• State Resolved
  • Locked Locked
  • Replies 29 replies
  • Answers 2 answers
  • Subscribers 32 subscribers
  • Views 1222 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.

MCT8329EVM: is this a normal early failure rate issue or a reliability issue or user issue

eric maundu
Prodigy 20 points
Part Number: MCT8329EVM

I got a mct8329 evaluation dev kit

I installed the application, connected the bldc motor, usb and power (48vdc) as per documentation.

when I tried to spin the motor by rotating the variable resistor ... it would start in a jerky manner before picking up speed then it would immediate(~5sec) stop with a fault error indicating that the motor was missing.

resetting the alarms on the application would not help restart the motor. the only way would be to turn off the 48vdc supply, disconnect the usb and pretty much restart the application to connect to the dev board again.

after a couple attempts at this, something went bang on the board and the mct8329 motor drive chip - went up in smoke/fire. the output stages of the board did not seem to suffer 

I tested the motor using another driver and it works ok.

Is this a normal early failure rate issue, a reliability issue or could I have done something wrong - there are only 5 connection points - the usb, the 3 motor wires and the motor dc supply,

  • 0
    TI__Mastermind 23870 points

    Hi Eric,

    Likely what occurred here is the voltage at some pin, likely the PVDD pin or GCTRL pin based on the location of the damage, was experiencing absolute max violations which over time led to the failure that you saw. 

    To check which pin failed you can check if any of the pins have internally become shorted to GND.

    If you still experiencing the NO_MTR fault after replacing the MCT8329 please try going though the steps outlined in the MCT8329 GUI:

    Regards,

    Joshua

  • 0 in reply to Joshua Horlander Cruz
    Prodigy 20 points

    hey Joshua,

    thank you for your message.

    I was looking at the evaluation board ... it seems like it has physical damage - this may sound silly, but how can I successfully remove and replace the chip 

  • 0 in reply to eric maundu
    TI__Mastermind 23870 points

    Hi Eric,

    I misunderstood, when I read you tested the motor using another driver I thought you had replaced the MCT8329 on the damaged board and it was able to spin the motor. I agree that there appears to be physical damage to the board. Instead of replacing the chip a new EVM will need to be used.

    To try and determine what occurred to this damaged board can you measure the pins of the MCF8329 chip and see if any of the pins are shorted to ground?

    Regards,

    Joshua

  • 0 in reply to Joshua Horlander Cruz
    Prodigy 20 points

    hey Joshua,

    I cannot find pins in the fiery side they vaporized 

  • 0 in reply to eric maundu
    TI__Mastermind 23870 points

    Hi Eric,

    Since the board and device was severely damaged trying to discover the reason for the damage to the device will be difficult. I would suggest getting another EVM and measuring the Voltage and Current of PVDD and the voltage of CPL and GCTRL when you attempt to start the motor.

    Regards,

    Joshua

  • 0 in reply to Joshua Horlander Cruz
    Prodigy 20 points

    what do I have to do to see if I can get a replacement board Joshua

  • 0 in reply to eric maundu
    TI__Mastermind 23870 points

    Hi Eric,

    Unfortunately we cannot provide a replacement board when the board damage was not caused by EVMs design.

    Regards,

    Joshua

  • 0 in reply to Joshua Horlander Cruz
    Prodigy 20 points

    depressing ... as i never did anything .. all i did was connect the three leads of the motor, the power and usb ... and tried to run it using the onboard variable resistor as per doc's ... it jerked around and fried .... if it was an output stage failure .. i would understand being responsible .. but the chip in the middle of the dev' board board ... so uggh - how i am supposed to trust this for the applications/products  we are trying to develop for 

  • 0 in reply to Joshua Horlander Cruz
    Prodigy 20 points

    i don't understand .... measure them then what ...

  • 0 in reply to eric maundu
    TI__Mastermind 23870 points

    Hi Eric,

    In the measurement of voltage and current of PVDD and the voltage of CPL and GCTRL, are you able to see any abs max violations on these pins or high current draw from PVDD when attempting to start the motor? Also are you still getting the NO_MTR fault with the new EVM board?

    Regards,

    Joshua

  • 0 in reply to Joshua Horlander Cruz
    Prodigy 20 points

    the last time i connected my motor to this as shown in the image and powered it up ... the motor jerked a bit when i rotated the potential meter and the board went up in smoke ... i have not powered anything yet, i dont want to repeat my experience. so i am trying to understand the doc's, what you mean and implications before i power it around this time. i did not understand your earlier notes .... where is cpl on the board. what are the absolute max violations i should watch out for  

  • 0 in reply to eric maundu
    TI__Mastermind 23870 points

    Hi Eric,

    You can find the absolute maximum ratings for our pins in section 6.1 of the datasheet if you measure a voltage at any of these pins that violates these then the device will likely become damaged and we cannot guarantee that the device will operate as defined in the datasheet.

    I was asking for you to look at this so we could determine if abs max violations were occurring which could have lead to damage to the chip and eventually the board damage that you saw. 

    The PVDD current was to determine if the current draw was higher than expected which would help indicate that the board is damaged.

    the CPL pin and trace is highlighted in the image below:

    I understand that you do not want to attempt to spin the motor without some confidence that the board will not brake to prevent this board from also becoming damaged. So to start, please make sure the potentiometer is fully clock wise rotated and/or that the break switch is flipped down so that the motor down no attempt to spin at start up:

    Then supply power to the device and please measure the voltage and current of PVDD and the voltage of CPL and GCTRL and provide me the measured values.

    Regards,

    Joshua

  • 0 in reply to Joshua Horlander Cruz
    Prodigy 20 points

    this time i am getting a a mtr_lck ... related to an abn_speed  when i read all registers.

    the potentiometer does not move the motor ... 

    however clearing the fault status moves the spins the motor briefly (but the flaut status remains)

    only a usb disconnect and application refresh removes the fault status

    i dont see any significant increase from the values below ~1v when the motor spins when i clear alerts

    motor current went to about 1A

    pvdd - 14v

    gctrl - 5.33v

    cpl - 0.01c

    i am currently using a 36v motor at 14v  ... i was able to move it with a different speed controller

  • 0 in reply to eric maundu
    TI__Mastermind 23870 points

    Hi Eric,

    The measured current and voltage values look good.

    Can you download the current register configuration and provide me the .json/.txt file? I want to check the register settings to see why the motor is spinning independent of the potentiometer. One item to note, SW1 must be set to the left or the motor will be provided a 3.3V signal and not the potentiometer setting.

    To try and resolve the ABN_SPEED fault please try going though the steps outlined in the fault handling section of the GUI:

    Regards,

    Joshua

  • 0 in reply to Joshua Horlander Cruz
    Prodigy 20 points

    i tried going through the abn_speed fault steps yesterday without any luck ... anything in particular i need to try

  • 0 in reply to eric maundu
    Prodigy 20 points

    also ... i saw your feed back on the other post ... my question was not  answered ... i asked ... was it a early failure, a reliability issue or a user operator error ....  .... i need to make a report so we can decide if this is an appropriate device for our needs or not ... i have evaluated other ti products without them going up on smoke as soon as i power them ... i get it you are helping me with my second board ... but why did the chip fry and why will it not happen again are my concerns

  • 0 in reply to eric maundu
    TI__Mastermind 23870 points

    Hi Eric,

    If you have followed the steps in the GUI to attempt to resolve the ABN_SPEED fault then please send me your register configuration and provide me the motor specifications for the motor you are using.

    eric maundu said:
    was it a early failure, a reliability issue or a user operator error

    This was most likely user operator error. We test our EVM boards and chips before they are shipped to customers. I could not determine the reason for the board burning as the device experienced damage to the pins and the pins could no longer be measured. I suspect that the NO_MOTOR fault was the root of the problem but I cannot say for sure since data cannot be collected.

    eric maundu said:
    why will it not happen again

    I cannot guarantee that this will not happen again, but making sure that the abs max values are avoided will help to prevent future damage to the board and chip.

    Regards,

    Joshua

  • 0 in reply to Joshua Horlander Cruz
    Prodigy 20 points

    Fullscreen regs.json.txt Download 
    {
  "signature": "oneui-register-data",
  "data": [
    [
      {
        "idx": 0,
        "id": "isd_config",
        "value": "0x7F404D02"
      },
      {
        "idx": 1,
        "id": "motor_startup1",
        "value": "0x6C9265B7"
      },
      {
        "idx": 2,
        "id": "motor_startup2",
        "value": "0x332D3295"
      },
      {
        "idx": 3,
        "id": "closed_loop1",
        "value": "0x1025E600"
      },
      {
        "idx": 4,
        "id": "closed_loop2",
        "value": "0x02A1A428"
      },
      {
        "idx": 5,
        "id": "closed_loop3",
        "value": "0x34C92001"
      },
      {
        "idx": 6,
        "id": "closed_loop4",
        "value": "0x001AC953"
      },
      {
        "idx": 7,
        "id": "const_speed",
        "value": "0x30010000"
      },
      {
        "idx": 8,
        "id": "const_pwr",
        "value": "0x1F400640"
      },
      {
        "idx": 9,
        "id": "150_deg_two_ph_profile",
        "value": "0x24DB7200"
      },
      {
        "idx": 10,
        "id": "150_deg_three_ph_profile",
        "value": "0x48DB6946"
      },
      {
        "idx": 11,
        "id": "ref_profiles1",
        "value": "0x032184A6"
      },
      {
        "idx": 12,
        "id": "ref_profiles2",
        "value": "0x34C80AFC"
      },
      {
        "idx": 13,
        "id": "ref_profiles3",
        "value": "0x4787D70C"
      },
      {
        "idx": 14,
        "id": "ref_profiles4",
        "value": "0x000C9932"
      },
      {
        "idx": 15,
        "id": "ref_profiles5",
        "value": "0x1F6BF200"
      },
      {
        "idx": 16,
        "id": "ref_profiles6",
        "value": "0x00708000"
      }
    ],
    [
      {
        "idx": 0,
        "id": "algo_ctrl1",
        "value": "0x00000000"
      },
      {
        "idx": 1,
        "id": "device_ctrl",
        "value": "0x00000000"
      }
    ],
    [
      {
        "idx": 0,
        "id": "ana_trim3",
        "value": "0x48004800"
      },
      {
        "idx": 1,
        "id": "ana_trim4",
        "value": "0x4800000000"
      },
      {
        "idx": 2,
        "id": "ana_trim5",
        "value": "0x00000048"
      },
      {
        "idx": 3,
        "id": "ana_trim6",
        "value": "0x00000000"
      },
      {
        "idx": 4,
        "id": "ana_trim7",
        "value": "0x00E9B8DB"
      },
      {
        "idx": 5,
        "id": "ana_trim8",
        "value": "0x000046E5"
      },
      {
        "idx": 6,
        "id": "ana_trim9",
        "value": "0x00A197DC"
      },
      {
        "idx": 7,
        "id": "ana_trim10",
        "value": "0x53790F3E"
      }
    ],
    [
      {
        "idx": 0,
        "id": "fault_config1",
        "value": "0x71783604"
      },
      {
        "idx": 1,
        "id": "fault_config2",
        "value": "0x7C85C00A"
      }
    ],
    [
      {
        "idx": 0,
        "id": "gate_driver_fault_status",
        "value": "0x00000000"
      },
      {
        "idx": 1,
        "id": "controller_fault_status",
        "value": "0x80900000"
      }
    ],
    [
      {
        "idx": 0,
        "id": "gd_config1",
        "value": "0x000600FC"
      },
      {
        "idx": 1,
        "id": "gd_config2",
        "value": "0x00000000"
      }
    ],
    [
      {
        "idx": 0,
        "id": "pin_config1",
        "value": "0x20600002"
      },
      {
        "idx": 1,
        "id": "pin_config2",
        "value": "0x3E0020AA"
      },
      {
        "idx": 2,
        "id": "device_config",
        "value": "0x27100008"
      }
    ],
    [
      {
        "idx": 0,
        "id": "sys_status1",
        "value": "0x008BFFFE"
      },
      {
        "idx": 1,
        "id": "sys_status2",
        "value": "0x80010000"
      },
      {
        "idx": 2,
        "id": "sys_status3",
        "value": "0x00000000"
      }
    ]
  ]
}

  • 0 in reply to eric maundu
    Prodigy 20 points

    changing the var's on the abn_speed ... may change the speed or duration .. but it stops after a short while (like sec's) with an abn_speed error

  • 0 in reply to eric maundu
    TI__Mastermind 23870 points

    Hi Eric,

    I will need some time to look over the register settings. I will aim to get back with you by 10/13 at the latest.

    Regards,

    Joshua

  • 0 in reply to Joshua Horlander Cruz
    Prodigy 20 points

    thanks appreciate ... getting the hang of it ... i am ok for now ...

  • 0 in reply to Joshua Horlander Cruz
    Prodigy 40 points

    I would like to mention that I have encountered the same failure mode as described in the original post. My board looks strikingly similar to the images of the failed board shown earlier. I was operating with a 48V PVDD, while following the configuration outlined in the quick start guide, which is within the specified rating.

    Upon examining my board, I noticed that the short occurred between Pins 5 and 4 of the MCT8239 (or their pcb traces), which represents PVDD and GND respectively. The spacing between these pins is only 0.2mm, whereas the IPC9592 standard recommends a 0.58mm spacing for a 48V system. Given that this module is rated for 60V, I would appreciate some assistance in understanding how the EVM (and IC) spacing is deemed suitable for the rated voltage. I would like to use this dev board but don't want to order another one if this is going to happen again.

  • 0 in reply to Nathan Hills
    Prodigy 20 points

    I too was testing a 48v bldc motor, following along the evm instructions when the mct839 chip went up in smoke for really no reason i can explain.

  • 0 in reply to eric maundu
    TI__Mastermind 23870 points

    Hi Eric,

    I will reach out to the engineer who designed this board and look deeper into problems with operating the EVM at 48V. I will aim to respond by 10/18 on this.

    Regards,

    Joshua

  • 0 in reply to Joshua Horlander Cruz
    TI__Mastermind 23870 points

    Hi Eric and Nathan,

    After aligning internally and preforming some testing it was determined that there is an error with the MCT8329 board which when operated above 48V will lead to the board damage that the both of you saw. The damage was caused by the voltage on the GCTRL pin exceeding its abs max values due to having VREG float which lead to the gate to drain capacitance of MOSFET Q7 pulling up the GCTRL pin past the abs max voltage when PVDD exceeded 48V.

    The solution for this issue is to either de-pop R56 or move the jumper J12 to the far right (closest to the nFault LED D2).

    We are working now to address this issue and will provide you both with a free board once this issue has been corrected on the boards.

    Regards,

    Joshua

  • 0 in reply to Joshua Horlander Cruz
    Prodigy 40 points

    Joshua,

    Thank you digging into this. Could you please send me a new board ASAP? It can be the same version -- I don't mind moving J12 jumper.

  • 0 in reply to Nathan Hills
    TI__Mastermind 23870 points

    Hi Nathan,

    I have sent you an email requesting a shipping address so we can get you a new EVM. This EVM will be the non-corrected version as they are not ready yet.

    Regards,

    Joshua

  • 0 in reply to Joshua Horlander Cruz
    TI__Mastermind 23870 points

    Hello Eric,

    I have sent you an email requesting a shipping address so we can provide you a corrected EVM.

    Regards,

    Joshua