How Hard Drive Platters Store Data

Hard drive platters store data as patterns of magnetic polarity on a thin cobalt-chromium-platinum alloy coating. Each bit is represented by a tiny magnetic domain whose north-south orientation encodes a 1 or 0. The read/write heads detect and alter these orientations as they fly nanometers above the spinning platter surface. A single modern platter can hold over 1 TB of data across billions of magnetic domains packed into concentric tracks. Platter integrity is what every stage of our hard drive data recovery workflow is built to preserve.

Magnetic Domains and Bit Encoding

The magnetic recording layer is divided into microscopic regions called grains, each containing thousands of atoms with aligned magnetic orientation. A single data bit occupies a cluster of these grains. The write head generates a localized magnetic field strong enough to flip the orientation of the grains in the target area without affecting neighboring clusters.

Modern drives do not use simple binary encoding. They use run-length-limited (RLL) coding schemes that translate raw data into patterns optimized for reliable magnetic storage. The read channel chip on the PCB applies partial response maximum likelihood (PRML) signal processing to extract the original data from the analog signal, which is noisy and attenuated at the bit densities used in current drives.

Each grain must maintain its magnetic orientation against thermal energy that tends to randomize it. This is the superparamagnetic limit: as grains get smaller to increase density, they become less thermally stable. Manufacturers combat this with perpendicular magnetic recording (PMR), which orients grains vertically rather than horizontally, allowing denser packing with larger effective grain volume.

Perpendicular Magnetic Recording and Recovery Implications

Before 2005, drives used longitudinal magnetic recording (LMR), where grains lay flat along the track surface with their magnetic poles oriented parallel to the platter. LMR hit the superparamagnetic wall at roughly 100 to 200 Gbit/in² because flatter grains needed to shrink below the threshold where thermal energy could flip them spontaneously. Shun-ichi Iwasaki at Tohoku University proposed perpendicular magnetic recording in 1976, but it took until 2005 for Toshiba to ship the first commercial PMR drive at 133 Gbit/in². By 2009, PMR powered roughly 75% of all hard drives shipped worldwide.

PMR orients the recording grains perpendicular to the platter surface, standing them upright like fence posts rather than laying them flat. A soft magnetic underlayer (SUL) beneath the recording layer acts as a magnetic mirror for the write head's field, creating a return flux path that roughly doubles the effective field strength at the grain surface. That stronger write field allows the recording layer to use higher-coercivity materials with finer grain boundaries, which resist thermal demagnetization at smaller grain footprints.

For recovery, PMR's tighter grain packing means the analog readback signal from each bit is weaker relative to noise. When read/write heads degrade or when donor heads from a matched donor drive fly at a slightly different height than the originals, the signal-to-noise ratio drops further. PC-3000 compensates by modifying the drive's service area adaptive parameters that control read channel equalization, retraining the read channel to extract usable data from a weakened signal. This process is described in the read channel section below, but it becomes critical on PMR media where the margin between a readable signal and noise is thin.

Areal Density, Track Pitch, and Recovery Tolerances

Areal density is the number of bits per square inch of platter surface. It is the product of two dimensions: linear bit density (bits per inch along a track) and track density (tracks per inch across the platter radius). Both have grown by orders of magnitude since the first hard drive.

The bit aspect ratio (BAR) describes the shape of each recorded bit as the ratio of its along-track length to its cross-track width. Older drives maintained a BAR of 4:1 to 6:1, producing elongated bits that were relatively easy for the read channel to distinguish. Modern high-density drives have compressed BAR toward 2:1 or even 1:1 (square bits) in two-dimensional magnetic recording (TDMR) designs. Square bits pack more tightly but generate stronger inter-track interference, which forces the read channel to filter signal from adjacent tracks using multiple readers per head.

At sub-70 nm track pitch, the physical tolerances for head positioning during imaging tighten to the point where any thermal or mechanical perturbation matters. Thermal expansion of the actuator arm by a fraction of a degree can shift the heads off-center by tens of nanometers. Micro-vibrations from the spindle motor or from placing the drive on an unstable surface have the same effect. When donor heads replace failed originals, even a slight difference in head geometry or fly height characteristic alters the effective read width, and the track margin error budget that was already near zero shrinks further. This is why high-capacity drives (16 TB and above) have lower tolerance for imperfect donor matches and why firmware-level adaptive parameters (head profile tables, zone-specific gain settings) must often be rebuilt in PC-3000 to get a clean image from a donor head set.

Tracks, Sectors, and Zones

Data on a platter is organized into concentric circular tracks. Each track is divided into sectors, typically 512 bytes or 4,096 bytes (Advanced Format) of user data plus error correction codes (ECC), sync bytes, and address markers.

Track

A single concentric ring of data. Modern drives have hundreds of thousands of tracks per platter surface, spaced fractions of a micrometer apart.

Sector

The smallest addressable unit of storage. Contains user data, a sector header with the logical block address (LBA), and ECC bytes for error detection and correction.

Zone

A group of adjacent tracks that share the same sectors-per-track count. Outer zones have more sectors per track than inner zones because the outer circumference is longer. This is zoned-bit recording (ZBR).

Zoned-bit recording means that outer tracks store more data per revolution than inner tracks. The read/write speed varies accordingly: outer tracks are faster because more data passes under the head per rotation. This is why sequential read benchmarks show higher throughput at the beginning of a drive (outer tracks) and lower throughput near capacity (inner tracks).

Anatomy of a Data Sector

A user data sector has its own internal field sequence, distinct from the servo wedge format described in the next section. As the platter spins beneath the read head, each sector passes the head in a strict temporal order: inter-sector gap, preamble (PLL field), data sync mark, modulated user data, ECC parity field, postamble. The read channel processes each of these fields with a different subsystem.

Preamble (PLL field)

A periodic high-frequency pattern, historically 12 to 30 bytes and around 100 bits at current densities. The read channel's voltage-controlled oscillator locks its phase and frequency to this pattern so the analog-to-digital converter samples the waveform at the correct bit-cell intervals. No PLL lock means no usable samples, which means no readable sector.

Data Sync Mark (DSM)

An asymmetric zero-phase pattern, typically around 20 bits, that marks the exact boundary where user data begins. The preamble establishes bit timing; the DSM establishes byte framing. This is a different signal from the Servo Address Mark in the servo wedge. A DSM read failure causes a single-sector read error without losing head position, because the servo tracking loop runs on the SAM from its own wedges and is not affected.

User data (modulation-coded)

4,096 bytes on modern Advanced Format drives, 512 bytes on legacy drives. The host payload is first encoded with a run-length-limited or maximum-transition-run modulation code to guarantee frequent magnetic transitions. Without enforced transitions, a long string of identical bits would let the PLL drift and corrupt every bit that followed.

ECC parity field

Approximately 50 bytes on legacy 512-byte sectors, expanded to roughly 100 bytes on 4K Advanced Format sectors. Modern drives store LDPC parity here; early-to-mid 2000s drives stored Reed-Solomon parity. The difference between these two coding families determines how much degradation a sector can sustain before it becomes unrecoverable.

Postamble

A short trailing pattern, 0.5 to 2 bytes, that flushes the final bits of the ECC field through the Viterbi detector and any trellis-based decoder pipeline so no tail bits are truncated before decoding completes.

Servo wedges are written into the platter surface at the factory and run radially across all tracks. When a single data sector straddles a servo wedge (common at inner zones with fewer sectors per track), the read channel freezes its data-path state into trap registers, processes the servo wedge with a separate control loop, then restores the saved state and finishes the data sector. From the host's perspective, the sector is read atomically; at the read-channel level, two independent signal paths are time-multiplexed across the same head.

Advanced Format 4K Sector Layout

The 2009 IDEMA Advanced Format specification replaced the 512-byte physical sector with a 4,096-byte sector. Eight legacy sectors, each requiring its own gap, preamble, sync mark, and 50-byte ECC field, collapsed into a single 4K payload with one set of overhead fields and a 100-byte ECC field. Seagate and Western Digital reported a 7% to 11% gain in usable platter capacity from this change alone, lifting overall format efficiency to roughly 97%. The longer ECC block also corrects burst errors more effectively than eight independent 50-byte blocks, because coding gain scales with block length.

Drives ship in two variants: 512e (4K physical, 512-byte logical) and 4Kn (4K physical and logical). A 512e drive that receives an unaligned 512-byte write from the host must read the full 4K sector into DRAM, modify 512 bytes, recompute the 100-byte ECC, and write the 4K sector back. This read-modify-write penalty shows up on misaligned partitions created by legacy operating systems and is a common cause of unexpectedly slow benchmarks on otherwise healthy Advanced Format drives.

Reed-Solomon to LDPC ECC Transition

Through the 1990s and early 2000s, hard drive sectors protected user data with Reed-Solomon codes. Reed-Solomon is an algebraic block code: it makes a hard binary decision about each bit (1 or 0) and then uses the Berlekamp-Massey or Euclidean algorithm to locate and correct errors up to a fixed algebraic minimum distance. Once the read channel quantizes an analog sample into a hard bit, the confidence information in the original signal is discarded.

As drives crossed the 500 Gbit/in² threshold and approached 1 Tbit/in² around 2009, grain sizes and track pitches shrank to the point where analog readback signals routinely fell close to the decision threshold. Hard-decision Reed-Solomon decoders became unable to keep bit error rates within acceptable limits. HGST (later absorbed into Western Digital) introduced the Iterative Detection Read Channel (IDRC) to replace Reed-Solomon with low-density parity-check (LDPC) codes. By the early 2010s, LDPC was the dominant ECC across the industry.

LDPC is a soft-decision code. Instead of forcing each sample into 1 or 0, the read channel emits a log-likelihood ratio (LLR) for every bit, which represents how confident the channel is in that bit's value. The LDPC decoder passes these LLRs around a bipartite graph of variable nodes (bits) and check nodes (parity constraints) using belief-propagation algorithms such as sum-product or min-sum. Each iteration, the check nodes flag parity violations and send feedback that nudges the least-reliable variable nodes toward a consistent solution. The decoder iterates until all parity checks pass or a maximum iteration count is hit.

For recovery, this change is the difference between a sector being readable or not on degraded media. HGST documented a 1 dB SNR gain on their first-generation LDPC channel and another 1 dB on the second generation, which translates into roughly an 8% increase in raw per-drive capacity achievable from the same magnetic media. In practice, LDPC can iteratively reconstruct bits made ambiguous by thermal asperities, adjacent-track interference, or media defects by leaning on the high-confidence probabilities of surrounding bits. PC-3000 exposes the read-channel parameters that govern LDPC iteration count, LLR quantization, and soft-decoder thresholds, so a technician working a drive with marginal heads or worn media can trade decode time for decode success on a sector-by-sector basis during hard drive data recovery imaging.

Servo Wedges and Head Positioning

Servo wedges are pre-written positioning markers embedded between data sectors on every track. They are written during manufacturing by a servo track writer, a precision instrument that programs the position reference data onto blank platters. After manufacturing, servo data is never overwritten by the drive.

Each servo wedge contains a preamble (synchronization pattern), a servo address mark (SAM), a track ID encoded in Gray code, and burst fields that provide sub-track positioning. The heads read servo wedges continuously during operation. The servo controller on the PCB uses this feedback to adjust voice coil motor current and maintain the heads on the target track center.

When a drive clicks repeatedly, it is often because the heads cannot read the servo wedges. Without servo feedback, the controller cannot position the heads, and the drive retries by sweeping the actuator across the platters searching for readable servo data.

Servo Sector Format in Detail

Modern drives embed roughly 200 to 500 servo wedges per track, with high-capacity enterprise helium designs at the upper end of that range, each consuming a small arc of the track between data sectors. Every servo wedge follows the same field sequence:

AGC Preamble

A repeating pattern that lets the phase-locked loop (PLL) lock onto the servo signal timing and the automatic gain control circuit normalize the signal amplitude. Without a stable AGC lock, every subsequent field in the wedge is unreadable.

Servo Address Mark (SAM)

A unique bit pattern that does not occur elsewhere in the servo data. It tells the servo controller where the track ID field begins. The SAM acts as a sync boundary; if it is corrupted, the controller misinterprets the track number and positions the heads on the wrong track.

Track ID (Gray Code)

The cylinder address encoded in Gray code, where adjacent track numbers differ by only one bit. Gray code prevents large position errors from single-bit misreads during seeks. A misread in standard binary could shift the decoded cylinder number by thousands of tracks; a single-bit Gray code error shifts it by one.

Position Error Signal Bursts (A/B/C/D)

Four offset burst patterns whose relative amplitudes tell the servo controller exactly where the head sits relative to the track center, at sub-nanometer resolution. The controller computes a position error signal (PES) from the amplitude ratios of these bursts and feeds it to the voice coil motor control loop to correct head position in real time. This is what keeps the head centered on a track that is narrower than a wavelength of visible light.

Servo Damage and Recovery with PC-3000

A hardware servo gate signal isolates servo wedge reads from data reads; the controller never attempts to interpret servo fields as user data or vice versa. When servo data is unreadable on one or more heads, the drive enters a BSY (busy) state and never reaches DRDY (drive ready) because it cannot establish the head position reference needed to begin reading user data.

In recovery, PC-3000 addresses servo and head positioning failures through service area module repair. The per-head adaptive parameters and translator modules in the drive's firmware map logical block addresses to physical head/cylinder/sector locations. If these modules are corrupted, PC-3000 can rebuild them from backup copies stored elsewhere in the service area. For drives where specific heads have failed, the technician modifies the RAM head map (the firmware table that tells the drive which physical heads are active) to disable the damaged head and allow imaging of data from the remaining good heads. The data that lived on the disabled head's platters is lost, but the rest of the drive becomes accessible.

Repeatable and Non-Repeatable Runout

Servo wedges define where each track sits, but the heads still have to follow those tracks against two distinct classes of mechanical disturbance. Drive engineers separate them by whether the disturbance is locked to spindle rotation or random with respect to it.

Repeatable Runout (RRO)

Positioning errors synchronous with spindle rotation, fundamental at 120 Hz on a 7,200 RPM drive and present as harmonics at integer multiples of that rate. RRO is caused by minor spindle eccentricity, fluid-dynamic-bearing geometry, and vibration during the factory Servo Track Writer pass that bakes a small per-revolution positioning bias into the servo bursts themselves.

Non-Repeatable Runout (NRRO)

Random asynchronous disturbances driven by platter flutter (aerodynamic shear inside the air or helium volume, turbulence around adjacent platters), fluid-dynamic-bearing imperfections, and chassis vibration. NRRO has no fixed phase relationship to the spindle index pulse, so it cannot be predicted from one revolution to the next.

RRO is compensated by per-drive adaptive tables learned at the factory. A Position Error Signal observer averages PES samples across many revolutions; the harmonic content that survives the averaging is the repeatable bias. The inverse of that waveform is stored as a feed-forward correction table and injected into the voice coil motor control loop on every wedge, so the actuator pre-compensates for the known eccentricity before the next burst is read. Those tables live in the service area as per-head adaptive modules.

NRRO cannot be learned because it does not repeat. It must be rejected by high-bandwidth feedback alone, which is why total trackable runout budget on modern drives is dominated by how much NRRO the closed-loop servo can suppress within its open-loop bandwidth.

For recovery, RRO compensation is mathematically tied to the original head stack's geometry and the original spindle's eccentricity profile. When a donor head stack assembly is transplanted, the donor heads inherit the patient's service area, so they receive feed-forward corrections computed for a slider mass, suspension stiffness, and air-bearing geometry they do not have. The donor head misinterprets the feed-forward signal and pushes itself off-track on the very revolutions the correction was meant to stabilize. PC-3000 Portable III is used to clear or recompute the RRO adaptive tables in the service area and load matching micro-jog parameters so the donor centers correctly on the patient's existing servo bursts.

Adjacent Track Interference and Wide Area Track Erasure

Beyond positioning, the act of writing itself degrades nearby tracks. Two distinct phenomena bound how aggressively a drive can re-write the same region without eroding neighboring data.

Adjacent Track Interference (ATI)

Cumulative signal-to-noise degradation on tracks immediately adjacent to a heavily written target track. ATI is driven by stray fields that bypass the wrap-around shield because the side shield's remanent magnetization interacts asymmetrically with the writer flux, leaking enough field into the neighbor track to weaken its recorded transitions over many thousands of write cycles.

Wide Area Track Erasure (WATE / Far Track Interference)

Degradation on tracks dozens of pitches away from the written track, caused by domain lock-up at the writer pole and trapped flux leakage from the main pole and return path. WATE manifests at distances where direct fringing fields are too weak to flip grains, but cumulative low-level perturbation still drives a measurable bit error rate increase. Dual Write Shield designs with a return-loop path back to the back-gap connection were introduced to constrain the leakage geometry that produces WATE.

For recovery, ATI and WATE explain why sectors that were never directly written can still go bad on heavily-written enterprise drives: the magnetic signal on those sectors fades below the ECC margin even though no host write ever touched them. PC-3000 read-channel parameter adjustments recover these faded sectors by extracting raw log-likelihood-ratio data before ECC checking. The operations include disabling background reallocation so the drive does not retire the sector before it can be imaged, modifying the PRML target polynomial to match the actual signal shape from the degraded track, deepening LDPC iteration counts so the soft-decision decoder gets more passes to converge, and tightening Viterbi state-metric thresholds so marginal samples are not discarded. These adjustments do not regenerate lost magnetic signal; they extract every usable bit of confidence information from what remains.

Servo Track Writer Evolution: External, Self, and Spiral

The Servo Track Writer process that creates the burst pattern has changed substantially across three generations. Each generation made the servo pattern more tightly bound to the individual drive that was being written, which has direct consequences for what a recovery technician can and cannot transplant.

1. External Servo Track Writer (External STW). A mechanical push-pin through a slot in the drive chassis positions an external write head, guided by a laser interferometer that measures radial position to nanometer accuracy. The interferometer drives the push-pin actuator while a clean-room STW machine lays down concentric servo tracks on blank platters one wedge at a time. The drive's own heads play no role; the chassis slot is sealed after the STW pass.
2. Self-Servo Write (SSW). The sealed drive writes its own servo pattern. A single reference track is written using a low-precision bootstrap (sometimes from a media-write head, sometimes pre-printed on the disk), and the drive steps outward and inward from that reference, using PES feedback against the previously written tracks to position each new track. SSW eliminates the external STW machine and the chassis slot, which lowered manufacturing cost and improved sealed-volume cleanliness for helium drives.
3. Spiral Self-Servo Write / Media Write Servo (Spiral SSW / MWS). A few reference spiral patterns are laid down rapidly across the full radius, then the drive writes its final concentric servo wedges using the spirals as a combined timing and radial reference. The spirals provide a coarse position-and-timing grid that constrains drift during the long concentric-write pass. This is what 18 TB and larger helium enterprise drives use today.

Because spiral SSW writes the final servo pattern while the heads are flying inside the sealed drive, every mechanical eccentricity of that specific spindle, that specific actuator pivot, and that specific suspension geometry is baked into the servo wedges as a fixed bias. The servo wedges on a modern enterprise drive are mathematically mated to that exact head stack assembly and that exact spindle. There is no master pattern that a donor would also match.

For recovery, this is why a donor head transplant on a spiral-SSW drive requires more firmware intervention than a transplant on a drive from the External-STW era. PC-3000 must manipulate actuator trajectory parameters during head swap work to bypass strict logical-to-physical alignment checks the firmware performs at startup, recompute micro-jog offsets so the donor read element lands on the patient's burst centers, and in some families clear the spiral-derived adaptive coefficients so the donor heads are not asked to follow a trajectory pre-compensated for a slider they no longer carry.

Dual-Stage Actuation and PZT Microactuators

The primary voice coil motor servo loop has a bandwidth ceiling set by the structural resonance of the actuator arm and the suspension. On current enterprise drives, that ceiling sits around 1 to 2 kHz. Above that frequency, the arm and suspension begin to flex rather than rotate as a rigid body, and the controller can no longer assume the head moves where the VCM commands it to move. Modern enterprise drives add a secondary actuator at the suspension tip to extend usable closed-loop bandwidth into the range where platter flutter and chassis vibration occur.

Primary Actuator (VCM)

The voice coil motor rotates the entire actuator arm assembly around the pivot bearing. High torque, long stroke (full radius of the platter), but bandwidth limited by arm and suspension structural resonance to roughly 1 to 2 kHz.

Secondary Microactuator (PZT / MEMS)

A piezoelectric (lead-zirconate-titanate) or MEMS element bonded at the suspension tip, between the gimbal and the slider. Stroke is on the order of hundreds of nanometers, but bandwidth extends to 4 to 5 kHz because the moving mass is just the slider and a small carrier rather than the full arm.

Closed-Loop Bandwidth

With dual-stage control, the VCM handles the low-frequency component of the position error and the PZT microactuator handles the high-frequency component, giving the combined system enough bandwidth to reject the platter flutter and chassis vibration that dominate NRRO at high areal density.

PZT material is brittle. It can fracture from thermal cycling at the head-disk interface, from physical shock to the drive, or from voltage transients on the microactuator drive lines. A fractured PZT presents in PC-3000 diagnostics as Dual Stage Actuator errors: servo loop instability above the VCM bandwidth that the controller cannot suppress, and in some firmware families an explicit DSA fault flag set during initialization.

During donor head transplants, the microactuator calibration data stored in the ROM and service area was learned against the original heads' PZT response (capacitance, actuation gain, resonance peak). The donor heads carry their own PZT with different characteristics. PC-3000 either clears the DSA adaptive parameters so the firmware re-learns against the donor PZT, or temporarily disables the microactuator entirely so the VCM alone performs degraded-but-functional track following long enough to image the drive. Imaging with the microactuator disabled is slower and less robust against vibration, but it is sufficient to extract user data when the alternative is no image at all.

Platter Materials and Coatings

A typical hard drive platter is a stack of thin layers deposited on a substrate:

1. Substrate: Aluminum alloy (most desktop/enterprise drives) or glass-ceramic (laptop drives, some helium-filled enterprise drives). The substrate is polished to sub-nanometer surface roughness.
2. Underlayer: A magnetically soft layer that helps orient the recording layer's magnetic grains perpendicular to the surface in PMR drives.
3. Magnetic recording layer: A cobalt-chromium-platinum alloy, 10 to 20 nanometers thick. This is where data is stored.
4. Overcoat: A diamond-like carbon (DLC) protective layer, approximately 2 to 3 nanometers thick, that protects the magnetic layer from corrosion and head contact.
5. Lubricant: A perfluoropolyether (PFPE) layer, roughly 1 nanometer thick, that reduces friction during head start/stop events and protects the DLC overcoat.

Platter Geometry Constraints on Donor Matching

Every physical and firmware-side attribute described above (substrate material, areal density, zone-bit-recording layout, servo burst geometry) becomes a hard constraint when a donor drive must be sourced for a head stack transplant. The donor cannot be selected by model number alone; the platter-side physics of the patient drive dictates a tighter matching framework. The six-criteria framework documented on how donor drives are matched is the procedural form of these constraints. The geometry-side reasoning is described below.

Head Fly Height: Sub-Nanometer Tolerances

On modern 7,200 RPM enterprise helium drives at 16 TB and above, the femto slider rests on a baseline air bearing of roughly 3 to 5 nm above the platter. During actual read and write operations, a Thermal Fly-height Control (TFC) element, also called Dynamic Flying-Height Control (DFH), drives a localized Joule heater near the trailing edge of the slider. The heater induces controlled thermal expansion of the read and write transducers, dropping operational clearance to about 1 to 2 nm at the instant of transfer. This is smaller than the diameter of a strand of DNA.

The fly-height profile is calibrated per head at the factory and stored in the service area. When a donor head stack flies even a fraction of a nanometer higher or lower than the patient's original heads, the readback amplitude shifts outside the window the read channel's adaptive equalizer was trained for. Automatic gain control begins overcorrecting, the timing-recovery loop loses clean zero crossings, and Viterbi samples stop landing on their partial-response targets. The drive throws servo errors and either retries by sweeping the actuator or fails to reach DRDY. PC-3000 Portable III addresses this by retuning the read channel adaptives and rewriting the per-head DFH offset in the service area so the donor heads can be coaxed to fly at the height the patient platters expect.

Glass vs Aluminum Substrate

Substrate material is not interchangeable across donors. Older 3.5-inch drives and most desktop-class drives use a polished aluminum-magnesium alloy substrate. Many high-capacity enterprise helium platforms (Seagate Exos X18, WD Ultrastar HC560) still use refined aluminum-alloy substrates; the industry has been transitioning to glass-ceramic substrates for designs packing ten or more platters into the same Z-height envelope and for HAMR-class media where thermal cycling at the head-disk interface demands a lower thermal expansion coefficient. The Toshiba MG10-D series and HAMR-class drives ship with glass substrates. Glass has a lower thermal expansion coefficient and higher rigidity than aluminum, which is what allows it to hold track registration at sub-70 nm pitch during long imaging runs.

For donor matching, this means a CMR aluminum-platter donor is not a valid substitute for a CMR glass-platter patient even when the model number, capacity, and firmware all match on paper. Sub-revisions within the same product family have crossed the aluminum-to-glass boundary mid-run. Mixing substrates puts the donor head stack into a thermal profile and a flatness profile it was not calibrated for, and head-to-platter contact during the first spin-up becomes likely.

Zone Tables Are Per-Drive, Not Per-Model

Zoned-bit recording divides the platter radius into bands with different sectors per track. The exact zone boundaries are not a fixed property of the model; they are calibrated at the factory against the measured recording capability of each individual head and stored in the service area. On Western Digital architectures, the zone allocation map lives in Module 03 (Format Select Data) and its shadow copies. On Seagate F3 architectures, it lives in SysFile 28 and the Zone Configuration modules. Two drives off the same production line with identical firmware strings can have different zone tables.

Because the zone table tells the translator how to convert logical block addresses into physical cylinder, head, and sector positions, importing a donor's zone table onto a patient drive corrupts the LBA-to-physical mapping. PC-3000 reads these modules out of the patient's surviving service area before any head work is attempted, so that after the transplant the donor heads can be told to read sectors using the patient's factory zone geometry rather than the donor's.

Servo Bursts Are Uncopyable: Why Heads Transplant, Not Platters

The A/B/C/D quadrature bursts and the Gray-code track IDs in each servo wedge are not written by the drive itself. They are written once, at the factory, by a Servo Track Writer (STW) that establishes the absolute concentric layout of every track and sector relative to the spindle motor's axis of rotation. The drive's own write heads cannot regenerate servo data; the data path is gated off from the servo path in hardware, and even if it were not, the heads have no external position reference to write against once the servo is gone.

This is why a platter transplant is a procedure of last resort and why standard mechanical recovery is always a head stack assembly (HSA) transplant. Moving platters to a new chassis introduces angular misalignment between the platters and the new spindle, and even a fraction of a degree of stacking error breaks the concentricity of the servo tracks relative to the new axis of rotation. The actuator cannot follow a track that wobbles radially across each revolution. Keeping the patient platters on their original spindle and importing donor heads preserves the absolute servo alignment; PC-3000 then dynamically retunes the donor heads' micro-jog adaptive (the radial offset between the read and write elements on each slider) so the new heads can lock onto the existing servo bursts.

Head Exchange Combs and the Clean Bench Environment

Lifting the head stack assembly off the patient platters without grazing the recording surface requires a precision tool, not a steady hand. Head exchange combs from HDDSurgery and Dolphin Data Lab are thin parallel blades sized for the specific inter-platter gap of each drive family. The comb slides between the suspension arms over the non-data parking zone or the outer edge of the platter stack, physically separating the sliders so they lift off the perfluoropolyether lubricant layer rather than dragging across it. The entire actuator assembly is then guided onto the parking ramp or out of the chassis with the sliders held off the magnetic surface by the comb teeth, eliminating the slider-to-platter contact that would otherwise score the cobalt-chromium-platinum layer and contaminate the donor heads before they ever fly.

All comb work is performed on a 0.02 micron ULPA-filtered clean bench. The filter is rated to ISO 29463 standards for particles down to 20 nanometers, which is the size threshold that matters when slider clearance over the platter is 1 to 5 nanometers during read and write. Any particle larger than the air bearing thickness becomes a potential head-strike when the donor heads are loaded and the drive is spun up for the first time after transplant. Helium drives are opened, head-swapped, and refilled with helium in this same environment in-house at the Austin TX lab; the chamber refill procedure restores the low-density internal atmosphere the suspension geometry was designed to fly in.

Six-Criteria Donor Matching in One Place

The constraints above feed directly into the six donor-matching criteria applied at the hard drive data recovery bench:

1. Model and family: establishes platter density, substrate generation, and read-channel architecture.
2. Firmware revision: determines service-area module layout (Module 03, SysFile 28), zone table format, and adaptive parameter structure.
3. Head map: a 10-head physical platform shipped with only 8 heads active is a different drive than one shipped with all 10 active; preamp channels and per-head calibration tables must match.
4. Site code: different manufacturing facilities source head stack assemblies and voice coils from different sub-contractors, producing mechanical incompatibilities even at identical model and firmware.
5. Date of manufacture window: a donor manufactured within roughly three months of the patient is more likely to share slider lithography and actuator mass with the original heads.
6. HSA part number and preamp revision: the preamplifier on the flex cable sets the read-channel bias and gain. A mismatched preamp (for example, a Seagate C202 substituted for an 8202) generates immediate servo errors on power-up and can corrupt the service area.

The full procedural form (including how site codes are decoded, how DOM windows are measured, and how the HSA part number is read off the flex cable) is documented on the donor matching reference page. All transplants are performed on a 0.02 micron ULPA-filtered clean bench, with DeepSpar Disk Imager handling bus-layer cloning once the mechanical work is complete and the firmware is stable.

HAMR, MAMR, and Bit Patterned Media: What Ships and What Does Not

The three recording technologies most often discussed as successors to conventional PMR are at very different stages of commercial readiness. Treating them as a single near-future roadmap obscures which drives are arriving on benches today and which remain research artifacts.

The Trilemma of Magnetic Recording

The reason successor technologies exist at all is the trilemma of magnetic recording: a three-way physical balance between signal-to-noise ratio, thermal stability, and writability that cannot be simultaneously optimized at conventional PMR materials. To raise areal density, the magnetic grains must be made smaller. A bit cell at modern PMR density spans a small cluster of weakly coupled ferromagnetic grains, because signal-to-noise ratio at the read channel scales with the number of grains per bit and falls off sharply once the grain count drops too low.

As grain volume shrinks, however, the magnetic anisotropy energy that holds each grain in its written orientation approaches ambient thermal energy. Below a threshold volume, thermal fluctuations flip grains at random; this is the superparamagnetic effect, and it sets a floor on how small the grains can be without losing the stored bit between writes and reads. The countermeasure is to use a higher-anisotropy alloy that holds its magnetization at smaller grain volumes, but higher anisotropy requires a stronger write field to flip in the first place. Current PMR write poles already operate near the practical ceiling of what an inductive write head can generate with the materials available for the main pole and the soft underlayer return path. Push the anisotropy higher and the writer can no longer flip the bit at all.

This three-way bind is what caps conventional PMR areal density at roughly 1 terabit per square inch. HAMR breaks the bind by using a near-field optical transducer to briefly heat the recording spot past the Curie temperature of an iron-platinum alloy, which has extreme room-temperature anisotropy but loses coercivity entirely while hot. The write pole flips the bit at near-zero coercivity, the spot cools in nanoseconds, and the bit locks back into a thermally stable orientation. MAMR variants use a spin torque oscillator to inject a microwave-frequency assist field that drives ferromagnetic resonance in the recording grains, lowering their effective coercivity at the instant of writing without the thermal cycle. Both technologies exist for the same reason: the underlying physics says PMR cannot scale further without one of the three constraints being relaxed.

1. Heat-Assisted Magnetic Recording (HAMR). Shipping in volume on Seagate Mozaic platforms at 30 TB and above as of 2025 to 2026. A laser diode mounted on the slider couples into a near-field transducer that heats a roughly 50 nm spot on the platter to over 400 C, briefly bringing the iron-platinum (FePt) recording layer past its Curie temperature so the write pole can flip the bit at a coercivity the cooled grains will then lock in. FePt's high room-temperature coercivity is what makes the smaller grain volumes thermally stable; the heat pulse is what makes them writable at all.
2. Energy-assisted recording variants (ePMR, FC-MAMR). Shipping on Western Digital and Toshiba enterprise drives in the 22 to 30 TB range. Toshiba's Flux-Control MAMR (FC-MAMR) places a Spin Torque Oscillator in the write gap that generates a 20 to 40 GHz AC field, driving ferromagnetic resonance in the recording grains to lower their effective coercivity at the instant of writing without raising platter temperature. Western Digital's ePMR uses a different mechanism: a DC bias current applied to the main write pole that sharpens the write field gradient and reduces jitter, with no microwave oscillator. Both approaches are mechanically gentler than HAMR at the head-disk interface because neither involves a high-temperature thermal cycle.
3. Bit Patterned Media (BPM) and Heated-Dot Magnetic Recording (HDMR). Roadmap items. BPM replaces the continuous granular recording layer with lithographic islands of magnetic material, one bit per island, eliminating grain-boundary noise entirely. HDMR combines BPM with HAMR-style heat assist. Neither has shipped in a volume commercial drive and the industry timeline targets the 2030s.

HAMR introduces extreme thermal cycling at the head-disk interface. The near-field transducer that focuses the laser energy is a known failure mode: NFT contamination and tip recession produce cold writes, where the write pole asserts its field but the recording layer never reached Curie temperature, so the bit is set in a marginal state that fails ECC on the next read. MAMR is mechanically closer to conventional PMR but adds the Spin Torque Oscillator as an additional electrical failure point in the head.

Our HDD recovery lab handles conventional PMR, CMR, SMR, and helium drives in-house, including helium-filled head stack transplants with chamber refill on our 0.02 micron ULPA-filtered clean bench. HAMR-class drives require specific PC-3000 module support and per-family service area knowledge that is still being released, and we evaluate HAMR jobs on a case-by-case basis before accepting them rather than treating them as routine.

What Platter Damage Looks Like

Platter damage manifests in several ways, each with different implications for data recovery:

Tribological Stack and Head-Disk Interface Physics

The visible damage modes above are end-state symptoms. The progression that produces them runs through three nanometer-scale layers stacked on top of the cobalt-chromium- platinum recording layer. Understanding what each layer does, and at what energy threshold it fails, is what separates a head event that ends in a clean head transplant from one that ends with a multi-platter contamination cascade and donor heads destroyed on the first spin-up.

Perfluoropolyether Lubricant (1 to 2 nm)

The topmost layer is a 1 to 2 nanometer film of perfluoropolyether (PFPE) polymer. At the slider velocities present in a 7200 RPM enterprise drive (linear contact velocities approaching 200 km/h at the outer diameter), the PFPE layer behaves as a non-Newtonian liquid: it exhibits shear thinning under load and transfers molecule-by-molecule onto the air bearing surface during near-contact events. Minor head-disk interference displaces this layer transiently without breaching the underlying overcoat. The displaced lubricant gathers into a localized ridge that the slider then strikes on subsequent revolutions, producing the thermal asperity signature described below.

Stiction Is Lubricant Failure, Not Motor Seizure

When a drive that previously spun up refuses to start and the spindle motor draws high current without rotation, the failure is most often stiction at the head-disk interface, not a seized fluid-dynamic bearing. Thermal cycling and ambient humidity changes can cause the PFPE lubricant to bond the polished air bearing surface of a parked slider directly to the platter. The bond exceeds the torque the spindle motor can apply during start-up. The diagnostic clue is that the drive will sometimes free after temperature equalization; a true bearing seizure does not free.

Diamond-Like Carbon Overcoat (2 to 3 nm)

Below the lubricant is a 2 to 3 nanometer diamond-like carbon (DLC) overcoat providing wear and corrosion protection. The DLC is elastic to a point and brittle past that point. When head-slap energy or sustained thermal asperity contact exceeds the elastic deformation limit, the DLC fractures rather than yielding. The fracture generates highly abrasive carbon debris in the 5 to 50 nanometer range, well below the size threshold the drive's own particle filter can capture from the internal air or helium volume.

Cobalt-Chromium-Platinum Recording Layer (10 to 40 nm)

Below the DLC sits the magnetic recording layer itself. When a slider breaches the DLC and contacts this layer on a spinning platter, the alumina-titanium-carbide (Al2O3/TiC) composite that forms the slider body acts as a cutting tool. It strips the magnetic alloy from the aluminum or glass-ceramic substrate and converts the data-bearing layer into ferromagnetic powder. The user data on the affected tracks is destroyed, not corrupted; there is no signal left to extract.

Ferromagnetic Particle Smearing and Multi-Platter Cascade

The abraded recording-layer particles are swept by the internal airflow across the entire platter stack. Particles embed into the air bearing surface of every surviving slider, turning each one into a secondary cutting tool that smears ferromagnetic debris across adjacent, previously healthy tracks. This is contact- mediated adjacent track interference, distinct from the magnetic-fringing ATI described earlier. On a multi-platter drive, a single-surface head event escalates into a stack-wide contamination cascade within minutes of continued power. The first power-down decision after a suspected head crash is the most consequential one in the entire recovery.

Two further physical effects appear in the field. On 2.5-inch drives using glass or glass-ceramic substrates, severe mechanical shock produces microscopic edge chipping at the platter rim rather than the ductile bending seen on aluminum substrates. Edge chips launch debris back into the head path and can render the affected platter surface permanently unflyable. Separately, drives that have been running with a localized lubricant ridge or particulate ledge accumulate thermal asperity tracks: continuous concentric rings where the slider was forced upward by an ongoing air-bearing disruption, producing degraded signal-to-noise on every revolution at that radius. The tracks themselves may still hold magnetic data, but the air bearing is no longer stable over them.

Damage Mode Recovery Mapping

The table below maps the tribological events above to the specific PC-3000 Portable III workflows used to extract data around them. All listed procedures are performed on the 0.02 micron ULPA-filtered clean bench at the Austin TX lab. None of them restores destroyed magnetic data; the function of each procedure is to image surviving sectors before continued operation enlarges the damaged region.

DeepSpar Disk Imager is used alongside PC-3000 once firmware stability is achieved, with the imager configured to skip over flagged-bad sectors on the first pass and revisit them under tighter controls only after the easily readable user data is secured. On helium drives, the head transplant and helium refill are performed in the same procedure on the clean bench; the suspension geometry of a helium drive is calibrated for the lower-density internal atmosphere and will not fly correctly in air.

Read Channel Signal Processing

The analog signal coming off a read head is not a clean sequence of ones and zeros. It is a continuous waveform where each magnetic transition produces a voltage pulse that overlaps with pulses from adjacent bits. The job of the read channel chip on the drive's PCB is to extract the original bit sequence from this noisy, overlapping analog signal.

Peak Detection vs. PRML

Before 1990, read channels used peak detection: the circuit looked for voltage spikes above a threshold and recorded a 1 for each peak. This worked at low bit densities where pulses were spaced far enough apart that they did not overlap. As densities increased, adjacent pulses began to merge (intersymbol interference, or ISI), and peak detection could no longer distinguish individual bits reliably.

IBM introduced partial response maximum likelihood (PRML) read channels around 1990. Instead of fighting intersymbol interference, PRML embraces it. The read channel samples the continuous analog waveform at discrete intervals and feeds those samples into a Viterbi detector. The Viterbi detector uses a trellis diagram (a graph of all possible bit sequence state transitions) and an add-compare-select (ACS) algorithm to find the most probable original bit sequence given the observed samples. PRML increased achievable densities by 30 to 40% over peak detection because it extracted usable data from signal that peak detection would have rejected as ambiguous.

Target Polynomials and Adaptive Filters

The “partial response” in PRML refers to a target polynomial that defines the expected shape of intersymbol interference. PR4 (class IV partial response) models each bit as affecting the current sample and the previous sample. EPR4 (extended partial response) extends that window to three samples. The read channel's equalizer filter shapes the raw analog signal to match the chosen target polynomial before the Viterbi detector processes it. Modern drives use adaptive finite impulse response (FIR) filters that continuously adjust their coefficients based on the actual signal characteristics of the media under the current head.

During recovery, these adaptive parameters matter directly. A degraded head or a donor head with different fly height characteristics produces an analog signal whose shape does not match the equalization profile the drive's firmware was calibrated for. PC-3000 modifies the service area firmware parameters that control read channel behavior, including the adaptive filter coefficients and equalization targets stored in the per-head calibration tables. The tool can also command repeated re-reads of the same sector while stepping through different read channel configurations until it finds a parameter set that produces a valid ECC decode. On drives with severe head degradation, PC-3000's physical block address (PBA) mode bypasses the translator module entirely, reading raw sectors in physical order rather than logical order to minimize head seeks and reduce the thermal cycling that accelerates further head deterioration.

CMR vs. SMR Recording Geometry

The track layout described above assumes conventional magnetic recording (CMR), where each track is written independently with a guard band between adjacent tracks. Shingled magnetic recording (SMR) eliminates most of that guard band by overlapping tracks like roof shingles. The write head writes a full-width track, then the next track partially overwrites the previous one, leaving only a narrow readable strip. This increases track density (and therefore areal density) without shrinking the write head, but it means random writes must rewrite entire shingle bands.

For recovery, SMR adds complexity at the firmware tier. The drive's firmware manages a shingled translation layer that maps logical writes to physical shingle bands, and corruption of that translation layer can make the drive appear empty even though the magnetic data on the platters is intact. PC-3000's firmware repair modules can rebuild the shingled translation tables for supported drive families, converting an unmountable volume into a firmware-tier recovery.

Write Head Geometry: PMR vs SMR

On a CMR drive, the write pole is sized to lay down a track approximately equal in width to the read sensor that will later read it back, with a small guard band of unwritten media between adjacent tracks. Track pitch on modern CMR helium drives sits in the 50 to 75 nm range, with the guard band consuming roughly 15 to 25 percent of that pitch. The read sensor is fabricated narrower than the written track to keep adjacent-track interference low.

SMR inverts that geometry. The write pole is left at a width that produces a strong, well-defined magnetic transition (effectively the same physical write head used on the equivalent CMR product), but each subsequent track is radially shifted by less than the full write width, so the cross-track edge of the new write partially overwrites the adjacent side of the previous track. After the band is fully written, only a narrow strip of each earlier track remains unoverwritten, and that strip is what the read sensor reads. The result is an effective written track that is narrower than the write head produced, which is what allows track density to climb without shrinking the write pole geometry. Modern PMR write heads use an all-wrap-around shield structure: a trailing shield that controls the down-track field gradient (sharp bit-to-bit transitions along the track) and side shields that control the cross-track gradient and constrain wide adjacent-track erasure. SMR depends heavily on the side shield on the shingling side to keep the surviving read strip clean.

One consequence of this asymmetry is that the read sensor on an SMR drive reads only the surviving narrow strip, so signal-to-noise at the read channel is lower per track than on a comparable CMR drive. The read channel compensates with longer LDPC iteration counts and tighter equalizer adaptation. During recovery on a degraded SMR drive, PC-3000's read channel parameter adjustments must respect this thinner readable strip; aggressive offset sweeps that work on CMR media can push the head into the overwritten portion of an adjacent track and produce nothing but noise.

Shingle Bands and Read-Modify-Write

Because writing track N also damages the readable strip of track N-1, an SMR drive cannot rewrite a single track in place. Tracks are grouped into physical bands separated by guard tracks. The USENIX FAST 2015 Skylight reverse-engineering of the first-generation Seagate Archive ST5000AS0011 and ST8000AS0002 measured internal band sizes in the 15 to 40 MiB range; band sizes are kept small to bound read-modify-write latency during background band cleaning. Internal physical band sizes on subsequent Seagate and Western Digital DM-SMR designs are not publicly documented, but the same latency constraint applies. The 256 MiB figure that often appears in SMR documentation is the host-visible zone size standardized for HM-SMR and HA-SMR drives under ZBC and ZAC, not the internal band size of a DM-SMR device. To modify any sector inside a band, the drive must read the entire band into a working area, modify the target sectors, and write the whole band back in sequential track order.

Drive-managed SMR firmware hides this read-modify-write penalty from the host by accepting random writes into a persistent media cache (a CMR-formatted region on the same platters, sometimes called an exception zone). Incoming writes land in the media cache at full random-write performance. A background band-cleaning task later moves the cached writes into their target SMR bands during idle time. The translator firmware maintains metadata that maps each logical block to its current physical location: still in the media cache, already cleaned to its band, or split between the two during a partial cleaning cycle.

That translator metadata is the single point of failure for DM-SMR recovery. If the metadata is corrupted (power loss during a cleaning cycle, firmware crash, or service-area damage), the drive can lose track of which media-cache writes have been committed to bands and which are still pending, and a host-visible logical block can point to stale, partially-overwritten, or unrelated platter content. The magnetic data is still intact on the platters; the index that locates it is gone. Recovery on these drives is firmware-tier work performed in PC-3000 against the service area: rebuilding the band-cleaning state, replaying the persistent media cache against its target bands, and reconstructing the logical-to-physical map before any user data is imaged.

DM-SMR, HM-SMR, and HA-SMR

The storage industry recognizes three SMR product classes; only two of them are defined by the SCSI Zoned Block Commands (ZBC) and ATA Zoned-device ATA Commands (ZAC) standards. Drive-Managed SMR (DM-SMR) predates ZBC/ZAC and presents a conventional SBC/ATA block interface that hides the shingled geometry behind firmware; this covers nearly all consumer Seagate Archive HDDs, WD Red SMR variants, and the SMR drives that have surfaced in CMR-labeled product lines. The ZBC and ZAC standards define the other two classes. Host-Managed SMR (HM-SMR) refuses non-sequential writes within a sequential-write-required zone and requires the host filesystem or application to issue writes in the prescribed order; these are typically enterprise drives used with f2fs, btrfs zoned mode, or object stores written for ZBC. Host-Aware SMR (HA-SMR) accepts random writes but exposes zone metadata to a cooperating host that can choose to write sequentially for predictable performance. From a recovery standpoint, HM-SMR drives have no translator metadata to corrupt because the host owns the placement decisions, but the host filesystem state becomes the point of failure instead. DM-SMR drives carry the full translator complexity inside the drive's firmware and are the SMR class that most often arrives for hard drive data recovery with a corrupted band map or unflushed media cache.

Frequently Asked Questions