How to Check Disk Health in Windows 10 and 11: SMART, PowerShell and CrystalDisk

Drives fail without warning. A quick health check takes two minutes and can save you from data loss. Here's how to do it properly.

Quick Check: SMART Status

SMART (Self-Monitoring, Analysis and Reporting Technology) is built into every modern drive. Windows can read it:

wmic diskdrive get status, model, size

OK means no SMART failures detected. Anything else — Pred Fail, Bad, Unknown — means the drive is reporting problems.

For more detail, use PowerShell:

Get-PhysicalDisk | Select-Object FriendlyName, MediaType, OperationalStatus, HealthStatus, Size

Detailed SMART Data with CrystalDiskInfo

The built-in tools only show pass/fail. CrystalDiskInfo (free, no install required) shows individual SMART attributes:

Download from crystalmark.info → run → look at the color indicators:

🔵 Good — healthy
🟡 Caution — some attributes degraded, monitor closely
🔴 Bad — replace soon

Key attributes to watch:

Reallocated Sectors Count — bad sectors the drive has remapped. Any value above 0 is concerning on an HDD.
Pending Sectors — sectors waiting to be reallocated. Should be 0.
Uncorrectable Errors — should be 0.
SSD: Wear Leveling Count / Media Wearout Indicator — shows remaining SSD lifespan.

Run chkdsk

chkdsk scans the filesystem for errors and marks bad sectors:

chkdsk C: /f /r /x

/f — fix filesystem errors
/r — find bad sectors and recover data
/x — dismount volume first (required for system drive)

On the system drive (C:), it can't run while Windows is active. It will schedule for next reboot:

Would you like to schedule this volume to be checked the next time the system restarts? (Y/N)

Press Y, then restart. Takes 20–60 minutes.

Check chkdsk Results

After it runs, find the results:

Get-EventLog -LogName Application -Source "Microsoft-Windows-Chkdsk" -Newest 1 |
  Select-Object -ExpandProperty Message

Or: eventvwr.msc → Windows Logs → Application → find source Chkdsk or Wininit.

Check Disk Performance

A sudden drop in speed can indicate a failing drive before SMART flags it:

# Built-in Windows speed test
winsat disk -drive c

Typical results:

HDD sequential read: 100–200 MB/s
SSD sequential read: 300–500 MB/s (SATA), 1000–7000 MB/s (NVMe)

Significantly lower numbers than expected = drive is struggling.

SSD-Specific: Check Total Bytes Written

SSDs have a finite write lifespan. Check how much has been written:

In CrystalDiskInfo, look at Total Host Writes or Total Bytes Written. Compare to the manufacturer's TBW (Terabytes Written) rating for your model.

For example: a 500GB Samsung 870 EVO is rated for 300 TBW. If yours shows 280 TBW written, it's near end of life.

HDD-Specific: Listen for Sounds

Mechanical drives give audio warnings before failure:

Clicking — read/write head failing. Back up immediately.
Grinding — physical damage. Stop using it.
Repeated seeking (constant activity light) — bad sectors, drive struggling.

SSDs fail silently — rely on SMART data, not sounds.

What to Do If the Drive Is Failing

Back up everything now — don't wait
For HDD: run chkdsk /f /r to mark bad sectors so they're avoided
For SSD: check if it's still under warranty (most SSDs have 3–5 year warranties)
Replace the drive before it fails completely — data recovery from a dead drive costs hundreds of dollars

⚡ Шукаєш потрібну команду?

→ PowerShell і CMD довідник — 40+ команд з пошуком за задачею. Введи "мережа", "диск" або "безпека" і одразу отримай готову команду.

Summary

Quick check: wmic diskdrive get status — if not OK, act immediately.
Detailed check: CrystalDiskInfo for SMART attributes.
Filesystem check: chkdsk C: /f /r scheduled on next boot.
Performance check: winsat disk -drive c

Check drive health every 6 months. Don't skip it for SSDs — they fail without warning too.