Drive not showing up? Don't power it on again.
Recovering Files From a Failed or Dead Hard Drive
If the drive doesn't appear, makes a clicking sound, or stopped after a drop or power cut, the most important thing is to not keep trying. Every extra power-on can increase the damage.
- +25
- Years of experience
- +50K
- Cases handled
- 99%
- Success rate after diagnosis
That strange noise is not a small detail.
When do you need a hard drive inspection?
- 01
Drive doesn't appear
The system doesn't recognise the disk or asks to format it on connection.
- 02
Slow reads or missing files
Files won't open, disappear suddenly, or the disk takes very long to read.
- 03
A drop or clicking sound
Signs that require powering off immediately and avoiding any home attempt.
Inspection comes before any recovery attempt.
Can files be recovered from a drive that won't work?
Recovery is possible in many cases, but the verdict isn't made from the outside. The fault may be logical, electronic or physical; correct handling starts with inspecting the case and choosing a path that doesn't stress the damaged disk.
Inspect the case
Choose the path
Safe recovery
Supported devices & brands
Drives and brands we work on
Each drive family behaves differently once it fails. This is what we account for before any read attempt.
Seagate
Seagate drives keep their firmware modules and translator table in the service area. A drive can spin up perfectly and still report the wrong capacity when that area is damaged, so the service area is read and assessed before we go anywhere near user data.
Western Digital
Many WD external drives encrypt the data inside the USB bridge board itself. Pulling the bare disk out and reading it directly can return unreadable data, so the original board stays with the drive throughout the work.
Toshiba and HGST
Toshiba 2.5-inch drives are common in laptops and external cases; HGST turns up more often as a 3.5-inch server disk. When replacement parts are needed, they are matched by family and firmware revision, not by the model number printed on the label.
2.5-inch and 3.5-inch
2.5-inch drives are the ones most often dropped while running. 3.5-inch drives carry more platters and more heads and run off a 12 V rail, so damage spreads across more surfaces and a power cut usually hits the board first.
SMR drives
SMR drives write overlapping tracks and manage their own internal map and media cache. Repeated retries push the drive into reorganising those bands, so it is imaged in a single calm pass and its map is interpreted rather than ignored.
External USB drives
Some enclosures hold an ordinary SATA disk; others carry a native USB board with no SATA connector at all. The fault is often in the board or the port rather than the disk, and inspection settles that before anything is opened.
How a hard drive is handled in the lab
- 01
Electrical check before any power-on
The drive arrives powered off. The board, the protection components, the spindle motor and the power path are checked first. The drive is not spun up until the electrical path is confirmed clean.
- 02
One short, observed spin-up
A single short power-on answers three questions: does the drive reach ready, does it report the correct model and capacity, and can the service area be read. If the heads sound wrong, power stops immediately and the work moves into a clean-air environment inside the lab.
- 03
Sector-by-sector imaging to a separate disk
The drive is imaged sector by sector onto a separate disk, with controlled read timeouts and a separate pass per head: healthy zones first, tired zones on a later pass. The point is never to leave the drive stuck on a damaged area.
- 04
Rebuilding the files from the image
Everything after that happens on the image, never on the original disk: the file system is rebuilt, samples are opened to confirm they are intact, and the data is handed back on a separate medium.
Start from the symptom
What kind of hard drive fault is it?
Each kind has a different route, and the common mistake is treating a physical fault as a software one.
- 01
Logical fault
The drive appears normally but the partition has vanished, or shows as RAW, or the system asks to format it, or files and folders were deleted. The priority is preventing any new write, then working from an image.
- 02
Unstable reading
It appears and disappears, or the speed drops sharply, or the machine freezes when opening a particular folder. Ordinary copying can consume whatever time the drive has left.
- 03
Electronic fault
It may stop after a power cut, an unsuitable adapter, or damage to the protection circuit. Swapping the board at random does not bring the data back: some calibration data belongs to that specific drive.
- 04
Mechanical fault
Repeated clicking, a grinding sound, no spin, or a stop after a fall. These call for powering down — not for running recovery software.
- 05
Water, fire or impact
Do not power it on, open it, or dry it with heat. Record what happened and leave it as it is. The casing, the circuit, or the internal components may be damaged to different degrees.
- 06
External drive not showing
The problem may be the cable, the enclosure or the USB port rather than the drive. A lit indicator says nothing about the drive's health. Send it with its original cable and adapter.
No single tool
How is the recovery route chosen?
If the drive is stable, a full read-only image is taken and the file system is then analysed on that image. If reading is unstable, the priority shifts to stabilising access and copying the important sectors before analysing files. An abnormal noise or a failure to spin calls for a physical assessment before any further power-on.
The first aim is not to open one folder quickly but to protect as much of the source as possible. Once the image exists, the partition and folder structure can be searched, then a sample of images, documents, videos and databases checked for integrity before any result is presented.
Between the extremes
Common cases and likely outcomes
Not promises — an account of what actually happens, depending on the drive's condition and what was done after the fault.
- 01
External drive not showing
The problem may be the cable, the enclosure or the port, and it may be the drive itself. Power source and circuit are examined before any other step.
- 02
The drive asks to be formatted
Damage to the partition table or file system may be the cause. Do not agree to format, and do not run CHKDSK if the files matter.
- 03
Very slow, files will not copy
The case may involve failing sectors. Repeatedly copying the same folder or running a long surface test puts the drive under strain. Recovery may be partial, so critical files are ordered first.
- 04
Clicking after a fall
A likely physical case. Stop powering it and do not try one cable after another. Every spin-up attempt can add damage, and the odds are established after examining the drive, not after hearing the noise over the phone.
- 05
Exposed to water or high heat
The circuit alone may be damaged, or the damage may have reached inside. Do not use rice or a hairdryer, and do not open the casing.
What we need for the initial assessment
- 01
Drive type
Internal or external, and whether it is an HDD or unconfirmed.
- 02
Make, model and capacity
With a clear photograph of the label if possible.
- 03
Does the system see it?
In BIOS or Disk Management, and whether it reports the right capacity.
- 04
Current noise
Normal, clicking, beeping, grinding, or not spinning.
- 05
The last event before the fault
A fall, power, water, slowness, formatting or deletion.
- 06
What was tried afterwards
Software, a board swap, or running CHKDSK.
- 07
Folder priority
Which folders or file types should come first.
- 08
Is the drive encrypted?
BitLocker or any other encryption, and the recovery key if you have it.
FAQ
Before you power it on again, read this.
The noise may indicate a physical fault, so the drive should be stopped and not powered on before inspection.
Yes, the cable, enclosure, electronic board and the disk itself can be inspected to find the cause.
It may help with simple deletions, but it can be dangerous with a physical fault or unstable reads.
Not always. The board may be healthy while the problem is internal, and some drives need calibration data tied to the original drive. Swapping at random can add a new fault.
No. Moisture and condensation can cause further damage, and this is not a safe recovery route.
If the data matters, do not start repairs before an image is taken. Repair tools are built to make a file system usable, and they may delete or alter records we need for recovery.
The structure may return intact if the file-system metadata survived. If it was damaged, files may be recovered by type without their original names or folder order.
It can be assessed, but opening the data needs the correct password or recovery key. Fixing a physical fault does not remove the encryption.
Writing to a damaged source is not advisable. Data is delivered on a separate medium after a documented sample is verified.
It depends on the fault type and disk size; logical cases can finish within a day or two, while mechanical faults take longer. We give a clear estimate after diagnosis.
In many cases yes—provided you don't power it on or try to dry it yourself. Leave it as is and bring it in for inspection quickly.
Yes. Every case is handled with full confidentiality and delivered on a separate medium, with an NDA available on request.
Next step
Send case detailsEvery extra power-on can change the outcome.
Stop the device and send your case details. We start by inspecting it and explain your recovery odds and the best next step before starting.