Solution Finder Get Tech Support

LNK3204 Input fusible resistor burns on rare occasions.

Posted by: mjansson on

Hello,

Our prototype that implements the LNK3204 regulator causes the input resistor to burn on rare occasions. Our design follows reference design RDR-506 with the difference of only using a single rectifier diode on the input side.

The power supply has been tested with dummy load with maximum current consumption of 120 mA on the secondary side and no issue was found. The secondary side was also measured and no anomalies were found. The current consumption was nowhere near the maximum capacity.

What could cause such an event? Replacing the resistor makes the power supply function again so no other parts are damaged in the process. Replacing the resistor with a value lower than 8.2 Ohm used in the reference design also helps reduce this issue.

Any suggestion on troubleshooting or causes would be appreciated! Thanks!

Best regards,
MJn

Comments

Submitted by PI-Kobe on 06/03/2020

Hi,

Could you send me the schematic and the BOM? I'd like to check how similar the prototype is to RD-506 before commenting. Thanks.

Regards,

Submitted by mjansson on 06/04/2020

Hi,

Thanks for your reply. Attached snippet of power supply section.

Attachment Size
BOM 144.7 KB
SCH 145.66 KB
Submitted by PI-Kobe on 06/04/2020

Hi,

Can you describe how often/how rare the RF1 burns? Does it happen at steady-state or during some tests? Have you checked the input current waveforms when this occurs? Thanks.

Regards,

Submitted by mjansson on 06/16/2020

Hi,

Sorry for late response. We noticed RF1 would burn when we repeatedly applied power, leading us to believe that the inrush current was the problem. Switching to a wire-wound resistor seems to have solved this issue. Thanks!