What to do right now
- If the drive clicks, disconnects or freezes, skip the scan and create a drive image first.
- Otherwise check the health indicators, then run a read-only surface scan, starting with a quick sample.
- Unreadable blocks mean copy important files now and plan a replacement; slow areas mean back up and scan again later.
Step by step
Decide whether the drive can take a scan
A full scan reads the whole disk, which takes hours on a large hard drive. If the drive clicks, drops off the system or makes Windows hang, do not scan it. Use Create a drive image (skip bad sectors) and save the image on a different physical disk.
Look at the health indicators first
Open Disk health & tools in FileFix Ultimate and select the drive. Reallocated sectors, pending sectors and uncorrectable errors are the drive's own record of trouble. A non-zero pending count is already a reason to back up first.
Set up the surface read scan
Choose Surface read scan, pick the drive and set the media type to Hard disk. For a first look choose Quick sample, which reads one of every 16 blocks. Choose Every block later for a complete map, and set a time limit if the drive is old.
Start the scan and watch the map
Press Start scan. The scan only reads and by default does not retry failed reads, so it does not hammer weak spots. Each map cell shows how long a block took to read. If read errors pile up, the scan stops by itself and recommends a drive image.
Read the result
The result names unreadable blocks, slow areas, or no read problems in the scanned range. A sample only covers the blocks it read, so a clean sample is a good sign, not a clean bill of health. Save the JSON report to compare with a later scan.
Act on what you found
With unreadable blocks, copy the files you need to another disk or image the drive, and plan to replace it. With only slow areas, keep a fresh backup and repeat the scan in a few weeks to see whether they grow.
Slow areas and unreadable blocks are different signals
A healthy hard drive returns each block in a few milliseconds. When a sector is weak, the drive's firmware retries and applies error correction before it answers, so the read still succeeds but takes far longer. Clusters of slow reads often appear before outright failures, which is why the scan draws a latency map instead of only counting errors.
An unreadable block is a read that failed even after the drive's own retries. The data there may already be lost, and any file using it can be partly damaged. A few unreadable blocks in one place can come from a single defect; many spread across the disk usually mean the drive is wearing out as a whole.
Why CHKDSK /r is the wrong first step on a weak drive
CHKDSK /r reads every sector, retries the ones that fail, marks bad clusters in the file system and moves readable data away from them. On a weak drive that means hours of heavy reading on exactly the weakest spots, plus writes to the file system. If the drive gives out halfway, you may be left with less than before. Copy or image first, recover from the copy, and only then repair or replace the drive. The guide on CHKDSK before recovery covers that decision in more detail.
Physical and logical bad sectors
A physical bad sector is damaged surface the heads can no longer read reliably. Software cannot repair it; the drive can only swap in a spare, which shows up as a reallocated sector in SMART. A logical bad sector has error-correction data that no longer matches, often after a power cut during a write, and it usually works again once overwritten.
Both look the same in a read-only scan. Tools that fix logical bad sectors do it by writing to them, which destroys what was there, so leave any write test until your data is safely elsewhere.
What a surface scan cannot do
- It cannot repair bad sectors or bring back data stored in them; it only shows where they are.
- A quick sample reads one block in 16, so problems between sampled blocks can go unseen.
- On SSDs the scan only detects slow areas; software cannot repair flash cells.
- A drive with mechanical damage, such as clicking, should not be scanned at all; if the data matters, that is a case for a recovery lab.
How FileFix Ultimate helps
- Surface read scan is read-only, does not retry failed reads by default and stops by itself when read errors pile up.
- Shows the result as a latency map and saves a JSON report for later comparison.
- Explains reallocated, pending and uncorrectable sector counts in plain language.
- Create a drive image (skip bad sectors) copies good areas first, keeps a .map file to resume, and lets you recover from the image.
- Free for personal use and offline, with no limit on how much you recover.
FileFix Ultimate is free for personal use, works offline and never writes to the drive you are recovering from. Business use needs a one-year licence per PC. Check the signature before you run it.
Frequently asked questions
How long does a full surface scan take?
It depends on capacity and speed. A hard drive often reads roughly 100 to 250 MB per second, so a 1 TB drive usually takes one and a half to three hours, longer with many slow areas. Quick sample finishes much sooner.
Can bad sectors spread?
They do not spread like an infection, but the cause often keeps working. Loose particles or a weakening head can damage more surface over time. A count that rises between two scans is the real warning sign.
Does a surface scan change anything on the drive?
No. Surface read scan only reads blocks and measures how long each read takes. It does not mark, remap or rewrite anything.
Is the Check button in drive Properties a bad sector test?
Not really. Error checking under Properties › Tools mainly checks the file system structure. It does not read every sector the way a surface scan does.