• Not Answered

problems with drv8818

Recently I designed a circuit board with a drv8818 chip.  I am attaching a document discussing the details of the layout and configuration.  I am trying to manually toggle the step bit to move the motor.  When toggling the bit the motor does not rotate.  My next step is to replace the chip because I possibly have a bad chip.  However I was curious if anyone had any ideas on some thing that is configured incorrectly or methods for determining if the chip has gone bad.


Thanks,
Brian

drv88818 configuration document.pdf

7 Replies

  • Hi Brian,

    Everything looks like it should be working. Are there two different grounds such as Power GND and Logic GND? That wasn't clear.

    Best regards,

    Jose Quinones

  • In reply to Jose Quinones:

    This board is primarily an analog board.  It will contain four motor drivers.  Currently I am testing this board through the header pins by manually toggling the STEP pin with a jumper wire.  The digital ground will be tied together on the micro processor board that will connect through the header in the future. 

    Is there a way to verify that the chip is working?  My working theory is that I have a bad chip.  Since it's a little more difficult to remove this chip I was wondering if there is a way to confirm this theory.

    Thanks,
    Brian

  • In reply to Brian Fast:

    Hi Brian,

    There are different ways to make sure the device is working. Setting signals manually may  make pre process cumbersome, but here is what I would test by using a scope:

    1. If the device is enabled, you should see the outputs toggling as they attempt to regulate current across the chip. If you see no output activity, then either the device is not enabled or is damaged. It is hard to believe both H Bridges will be damaged, so make sure you check the four outputs.

    2. If the device is working properly, you should be able to see the charge pump voltage being VM + 10V (can be as high as VM + 12V). Check the VCP voltage and if it is higher than VM, then the charge pump is working. BTW, if the charge pump is not working, then it could be that the device is in SLEEP mode, or the charge pump is damaged. If the charge pump is damaged, then none of the outputs will work.

    3. As you toggle the STEP input, you should be able to see transitions on the nHOME output. In the case of full step (as configured) you should see an nHOME assertion every four successful steps. If you don't see the nHOME being asserted periodically, then either the device is on SLEEP, on RESET or the STEP inputs are not being accepted. Have under consideration that a STEP input must have a duration as specified on the datasheet. If the STEP pulse is not long enough, the rising edge is discarded and the STEP command will not take place.

    Since you are seeing the motor is easy to rotate with your hand, this seems to indicate the H Bridges are not enabled. It can be anything (SLEEP, RESET, Disabled, defective). So first thing is to make sure the device is getting enabled. Are you seeing the same problem on the four chips?  

    One last note. Your VREF is 3.3V which implies maximum current. It is hard to see what your SENSE resistor is, but it is very possible you are pushing too much current through this device in which case you will start seeing Thermsl Shutdown Protection and Over Current Protection taking place at all times. I would check on this as well.

    Hope the info helps. Best regards,

    Jose Quinones

  • In reply to Jose Quinones:

    Jose, Thank you for your feedback.  Based on your feedback I did some further testing.

    - First I went and populated a second board, with a new DRV8818, to make sure I wasn't dealing with a chip issue.  I am only populating 1 of the 4 motor drivers, at this point, to verify functionality.

    - I went ahead and decreased the current limits.  I am using a Vref = 0.25V with a 0.22 ohm resistor.   I also tried a Vref = 1V

    - I also decreased the Decay voltage to 0.25V and 1V as modified above (I only have one pot so this is currently tied to Vref)

    - I changed the RCx resistor and capacitor values to 47kohm and 1000pF which are between the recommended limits. 

    - The remainder of the I/O configuration remained the same as stated above.

    - I verified the charge pump is working by measuring VCP and it was 12V greater than VM.

    - Before testing anything else I verified the voltage of RCA and RCB.  There was a DC voltage of 900mV on each of the signals.  However neither signal had a transient as indicated on page 10 of the DRV8818 specification.  I am assuming if the timing circuit isn't oscillating then the chip isn't configured properly?  Is this true?  What would keep RCx from transitioning as seen in the specification. 

    - Do you have any more suggestions?

    - The motor still doesn't rotate one I toggle the step pin.  I am manually toggling the pin to Vcc and I am leaving it high for multiple seconds.

    Thanks,
    Brian

  • In reply to Brian Fast:

    Hi Brian,

    Are you continuously toggling the STEP input or just once? The STEP input works by taking rising edges and commutating steps henceforth. You will get one step/microstep every time a STEP rising edge is detected.

    Best thing is to put a square wave into this input. Attempting to do this by hand is not going to give you the results you require.

    If you do not see activity on the RC pins that means you never reached an ITRIP. What motor are you using?

    Best regards,

    Jose Quinones

  • In reply to Jose Quinones:

    I am toggling the step input multiple times.  I toggled it 100 times and saw no rotations on the rotor.  I am also monitoring the voltage to insure a good connection is made.  Since it has a pulldown resistor I toggled this by connecting and disconnecting the wire to Vcc.  I have been manually toggling the pin for convenience right now. The next step will include connecting a micro processor to precisely control the steps.

    The motor connected to the circuit is found at http://www.interinar.com/public_docs/23KM-K723-23V.pdf

    Motor Pin         Board Connection
    Pin 6                  A1 (Orange)
    Pin 4                     A2 (Red)
    Pin 3                     B1 (Blue)
    Pin 1                   B2 (Yellow)

    I am assuming even if I'm not toggling the STEP bit there should be some sort of holding torque to keep the rotor from spinning.  Thus I am assuming that there is a problem that's causing the RCx pin from transitioning. 

    Thanks,
    Brian

  • In reply to Brian Fast:

    I am still struggling to get this board functioning.  I was wondering if you had any other suggestions.

    At this point I am just trying to have a holding torque.  Basically I just want to have the motor driver hold the motor shaft at a single position.  I also purchased the drv8818evm to have a system to compare against.  The motor I purchased works well with the evm.  I have configured the inputs on my board to mimic the evm however my board is not working. The main difference is the signals on RCA and RCB.

    When I monitor the RCA and RCB voltages, on my board, with a scope the value is always 0.9V.  There is no waveform on the signals.  If the chip is working and enabled then the saw tooth wave should be present. 

    Other things:

    I have verified the charge pump is working by monitoring the voltage which is 12V greater than Vm.

    I have verified the Sleep = 5V, Enable = 0V, Reset = 3.3, USM0 = 0V, USM1 = 0V, SR = 0V, Decay = 0V, Vref = 0-0.25V.

    Timing circuit is using 1000pF and 38kOhm.