EHD / Whatever it is doing now / I/O device error and CRC

Data error (cyclic redundancy check) · I/O device error · semaphore timeout

I/O device error, and the cyclic redundancy check. Windows is telling you the drive could not read a sector, and the drive is telling you it tried very hard.

Data error (cyclic redundancy check) means the drive returned a sector whose checksum did not match, or could not return it at all. The request could not be performed because of an I/O device error means the drive did not answer a read in the time Windows allowed. The semaphore timeout period has expired is the same thing one layer down. All three are the language Windows uses for a drive with bad sectors or weak heads, and all three tend to appear while you are copying the very files you most want. The drive is still readable; it is readable in the way a drive is readable once, on equipment that can retry a sector without a whole computer timing out, and it is readable less each time it is asked. One drive is £300 + VAT after the free look, fixed in writing, 3–4 days at the bench.

Free first lookOne fixed figure in writingNo data, no bill on most jobsReturn postage paid

Rather talk it through? An engineer answers the bench line
0800 6890668

What a CRC error is, and why copying it again makes it worse.

Every sector on a hard drive carries error-correction data alongside its contents. When the drive reads a sector it checks one against the other, corrects small errors silently, and retries the read when it cannot: moving the head, re-reading, adjusting, sometimes for seconds. If it still cannot get a clean sector, it tells the computer, and Windows reports a cyclic redundancy check error. If the retries take longer than Windows will wait, Windows reports an I/O device error or a semaphore timeout instead. The message differs; the drive is doing the same thing.

The drive is doing that because the surface has worn or the head has weakened, and every retry is another pass of a weak head over a weak surface. Copying the file again asks the drive to do it all again. chkdsk /r asks it to do it for every sector on the drive. A drive that throws one CRC error on a Monday and is worked hard all week can be a drive that throws nothing but CRC errors by Friday.

On the bench the drive is imaged on equipment that controls how long a read may take and how many times it is retried, reads the healthy areas first at full speed, and comes back to the weak sectors last, in short passes, until they give up what they hold or are marked as lost. Where the heads themselves are the problem, they are replaced first. The file you were copying is recovered from the image, along with everything else.

What you see, and what it means.

Describe yours to us →

What you see	The usual reason	Where that leaves you
Data error (cyclic redundancy check)	A sector the drive could not read cleanly	Stop; image
The request could not be performed because of an I/O device error	A read that timed out	The same
The semaphore timeout period has expired	The same, reported lower down	The same
Copy stalls at the same percentage every time	A run of bad sectors in one file	Imaged with the weak area last
CRC errors, and a soft ticking	Weak heads	Head work, then imaging
chkdsk found bad sectors and the drive got worse	chkdsk hammered a weak surface	Stop; image what remains

From the box arriving to your files going back.

Work we have closed →

Logged the day it lands, and the first look costs nothing Free

A case number goes on it the day the parcel is opened, and an engineer settles what has actually failed before anything else happens. A drive that clicks or beeps is not powered again; one that spins is read on our own imaging equipment, never in a computer that will retry and time out and retry. Back to you come two things together: a straight note of what is liftable and what is not, plus one figure, fixed and written down. Accept it, or decline and owe us nothing.

Nothing to pay for lookingA single figure, put in writingNo hopeful power-ons

The enclosure and its board

Every external drive is two things: the drive, and the USB board that talks to it. The board is checked first, because a failed bridge, a snapped socket or a dead regulator is the whole fault on a good share of jobs, and because on WD's encrypted drives the board holds the key. Where the drive itself has stopped, it comes out of its case onto the bench.

Bridge, socket and power checked firstWD keys kept with their boards

The mechanics, in clean air

Stuck or failed heads are replaced with a matched set from a donor, seized spindles freed, and firmware faults corrected in the drive's service area from the terminal. The drive is not repaired to be used again; it is made to read once, in a single careful pass, and that pass is planned before it begins.

Head and spindle work on a clean benchFirmware and service-area repair

A full image, then the volume

Nothing is worked on live. The whole drive is imaged once, head by head and weak areas last, with reads timed to what the platters can stand, and the file system is put back together on the image. Encrypted volumes are unlocked there too, with the original board's key or the password you supply, and never on the original.

One careful image, head by headThe volume verified end to end

You see the file list before you pay

What was recovered is listed for you first, and only then does a bill exist. Approve the list and it is invoiced; turn it down and it is not — and where nothing has come back, most jobs carry no charge at all. Recovered data travels home on fresh media bought in for your job, with the postage at our end. Your case is not closed until you have opened the files on a machine of your own.

No charge until you accept the figureFresh media, supplied with the job3–4 days at the bench

From the bench

chkdsk /r is the worst thing to run on a drive with CRC errors. It reads every sector on the drive, and on a weak surface that is the end of it.
The file that fails is not the only file at risk. The drive that could not read it is a drive that is failing; copy nothing else, and send it.
Imaging is not copying. The bench reads sector by sector with its own limits on time and retries, healthy areas first, and comes back to the weak ones when nothing else is at stake.
An external drive on a USB bridge hides the detail. The bridge times out before the drive has finished retrying, which is why the same drive on a bench interface can read what it could not through the box.

One job, followed all the way through.

UK · EHD-2026-0479JOB LOGGED ✓

A WD Elements 4TB that threw CRC errors on the one folder that mattered

A wedding videographer's edit drive, failing on the current project and nothing else. The owner stopped copying after the second error, which is why the drive still read at all. On the bench a weak head was replaced and the drive imaged healthy areas first, then the project's sectors in short passes over a night. The project came back complete; two sectors in an old cache file did not.

99.99% of it recovered4 days here, and back by post

Illustrative example — replace with a genuine case

What helps, and what harms.

Do this much first

Stop copying at the first error
Note which file it failed on
Unplug it and send it whole
Say what matters most, so it is imaged first

What sets us back

Running chkdsk /r or a full surface scan
Copying the same file again and again
Running recovery software through the USB box
Leaving it plugged in overnight

Questions answered before you commit.

What is a cyclic redundancy check error?

The drive read a sector whose checksum did not match, or could not read it at all, after retrying. It means bad sectors or weak heads, and it means stop.

Should I run chkdsk to fix it?

No. chkdsk /r reads every sector on the drive, and on a weak surface that finishes the drive off. The bench images it with controlled retries instead.

The error only happens on one file. Is the rest safe?

The rest is on the same drive with the same heads. Copy nothing more; send it and have the whole drive imaged.

What does it cost?

£300 + VAT for one drive after the free look, fixed in writing.

How long does it take?

3–4 days at the bench after the free look; a drive with many bad sectors is at the longer end.

Where this usually leads.

Slow, freezing, disconnecting →Asks to format, or not readable →Clicking or ticking →SMR portable drives →

Nothing gets worse while it is unplugged.

Looking at it is free. Back comes a list of what opened and what did not, together with a single price to finish, set down in writing while you are still free to say no. Until that list reaches you, leave the drive unplugged.

0800 6890668

Price my job Packing and posting

Every kind of external drive

Manufacturers

Whatever it is doing now

What it costs, and what follows