Read/Write Head
The read/write head is the business end of a hard drive. A tiny electromagnetic transducer (between 0.1 and 0.3 millimeters across) sits at the end of an actuator arm, flying just nanometers above the spinning platter surface. The read element detects existing magnetic patterns; the write element changes them. When heads fail, the drive typically becomes unreadable instantly, often producing the rhythmic clicking sound known as the click of death.
SERT ¡ Cheadle
Single-digit nm range
TMR-default era
A read/write head is a tiny electromagnetic transducer at the end of a hard drive’s actuator arm that detects (read) and modifies (write) the magnetic polarization of the platter’s recording surface. Each platter recording surface in an HDD has its own dedicated read/write head, all mounted together on a single moving structure called the Head Stack Assembly. Modern read/write heads are extremely small (0.1 to 0.3 millimeters across) and operate at a flying height of 3 to 6 nanometers above the platter surface, supported by a thin air bearing created by the platter’s rotation.
What a Read/Write Head Is
The read/write head is the component that turns the rotating platter into useful storage. Without the head, the platter is just a spinning disk with magnetic patterns on its surface that nothing can detect or change. The head provides the bridge between the magnetic patterns on the platter and the electrical signals the rest of the drive’s electronics can process.1
The scale of a head
The Gillware Data Recovery 101 documentation describes the size: read/write heads are tiny sensors about 0.1 to 0.3 millimeters across. For comparison, a typical grain of sand is around 0.5 millimeters; a head is smaller than a sand grain. The head is mounted on a slider (a small carrier that controls the head’s flying behavior) which is in turn mounted on a suspension (a thin metal arm) that connects to the rotating actuator structure.
The flying height
Read/write heads do not touch the platter during normal operation. Instead, they fly above the platter surface on a thin cushion of air created by the platter’s rotation, called an air bearing. The ACS Data Recovery documentation notes the modern flying height: “In modern hard drives the heads fly above the disk surface with clearance of less than 6 nanometers.” The Cheadle Data Recovery analogy makes the scale tangible: “If the read-write head were the size of a jumbo jet it would fly 1.5mm off the ground.” The clearance is so small that any disturbance (vibration, dust particle, manufacturing imperfection) can cause the head to contact the platter, producing damage on both surfaces.
One head per surface
Each platter has two recording surfaces (top and bottom), and each surface has its own dedicated read/write head. The relationship is fixed: head 0 reads platter 0’s top surface, head 1 reads platter 0’s bottom, head 2 reads platter 1’s top, and so on. A three-platter drive has six heads; a four-platter drive has eight heads. The heads are mounted at different vertical positions on the actuator structure to align with their respective platter surfaces.
Higher capacity drives have more heads
The Gillware documentation captures the scaling: “The number of heads depends on the storage capacity of the drive. Higher capacity drives have more heads than lower capacity drives.” This is true at the platter level too: each platter contributes capacity, and more platters mean more heads. The relationship matters for recovery because head failures can affect different parts of the drive’s data depending on which surface’s head failed.
How Read/Write Heads Work
Modern read/write heads have separate elements for reading and writing, co-located on the head but operating through different physical mechanisms. Understanding the difference clarifies why the heads can fail in distinct ways for read versus write operations.2
The read element
The Gillware Data Recovery 101 documentation describes the read mechanism: “When reading data, the platters magnetic field induces an electrical current in the read/write head coil. This current is interpreted by the HDD’s controller.” Modern drives use Tunneling Magneto-Resistive (TMR) sensors or Giant Magneto-Resistive (GMR) sensors, both of which:
- Change electrical resistance based on the magnetic field directly beneath the sensor.
- Produce a varying voltage signal as different magnetic patterns pass underneath during platter rotation.
- Are extremely sensitive (necessary because the magnetic fields they’re detecting are very weak).
- Are also extremely fragile (the sensitivity comes from delicate nano-scale structures).
The write element
The write element is a small inductive coil. When the controller wants to write a bit, it energizes the coil with electrical current, which produces a magnetic field strong enough to align the magnetic domains on the platter directly beneath the head into a specific orientation. The orientation encodes the bit value: one direction is a 0, the opposite direction is a 1. Modern write heads use perpendicular magnetic recording, which orients the magnetic domains vertically (perpendicular to the platter surface) for higher density.
The preamplifier
The signals the read element produces are extremely weak, far too weak to travel through the drive’s flex cable to the main controller without amplification. The Cheadle Data Recovery documentation explains: “The pre-amplifier (or preamp) is a chip which controls the read-write heads and amplifies signals to and from them. The signals generated by the read-write heads are very weak which is why this component is necessary.” The preamp sits on the head stack assembly itself, very close to the heads, to keep the signal path short. Preamp failures cause symptoms similar to head failures (the heads themselves work fine but their signals can’t be processed).
Static sensitivity
The Cheadle documentation notes: “Preamps cannot withstand static discharges so precautions are required when there is any handing or repair work.” This is one of the reasons DIY drive opening is so destructive; a single static discharge from the user’s hand can permanently destroy the preamp, eliminating any chance of recovery without specialized equipment to replace it.
Servo information
Beyond user data, the heads also read servo information embedded on the platter that tells the drive where the heads are positioned. Servo bursts are short magnetic patterns at known intervals around each track; by reading them, the drive’s firmware knows exactly which track the head is over and how to keep the head correctly centered on the track. When servo information can’t be read (because of a head problem or platter damage at the servo location), the drive can’t position the heads correctly and behaves erratically.
The Head Stack Assembly
All of an HDD’s read/write heads are mounted together on a single moving structure called the Head Stack Assembly (HSA). The HSA is the unit that gets replaced during head failure recovery; individual heads can’t typically be replaced without disturbing the precise alignment of the entire assembly.
Components of the HSA
A complete HSA includes several subassemblies:
- Actuator pivot bearing: the rotating bearing at the base where the entire assembly pivots to position heads over different tracks.
- Actuator arm: the rigid arm that extends from the pivot toward the platter; carries the suspension and head at its tip.
- Suspension: the thin flexible metal arm that holds each head at the correct angle and applies the small downward force that, balanced against the air bearing’s lift, keeps the head at the right flying height.
- Slider: the small carrier that the head is mounted on; controls the air bearing geometry that determines flying behavior.
- Read/write head: the actual sensor at the slider’s tip.
- Flex cable: the ribbon cable that carries signals between the heads and the controller.
- Preamplifier chip: mounted on the flex cable, amplifies head signals before they leave the HSA.
All heads move together
The HSA pivots as a single unit; all heads move together. When the actuator pivots so that head 0 is over track 1000, head 1 is also over track 1000 of its surface, head 2 is over track 1000 of its surface, and so on. The drive can switch between heads electronically without physical movement; the head selection is done in the preamp by activating one head’s electrical path. This is why drives can read from different surfaces in rapid succession without seeking; only switching tracks requires physical movement.
The head stop
The actuator’s range of motion is limited by mechanical stops at both ends. When a drive can’t successfully position the heads (because of a head failure or other servo problem), it may sweep the heads from one extent of motion to the other; the heads hitting the stops produces the rhythmic clicking sound. The Gillware documentation describes this: “When heads fail, the HDD will often times make a clicking sound, which is the result of the head stack assembly flying blindly from one extent of the platter to the other. When the heads encounter the head stop, they make the clicking noise.”
These sounds typically indicate read/write head failure. Continuing to power the drive risks the failed heads contacting the platters, scraping off magnetic coating that holds your data. The dust kicked up by a head-platter collision can also damage other heads and surfaces in the drive, expanding a recoverable failure into an unrecoverable one. Disconnect the drive from power and do not attempt repairs at home; head replacement is exclusively a cleanroom procedure.
How Read/Write Heads Fail
Read/write head failures are the most common physical failure mode in hard drives, and the failure modes vary depending on what specifically went wrong. Understanding the failure modes helps explain the diagnostic patterns and the recovery prospects.3
Why heads fail first
The Gillware documentation captures why heads are typically the first components to fail: “Read/write heads are the most sensitive and delicate part of your hard drive… These components work harder than any other part of the hard drive. Unfortunately, they can also be the first to wear out as your hard drive ages.” The combination of microscopic scale, continuous mechanical operation, magnetic and electrical sensitivity, and proximity to the spinning platter makes heads uniquely vulnerable. Any other component can fail too, but heads fail more often.
Common failure modes
| Failure mode | Symptoms | Recovery prospects |
|---|---|---|
| Stuck heads | Drive doesn’t spin up; sometimes whining sound | Often recoverable with cleanroom unsticking procedures |
| Weak heads | Drive reads but with errors; slow performance; bad sectors | Recoverable; head replacement or imaging-around-weakness |
| Failed heads | Drive doesn’t initialize; clicking; unrecognized by computer | Recoverable via head replacement if no platter damage |
| Bent suspension | Drive operates erratically; intermittent failures | Difficult; suspension can’t be straightened, requires HSA replacement |
| Contamination | Various symptoms; often clicking | Variable; depends on contamination source and severity |
| Preamp failure | Drive recognizes but can’t read; like failed heads | Requires HSA replacement (preamp is part of HSA) |
The contamination cascade
The SERT Data Recovery documentation describes a critical failure pattern: “When there is a head failure it is likely to cause platter damage and kick up dust that will float throughout the drive and cause the other heads to fail.” A localized initial failure can become a total failure as the dust contaminates additional surfaces and heads. This is why operating a drive after the first signs of head failure (unusual noises, performance degradation, error messages) typically makes recovery harder; each minute of additional operation potentially expands the damage.
Multiple heads can fail independently
The SERT documentation notes: “Just because there is a read/write head failure, doesn’t mean they have all failed. One or more can fail at any time, and the rest may operate just fine.” Some drives may have one failed head with the others functioning normally; in such cases, parts of the drive’s data may be readable while other parts (the surfaces served by the failed heads) are not. This pattern is often the first sign of impending total failure; once one head has failed, the contamination cascade often takes the rest within hours or days of continued operation.
The 2007 sound change
The Cheadle Data Recovery documentation notes a generational shift: “In general in hard disk drives from 2007 onwards the clicking sound is much quieter, and some models of hard disk drive will spin down after a few seconds if it has failed to initialize.” Modern failure-detection logic shuts down the drive proactively rather than continuing to click indefinitely. Older drives clicked loudly and continuously; modern drives may produce only a few quiet clicks before shutting down. The sound change doesn’t reflect any improvement in head reliability; it reflects better failure handling firmware.
Head Replacement and Recovery
Head replacement is the most common physical recovery procedure performed in cleanroom labs. The procedure is delicate and requires specialized equipment, but when done correctly it allows recovery of data from drives that would otherwise be unreadable.4
The donor drive requirement
Head replacement requires a donor drive that matches the failed drive’s exact specifications. The Cheadle Data Recovery documentation captures the procedure: “To recover the data when there is a failure of the read-write heads it is necessary to replace the head assembly of the failed hard disk drive with that of a working donor.” The match must be exact: same manufacturer, same exact model number, same firmware revision, ideally same date code. A 1 TB drive’s HSA won’t work in a 2 TB drive of the same family; firmware revision changes can break compatibility even within identical model numbers.
The replacement procedure
Inside a cleanroom (typically Class 100 or better), the recovery engineer:
- Opens both the failed drive and the donor drive.
- Carefully removes the donor’s HSA using specialized fixtures that prevent the heads from contacting the platters during removal.
- Removes the failed drive’s HSA using similar fixtures (the failed heads need to come out without scraping the platters).
- Installs the donor’s HSA into the failed drive, maintaining alignment of all heads relative to all platter surfaces.
- Closes the drive in cleanroom conditions to prevent contamination.
- Powers up the drive to test for functional reads.
- If reads work, immediately images the drive to a known-good destination (the replacement heads typically have limited remaining lifespan in the failed drive).
Why DIY replacement fails
Multiple recovery sources note unanimously that DIY head replacement essentially never succeeds. The Gillware documentation states: “There are no DIY solutions to hard drive heads failure, and chances are extremely slim that anybody without the proper tools, even if they know computers backward and forwards, can reliably repair HDDs with failed heads.” The reasons:
- Cleanroom contamination control isn’t possible in normal environments.
- Heads almost certainly contact platters during DIY removal/installation, producing scratch damage.
- Static discharge from human hands typically destroys the preamp.
- Donor drive matching is harder than it appears (firmware revisions, date codes).
- Specialized fixtures aren’t available outside professional recovery labs.
- Even successful DIY swaps often produce more damage than they fix.
Imaging quickly after replacement
Replacement heads in a failed drive typically have limited operational lifespan because the slight remaining contamination from the original failure tends to wear the new heads faster than normal. Cleanroom recovery workflows therefore prioritize fast imaging immediately after head replacement: get a complete byte-by-byte image of the drive while the new heads are still functional. The image becomes the source for actual file recovery; the drive itself may not survive long after replacement.
Tools used in the procedure
Cleanroom labs use specialized tooling for head work:
- Head combs: small fixtures that hold the heads safely retracted away from the platters during HSA removal and installation.
- Platter lifts: tools for raising platters off the spindle when access to lower heads is needed.
- Vacuum chucks: for handling components without finger contact.
- Imaging hardware: specialized imagers (PC-3000, DeepSpar Disk Imager, Atola Insight) that can image drives with weak or partially-failed heads more reliably than standard hardware.
Read/write head failures are the dominant scenario in cleanroom data recovery. Most physically failed drives sent to professional recovery labs are head-failure cases, and head replacement is the most common procedure performed. The recovery prospects depend heavily on whether the platters are intact: drives where the heads failed cleanly without contacting the platters typically recover well, while drives where head failure caused platter damage have substantially lower recovery rates.5
For users facing potential head failure (clicking sounds, drives that don’t initialize, drives that report bad sectors and refuse to mount), the practical guidance is consistent: stop using the drive immediately, don’t attempt DIY repair, and bring it to a cleanroom recovery service. The reasons are compounding: continued operation risks the failed heads contacting platters and causing irreversible damage; DIY attempts almost always destroy any remaining recovery prospects; cleanroom services have the tools and expertise that home environments simply can’t replicate. The cost differential is real (cleanroom recovery costs hundreds to thousands of dollars; DIY costs nothing in materials), but the success rate differential is also real (cleanroom recovery often succeeds; DIY essentially never does on physical head failures).
For users wondering whether to choose HDDs or SSDs based on physical reliability, the head-failure failure mode is uniquely an HDD concern. SSDs have no moving parts, no air bearings, no microscopic mechanical tolerances; their failure modes are different (controller failures, NAND wear-out, TRIM-related data loss) and have different recovery paths. For users with workloads that involve frequent drive movement or potential physical shock (laptops in active use, external drives in transit), SSDs largely eliminate the head-failure scenario. For static desktop use, modern HDDs remain reliable enough that head failure isn’t an everyday concern, but the failure mode exists and explicit backups remain the best protection. Recovery software handles logical failures; physical head failures require the cleanroom path that no software product can substitute for.
Read/Write Head FAQ
A read/write head is a tiny electromagnetic transducer at the end of a hard drive’s actuator arm that detects and modifies the magnetic polarization of the platter’s recording surface. The read function detects the existing magnetic patterns on the platter; the write function changes those patterns to encode new data. Modern read/write heads are extremely small (0.1 to 0.3 millimeters across) and operate at a flying height of 3 to 6 nanometers above the platter surface, supported by a thin air bearing created by the platter’s rotation. Each platter recording surface in an HDD has its own dedicated read/write head; a three-platter drive has six heads.
Modern read/write heads have separate elements for reading and writing. The read element is typically a Tunneling Magneto-Resistive (TMR) sensor that changes its electrical resistance based on the magnetic field of the platter directly beneath it; as the platter rotates and the magnetic patterns pass under the head, the sensor’s resistance changes accordingly, generating an electrical signal that the drive’s electronics decode into binary data. The write element is a thin-film inductive coil that, when energized with electrical current, produces a magnetic field strong enough to align the magnetic domains on the platter into specific patterns representing 0s and 1s. The two elements are co-located on the head but are separate transducers with different electrical paths. The signals from both elements are amplified by a preamplifier chip on the head stack assembly before being processed by the main HDD controller.
Read/write heads are the most sensitive component in an HDD and one of the most common failure points. Several factors contribute to their fragility: the heads operate at extraordinarily small flying heights (3-6 nanometers, comparable to a jumbo jet flying 1.5 millimeters off the ground), making them vulnerable to physical shock and vibration; the heads work continuously during drive operation, accumulating wear over time; the heads are sensitive to contamination from any debris that gets inside the drive’s hermetic enclosure or that’s kicked up by previous platter damage; and the read elements use delicate magneto-resistive structures that can degrade with age or exposure to magnetic interference. As Gillware’s documentation notes, read/write heads work harder than any other part of the hard drive and are often the first components to wear out.
Rhythmic clicking from a hard drive (the ‘click of death’) is most commonly produced by failed or failing read/write heads. The mechanism is straightforward: when the heads can’t successfully read the servo information that tells them where they are on the platter, they execute a recalibration sweep that moves the head stack assembly from one extent of the platter to the other; when the heads hit the mechanical head stop at the limit of motion, they produce the characteristic click. The drive may attempt this sequence repeatedly, producing the rhythmic clicking pattern. Older HDDs (pre-2007) produced louder clicks; modern drives are quieter but the pattern remains recognizable. Clicking sounds are a strong signal that the drive needs immediate attention from a cleanroom recovery service; continued operation typically makes the situation worse as failed heads can damage the platters.
Effectively no, despite some online tutorials suggesting otherwise. Read/write head replacement requires a cleanroom environment to prevent contamination, specialized fixtures to handle the head stack assembly without bending the suspension, and a matching donor drive with not just the same model but the same firmware revision and often the same date code. Without all three, the replacement either fails immediately (heads bent during installation) or appears to work briefly before causing additional platter damage that destroys remaining recoverable data. Multiple recovery services note that DIY head replacement attempts before professional intervention are among the most common reasons recoverable drives become unrecoverable. Cleanroom recovery costs significantly more than DIY attempts but has a vastly higher success rate; for any drive containing important data, professional recovery is the only reasonable approach.
The Head Stack Assembly (HSA) is the complete moving structure that holds all of an HDD’s read/write heads. It includes the actuator pivot bearing, the actuator arm, the suspension that holds each head at the correct position, the heads themselves at the end of each suspension, the flex cable that carries signals between the heads and the controller, and the preamplifier chip that amplifies the heads’ weak signals. All heads in the HSA move together; when the actuator pivots to position one head over a specific track, all the other heads move to corresponding tracks on their respective platter surfaces. The HSA is typically the component replaced during head failure recovery, because it’s not feasible to replace individual heads without disturbing the precise alignment of the entire assembly. Replacement HSAs come from donor drives that match the failed drive’s exact specifications.
Related glossary entries
- Platter: the data-bearing surface that the heads read from and write to.
- Head Crash: the catastrophic failure where heads contact the platter and cause damage.
- Click of Death: the characteristic clicking sound produced by failing heads.
- Donor Drive: provides the replacement HSA for head transplant procedures.
- Cleanroom Data Recovery: the only environment where head replacement can be performed safely.
- HDD: the broader storage device; head failure is the most common physical failure mode.
- Firmware Corruption: separate failure mode that can produce similar symptoms but doesn’t require head replacement.
Sources
- ACS Data Recovery: Hard Drive Read / Write Heads (accessed May 2026)
- Gillware: Data Recovery 101: Read/Write Head Assembly
- Gillware: Hard Disk Head Repair: Data Recovery After Heads Failure
- Cheadle Data Recovery: Mechanical failure in HDDs – Head assembly & spindle motor
- SERT Data Recovery: How To Fix A Hard Drive With Bad Read/Write Heads
About the Authors
Data Recovery Fix earns revenue through affiliate links on some product recommendations. This does not influence our reference content. Glossary entries are written and reviewed independently based on documented research, vendor documentation, independent testing, and recovery-engineer review. If anything on this page looks inaccurate, outdated, or worth revisiting, please reach out at contact@datarecoveryfix.com and we’ll review it promptly.
