Determining if a power supply unit (PSU) is faulty can be tricky. A bad PSU can cause a variety of issues including random crashes, blue screens, sudden restarts, and failure to boot. Here are some ways to help diagnose PSU problems.
Visual Inspection
Start by carefully inspecting the power supply unit. Look for any signs of damage like burnt or bulging components. Also check that the fan inside the PSU is spinning properly. If the fan is not working or is making strange noises, the PSU could be overheating.
Examine all the power cables to make sure they are not damaged. Look for any cuts, exposed wires, or frayed/melted insulation. Faulty power cables can prevent the PSU from delivering stable power to components.
Make sure the power connector ports on the back of the PSU are not damaged or bent. The PSU connectors need to make solid contact to work properly. Loose, bent, or broken connectors can cause power delivery issues.
No Power
A sure sign of PSU failure is no power at all. When you press the power button, if nothing happens – no fans spin, no lights turn on, no POST beep – the PSU should be suspected. Here are some things to check:
- Make sure the PSU is turned ON and the power cable connected firmly to the back of the PSU.
- Verify the power cable is plugged into a working wall outlet. Try a different power cable if possible.
- Check that the power switch on the back of the PSU is turned to the ON position.
- Make sure the main 24-pin motherboard power connector is fully inserted into the motherboard.
- Check all other PSU connectors like the 8-pin CPU power, PCIe cables etc are properly connected.
- Bypass any power strips and connect the PSU directly to the wall outlet.
If there is still no power after checking all these points, the PSU has likely failed.
Power Up Then Immediate Shutdown
Another common PSU fault symptom is the system starting up for a split second but then abruptly shutting down. This immediate power off indicates the PSU is not able to maintain stable power delivery.
Causes for this problem include:
- Overcurrent protection kicking in due to short circuit or power overload
- Unable to deliver expected wattage due to age, heat damage etc
- Incorrect cable connections causing overload
- Defective PSU components like capacitors, resistors, diodes etc
If the system crashes immediately on startup, disconnect all unnecessary components and peripherals. Try booting with just the motherboard, CPU, and one RAM stick. If it powers up normally, there is likely a short circuit or faulty component in one of the disconnected parts.
Intermittent Power Cycling
Intermittent power cycling or random system reboots are also commonly caused by failing power supplies. This happens when the PSU can power the system initially but is unable to maintain stable voltage delivery over time.
As components heat up during use, the demands on the PSU increase. An aging or defective PSU starts to struggle with power regulation, causing the system to abruptly crash and restart. Problems include:
- Insufficient wattage – Can’t meet increased power demands when components get hot.
- Capacitor degradation – Stability issues due to bulging/leaking capacitors.
- Bad fan – Fan failure can lead to PSU overheating and instability.
- Voltage fluctuation – Unstable DC output voltage from PSU rails.
Intermittent power cycling typically gets worse over time as the PSU degrades further. Stress testing tools like Furmark and Prime95 can be used to simulate maximum system load and detect any voltage regulation issues.
Computer Frozen or Unresponsive
A computer that freezes or becomes unresponsive can also point toward PSU issues in some cases. When the PSU is unable to maintain steady power delivery, components may start to act erratically or crash as they don’t get stable power.
Sudden system lockups and BSOD errors can be caused by even small voltage fluctuations or ripple from a bad PSU. If system instability keeps getting worse, the PSU should be suspected as the culprit.
PC Won’t Turn On After Upgrading Components
Upgrading components like the CPU, RAM, GPU etc places greater power demands on a PSU. An undersized or lower quality PSU that was working fine earlier may start to exhibit problems when asked to deliver more power.
If a PC refuses to POST or boot after new parts are installed, the most likely reason is insufficient wattage from an aging PSU. Always make sure the PSU can provide enough power headroom for any upgraded components. Replacing the PSU with a higher wattage model should resolve the no power issues after an upgrade.
Odd Noises
Unusual noises from a PSU like high pitched whining, buzzing, humming or grinding noises often indicate issues:
- Coil whine – High frequency whining caused by vibrating coils as power passes through.
- Capacitor hum/buzz – Buzzing or humming noise caused by failing capacitors.
- Bearing noises – Grinding, scraping noise from a faulty PSU fan bearing.
These types of anomalous noises point to component problems within the PSU that can impact performance and reliability. So investigate and replace the unit if strange noises are heard.
Overheating
A PSU running excessively hot can also indicate issues. Check for overheating if you notice any of these symptoms:
- Loud fan noise from PSU
- Hot air being exhausted from PSU vents
- Smell of burning plastic from PSU area
- High internal case temperatures
Overheating can be caused by dust buildup on the PSU’s heatsink, a clogged fan, or simply an overtaxed PSU being forced to deliver more power than it can handle.
Prolonged overheating degrades PSU components and shortens its lifespan. A thermal shutdown can also suddenly occur if temperatures get too high.
Testing Voltages
One way to directly check for PSU issues is to measure the DC output voltages using a digital multimeter. Compare the measured voltages to the rated specifications for your particular PSU model.
Major rails that can be tested include:
| Rail | Normal Voltage |
|---|---|
| 3.3V | 3.3V ±5% (3.135V – 3.465V) |
| 5V | 5V ±5% (4.75V – 5.25V) |
| 12V | 12V ±5% (11.4V – 12.6V) |
Any voltages outside the normal range likely indicate a faulty PSU. Use a load tester to simulate stress conditions for a more realistic voltage measurement.
Testing Under Load
PSU testers are a more reliable way to simulate actual system loading conditions. They put a sustained power draw load on the PSU to test whether the voltages stay within specifications when stressed.
Using a PSU load tester can reveal instabilities and deviations not apparent when idling. Portable load testers can be purchased quite cheaply online. Some other options include:
- Run Furmark + Prime95 together to max out GPU and CPU power draw.
- Remove PSU from system and use a paperclip to short the 24-pin connector and test.
- Use a multimeter to check voltages during high system load.
If voltage regulation fails during stress testing, it confirms the PSU needs to be replaced.
OEM Power Supplies
The power supplies (PSU) that come with pre-built OEM desktops are often cheap, lower quality units. They frequently lack sufficient wattage ratings for future upgrades and use lower-grade internal components.
For gaming systems, workstations, overclocked systems, or PCs with high-end components, an OEM PSU should be one of the first parts replaced. Quality aftermarket PSUs from established brands provide better voltage regulation, efficiency, cooling, and long-term reliability.
Conclusion
Determining PSU faults requires systematic troubleshooting. Visually inspect for damage, test voltages at rest and under stress, check for overheating, and listen for any unusual noises. If multiple symptoms point to PSU failure, replacement is likely the solution versus further diagnosis. When selecting a new PSU, don’t skimp on quality – a good power supply protects all your other components.