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.

TPS92518: TPS92518 Eval kit getting damaged with low current sense resistor value

Part Number: TPS92518

I am using TPS92518EVAL kit with below configuration

1) Vin=24V

2) Load resistance is 8 ohms.

2) Current sense changed to 0.075 ohms (On original kit its 0.15 ohm) to increase output current

3) Duty cycle applied to PWM2 input on Eval kit.

4) No SPI configuration is used, so default LEDx _PKTH_DAC register value is 127.

5) Expected current is 127/(1000*0.075)= 1.69A. It is observed that as soon as PWM duty cycle (any value) is applied to input the VLED1 output is damaged. The output resistance is 5 ohms. The external diode D3 is OK, so IC output pin is getting shorted to ground.

What can be reason for this problem. Is it related to below specification?

 In my solution I am not using SPI control to adjust LEDx _PKTH_DAC register to 255. So how to set 1.7A current on Eval kit?

  • Hello Guruprasad,

    What resistor are you using for the load?  Is it a wirewound resistor?  Check the section of the section of the datasheet that shows diodes from the output to Vin to protect the TPS92518 from load disconnect and inductive load characteristics.  Section 8.3.3.1 on page 16 of the datasheet.

    I assume you are doing this:  "Device enable. If not configured as under voltage lock out or enable, tie to VCCx. Tie to >23.6V to bypass SPI
    communication and enable default register values." 

    5)  The output resistance is 5 ohms, is this with or without the load?  Changing the current sense resistor to 1/2 the value and running LEDx_PKTH_DAC at 127 should be okay since it returns the peak current back to the EVM maximum when LEDx_PKTH_DAC is set to 255.

    Is the MOSFET damaged?  Any other circuit damage?

    Do you have an oscilloscope to look at waveforms such as the switch node?

    The EN bypass to not use SPI is only for default values (one setting).  To change LEDxPKTH_DAC to 255 SPI communication is needed. 

    Note that if you plan on running higher current than the EVM is designed for the design should be calculated and components checked such as the inductor and power components to make sure it can handle the higher current.

    Were there any other modifications to the EVM besides the current sense resistor?

    Best Regards,

  • Please find below response

    What resistor are you using for the load?  Is it a wirewound resistor?  Check the section of the section of the datasheet that shows diodes from the output to Vin to protect the TPS92518 from load disconnect and inductive load characteristics.  Section 8.3.3.1 on page 16 of the datasheet.   -> I am using wire wound resistor

    I assume you are doing this:  "Device enable. If not configured as under voltage lock out or enable, tie to VCCx. Tie to >23.6V to bypass SPI
    communication and enable default register values." -> On kit resistor R5 is shorted, and EN pin I tied  to 24V externally. So VCC1 (output pin of IC) is tied to 24V supply this may be reason for IC failing at higher current. I shall remove remove R5 and check again

    The output resistance is 5 ohms, is this with or without the load?  Changing the current sense resistor to 1/2 the value and running LEDx_PKTH_DAC at 127 should be okay since it returns the peak current back to the EVM maximum when LEDx_PKTH_DAC is set to 255. -> This is without load, this indicates that some damage is there on channel.

    Is the MOSFET damaged?  Any other circuit damage? -> I will check

     

    Note that if you plan on running higher current than the EVM is designed for the design should be calculated and components checked such as the inductor and power components to make sure it can handle the higher current. -> To run kit at high current I need to change inductor only. The existing inductor on kit shall work upto 2 A current. Rest components like schottky diode and MOSFET have sufficient margin. I checked with other inductor with higher rating but same issue observed.

  • Hello Guruprasad,

    If VCC was pulled up to 24V that most likely caused TPS92518 damage.

    The TPS92518 is intended for driving LEDs, inductive loads need to have the added protection as stated above.  It would be better to run a non-inductive load to prevent issue if that can be done.  Also keeping the output capacitors will help with this (not sure if you planned on removing them).

    Best Regards,