• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 31 subscribers
  • Views 1984 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.

LDC1000: Not responding over SPI

Nishant Pol
Prodigy 170 points
Other Parts Discussed in Thread: TLV1117

I have the LDC1000 mounted on a custom PCB with a 2" coil.  The SPI pins are brought out on an IDC header.  I drive the SPI bus using a Sparkfun Bus Pirate (3.3V logic).  The logic side of the LDC1000 is powered by 3.3V, and the analog side of the LDC1000 is powered by 5V, both from regulators (TLV1117) on the PCB, which have a 6V input from a bench power supply.  The power output looks clean on the scope, and at the correct voltage. 


I was testing these boards for a few weeks, and was able to read the device ID, and valid proximity data.  However, between the last time I powered the system down, and this morning, when I first tried to test the sensors (using an MCU), something happened and now the LDC1000 does not respond.  Testing the the Bus Pirate shows that MOSI and CS are as expected, but MISO looks very odd:  I am sending 0x80 and 0x00 to read the device ID register (which should read 0x80, although the datasheet lists 0x84 and 0x80).

1: SCK, 2: MOSI

1: SCK, 2: MOSI (zoomed in)

1: SCK, 2: /CS

1: SCK, 2: MISO (failure 2)

1: SCK, 2: MISO (failure mode 2)

1: SCK, 2: MISO (failure mode 3)

Thanks in advance for any help you can give.  This is for a trash sorting robot due in 1.5 weeks, and oddly we are only seeing this now. 

  • 0
    TI__Genius 13650 points

    Hello Nishant,

    please ensure that you are completely within the SPI timing requirements that are specified in the LDC1000 datasheet. If all timing specifications are satisfied, please swap the LDC1000 with a different device to ensure that the IC has not been damaged. Also, I recommend keeping all cables between the Bus Pirate and the LDC1000 as short as possible. 

  • 0 in reply to Ben Kasemsadeh
    Prodigy 170 points

    I was initially using 1MHz SPI, then switched to 30kHz, but the same behavior persisted, so I suspect the IC's were damaged by ESD or from being backdriven (supply is off but SPI SCK, CS, and MOSI toggle). 


    I'm assuming that the SPI and VDD/VIO pins are protected from ESD per the datasheet specifications.  Are the coil pins (INA and INB) also protected?

  • 0 in reply to Nishant Pol
    TI__Genius 13650 points

    Hi Nishant,
    did you verify that the devices were damaged by replacing them with new ones?

    The ESD ratings are the same on all pins for this device.

  • 0 in reply to Ben Kasemsadeh
    Prodigy 170 points

    Hi Ben,


    Thanks for your help.  Unfortunately, after we replaced the IC's, we could not get reliable communication over SPI. 2/4 boards did not communicate, and 2/4 would send the device ID only occasionally.  We saw this issue before, and it is probably caused by manufacturing (we have limited SMD equipment, and install the IC using a hot air station, with solder on the footprint).  I am spinning a new board with the Q1 package which will hopefully fix both issues.

  • 0 in reply to Nishant Pol
    TI__Genius 13650 points
    Hi Nishant,
    additional to your proposal, I recommend verifying that it is not a signal integrity issue caused by long wires or timing violations from the host.
  • 0 in reply to Ben Kasemsadeh
    Prodigy 170 points
    The problem has been fixed with new PCBs that have the LDC1000Q1 IC along with an external ESD diode array on SPI and power pins. I am suspecting that the previous ICs were damaged by ESD (our university lab does not have any ESD handling equipment, and chairs are known sources for static), or manufacturing (using a hot air rework station to solder the IC). Thanks for your help.
  • 0 in reply to Nishant Pol
    TI__Genius 13650 points
    Hi Nishant,
    thank you for following up. I am glad you managed to resolve the issue.