How to Check Your PC PSU (Power Supply Unit): A Complete Guide

Your power supply unit is one of the most overlooked components in a PC — until something goes wrong. Knowing how to check your PSU can save you from misdiagnosing a failing GPU, mysterious crashes, or a system that simply won't boot. Here's what you need to know.

Why Checking Your PSU Matters

The PSU (Power Supply Unit) converts AC power from your wall outlet into the DC voltages your components need — typically 3.3V, 5V, and 12V rails. When a PSU degrades or fails, it rarely announces itself cleanly. Instead, you get:

• Random system shutdowns or restarts
• Failure to POST (the system doesn't boot at all)
• Instability under load (crashes during gaming or rendering)
• Burning smells or coil whine
• Other components appearing to fail

Because PSU symptoms mimic so many other hardware issues, testing it directly is often the fastest way to isolate the problem.

Method 1: The Paperclip Test (Basic Power-On Check)

The paperclip test lets you check whether a PSU can power on at all — without connecting it to a motherboard.

What you need: A paperclip or short piece of wire, and ideally a multimeter.

Steps:

1. Unplug the PSU from all components and from the wall.
2. Locate the 24-pin ATX connector (the large main motherboard connector).
3. Bend your paperclip into a U-shape and insert one end into the green wire pin (PS_ON) and the other end into any black wire pin (ground).
4. Plug the PSU back into the wall and switch it on.

If the PSU fan spins, the unit is receiving power and responding to the on-signal. If nothing happens, the PSU is likely dead or severely degraded.

⚠️ Important: This test only confirms the PSU can turn on — not that it's delivering correct, stable voltages.

Method 2: Multimeter Testing (Voltage Rail Check)

A multimeter gives you real data on whether your PSU is outputting the correct voltages. This is the most reliable DIY method.

Target voltage tolerances under ATX specification:

How to test:

1. With the PSU connected and running (paperclip test or in-system), use the multimeter's DC voltage setting.
2. Place the black probe on any ground (black wire) and the red probe on the wire color matching the rail you're testing (yellow = 12V, red = 5V, orange = 3.3V).
3. Read the voltage and compare it to the table above.

Voltages consistently outside these ranges — especially under load — indicate a PSU that's failing or already underperforming.

Method 3: Software Monitoring Tools

Several tools can read voltages reported through your motherboard's sensor chip:

• HWiNFO64 — detailed and widely trusted for hardware monitoring
• HWMonitor — simpler interface, similar data
• Open Hardware Monitor — open-source alternative

🖥️ These tools display real-time voltage readings for the main rails. However, there's an important caveat: motherboard sensor accuracy varies significantly. Budget boards in particular may report voltages with meaningful error margins. Software monitoring is useful for spotting trends and major deviations, not for precision verification.

Method 4: PSU Tester Devices

Dedicated PSU tester units are inexpensive tools that plug directly into the PSU's connectors and display voltage readings via LCD or LED indicators. They're faster than a multimeter for a quick pass/fail check and don't require knowledge of which probe goes where.

They're popular for technicians and enthusiasts who test multiple units. That said, they share the same limitation as basic multimeter tests — they typically check voltages under no load or light load, which may not expose issues that only appear when the PSU is under stress.

Testing Under Load: The Critical Variable

Many PSU failures are load-dependent. A unit can pass all basic tests at idle but sag or shut down when your GPU spools up during a game or render.

To stress-test your PSU under realistic conditions:

• Run Prime95 (CPU load) and FurMark (GPU load) simultaneously
• Monitor voltages in real time with HWiNFO64 during the stress test
• Watch for voltage drops, system instability, or unexpected shutdowns

A PSU that holds voltages within spec at idle but shows significant sag under full load is a PSU that may need replacing — even if it "passed" simpler tests.

Factors That Affect What You'll Find

How you interpret your results depends on several variables unique to your setup:

• PSU age and brand tier — a five-year-old unit from a budget brand degrades differently than a newer unit from an established manufacturer
• Total system wattage draw — a PSU running near its rated capacity ages faster and shows stress sooner
• Ambient temperature and airflow — poor case airflow accelerates PSU degradation
• Modular vs. non-modular design — doesn't affect testing method, but affects connector access
• 80 PLUS certification tier — higher efficiency ratings generally correlate with better voltage regulation, though not always

When Software Says One Thing and a Multimeter Says Another

Trust the multimeter. Motherboard sensor chips are designed for general monitoring, not precision measurement. If your software shows a 12V rail reading 11.2V but your multimeter reads 11.8V, the multimeter is almost certainly closer to reality.

That said, if both tools agree that voltages are out of spec, that's a much stronger signal. 🔍

What the Tests Won't Tell You

No single test gives you the complete picture. A PSU can pass the paperclip test, show reasonable voltages at idle, and still fail catastrophically under a sustained load — especially if the capacitors are aging (a common failure mode in older units). Capacitor degradation typically shows up as instability, ripple on the voltage lines, and poor performance under load rather than clean pass/fail behavior.

This is why the combination of methods — basic power-on check, voltage measurement, and load testing — gives you the most complete view. Which combination makes sense for your situation depends on what symptoms you're seeing, what tools you have access to, and how much of the diagnostic you're comfortable doing yourself.