Another Melted 16-Pin Power Connector Reported on High-End NVIDIA GPU Setup
A recent incident has brought renewed attention to the reliability of 16-pin power connectors in high-performance workstation builds. A repair technician has reported a catastrophic failure involving an ASRock Taichi TC-1650T power supply unit (PSU) and a $15,000 NVIDIA RTX PRO 6000 Blackwell Workstation Edition graphics card. The failure resulted in a melted 16-pin connector on the PSU side, with the damage confined to the power supply and the GPU reportedly unharmed.
Details of the Connector Failure
According to the technician’s account, shared on Reddit, the ASRock Taichi TC-1650T PSU was equipped with native 12V-2x6 cables featuring ASRock’s proprietary NTC temperature sensor. Despite both ends of the cable being fully seated, the connector failed and melted into the PSU socket. The technician noted a critical issue: the temperature sensor was installed on the ground line instead of the 12V line, which is the area most susceptible to overheating. This misplacement likely prevented the sensor from detecting the rising temperatures in time to trigger a system shutdown.
TempGuard System: Intended Protection and Real-World Performance
ASRock’s TempGuard system, which utilizes the NTC sensor, is designed to monitor cable temperatures and shut down the system before a connector can overheat and melt. While this safety feature has successfully prevented damage in the past—such as in a case where a shunt-modded RTX 5090 drew over 1,300 watts—this latest incident highlights a potential flaw in sensor placement. If the sensor is not correctly positioned on the 12V line, it may fail to detect dangerous temperature spikes, undermining the effectiveness of the protection system.
Recurring Issues with High-End GPU and PSU Pairings
This is not the first time the combination of the ASRock Taichi TC-1650T PSU and the NVIDIA RTX PRO 6000 Blackwell Workstation Edition GPU has experienced connector melting. A similar incident involving the same hardware pairing was reported in March, raising concerns about the reliability of these components when used together. Given the high value of the GPU, any risk of power connector failure is a significant concern for professionals relying on workstation stability.
Implications for Workstation Builders and Enthusiasts
The repeated failures suggest that the issue may extend beyond isolated defective units and could be related to design or manufacturing inconsistencies, particularly in sensor placement on the 12V-2x6 cables. For users investing in premium workstation hardware, ensuring proper cable and sensor installation is critical to system safety and longevity. As high-wattage GPUs become more common, robust and reliable power delivery solutions are essential to prevent costly hardware damage.
