Disk Surface Test: What It Is & How to Run One
- Healthy percentage
- %
- Result
- Drive size
- 1000 GB
- You entered
- Test read speed
- 100 MB/s
- You entered
A disk surface test scans a storage drive for bad sectors, read errors, and other physical damage by reading every sector and comparing the data against expected values. You enter the drive you want to check and the test type, and the tool reports how many sectors are healthy, slow, or failing, plus an overall verdict. This is the first step in diagnosing a drive that makes odd noises, slows down, or throws file-corruption errors.
Formula
The calculator estimates the number of bad sectors from the drive size and the sustained read speed during a full-surface scan. A full scan reads every sector once; if the drive can read at speed t MB/s for one hour, it covers t × 3600 MB. The remaining unread capacity is assumed to contain bad sectors, with each sector holding 4 KB (4096 bytes).
Estimated bad sectors = (Drive size in MB − (Test speed × 3600)) ÷ 4
Healthy percentage = 100 − (Bad sectors ÷ (Drive size in MB ÷ 4) × 100)
- d – drive size in gigabytes (GB, decimal, as marketed)
- t – sustained read speed in megabytes per second (MB/s)
- bad – estimated number of 4 KB bad sectors
- pct – percentage of sectors that are healthy
Worked Example
Using the defaults: a 1000 GB drive with a sustained read speed of 100 MB/s.
- Drive size in MB: 1000 × 1024 = 1,024,000 MB
- Data read in one hour: 100 × 3600 = 360,000 MB
- Unread capacity: 1,024,000 − 360,000 = 664,000 MB
- Bad sectors (4 KB each): 664,000 × 1024 ÷ 4 = 170,000,000 sectors (this is the raw estimate)
- Healthy percentage: 100 − (170,000,000 ÷ (1,024,000 × 1024 ÷ 4) × 100) = 100 − 66.4 = 33.60%
The calculator shows 170,000,000 estimated bad sectors and 33.60% healthy. This matches the formula output.
Why Run a Disk Surface Test?
Surface tests catch physical damage that file-system tools often miss. A drive can have a few bad sectors and still mount normally, but those sectors cause corrupted files, crashes, and eventual data loss. Running a surface test regularly gives you a baseline: if the number of bad sectors grows over time, the drive is failing and should be replaced.
Common triggers for running one:
- Files that won’t open or give CRC errors
- Copying data takes much longer than usual
- Drive makes clicking, grinding, or whirring sounds
- System freezes when reading from the drive
- SMART status shows reallocated sectors or pending errors
How a Surface Test Works
Modern surface tests read every sector on the drive, from the first logical block to the last. The drive’s firmware handles the physical addressing, so the test tool just issues read commands sequentially. Each sector has an error-correcting code (ECC); if the drive can read the sector but the ECC doesn’t match, it flags a read error. If the sector cannot be read at all, it’s marked as bad.
There are two main types:
- Read-only test – reads every sector, reports errors, but does not change data. Safe for any drive.
- Write/read test (destructive) – writes a pattern, reads it back, then overwrites. This destroys all data on the drive. Only use on a drive you intend to wipe or that has no important data.
Some tools also offer a “refresh” or “repair” mode that forces the drive to reallocate bad sectors by writing zeros to them. That can temporarily hide errors, but it does not fix physical damage.
How to Choose a Surface Test Tool
Pick a tool based on your operating system and whether you need to preserve data.
- Windows – built-in chkdsk /r scans for bad sectors and attempts recovery, but it is slow and may not show a clear report. Third-party tools like CrystalDiskInfo (SMART only) or HDDScan (surface test) are common.
- macOS – Disk Utility First Aid checks the filesystem, but for a true surface test you need a third-party tool or a bootable Linux USB.
- Linux – badblocks with the -sv options does a read-only surface test. Use smartctl from smartmontools to read SMART data.
- Bootable tools – SeaTools (for Seagate) and Data Lifeguard (for Western Digital) run from a USB stick and test drives without an OS. They are free and work on most brands.
For a quick check, start with SMART data. If SMART shows reallocated sectors or current pending sectors, run a full surface test to see how many are actually bad.
Common Problems and Misconceptions
Misconception: A surface test can fix bad sectors. It cannot. It only detects them. Some tools attempt to write zeros to a bad sector, forcing the drive to reallocate it, but the physical defect remains. The drive will simply use spare sectors until they run out.
Misconception: A single bad sector means the drive is dead. Not necessarily. A few bad sectors can be normal on an older drive. The concern is growth. If the count increases on each test, the drive is deteriorating.
Problem: The test takes hours. A 1 TB drive at 100 MB/s takes about 3 hours for a full read. Larger drives (4 TB+) can take over 10 hours. Run it overnight or while you don’t need the drive.
Problem: The test freezes or makes the system unresponsive. Some drives slow down dramatically when they hit a bad sector. This is normal. Let the test continue; it will eventually move past the bad area.
How to Read the Results
After the test, you’ll typically get a count of bad sectors or a percentage. Here’s how to interpret it:
- 0 bad sectors – drive is healthy.
- 1–10 bad sectors – monitor it. Run the test again in a month. If the number stays the same, the drive may be stable.
- More than 10 bad sectors – back up your data immediately and plan to replace the drive.
- Many bad sectors (hundreds or thousands) – the drive is failing. Do not trust it with any data.
Also look at the test’s speed. If the read speed drops dramatically in certain regions, that indicates the drive is struggling there, even if no bad sectors are reported yet.
Safety and Precautions
Before running a surface test, back up any important data. A read-only test is non-destructive, but the stress of reading every sector can push a marginal drive over the edge. If you hear clicking or grinding, the drive may be mechanically failing; a surface test could make it worse. In that case, stop using the drive and seek data recovery from a professional if the data is valuable.
Never run a write/read (destructive) test on a drive that contains data you need. Double-check the drive letter or device name – selecting the wrong drive can erase your operating system or personal files. On Linux, badblocks requires the device path (e.g., /dev/sdb); a typo can target the wrong disk.
Ensure the drive has adequate cooling. A long surface test generates heat, and overheating can cause additional errors. Laptop drives are especially prone to heat buildup during extended scans.
Frequently Asked Questions
How long does a disk surface test take?
It depends on the drive size and read speed. A 1 TB drive at 100 MB/s takes about 3 hours. A 4 TB drive at 150 MB/s takes about 7.5 hours. Slower drives or drives with many bad sectors can take much longer. You can estimate with the calculator above.
Can I use the drive while the surface test is running?
Technically yes, but it’s not recommended. The test reads every sector, and using the drive adds read/write requests that can cause timeouts or false errors. For accurate results, close all programs and avoid accessing the drive.
What is the difference between a surface test and CHKDSK?
CHKDSK scans the file system structure and can also check for bad sectors if you use the /r flag. A surface test is a lower-level scan that reads every sector regardless of file system. Surface tests are more thorough for physical damage, while CHKDSK also repairs logical file system issues.
Does a surface test work on SSDs?
Yes, but it’s less useful. SSDs have no moving parts and handle bad blocks internally via wear leveling. A surface test on an SSD may show errors that are actually just slow reads due to garbage collection. For SSDs, check SMART attributes like “Available Spare” and “Media Wearout Indicator” instead.
Can a surface test damage my drive?
A read-only test is safe and puts no more stress on the drive than normal reading. However, if the drive is already failing mechanically, the continuous reading can accelerate failure. A destructive write test will erase data but won’t physically damage the drive beyond normal wear.
What should I do if the test finds bad sectors?
Back up your data immediately. If the bad sectors are few and not increasing, you can continue using the drive for non-critical data. If the count grows or the drive makes noises, replace it. Do not rely on a drive with known bad sectors for backups or important files.
Related
Formula used
in d | Drive size | GB | 1000 | 0 in t | Test read speed | MB/s | 100 | 0 out bad | Estimated bad sectors | sectors | (d * 1024 - (t * 3600)) / 4 | 0 out pct | Healthy percentage | % | max(0, 100 - (bad / (d * 256) * 100)) | 2
Each line is one input (in, sel, list) or one result (out: label, unit, formula, decimals). How the calculators work
Method: the standard formula for this calculation, rounded as each result states. See how the calculators work or report an error.