The PSU cables ignited.
The PSU cables ignited.
Hello everyone, recently I assembled my ideal watercooled PC with these components: Intel 19-12900K (NEW), ROG Maximus Z690 (NEW), 32GB Dominator DDR5 5600MHz (NEW), RM1000x Corsair PSU (1 Year Old) 2TB PCIe4.0 WD black m.2 (NEW), and Gigabyte Aorus 3090 Xtreme Waterforce. After about a week, I left the room for 15 minutes and the PSU wires caught fire. Since I live on campus, I quickly left the building when the alarm sounded, hoping the fire hadn’t started yet. During that time, three cables ignited—GPU power, CPU power, and RGB controller cables. I’ve attached photos for reference. Someone might be able to assist me in figuring out the cause and preventing future issues. Also, note that there was a loud bang 30 minutes before this happened; I initially thought it was pipes leaking coolant, but there were no visible signs. It could have been a capacitor failure, though I’m not sure. The security team who entered the room later said the PC was still on, possibly because it couldn’t detect the wires burning. If anyone can help me check the CPU, RAM, and SSD, I’ll be very thankful. In the meantime, I’ve ordered a new motherboard and PSU to test them once they arrive. The black cable connects to the white power cable for the GPU, while the exposed copper cables and one melted wire are for the CPU, and the RGB controller cable is stuck to the CPU cable.
This damage seems significant. The melted CPU cable suggests it was connected directly to the motherboard or perhaps routed through an extension cable. It’s likely the other power cables were hidden in the case bottom with the extension wires nearby. I’m trying to pinpoint exactly where it failed. From the photos, the melted PCIe power cables might be the starting point, with any melting at connector junctions between the extension and PSU cables indicating the issue. Did the PC have a high power demand at the time? Was it running a demanding task like gaming, rendering, or mining? Or was it idle? Also, what type of cable extensions were you using?
Did you go overloading the connectors with too much current going through them? Usually that only happens if you split an 8 pin into 2 8 pin or too many fans for 1 cable I would think cables catching on fire would happen to me first since after all my psu and cables is of dubious quality (also over a decade old) and i overcurrent them anyway by hooking up a 72w tec on molexes that arent even supposed to draw 6a on 12v xD Did you see any smoke? If the psu starts smoking swap it out right away cause if it malfunctions and dies it could take out other parts too If its this high end of a build id assume you wouldnt be stupid enough to cheap out on cables, adapters, hubs, etc. Right?
It might have been a fault in one of the PC parts—like the graphics card or motherboard—causing a surge and breaking the connection, leading to overheating and melted cables. I’d begin by taking off the waterblock on the GPU and checking the card for any signs of damage.
The CPU cable ran directly from the power supply to the motherboard, though it showed some strain due to the longer distance. I had to bend it tightly to fit it in, but it performed well under all stress tests. I used the system for intensive gaming, and it stayed cool—only about 5% CPU usage and 2% GPU usage, with just a few background tasks. The cables were safely stored inside the PSU housing when I left the PC.
From your photos, it seems you included various extension cables—black ones (not the damaged ones), white ones, and possibly flat ribbon types. The extension cables are generally cheaper and simpler, but they can cause issues and lead to problems if not handled properly. It’s great you noticed them early, which helped prevent any bigger issues. That’s the key point.
I didn’t overload them. The GPU cables were connected via two PCIe power cables from the PSU. The PSU cables came from eight pins and then branched into six-pin connections, so I required two eight-pin ports. There was no smoke when I heard the sound, and none of the PSU cables were from Antec, meaning they were premium quality.
The clear reason turned out to be a small issue, though where exactly? It might be a component such as the RGB/fan hub randomly failing or the cables simply breaking.
The only extension cables used were on the 24-pin port, which worked without issues. The GPU cable that failed at the connection point was also affected, along with the RGB controller cable. Keep in mind that the CPU cable that got burned, the GPU cable’s connection point, and the RGB controller cable all suffered damage at the same spot, making it hard to pinpoint exactly which one started it off.