Power Distribution Board Failure
Summary
Power distribution board failure appears as complete no-power conditions, intermittent no-boot behavior, dead GPU or fan power paths, burning or smoke near the chassis power area, and repeated symptoms that persist after PSU swaps. In this set, the durable fix was usually barebone repair, chassis replacement, or direct PDB replacement rather than repeated PSU-only troubleshooting.
Frequency
17 tickets.
Common Causes
- Electrical damage or abrupt power events damaging the chassis power-distribution path. Systems failed after emergency power-off testing, over-current events, suspected shorts, or smoke/burn evidence near the board. Examples: #12037, #12039, #30716, #3313, #9217.
- PDB or power-port failure initially mistaken for PSU, motherboard, or PCIe-slot failure. PSU swaps sometimes passed, but the real fault stayed in the chassis distribution layer or a specific power output port. Examples: #11973, #20738, #33686, #5530, #8835, and 2 more.
- Chassis-level aging or barebone design issues causing wider subsystem failures. Some cases affected fan power, repeated no-boot behavior, or multiple systems from the same platform, pointing to the barebone rather than one attached component. Examples: #7577, #35458, #35690, #37336, #2914.
- Manufacturing or assembly defects in power-distribution hardware. A smaller subset involved incorrect installed power components or direct faulty-board replacement without much ambiguity. Examples: #23857, #4957.
Diagnostic Steps
- Rule out simple PSU-only failure first. Try known-good PSUs or single-PSU operation, then check whether the same no-power or over-current behavior remains. Examples: #11973, #3313, #5530, #30716.
- Check whether the failure localizes to the chassis power layer. Look for dead fan banks, dead GPU power, non-working front power behavior, smoke smell, burn marks, or one bad power-distribution port. Examples: #20738, #33686, #35458, #35690, #9217.
- Reduce the system and swap peripherals before condemning other parts. Customers and depot teams commonly swapped DIMMs, GPUs, drives, or slots to prove the failure did not follow those components. Examples: #7577, #11973, #33686, #8835.
- Escalate to depot or manufacturer validation when power symptoms persist. Many confirmed cases required barebone-level diagnosis, because the failing part was inside the chassis power path rather than a field-safe replaceable component. Examples: #12037, #12039, #2914, #30716, #8835.
Solutions
- Replace the PDB or send the barebone/chassis for manufacturer repair. This was the most common successful path for confirmed no-power and shutdown failures. Examples: #11973, #12037, #12039, #30716, #7577, and 4 more.
- Replace the entire chassis or barebone when the platform had broader power-path risk. Exxact used chassis swaps when direct repair was slow, insufficient, or repeated failures suggested the whole enclosure should be retired. Examples: #2914, #8835, #35690, #37336, #9217.
- Replace a bad power-distribution board or port directly when depot testing isolated it. Some tickets reached a cleaner component-level fix once Exxact proved one port or board was the true root cause. Examples: #33686, #35690, #4957.
- Move out-of-warranty or EOL cases into paid replacement or migration planning. When parts were no longer available, support shifted to sales-led barebone replacement instead of repair. Examples: #35458.
Edge Cases
- Symptoms may look like GPU, motherboard, or PCIe-slot failure first. One repeat RMA that seemed like a bad PCIe slot was ultimately traced to a defective power-distribution port and then the PDB itself (#33686).
- PDB failure can coexist with other damage. Depot testing sometimes found separate GPU, thermal, or main-board issues after the original power fault was fixed (#11973, #30716, #35690).
- The issue can affect many systems from one chassis family. One later program replaced five Chenbro systems after ongoing problems were traced to Chenbro PDB issues across the fleet (#37336).
- Advance replacement decisions depend on risk and logistics. High-risk RAID or international cases sometimes pushed toward cross-ship or special handling rather than standard return-first processing (#4957, #9217).
Related Issues
Referenced by
- RTX 3090 — product affected by this issue (×3)
- Allen Huynh — handled tickets on this issue (×1)
- Ian Dicarlo — handled tickets on this issue (×5)
- Overheating — related issue (×2)
- Jason Chen — handled tickets on this issue (×3)
- David Nguyen — handled tickets on this issue (×1)
- RMA Workflow — co-occurs with this issue (×16)
- A100 — product affected by this issue (×1)
- Rtx A6000 — product affected by this issue (×1)
- System Boot Failure — related issue (×2)
Comments
0 comments
Please sign in to leave a comment.