Member Avatar for polarpandabear

Hello Everyone!

I had a consistent problem with my one year old PC. it randomly shuts down and I would like to clarify and confident enough that overheating in my GPU is not the cause of shutting down. I'm still trying my best (until now) to have solutions.

My PC brief specs:
AMD Athlon II x2 3.0ghz (not OC)
Powercolor Radeon HD 7850 fling force
Simple MSI motherboard
Kingston 2gb ddr3 memory
500w thermaltake PSU

Solutions that I had made:
1. cleaning my PC
2. replaced thermal compound on CPU and GPU
3. updated device drivers
4. updated BIOS
5. check airflow

Any solutions? thanks :D

  • 3 Contributors
  • 7 Replies
  • 180 Views
  • 1 Day Discussion Span
  • Latest Post Latest Post by rubberman

Dani AI

Generated

Brief diagnostic summary and where the evidence points

The pattern in the thread — long, CPU-only renders running fine while light GPU activity (games or browser video) sometimes causes an abrupt power-off — makes a power-delivery or GPU-path fault the primary suspect rather than overheating. System logs and driver-fault entries are the best objective indicators: Windows System events commonly show an unexpected power loss (Kernel‑Power) or record GPU/driver failures (AMD driver module names) at the time of the shutdown, and those entries narrow the cause quickly.

Practical, prioritized checks that add to the existing replies from and

  • Swap or bench-test with a known-good PSU (best single test). If the stops, the old PSU is the culprit.
  • When a swap is not immediately possible, monitor voltages and sensors under load (HWMonitor/GPU‑Z/OCCT) and stress the GPU while watching the 12V/5V/3.3V rails. ATX tolerances are useful reference: 12V ≈ 11.4–12.6V, 5V ≈ 4.75–5.25V, 3.3V ≈ 3.135–3.465V; values outside those ranges indicate failing power delivery.
  • Check cable/connectors and avoid molex-to‑PCIe adapters; reseat the GPU power plug and test with minimal hardware (single RAM stick, no extra drives/USB) to rule out peripheral current draw.
  • Run a memory test (memtest86) and capture the exact Event Viewer entries for any shutdown—those logs guide whether it’s hardware, driver, or filesystem-related.
  • For browser crashes only, disabling browser hardware acceleration isolates GPU-accelerated decoding as a factor.

Safety notes and short-term mitigation

Never open a PSU (dangerous capacitors); external inspection for bulging caps, fan failure or burnt smell is safe after unplugging. Sudden shutdowns risk filesystem corruption and can shorten drive life, so avoid long unattended renders until power stability is confirmed. If replacement is needed, choose a reputable PSU with solid 12V capacity and headroom rather than one sized only to the minimum wattage.

Recommended Answers

All 7 Replies

Member Avatar for rubberman

How many and what type of disc drives are you running? Devices that are USB powered? Other cards besides the video?

FWIW, the Radeon 7850 draws approximately 300 watts when not at idle or just doing simple desktop stuff (provided your system can engage with the 7850's low-power control system). With everything else, your power supply may be the culprit here. Try replacing it with a 750W unit. Thermaltake units are, AFAIK, perfectly fine supplies, but I think that 500W is a bit weak for your needs. I have seen more systems fail from inadequate PS units than just about anything else, other than RAM or CPU overheating.

Edited by rubberman
Member Avatar for polarpandabear

thanks rubberman for reply :D

I forgot to tell that I have 2 HDD and 1 combo drive. I recently installed 2 generic 80mm fans.

I also suspected that PS is the main problem. the average temp of my CPU and GPU ranges between 40 - 50C.

I am a 3D artist with a little knowledge on troubleshooting and I do a lot of rendering in my machine. the worst is 7 days straight using mental ray (which computations only occur in CPU) but the problem starts when it comes on playing games like league of legends. I'm currently using a regular thermaltake 500W.

too bad I can't search for any whitepaper or other stuff about my VC.

please enlightened me on the perks of having low wattage on my machine. I don't know the consequences on having that problem

I would also like to ask if this problem will also affect the health and performance of the other parts of the machine.

Thanks

Edited by polarpandabear
Member Avatar for DistantGalaxy

Did the problem arise before or after installing the fans?

If you haven't tried already, install them so that one fan expels air while the other draws air in. This should increase airflow through the case. If both fans are expelling air it might cause a reduction of pressure inside the case and prevent the PSU fan from working effectively.

Edited by DistantGalaxy
Member Avatar for polarpandabear

The problem Already arise before I installed the fans. My set up before was an open case to watch the temp of my vcard. But the frequency of shutting down was not worse than I am experiencing today. I am just using google chrome and when I stream a video on youtube it shuts off Which I could do before.
Btw I already tried to flip the other Fan still it crashes

Member Avatar for rubberman

As we both suspect, your 500W PS is not adequate, which assumption is reinforced with your newer comments about it failing when game playing (forces the GPU into high-power mode). Get a 750 or 1000W PS and see if that fixes the problem. Myself, for my workstation I always go for the higher capability PS (1000W). In this case, more is better. It costs only incrementally more, but it will cover all reasonable power requirements without problems.

Member Avatar for polarpandabear

I have already temporary solutions as I earn enough monry to buy new PS. I need to unmount secondary disk drives and cd drive. I'm hoping thwt would help. i would like to ask if the other parts are still fine in this situwation.

Member Avatar for rubberman

They should be. YOu need to reduce the current draw. You need to disconnect the secondary disk drives and cd drive, including the power plug. IE, shut down the system, unplug the drive(s), and then restart the system. Whether this works or not will depend upon the power draw of the CPU, which can be significant. I haven't checked, but if it is over 100W then you may still have issues when the video is operating at max.

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.