Key Takeaways
1. The Asus RTX 4090 TUF faced shutdown issues and GPU detection problems after the installation of a WireView Pro power monitoring device, which partially melted into the power socket.
2. Unusual resistance measurements across the 12VHPWR pins were found, indicating poor contact and potential current imbalance, raising doubts about Nvidia’s safety claims regarding their adapters.
3. Der8auer successfully repaired the damaged graphics card by replacing the melted power socket and updating thermal pads, restoring its functionality without overheating.
4. The incident highlights the unpredictability and fragility of Nvidia’s high-power 12VHPWR ecosystem, despite built-in design protections.
5. Current connector standards remain unreliable, suggesting that future “next-gen” connectors may still be prone to failures in real-world usage.
In a fresh teardown and repair video, hardware expert der8auer takes a look at a malfunctioning Asus RTX 4090 TUF. The owner, Aaron, faced unexpected shutdowns and issues with GPU detection. The graphics card had been working well for over a year, but problems arose a few weeks after installing a WireView Pro power monitoring device. When inspecting the system, it showed no power draw on boot, even though there was 12V voltage present. Attempts to remove the device showed that the WireView had partially melted into the power socket of the graphics card. A closer look revealed deformation, plastic cracks, and bent sense pins.
Unusual Resistance Measurements
Using a precise milliohm meter, der8auer checked the resistances across the 12VHPWR pins and found them to be unusually high—up to 177 milliohms—compared to just over 1 milliohm on a reference RTX 5090. Although these values might be skewed due to thermal damage, the difference suggests poor contact and potential current imbalance. This raises questions about Nvidia’s earlier statements claiming that their official adapters are safe due to built-in pin bridging. Der8auer emphasizes that bridging on the connector body does not prevent uneven current flow at individual pins, where thermal damage can still occur.
Successful Repairs
Even with the significant damage to the connector, the graphics card itself was still electrically operational. Der8auer removed the melted power socket, cleaned the PCB, and put in a new 12VHPWR connector. He also swapped out the thermal pads on the memory and VRMs with Minus Pad Advance and added a new thermal interface for the GPU core. After putting everything back together, the card powered on without any issues and ran an hour-long 3DMark Speedway loop at normal temperatures: 63°C for the core, 76°C for memory, and a stable power draw measured via WireView.
In the end, while the card was back to working fully, der8auer wrapped up the video by emphasizing how problematic and unpredictable Nvidia’s high-power 12VHPWR ecosystem remains. Even with official adapters and designs that are supposed to be safe, incidents of melting still happen. A new WireView will be sent to Aaron, and der8auer hinted at more experiments to gain insight into the conditions that lead to such failures.
Concerns About Connector Standards
This recent situation underscores the fragility of the current high-power GPU connector standards and contradicts Nvidia’s earlier reassurances. Despite design protections like integrated pin bridging, real-world usage continues to result in failures. The absence of a completely reliable solution indicates that the “next-gen” connector may still not be safe in all circumstances.
Source:
Link
Leave a Reply