KitGuru KitGuru.net – Tech News | Hardware News | Hardware Reviews | IOS | Mobile | Gaming | Graphics Cards
Roman Der8auer has been busy testing RTX 5090 graphics cards following the first reports of melting 12VHPWR connectors. Much like when the RTX 4090 launched, at least one early RTX 5090 buyer has encountered an issue with the connector overheating and melting.
An RTX 5090 Founder's Edition buyer in Germany posted on Reddit earlier this week showing that their connector had fried after using a third-party Moddiy 12VHPWR cable. This person used this same cable with their RTX 4090 without issue, but it quickly melted when being used with the RTX 5090 Founder's Edition graphics card. Der8auer reached out to the user after seeing their story and has since collected the graphics card for analysis.
In a video, Der8auer examines the damaged card under a microscope and explains that he himself uses Moddiy cables for products, finding them to be of high quality. Still, when used with the RTX 5090, this particular 12VHPWR cable melted at both the GPU side and the PSU side.
Using thermal imaging, two specific 12V cables are getting hotter than the others due to what seems to be uneven power distribution. The connector at the PSU side reached a dangerously high 150 degrees Celsius. According to the investigation, over 22 amps was being pushed down one wire, when each wire is normally only supposed to carry around 6 amps. Some of the other wires transported as little as 2 amps while the other wire was overloaded.
Der8auer does not take the extra step of taking apart the cable or the connector. However, Nvidia will likely need to investigate this further and as part of that, they will no doubt want to see the melted cable for themselves to conduct their own analysis. Nvidia has not yet publicly commented on the situation.
Discuss on our Facebook page, HERE.
KitGuru Says: Fortunately, there only appears to be one case of a burning RTX 5090 connector so far. In the meantime, those with an RTX 5090 may want to exercise caution until Nvidia is able to clarify what the issue was here, as it may not have been a case of user error as initially thought.
