A very confusing yet interesting PSU problem on a new PC build


Posts: 4,511   +5,358
Sorry if this isn't in the right spot. I believe it belongs in general discussion because it's more about an interesting issue I encountered rather than the building of the PC. The purpose of this post is more about a fascinating failure rather than advice in building a PC. However every time I make a post like this it gets moved.

So, today I was building a new PC for my friend. He ordered the parts but didn't have a PSU, I just so happen to have a very powerful PSU laying around that I wasn't using. It was expensive and I'm interested in fixing it(if there is something wrong with it).

here it is if you're interested

so on to what happened. I hadn't used it in awhile so I plugged it in to a working PC(but not the harddrives, this will be important later) and everything turned on no problem. Videocard, motherboard, cpu. It turned on, posted and saw everything on the test PC no problem. So we assemble the new PC and installed it. When everything is plugged in the lights turn on when the power ISN'T turned on. I have other RGB gaming hardware and I see it has random LEDs on the graphics card and motherboard when it isn't powered on. I don't know what this is called but I'm doing my best to describe what they are. As soon as I hit the power button I hear I click coming from the PSU and all ofthe other RGB lights turn on for a fraction of a second then everything shuts off. If I hit the power button again I get no response from the system. I have to cycle the switch on the back of the PSU before anything happens. once cycled, the same thing will happen, short power on then no response from the power button.

At this point we're kinda annoyed and I tell my friend about the power on trick of shorting pins 15 and 16 on the 24 pin connector to force the PSU on. I told him there was a significant risk associated with trying this and he told me to go ahead with it. I started to suspect the motherboard was bad because we had just tested in another system not 20 minutes before and it worked. I had trouble getting access to the 24 pin connector while plugged into the mobo so I unplugged it from motherboard and then shorted them for force the PSU to turn on. the CPU 8in power connector was still plugged it, the 2x8pin pci-e connectors were still connected to the videocard and the power was still connected to the harddrives and fans. Yes I should have done a number of things differently here but this is what happened

so I short the pins and the PSU turns on powering up the system. It immediately causes the SSD to let the magic blue smoke out. I thought I fried the system but I take a known working PSU out of another system and plug it in, everything works fine(thank effing god!)but all the harddrive are fried. I can feel you guys cringing just writing this but I've done this trick tons of times and never had an issue.

So on to what I find interesting and what I think might be wrong with the PSU. On account of this same PSU working and not killing another system minutes before this, I don't think the PSU is the problem. I do feel that the motherboard or PSU had some sort of protection in place to identify something is wrong and prevent it from turning on. I believe it was the motherboard doing this because shorting pins 15 and 16 did not prevent the PSU from powering on. After thinking about it for awhile I am starting to think that one of the drives, specifically the SSD, had some type of internal short that shorted the 5v and 12v rails frying everything connected to that SATA cable. A visual inspection of the PSU seems find. None of the traces on the PCB looks burnt, the caps still look new and it's very clean, no dust.

It's late and I can't find my multimeter as I want to test the rails before I throw it into some type of throwaway system. I mostly wanted to share what happened as I find it incredibly interesting and some insight into what might have happened would be a bonus. I also enjoy issues that stump me. I've been doing this for 20 years and I'm sure some of you know what it feels like to start to think you know everything. Then a problem comes around and you have to just stand back and appreciate the, "what the eff is going on here?" moment
Last edited: