• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 301 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.

TLC6C5712-Q1: LED failure due to Electrical Over stress

Hemaraj Thirupathi
Prodigy 10 points
Part Number: TLC6C5712-Q1

Tool/software:

Hello Team, 

I am Hemaraj, working as a electronic engineer in Duratronics GmbH, Daun, Germany.

We have used TLC6C5712QPWPRQ1 IC in our automobile gear light indicator application

There is an failure occurred in field vehicle. one of the LED in the system was destroyed due to overcurrent ( limit above the LED max current 20mA)

i would like to understand what would be reason for this failure as we are driving thought the LED driver IC(TLC6C5712-Q1) which has current limit set through Dot correction bit.

Dot correction bit was set in the register for all channels to limit current to 10mA max

I would like to know the failure conditions which would allow its IC max current (75mA ) to destroy the LED.

Kindly extend your support. I will provide all relevant details related to the project during further discussion. 

  • 0
    TI__Genius 11635 points

    Hi Hemaraj,

    How do you know the LED was damaged due to overcurrent? Is is possible the damage was from some other fault such as high junction temperature or short to battery event?

    Have you verified the TLC6C5712-Q1 device operation through functional test? Or was this device damaged also?

    What current is measured from the TLC6C5712-Q1 channel? What are the register settings?

    We need a full schematic of the TLC6C5712-Q1 system and the LED load in order to determine if the failure could be related to an applications issue.

    Please address the questions above and provide the detailed schematics so we can help support.

    Thanks,

    Zach

  • 0 in reply to Zach Olson
    Prodigy 10 points

    Hello Zach , 

    Thanks for your reply. 

    Please note.

    LED was damaged due to overcurrent? Is it possible the damage was from some other fault such as high junction temperature or short to battery event?

    1).... Failure analysis report showed burnt mark on the LED semiconductor die. (FAR report available - no option to attach - require support)

    2).... No chance of Battery voltage or short circuit to LED contact. the LED is driven directly by the TI driver IC. (VCC - 5V is supplied)

    3).... No high Junction temperature as the product qualified the Thermal validations. 

    What current is measured from the TLC6C5712-Q1 channel? What are the register settings?

    1).....Hardware settings: 33K connected to Iref. 85 is set in Dot correction bit. so as per calculation 10 mA is set.

     We need a full schematic of the TLC6C5712-Q1 system and the LED load in order to determine if the failure could be related to an applications issue.

    1).....Available - no option to attach - require support.

  • 0 in reply to Hemaraj Thirupathi
    TI__Genius 11635 points

    Hi Hemaraj,

    Thanks for the additional information. To attach files, you can simply drag and drop them into the message.

    Hardware settings: 33K connected to Iref. 85 is set in Dot correction bit. so as per calculation 10 mA is set.

    Did you actually read the registers and measure the output current? or is this all "per calculation". We need to verify the register settings and current output by bench measurement for every channel.

    I would also reconsider your calculations. For 33-KΩ R(iref) resistor and dot correction of 85, I calculate a max current of 18.6mA and a channel current of 6.2mA with dot correction applied. (see attached calculations).

    TLC6C5712_Duratronics.xlsx

    For an R(iref) resistor of 33kΩ, it is not possible for the output current to be greater than ~18.6mA unless additional current is pulled out of the IREF pin. If this were the case, I would expect all LEDs to be damaged as the current for all channels would have increased due to the higher reference current.

    In this case, it appears the the single LED may have been at fault. However, we should verify the bench measurements of the TLC6C5712-Q1 to be sure. If the TLC6C5712-Q1 device is confirmed to be operating correctly, I suspect the LED to be at fault.

    Regards,

    Zach

  • 0 in reply to Zach Olson
    Prodigy 10 points

    Hello Zack, 

    Thanks for the excel calculation sheet.

    Sorry. There is a small mistake in my statement: Hardware settings: 21.5K connected to Iref. 85 is set in Dot correction bit. so as per calculation 10 mA is set.

    Here is the Schematic ( We have main PCB and Indicator PCB ). - 21.5K resistance is achieved by activating 33K(R18) resistor in parallel to 62K(R16) resistor

    . /cfs-file/__key/communityserver-discussions-components-files/196/Schematic_5F00_LED_5F00_failure_5F00_DURA_5F00_00.pdf

    Here is FAR report: 

    /cfs-file/__key/communityserver-discussions-components-files/196/FAR250587_5F00_r1.pdf

    Currently there are two LED failures. but i feel only one is relevant to EOS (Electrical Over stress) , which is having the burnt die. 

    Teh dealer have replaced with new the indicator PCB and it is working fine now. The TI LED driver is functioning. 

    Could you please check and provide your feedback? Is there in any event the DOT correction bit not considered for this particular channel? 

  • 0 in reply to Hemaraj Thirupathi
    TI__Genius 11635 points

    Hi Hemaraj,

    I have received the TLC6C5712-Q1 schematic. Can you also provide the schematic for the LED load board for the damaged LED?

    I notice that there is an NMOS switch that is used to change the reference current by adding resistors in parallel as discussed. This is not a typical configuration. Typically the reference resistor is placed close to the device, often with a 10nF capacitor in parallel to reduce noise. Is the NMOS and 33kΩ resistor located on the driver board close to the IREF pin? Or is this an off-board control?

    Can you provide a scope capture of the gate and drain voltage (pins 1 and 3) of the NMOS refenced to the GND pin (pin 28) of the TLC device during switching of the NMOS?

    Were you able to read the fault/status registers on the unit? What faults were reported? Is the REF_MASK bit enabled? if so, any reference current fault will not be reported to the fault register.

    Is there in any event the DOT correction bit not considered for this particular channel? 

    During certain error conditions, the dot correction can be set to the default value (00h) which sets the current to zero. From my understanding, there is no way for the channel to output full-scale current unless the dot correction register is changed. Are the registers locked using the LOCK_CORR command? This is the recommended way to prevent unintended modification of registers. See Register Protection section of the datasheet.

    Regards,

    Zach

  • 0 in reply to Zach Olson
    Prodigy 10 points

    Hello Zack, 

    1. Can you also provide the schematic for the LED load board for the damaged LED?

    --> the LED load board ( Indicator PCB ) is attached. we have only mechanical switch and LED in the indicator PCB. 

    The indicator board (LED) and Main PCB (TLC IC) are connected through cable harness. 

    /resized-image/__size/640x480/__key/communityserver-discussions-components-files/196/R16_5F00_R18_5F00_Placement_5F00_IC_5F00_M110.JPG

    2. Is the NMOS and 33kΩ resistor located on the driver board close to the IREF pin? Or is this an off-board control?

    ---> the resistors are close to driver (6 ~ 7 mm) and there are in same board.( not off-board control)

    /resized-image/__size/640x480/__key/communityserver-discussions-components-files/196/LED_5F00_Indicator_5F00_PCB_5F00_schematic.JPG. 

    3. Can you provide a scope capture of the gate and drain voltage (pins 1 and 3) of the NMOS refenced to the GND pin (pin 28) of the TLC device during switching of the NMOS?

    ->I will update next week (06.06.2025) regarding the scope capture of NMOS gate and drain.

    4. Were you able to read the fault/status registers on the unit? What faults were reported? Is the REF_MASK bit enabled? if so, any reference current fault will not be reported to the fault register.

    --> Unfortunately due to service dealer's mistake, logs are not captured. 

    5. Are the registers locked using the LOCK_CORR command? This is the recommended way to prevent unintended modification of registers.

    ---> Yes. Registers are locked to avoid unintended modifications.

  • 0 in reply to Hemaraj Thirupathi
    TI__Genius 11635 points

    Hi Hemaraj,

    Thanks for the additional information. From everything I have seen so far, I do not see any issue with the TLC6C5712-Q1 device. It seems likely that the LED itself is at fault, especially since you have seen other quality issues with these LEDs.

    If you want to perform more rigorous testing, you can monitor the output current, drain voltage, and gate voltage while switching the NMOS under various conditions to see if the current increases past the abs max of the diode. I would also poll the status registers to see if any faults are triggered during this testing.

    Feel free to post results from the test above if you find anything of note. Otherwise, I cannot provide much additional guidance as the TLC6C5712-Q1 device is functional and operating as expected.

    Regards,

    Zach