• State Resolved
  • Locked Locked
  • Replies 25 replies
  • Subscribers 29 subscribers
  • Views 2513 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.

DRV8828EVM / not to reach desired speed

K.Yaita
Guru 29720 points
Other Parts Discussed in Thread: DRV8828, MSP430F2617
Hi Team,
 
My customer evaluates DRV8828EVM and have a question.
 
He set the GUI to values shown and repeated the following sequence(1→2→1→2・・・), he found sometimes stepping motor didn't reach "Desired Speed PPS" and then stayed "Start Speed PPS".
The recall factor of this behavior is about 60%.(12 times out of 20)
1) select "Start Steps"
2) select "Stepping"
 
 
When this issue occured,  to select "UpdateSpeed" help him to to reach desired speed.
He found the waveform of Phase pin that is input from MSP430 is different from normal one and this issue also occures even if motor isn't connected to DRV8828EVM.
 
Is it possible case of DRV8828EVM?
 
Best Regards,
Yaita/ Japan disty
  • 0
    TI__Mastermind 26065 points

    Hi Yaita

    Will check this on bench soon.

    Thanks

  • 0 in reply to Wilson Zuo
    TI__Mastermind 26065 points

    Hi Yaita

    I confirmed you finding. I think this is because of some bugs during the command  receiving process of the MCU firmware from the USB simulated COMx. So there will be some time the MCU didn't outputs correct "speed info" to DRV8828. But everything of  DRV8828 is working as expacted.

    We will run further correction of the firmware but may take sometime. 

    Thanks.

  • 0 in reply to Wilson Zuo
    Guru 29720 points
    Hi Wilson-san,
     
    I really appreciated for your support. I understood. 
    I'd like to ask additional question.
     
    (1)
    I also heard setting "Number Of Steps (Move Steps)" sometimes didn't work as expected.
     
    (2)
    It seems "PHASE x Control Signals" didn't work as expected too.
    For example)
     - Motor goes to operate even if ENABLE is unchecked.
     - Motor current isn't 0A even if I0-I4 are all unchecked(0%).
     
    Is these because of some bugs on EVM?
     
    Best Regards,
    Yaita
  • 0 in reply to K.Yaita
    TI__Mastermind 26065 points

    Hi Yaita

    (1)This should be the same reason of the command transfer.

    (2) The stepper running function of the lower side of the GUI will total over ride the upper side settings.

    You will find when you only press the start running of motor, the ENABLE is automatically ON, the VREF is set to 2.5V MAX, and Ix all are set to generate MAX current level. And at the running time, if you change anything in upper side of the GUI, the motor will stop. So maybe when the time we build this GUI (long time ago), the function definition is not planned so well.

    Sorry to bring you the confuse.

    Best Regards.

  • 0 in reply to Wilson Zuo
    Guru 29720 points
    Hi Wilson-san,
     
    Thank you for your great support.
    I understood.
     
    Best Regards,
    Yaita
  • 0 in reply to K.Yaita
    Guru 29720 points
    Hi Wilson-san,
     
    My customer asked me about schedule of modifying firmware(EVM).
    Do you have any plan?
     
    Best Regards,
    Yaita
  • 0 in reply to K.Yaita
    TI__Mastermind 26065 points

    Hi Yaita

    I will look into this but an official release of a new version really take times. 

    I will try to provide a revised version here for your customer after I get the issue fixed. Probably within two weeks. Is that make sense?

    Best Regards

  • 0 in reply to Wilson Zuo
    Guru 29720 points
    Hi Wilson-san,
     
    Yes, I will inform my customer accordingly.
    Your continuous support would be appreciated.
     
    Best Regards,
    Yaita
  • 0 in reply to K.Yaita
    Guru 29720 points
    Hi Wilson-san,
     
    Is there any update about revised version of firmware?
     
    If firmware is revised, how do I need to update after that?
    Could you send email to me if possible?
     
    Best Regards,
    Yaita
  • 0 in reply to K.Yaita
    TI__Mastermind 26065 points

    Hi Yaita

    I'm working on this but still not finished.

    After it was revised, I will attached the new firmware package here as well as send you an email.

    You can use CCS and the MSP-FET430UIF tools to download the new firmware into the MSP430F2617 MCU.

    Sorry for keeping you waiting. I will try finish it sooner.

    Thanks and best regards,

  • 0 in reply to Wilson Zuo
    Guru 29720 points
    Hi Wilson-san,
     
    Thank you for your continuous support.
    My customer has a hard time to evaluate DRV8828EVM.
    I will wait for your reply.
     
    Best Regards,
    Yaita 
  • 0 in reply to K.Yaita
    TI__Mastermind 26065 points

    Hi Yaita

    Glad to say the problem is just found. The issue should be caused by unproper enter and leave the LPM0 mode for the MCU. After disable entering into the LPM0 mode. It is function good and no exception bevavior on your clicking on GUI.

    I just attached the new CCS project here. 7367.DRV8828EVM_FIX.zip

    We will start the flow on the update of the website contents.

    Thanks of your pointing out the problem.

    Best regards,

  • 0 in reply to Wilson Zuo
    Guru 29720 points
    Hi Wilson-san,
     
    I really appreciate for your quick support.
    I will confirm your new CCS project.
     
    Best regards,
    Yaita
  • 0 in reply to K.Yaita
    Guru 29720 points
    Hi Wilson-san,
     
    I have firmware updated by my customer, however I heard the issue remain the same.
    So, could you send me the original CCS project?
    I'd like to confirm the difference between these.
     
    I also concern if update is completed in customer's EVM, Is there any method to confirm? 
     
    Best Regards,
    Yaita
  • 0 in reply to K.Yaita
    TI__Mastermind 26065 points

    Hi Yaita

    The original CCS project was attached. You can download the

    7367.DRV8828EVM_FIX.zip package. All the source files included.

    It is strange that customer found the issue the same. I really checked the performance repeatedly.

    It is possible that customer didn't really reflashed the firmware?

    Could you help to verify the new code? I will try to get another board to verify too.

    Best regards,

  • 0 in reply to Wilson Zuo
    Guru 29720 points
    Wilson-san,
     
    I'd like to confirm just in case.
    Is the file "7367.DRV8828EVM_FIX.zip" that you attatched at Jul 24 previous(original) CCS project?
    This file name seems the same as the file that you attatched at Jul 18.
     
    I'd like to compare new(revised) CCS project and previous(original) CCS project.
     
    Best Regards,
    Yaita
  • 0 in reply to K.Yaita
    Guru 29720 points
    Wilson-san,
     
    I confirmed the file that you attatched at Jul 18 and Jul 24 are the same content.
    Could you please give me the previous(original) CCS project?
     
    Best Regards,
    Yaita
  • 0 in reply to K.Yaita
    TI__Mastermind 26065 points

    Hi Yaita

    This is the original source code.

    8054.CPG006_RA_MSP430F2617_Firmware_LMD_TI.zip

    Best regards,

  • 0 in reply to Wilson Zuo
    TI__Mastermind 26065 points

    Hi Yaita

    I checked the revision firmware on another board. The result to me is fine. So I'm not sure whether the new firmware was correctly reflashed into the board at customer side. A MSP430-FET430UIF tool is must need to do the programming.

    Best regards,

  • 0 in reply to Wilson Zuo
    Guru 29720 points
    Hi Wilson-san,
     
    The reason is my customer didn't choose correct dialog in update.
    He confirmed the function is no problem for now.
     
    Thank you for your continuous support.
     
    Best Regards,
    Yaita
  • 0 in reply to K.Yaita
    Guru 29720 points
    Hi Wilson-san,
     
    My customer also pointed out the following.
    In "Current Control(VREF) Phase B " of GUI,  it seems that the value of (2) is inconsistent with (1), and (2) indicate incorrect value.
    I heard "Current Control(VREF) Phase A " is OK.
     
     
    Is it just a improper indication of GUI? 
     
    Best Regards,
    Yaita
  • 0 in reply to K.Yaita
    TI__Mastermind 26065 points

    Hi Yaita

    Thank you for the feedback. We will check where the issue comes from.

    Best Regards,

  • 0 in reply to Wilson Zuo
    Guru 29720 points
    Hi Wilson-san,
     
    In addition, it seems "Current Control (VREF) Phase x" slider doesn't have effect on actual VREF voltage of EVM.
    Does your EVM operate well?
     
    Best Regards,
    Yaita
     
  • 0 in reply to K.Yaita
    TI__Mastermind 26065 points

    Hi Yaita

    I checked the both the slider A and B are working correctly.

    But we should pay attention to our test about the upper side (PHASE and Curent Control function) and the bottom side (Microstepping Demo function) of the GUI.

    Usually we test the upper side function first. We can check the ENABLE, PHASE, Ix and Decay. Note that the VREF slider control the DAC outputs to the VREFx pins (you can check the voltage with simply a multimeter). Note that the real current will depend on the Ix, decay setting and the VREFx.

    Once we start the Microsteppering Demo control, all the ENABLE, PHASE, (Ix are exclude), Decay, VREFx are taken over by the DAC-VREFx-PHASE micro-stepping algorithm. In this algorithm (Please refer to http://www.ti.com/lit/an/slva443/slva443.pdf ), VREFx are fed with fixed range sinewave voltage from the DAC. When the motor is running and controlled by the bottom side GUI function, any change on the upper side will still affect the outputs (I should correct my last saying that they are independently). Some parameters are continously updating in the micro-stepping algorithm, such as the VREFx voltage and PHASE and DECAY, so your mannully change on the slider in upper GUI will only take a transient effect because it will be override soon. But some IO can disturb the microstepping totally, such as ENABLE and Ix.

    We are sorry that these mixed effects of upper side and bottom side GUI may cause lots of confuse to the verification. A good thing is that the VREFx range of the micro stepping algrithm is fixed (normally too big for the full scale current level). But we still can change the current level by changing the Ix setting of A and B together in the upper side GUI.

    I noticed your customer may get a lot of questions about the DRV8828EVM and GUI. I think they must be very interested with the EVM's micro stepping method. As I shown the link of slva443.pdf, this is a very good method to do high level micro stepping with our DRV8828/29/12/13/41. Also you can find http://www.ti.com/lit/an/slva641/slva641.pdf to optimize this method if higher performance is needed.

    We should said our DRV8828GUI is far from perfect to do the application job. We are keeping improving on all our new products, EVM, and GUI. Sorry for all the confusion. Your experience give us a lot feedback. Thank you!!

    Best regards,

  • 0 in reply to Wilson Zuo
    Guru 29720 points
    Hi Wilson-san,
     
    Thank you for your detail explanation.
    I will talk to my customer about this GUI function.
    I will contact you again if I am asked additional question, then please give me your advice.
     
    Thanks and Best Regards,
    Yaita