I have an old HDD that started making errors, and now some files won’t open because of bad sectors. It has important photos and documents I never backed up, so I need help with the safest way to recover data without making the drive worse. Looking for advice on data recovery tools, cloning options, or whether this needs a professional service.
Hard drive throwing bad sectors, what I did first
I had this happen once, and the first mistake I almost made was keeping the drive online while I looked stuff up. Don’t do tht. If the disk is failing, every extra read or write adds stress. If it’s an external unit, unplug it. If it’s inside your PC, leave it alone and use another machine to prep your recovery steps.
Two kinds of bad sectors, and they are not the same problem
Some bad sectors are logical. Those come from corrupted data or filesystem errors. Annoying, yes, but sometimes recoverable with software.
The other kind is physical. That means the media itself has damage. Age, impact, heat, wear, any of those. If you hear clicking, scraping, grinding, or odd beeps, I would stop right there. At home fixes tend to make a bad disk worse. At that point, a recovery lab is the lane.
If the drive still shows up, don’t scan it first
This part matters more than people think.
A normal recovery scan hits the disk hard. On a weak drive, that load can finish it off. I learned to image the drive first, byte for byte, onto a healthy disk. Then I work from the copy. You get one cleaner shot at the source drive, and after tht you leave it alone.
A byte-to-byte image is a raw copy of the whole disk, including damaged areas where reads fail. The point is simple. You reduce repeated stress on the dying hardware.
The tool I’ve seen work for this
If you want one route people keep using for this, there’s Disk Drill. The useful part here is its byte-to-byte backup option. It was built for unstable drives, and newer versions handle bad blocks more carefully by skipping over trouble spots and retrying with smaller block sizes. On a flaky disk, tht behavior matters.
It also works with common filesystems like NTFS and FAT32, so you’re not boxed into one setup.
The short version of the process
- Install the recovery software on a healthy computer, not the one tied to the failing drive.
- Connect the bad drive and a good drive with enough free space for the image.
- Run the drive backup or byte-to-byte image feature.
- When the image finishes, disconnect the failing disk.
- Scan the image file, not the original hardware.
That last step is the whole point. You do your digging on the copy.
About trying to reuse the drive later
If you get your files off and you’re tempted to repair the disk, you can try. On Windows, chkdsk /r in PowerShell checks the surface and marks damaged sectors so the OS avoids them. A full format, not a quick one, also forces a sector check.
Still, once a drive starts showing bad sectors, I stop trusting it. Maybe it runs for a while. Maybe it doesn’t. I would keep nothing important on it. Temporary junk, duplicate data, stuff you already store elsewhere, fine. Primary storage, no.
When software stops helping
If the drive is missing from the system entirely, or the image gives you nothing usable, then it turns into a money decision. Pro recovery labs are expensive. I’ve seen numbers from about $500 up to $3,000, sometimes more, depending on the damage and the work involved. They have the gear to open drives safely and replace failed parts in a controlled space. That is not a kitchen table job.
Do this after the crisis passes
Set up 3-2-1 backups. Three copies of your data. Two storage types. One copy off-site.
It sounds fussy until a drive starts dying and your desktop freezes while you watch. Then it feels cheap.
3 Likes
Stop trying to open the bad files. That keeps hitting the same weak spots on the platter.
I agree with @mikeappsreviewer on one thing, don’t run repair tools first. I disagree on running CHKDSK later on a drive you care about. On a failing HDD, CHKDSK often makes a mess becuase it forces lots of reads and writes while “fixing” metadata. Save repair for after recovery, or skip it.
What I’d do instead:
-
Check SMART first with CrystalDiskInfo or smartctl.
If you see rising Reallocated Sectors, Current Pending Sector, or Uncorrectable Sector Count, treat the drive as dying. -
Keep the drive cool and stable.
Use a direct SATA connection if possible. USB bridges sometimes reset on read errors. Put a fan on the drive if it’s overheating. -
Copy the easiest files first.
Start with your photos, docs, and desktop folders. Use robocopy on Windows with limited retries:
robocopy X:\Users\YourName\Pictures D:\Rescue\Pictures /E /R:1 /W:1
This skips stubborn files faster. -
If normal copying fails, switch to image-based recovery.
Disk Drill is fine for this, especially if you want to recover files from a disk image instead of hammering the original drive over and over. -
Avoid defrag, full scans, “repair bad sectors,” and any write test.
Those are for throwaway disks, not rescue jobs. -
If the drive clicks, spins down, or disappears mid-copy, stop. Lab time.
For extra reading, this thread on recovering data from hard drive bad blocks is worth a look.
After you get the data off, retire the HDD. Don’t trust it agian.
I’d add one thing neither @mikeappsreviewer nor @sterrenkijker really emphasized enough: prioritize by irreplaceability, not folder structure.
If the drive still mounts, don’t start with a whole-disk job unless the drive is actively getting worse by the minute. I know that goes a bit agianst the “image first always” crowd, but on some dying drives you can waste the last readable hours cloning junk sectors and old installers while your family photos sit untouched. I’d manually grab the absolute must-have stuff first: photos, tax docs, scans, anything impossible to replace. Then do imaging.
Also, avoid Windows Explorer for bulk copying if it hangs the whole system. Something like TeraCopy or even robocopy with low retries is less painfull. If the drive starts slowing hard, vanishing, or making new noises, stop. That’s the point where DIY turns into data destruction.
After the priority copy, then yes, Disk Drill makes sense because you can create a byte-level backup and scan that instead of repeatedly poking the original disk. That’s the safer recovery workflow once the easy wins are gone. If you want a simple walkthrough, this Disk Drill data recovery guide for damaged hard drives is easier to follow than most written guides.
One more thing: don’t trust SMART too much if it says “good.” I’ve seen drives die very politely while reporting everything was fine. If files are already unreadable, the drive is already on probation. Recover what you can, then retire it.
