. I have found that the unit gets very hot and has very poor microstepping as compared to the A4988, I was wondering if this was expected or if it might be a knock off chip in these modules?

Specifically on the microstepping I have connected an encoder to the motor and monitored the motor angle and found that the 16x microstepping causes motor to jump forward or lag behind 2-4 microsteps.  I have tried increasing current and when I do the motor starts humming and getting really hot, but the performance does not improve. 

When I replace the module with an A4988 module the error is less than 1 microstep with lower current and no noise.   I had expected that the TI chip would far exceed the A4988 chip and hence wanted to see if this is expected behavior of the DRV8825?

  • Hi Trampas,

    What decay mode setting are you using? Have you tried fast decay?

    The two devices have different current regulation, blanking times, and pwm frequencies.

    Please consider the DRV8880. This device has additional current regulation setttings and adjustable off time.
  • The decay mode is set to the mixed (floating pin). 

    Here is what I am seeing with 16x microstepping. 

  • Hi Trampas,

    Can you confirm this lag or jump is occurring as the current in one winding approaches zero or increases from zero? The blanking time may account for this. The blanking time of 4us typical can inject more current than desired at lower current settings.

    The previously mentioned DRV8880 adjusts the blanking time based on the percentage of current.
  • Hi Rick,

    I have just fallen into the same trap...

    Can't you withdraw the Figure8/9/10 in the datasheet -- they are surely not taken with a DRV8825 -- and make a remark, that microstepping is only possible in fast-decay mode (at the cost of higher current-ripple) ?

    Thx

    Martin

  • Hi Martin,

    The figures were taken with the DRV8825. There are several factors that determine the ability to control the current.

    The motor inductance, resistance, step speed, and system voltage can affect the current control. As the system voltages have been increased to increase motor speed, using slow decay for increasing steps has been more difficult.
  • Hi Martin,

    Sorry, I hit reply too soon.

    Our newer steppers have the ability to better regulate current across various motors and voltages. Please look at the DRV8886 and DRV8886AT. These devices also have internal current sensing to eliminate the need for current shunt resistors.

    The DRV8886 and DRV8886AT are not pin to pin compatible with the DRV8825.
  • Hi Rick,

    pls can you tell me the conditions of your setup?

    Mine are: 24V, 2,8Ohm/4,2mH, and I'm getting the following results:

    Even if I reduce the input-voltage to 10V, slow-decay isn't working:

    Switching to fast-decay solves the prob (w/ very high current-ripple)

    And yes, I suppose the DRV8880/DRV8886 would have been a better choice...

    Regards

    Martin

  • Hi Martin,

    I don't have the motor parameters available. The motor is 1.8 degrees per full step.

    From the datasheet image, the motor speed is much higher (3200 steps/second at 1/8 steps = 120 RPM, 2 revolutions/sec).

    The image you provide shows the motor running at ~1/3 revolution per minute. Is that correct?
  • Hi Rick,
    correct, the motor is turning VERY SLOW. But the situation does not really change a lot at higher speeds (until the back emf enters the game...)
    Regards
    Martin