GN’s commentary is more nuanced than that. YouTubers say they’ve identified two specific causes for melting links. One is foreign object debris (FOD) in the link. This can happen due to manufacturer error or the user’s environment, but in any case, debris in the socket could cause the current to take a “parallel path”. This path will likely have very high resistance compared to the intended path, and this will generate too much heat. A link that looks correct but is still not fully inserted. The other issue, which we believe (and GN states) is drastically more common, is “serious” user error. As Jon Gerow (aka Jonnyguru) stated last week, if you don’t fully insert the connector, it’s likely to make a bad connection and overheat. GN actually tried this in a previous video on the problem and couldn’t get it to cause a link melt, but as Steve himself admits, he just underestimated the extent to which the link had to be a “poor installation”. The link has a 6.5mm length that can be inserted and we can confirm that both Steve and Jon are right when they say that it is deceptively difficult to fully install. It’s very easy to get it into a state where it looks fully seated but isn’t, and this state can allow the connector to move out of place, especially if you’re fiddling with the cable management on the other side of the chassis . This poor installation can cause the plug to be angled, which greatly increases the chance of an improper connection, which in turn could cause overheating and melting. The GN diagram for the most likely cause of the majority of failures. Steve is adamant that the failures are not due to one type of adapter or the other, as these failures can affect either Astron or NTK sourced adapters and have even occurred with native 12VHPWR cables—no adapters required. He says that “all 12VHPWR connectors can fail by one method or another,” but also that his objective opinion is that the connector itself is perfectly safe. As he points out, failures are extremely rare. After all, the video concludes, a failure could happen to anyone if you get a faulty socket, but that goes for any product. To protect yourself from user-induced damage, it’s important to insert the plug fully into the outlet—try pulling it gently afterward to make sure—and then not stretch it with your cable routing. This will allow you to avoid a situation where it “starts acceptable but erratic and then melts down” either as cable routing is managed or from system shock (eg case fans) over time. We recommend watching the video if you have the time, as it is very interesting to see the lengths GN went to in trying to find the real cause of this problem. The channel sent user-supplied links to a failure analysis lab, where they were x-rayed, dissected and scanned with an electron microscope. Kudos to Gamers Nexus for going the extra mile, although it really seems like NVIDIA should have done it themselves.
