CHK File Documentation
Summary
A .chk file is a fragment of recovered data that Windows’ CHKDSK or the older ScanDisk utility salvaged from a damaged or improperly disconnected disk. The fragments appear in hidden folders named FOUND.000, FOUND.001 and so on, as files like FILE0000.CHK. A .chk has no format of its own: it is the raw bytes of part of a file that lost its name. To recover one you identify what the original file was and rename it to the right extension. Many fragments are incomplete, and if your real files are intact they are usually safe to delete.
Technical details
| Feature | Value |
|---|---|
| Full name | Recovered file fragment (CHKDSK / ScanDisk) |
| File extension | .chk |
| MIME type | application/octet-stream |
| Format type | Not a format — raw recovered data from lost clusters |
| Created by | Windows CHKDSK; MS‑DOS / Windows 9x ScanDisk |
| Developer | Microsoft |
| Category | Misc / recovered file |
| Typical location | Hidden FOUND.000, FOUND.001 ... folders at drive root |
| Typical naming | FILE0000.CHK, FILE0001.CHK ... |
| Magic number | None of its own; header is the original file’s (e.g. FF D8 FF, %PDF, PK) |
| Metadata retained | None — original name, extension and path are lost |
| How to recover | Identify the type from the header, then rename to that extension |
| Recovery tools | UnCHK, FileCHK, CHK‑Mate, TrID |
| Safe to delete | Usually, once your real files are confirmed intact |
| Related extensions | .bak, .tmp, .dat, .000 |
What is a chk file?
A .chk file is a fragment of recovered data produced by Windows’ disk-repair tools: CHKDSK on modern Windows, and the older ScanDisk on MS‑DOS and Windows 9x. When those tools scan a FAT or NTFS volume and find lost clusters — chunks of data the file system no longer links to any named file — they gather the orphaned clusters into files named FILE0000.CHK, FILE0001.CHK and so on, placed in hidden folders called FOUND.000, FOUND.001 at the root of the drive. The point is to give you a chance to recover data that would otherwise be discarded.
Lost clusters usually appear after a crash, a power loss, or a USB drive pulled out mid-write, which leaves the file system with allocated space that no directory entry points to. Because a .chk is just raw bytes that lost their original name and extension, it has no inherent type and no dedicated viewer. Sometimes a fragment is exactly one whole original file, so you only need to rename it; often it is a partial piece that cannot be fully reconstructed.
Lost clusters and the FOUND folder
On a FAT volume, every file is a chain of clusters recorded in the File Allocation Table, and a directory entry names the first cluster. If the directory entry is lost but the allocation chain survives, the clusters are still marked “in use” yet belong to no file: they are cross-linked or orphaned. CHKDSK’s job is to reconcile the allocation table with the directory, and when it finds these orphans it does not simply free them by default; it collects each chain into a .chk file inside a FOUND.nnn directory. NTFS keeps equivalent metadata in the Master File Table, and CHKDSK performs the same salvage there. The FOUND.000 folder and its FILE####.CHK contents are the visible result of that reconciliation, which is why a mysterious FOUND folder appears after a rough shutdown.
A fragment keeps the original bytes but loses everything else
The recovered clusters are copied verbatim, so a .chk fragment keeps the original file’s data, including its opening bytes, but loses the name, extension, path and any file-system timestamps. That surviving header is what makes recovery possible: if the fragment starts at the beginning of the original file, its magic number still identifies the format.
| First bytes of the fragment | Original file was |
|---|---|
FF D8 FF | A JPEG image — rename to .jpg |
25 50 44 46 (%PDF) | A PDF document — rename to .pdf |
50 4B 03 04 (PK) | A ZIP, or an Office .docx/.xlsx — rename accordingly |
89 50 4E 47 | A PNG image — rename to .png |
Two caveats follow from how the fragments are built. First, if a fragment begins in the middle of a file rather than at its start, there is no header to read and the type cannot be identified from the first bytes. Second, even a fragment with a correct header may be truncated, so it opens as a corrupt or half-rendered file.
Recovering a fragment: header check and rename
The whole recovery task is really one operation: identify what the fragment originally was, then restore its extension. Manually, you open the .chk in a hex or text editor, read the first bytes, match them to a known signature, and copy the file to that extension (FILE0000.CHK to photo.jpg). If you cannot recognise the header, a file-type identifier such as TrID, or file FILE0000.CHK on Linux, will name the format from its signature. For a folder full of fragments, batch tools automate the same logic: UnCHK and FileCHK (from Eric Phelps) and CHK‑Mate (DIY DataRecovery) scan a FOUND folder, detect common types by header, and restore the correct extensions in bulk. None of this transcodes anything; it only puts back the extension the fragment lost.
Can I delete the FOUND folder?
A .chk fragment is just recovered data, so inspecting one is low risk. Two cautions apply. First, a fragment could be part of any file, including a malicious one, so scan anything you reconstruct into an executable or Office document before opening it. Second, before deleting a FOUND.000 folder to reclaim space, make sure you are not discarding the only copy of data you still need: if all your real files are present and working, the fragments are usually redundant and safe to delete. Avoid sketchy “CHK file opener” download sites, which for this obscure case often bundle adware; a manual rename or the established UnCHK/FileCHK/CHK‑Mate tools are enough.
Frequently asked questions
Why do I have a FOUND.000 folder?
Windows created it when CHKDSK or ScanDisk found lost clusters — data not linked to any file, typically after a crash, power loss, or unplugging a drive mid-write. The salvaged chunks are stored there as .chk files.
How do I open a .chk file?
There is no .chk viewer; you have to find what the file originally was. Check its first bytes in a hex or text editor (or use a tool like TrID), then rename it to the right extension (.jpg, .pdf, .docx) and open it normally.
Can I recover my original files from .chk fragments?
Sometimes. A fragment that is a complete file just needs renaming to the right extension. A partial fragment may be only part of a file and not fully recoverable. Tools like UnCHK and FileCHK automate detecting and renaming the recoverable ones.
Can I delete .chk files and the FOUND.000 folder?
Usually yes. If all your real files are present and working, the fragments are leftover orphaned data and safe to delete to free space. Only keep them if you are trying to recover something you actually lost.
References
- Microsoft — chkdsk command reference
- UnCHK / FileCHK (Eric Phelps)
- TrID — file-type identifier by signature
Feedback
Was this page helpful?
Glad to hear it! Please tell us how we can improve.
Sorry to hear that. Please tell us how we can improve.