Burned TNY275 chip

Hi,

I have been asked to perform a failure analysis on an AC/DC power supply design utilizing the TNY275 (TinySwitch-III family).

This is a legacy product (5–6 years old) that has recently experienced failures across multiple, separate field locations. 

The product operated reliably for several years prior, and no other electronic equipment at these locations showed defects.

1. Observed Failure Modes (Multiple Units):

The board defects show consistent patterns:

• TNY275 (U1) Component Failure:
  • Multiple chips show a burned Pin 4 (DRAIN).
  • Some packages are cracked or have the lid completely missing (indicating a catastrophic thermal or electrical event).
• Bulk Capacitor (C5):
  • The primary side bulk/buffer electrolytic capacitor (C5) is bloated on several units.Bulk Capacitor (C5)
• Fusible resistor (R9):
  • The resistor is burned of heavily damaged on these boards.
• TNY275 (U1) not burned but not working anymore:
  • I did a few diode measurements between pin 5 and pin 4 with multiple failing products:
    • 1.032V - 1.044V - 1.041V - 1.031V (a new product measures as 0.521V)

(Note: I have attached photos of the burned-out components for reference.)

2. Questions & Hypotheses:

Given the nature of the failures and the simultaneous occurrence after a long period of stable operation, I have the following questions regarding potential root causes:

1. What is the most probable cause of these failures, considering the combination of the bloated C5 and the apparent breakdown of the TNY275 (specifically the DRAIN pin)?
2. Could long-term operation at the higher limit of the mains voltage (e.g., 250 VAC, potentially caused by local power grid conditions or solar panel back-feeding) have caused material degradation over time?
3. Does this pattern suggest a latent design flaw in the protection or component selection that only manifests under stress after several years in the field?

Any insights or thoughts from your team would be highly appreciated.

Best regards,

J.