Check 'CrystalDiskInfo' to See If an SSD Is Dying

Why SSDs Fail Differently Than Hard Drives

Mechanical hard drives usually give you warning signs before they die: clicking noises, slow reads, bad sectors that spread visibly. SSDs fail more quietly. NAND flash cells wear out after a finite number of program/erase cycles, and once a drive's controller runs out of healthy blocks to remap around, it can go from "working fine" to "not detected" in a single power cycle. The only reliable early-warning system is S.M.A.R.T. (Self-Monitoring, Analysis, and Reporting Technology) telemetry that the drive's own controller tracks internally. Windows doesn't expose this data in a readable way anywhere in the UI, which is where CrystalDiskInfo comes in — a free, open-source utility that pulls raw S.M.A.R.T. attributes and translates them into a plain health percentage and color-coded status.

Installing and Running CrystalDiskInfo

1. Download CrystalDiskInfo from the official crystalmark.info site or via winget. Avoid third-party download mirrors — this tool is a common bundleware target.
2. Prefer the "Standard Edition (Installer)" over the sponsored/bundled variants if you use the installer, or grab the portable ZIP if you'd rather not install anything.
3. Launch the app. It scans every connected drive automatically — internal NVMe/SATA SSDs, HDDs, and most USB enclosures that pass through S.M.A.R.T. data.

# Alternative: install via winget from an elevated PowerShell prompt
winget install --id CrystalDewWorld.CrystalDiskInfo

Reading the Health Status

Each drive gets a colored status in the main window:

• Blue "Good" — health is 100% or close to it, no concerning attributes.
• Yellow "Caution" — one or more attributes have crossed a warning threshold. Not an emergency, but back up your data and keep an eye on the trend.
• Red "Bad" — a critical attribute has failed its threshold. Treat this as imminent failure and move your data off immediately.

The health percentage itself is a vendor-calculated estimate, not a precise countdown — different manufacturers weight it differently. Don't take "91%" as "91% of the drive's life remains" literally; treat it as a relative indicator and pay closer attention to the specific attributes below.

The Attributes That Actually Matter

Click a drive in the list and look at the raw attribute table. For SSDs, these are the ones worth watching:

• 05 Reallocated Sectors Count — the number of flash blocks the controller has already retired because they went bad. Zero is normal. A number that's climbing over time, even slowly, means cells are wearing out faster than expected.
• C4/197 Reallocation Event Count / Pending Sector Count — sectors currently flagged as unstable, awaiting reallocation. Any nonzero and rising value here is a warning sign.
• 05/AA/B1 Wear Leveling Count or SSD Life Left — some vendors report this directly as a percentage of rated write endurance consumed.
• 09 Power-On Hours and 0C Power Cycle Count — mostly informational, useful for knowing how old/heavily used a drive really is regardless of purchase date.
• 241/242 Total LBAs Written/Read — used to calculate Total Terabytes Written (TBW).

Understanding Total Bytes Written (TBW)

Every SSD is rated by its manufacturer for a maximum number of terabytes it can write over its lifetime before flash wear becomes a real risk — commonly somewhere between 150TBW and 1200TBW depending on capacity and NAND type (QLC drives sit at the low end, enterprise NVMe with SLC caching at the high end). CrystalDiskInfo shows "Total Host Writes" pulled from the LBAs Written attribute. Compare that number against your drive's published TBW rating (check the manufacturer's spec sheet or product page) to estimate how much endurance you've used. A drive at 40TBW written against a 300TBW rating still has plenty of life left even if it's a few years old.

What to Do If a Drive Shows Caution or Bad

1. Back up everything immediately — don't wait to see if it gets worse.
2. Run chkdsk /f on the affected volume to let Windows attempt to work around any bad sectors it can detect at the filesystem level.
3. Check the manufacturer's own diagnostic tool (Samsung Magician, Crucial Storage Executive, WD Dashboard, etc.) — these sometimes surface additional context or firmware updates that resolve false-positive warnings.
4. If the drive is still under warranty, most manufacturers will honor an RMA based on a CrystalDiskInfo or their own tool's health report showing reallocated sectors or a failed attribute.
5. Once you've migrated off a failing drive, stop using it for anything beyond disposable, non-critical data.

Wrap-Up

CrystalDiskInfo takes under a minute to install and run, and it's the fastest way to turn an SSD's opaque internal telemetry into something you can actually act on. Make checking it part of your routine maintenance, especially on drives that are a few years old or that have been used heavily for things like video editing scratch disks, VM storage, or swap/pagefile duty — workloads that chew through write endurance far faster than typical desktop use.