The CPU receives power immediately, while other components remain inactive—likely the PSU is the issue. Please check it.
The CPU receives power immediately, while other components remain inactive—likely the PSU is the issue. Please check it.
Hey everyone, I’ve been a hobbyist for quite some time, but after a long hiatus things shifted around. Eventually, I decided to dive back into the PC world and rebuilt my setup using these components:
- ASRock a620i lightning wifi am5
- Amd 7600x processor
- Thermalight copper CPU cooler
- 32gb DDR5 RAM (t-create)
- Chromax fan mounted on the case bottom
- 1 tb SSD
- Corsair SF750 power supply
- Some SFF case with white finish and leather handle
This build was made about a year ago, and it performed well. Although I hadn’t added a GPU yet because of high prices, I managed to play several of my older games directly onboard.
A few days back I woke up to find the computer off. It seemed to be in sleep mode, which I thought might be the reason. While running a virtual machine with Home Assistant, I powered it on and got power but no display over HDMI. That marked the beginning of troubles. After several attempts to re-seat RAM, check cables, and even using a multimeter on the connector wires, everything seemed fine. So I started wondering what I had done the night before.
The only significant action was signing up for GeForcenow to test its effectiveness compared to buying a good GPU. I remembered there were prompts during installation about system changes. I decided to clear CMOS and proceed. But that didn’t help. Then I found the CMOS battery, unplugged it for a day, and reset the pins again just to be sure. Now I noticed the power light on the case and the fan spinning, but nothing else was active. After waiting about five minutes for the BIOS to boot, there was still no display. The HDMI port had been disconnected multiple times while sharing the cable with other devices. I became convinced that the HDMI was likely broken at this stage. My quick thinking suggested either the output was redirected to the GPU or the HDMI port was damaged. I admitted I’d stopped searching online for solutions.
At that moment, I checked the motherboard closely and found bent pins on the CPU. I didn’t like handling them, but they weren’t causing any issues. I put the CPU back in place and after a day passed, I still saw only the power supply fan spinning and the case fan shutting down, while the main fan stopped working. This was unexpected since the PSU should handle that. I then checked the MOSFETs on the CPU and found they were damaged—3-4 pins bent. I tried using a razor to fix them but had no success. Eventually, I realized I wouldn’t need a GPU anymore and decided to buy an affordable motherboard.
I replaced my old board with an ASRock B650i Lightning Mini, keeping the cooler as it was needed for testing. After pressing the power button, the main case fan started and the PSU activated its shutdown sequence normally. I thought everything was fine until power cut off after about 3-4 seconds. On a new board, I got nothing—no response at all. I tried adding extra RAM but had no luck. I considered using a cheaper PSU but thought it might work. I also wanted to test the motherboard without cables, so I used a multimeter to check for power on the supply side and pin configuration.
Anyway, my thoughts are that the CPU must be damaged somehow, but the power supply seems intact. Thanks for your patience!
The main header of the post is accurate now after testing both methods. The CPU receives power immediately when the power supply switch is activated. This behavior seems unexpected, as it doesn't appear to occur during a normal start-up sequence.
Did you operate the CPU without a cooler connected?
Lacking a cooler, numerous desktop PC CPUs may overheat and shut down within seconds. Even without TIM (heatsink paste), attaching the cooler is advisable. Any cooling method is preferable to none. Without TIM it won't perform as well, but direct metal-to-metal contact with the cooler will still manage to cool the CPU.
Did your thermometer read 104 degrees Fahrenheit or 104 degrees Centigrade? If so, it exceeds the threshold where CPU thermal shutdown happens.
The 7600X has a TDP (Thermal Design Power) of 105W, meaning the CPU can release several tens of Watts when the system starts. Without a heatsink, the CPU will quickly heat up and likely surpass its maximum operating temperature of 95°C in seconds, triggering an automatic shutdown.
So your PC functioned normally for a year before you discovered 3 to 4 bent pins in the CPU socket. Could it be that you accidentally reinserted the CPU into the socket during a recent repair? It's unlikely the original issue was bent pins, though it's not impossible if all relevant pins were bent. Were the pins severely distorted?
By "CPU gets power" do you mean the screen displays POST messages from the BIOS?
The motherboard will remain active to some degree; the +5VSB rail should stay at +5V as soon as AC power is connected to the ATX PSU and the rear switch is in the on position. This is normal.
Regarding your concern, with a fault condition, any outcome is conceivable if the system has suffered damage.
Certainly, here is the rewritten version:
It’s great that you replied and clarified some details. The main concern was not bent pins on the motherboard, which I had experienced after several removals and installs of the CPU. At that time, I attempted to test the board by removing all components to see if it would work. Now I’m struggling to understand what happened because I can’t even reach the post. This is why I’m becoming very anxious about figuring out the cause, as I initially suspected a hardware problem like a short circuit or power supply issue.
I’m uncertain how to test the power supply itself since it has a power sensing feature. However, I’m confident in using a multimeter to gather more information. Also, I only booted up the computer without a cooler, so I’ll try your advice about mounting it directly on metal to prevent any issues.
The situation began when my power button displayed power and a fan was running, but the computer wouldn’t respond. My only visible port was HDMI, which led me down this path. Yes, I’m sorry for not including everything. I was really surprised to notice that the CPU receives power immediately once the power switch is flipped, even though no call had triggered it before.
From what I understand, the power button on the PC should send a signal to the start pins, allowing the power supply to deliver electricity to the CPU and the rest of the system. This sudden CPU power seems to happen regardless of whether I route the power through the case or directly to the supply. Maybe I’m mistaken, but why would a power supply keep feeding the CPU in this way?
Thank you for your assistance. I hope this helps me understand what might be wrong. I’m just trying to be a parts swapper since you don’t seem to learn much.
Hi everyone, I've been working on fixing a home build I was managing for a year without any problems. The build includes: an Asrock a620 lightning WiFi mini (board 1 that failed), a current board of Asrock lightning b650 mini (still having issues), a Ryzen 7600x processor, 32GB T-Create DDR5 memory, a 1TB M.2 drive, a thermal right copper cooler, one case fan, and a Corsair 750sf case fan. Recently, the system stopped working completely. I suspect the problem might be with the motherboard or power supply. After ordering both after reaching a breaking point, I tried simplifying everything to just the essentials but still couldn’t get it running. Now I’m constantly noticing the CPU getting very hot when left on, and power is always flowing to it even when there’s no load. I’ve seen some discussions about this issue and removed the CPU power cable in an attempt to salvage it. I disconnected the power supply to see if it would stop heating up, but it continued to get hot without any power being applied. I’m now considering replacing the CPU or RAM, as I feel more like a parts person than someone who can troubleshoot effectively. I’m capable with a multimeter, but I don’t think that’s enough to test the CPU properly. Any advice would be greatly appreciated.
From all the details provided, it seems the CPU is likely experiencing an internal short circuit. This is supported by the fact that it becomes warm even when the EPS power cable is removed under typical conditions, which would normally cut off power to the CPU. The consistent results from swapping in a new PSU and a second motherboard, along with the absence of fan spin or POST attempts, point to the CPU being the source of the issue. RAM problems wouldn’t explain the overheating without EPS power, so they can be dismissed. Since diagnosing an internal short requires specialized tools and there’s no way to test it directly, the next step would be to try a known good AM5 CPU or replace the 7600X entirely.
As far as I'm concerned, when you've bent pins in the CPU socket, you're heading towards disaster and a new motherboard. You might be able to straighten the pins without them breaking off, or you might ruin the board.
I've been lucky so far and never bent pins in a CPU socket, but I have bent pins on old AMD CPUs and managed to straighten them.
The last time I booted a PC without a CPU cooler was back in the days of 80286/386SX CPUs. I would never boot a modern computer without a heatsink, if one was required. Far too much risk of instantly killing the CPU.
Modern CPUs are designed to thermal throttle when they get too hot, but that's when there's a cooler attached. I have no idea what would happen with no cooler, because I'm not prepared to experiment on a working system.
With a suspect CPU at the beginning, followed by bent pins in the socket, plus a PSU in an unknown state, I think I would play safe and buy a new mobo, CPU, RAM and PSU. If you test a faulty component with known good components, you risk destroying the good components. More money down the drain.
An old LGA1150 mobo died on me recently. At first it was intermittent, now it fails to POST. The CPU, RAM, SSD, GPU and PSU are fine, so I've abandoned the mobo as scrap. It's not always the PSU that fails.
Life isn't always fair.