LP5036: red LED stuck on after ESD

Ulrich Gerhardt

Part Number: LP5036
Other Parts Discussed in Thread: LP5030

Hi,

I do have similar problem with LP5036 / LP5030 LED driver like in post:

https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/969250/lp5036evm-lp5036evm-with-red-led-stuck-on-no-longer-able-to-talk-to-chip?tisearch=e2e-quicksearch&keymatch=LP5036%20esd

After precompliance testing of our product with ESD contact discharge (15kV) to a nearby metal part, the red LEDs are stuck on.
The metal part hast no direct connection to the pcb.
Communication to the LED driver via I2C is still possible. Also the red LEDs are on even after a reset or power cycle.

To one DUT it had happened that this failure was magically gone after about one week. After ESD testing it once more, the failure could be reproduced.

Can you assist us and give clues as to what the defect is and how the noise immunity could be improved.

over 1 year ago

0 Dave Yaeger over 1 year ago

TI__Intellectual 2925 points

Our experts for this issue are currently celebrating the Chinese national holiday, and they will respond next week.

0 Aaron Bing over 1 year ago

TI__Expert 4120 points

Hi Ulrich,

Is LP5030/6 used to drive all single-color red LEDs or there are green/blue LEDs also on the load side? The success of I2C communication confirms that the LED driver works fine, such that the faulty condition is from these red LEDs. Would you provide the PCB layout and detailed ESD contact sequence for reference?

Best Regards,

Aaron

0 Ulrich Gerhardt over 1 year ago in reply to Aaron Bing

Prodigy 10 points

Hi Aaron,

we are using 9 RGB LEDs (OSRAM LRTBGVSR).

Unfortunately we can't provide the PCB Layout, but I made a drawing of the joystick application we have.

We performed only one contact discharge with +15kV / 330pF, 2kOhm.
I don't think that the LED causes this fault, because i think the failure mode of the LED would be either working or defective. There would me no leakage current in the LED itself, which lights the red LED.
Also i did further investigations on the LP5030:

If I leave the IREF pin from LP5030 unconnected or short it to ground the brightness of the red LEDs don't change.
This looks to me like the current control circuit for the red LEDs is defective.
I also measured the voltage on IREF pin with the oscilloscope. The voltage on IREF has short interruptions. A working LP5030 does not show this behaviour.

Hopefully this helps to find the problem.

Best Regards,
Ulrich

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/196/joystick_5F00_lp5030_5F00_ti.PNG

/resized-image/__size/640x480/__key/communityserver-discussions-components-files/196/IREF_5F00_voltage_5F00_LP5030_5F00_defect.PNG

0 Ulrich Gerhardt over 1 year ago in reply to Ulrich Gerhardt

Prodigy 10 points

sorry for the poor quality of the pictures above...

/resized-image/__size/960x720/__key/communityserver-discussions-components-files/196/4743.joystick_5F00_lp5030_5F00_ti.PNG

/resized-image/__size/1280x960/__key/communityserver-discussions-components-files/196/2185.IREF_5F00_voltage_5F00_LP5030_5F00_defect.PNG

0 Aaron Bing over 1 year ago in reply to Ulrich Gerhardt

TI__Expert 4120 points

Hi Ulrich,

Are all 9 red LEDs stuck on in this case? The output current should be close to zero when the IREF pin is floating. If the IREF control circuit is not properly functioning, all the current sink output will be affect. The red LEDs are stuck on due to its relatively small parasitic capacitance.

Would you probe the stuck-on LEDs for a continuity test and swap the LP5030 with another? Please also measure the current data on output channel when the IREF pin is floating, configure color-mixing registers as usual, and examine the current adjustment in full range with green and blue LEDs to see if it works ok.

Best Regards,

Aaron

Power management

Power management forum

LP5036: red LED stuck on after ESD