Reseat the Drive
Power off the computer. Remove the M.2 retaining screw with a Phillips screwdriver. Pull the drive straight out of the slot. Clean the gold edge connector with 90%+ isopropyl alcohol and a lint-free cloth to remove any oxidation or dust. Let it dry for 30 seconds, then reseat the drive firmly until the connector is fully inserted and the screw holes align.
A loose or partially seated NVMe drive is the most common reason for a BIOS detection failure that is not caused by hardware damage. The M.2 connector uses a high-density layout with up to 75 pin positions. Even a fraction of a millimeter of misalignment... A fraction of a millimeter of misalignment is enough to lose signal integrity on the PCIe lanes.
Check the M.2 Standoff Position
M.2 drives come in four standard lengths: 2230, 2242, 2260, and 2280. The number represents the width and length in millimeters. Most desktop and laptop NVMe drives are 2280 (22mm wide, 80mm long), but compact laptops and the Steam Deck use 2230.
If the standoff is in the wrong position, the screw forces the PCB to flex at the connector end. This can crack solder joints on the controller or NAND packages, or prevent the gold fingers from seating at the correct depth. Verify the standoff is in the hole that matches your drive's length before screwing it down.
Common mistake: Some motherboards ship with the standoff pre-installed at the 2260 position. A 2280 drive screwed into a 2260 standoff will flex upward at the far end, lifting the connector pins out of the slot.
Check Disk Management vs. BIOS
This step determines whether you have a software problem or a hardware problem. Boot into your BIOS/UEFI setup (typically by pressing F2, DEL, or F12 during POST) and look for the drive in the storage or NVMe device list.
Visible in BIOS, Not in Windows
The drive hardware is functional. The issue is a missing partition table, an uninitialized disk, or a file system Windows does not recognize. Open Disk Management (diskmgmt.msc). If the drive appears as "Unallocated" or "Not Initialized," this is recoverable with software tools or by assigning a drive letter. Do not format if you need the existing data.
Absent from BIOS Entirely
The motherboard cannot detect the drive at the hardware level. No operating system, driver, or software tool can access a device the BIOS does not see. This points to a connection problem (Steps 1-2), a dead M.2 slot (Step 4), or a failed controller/power component on the SSD itself (Step 5).
Try a Different Slot or System
Test the drive in a second M.2 slot on the same motherboard if one is available. Many motherboards have two or more M.2 slots, and they connect through different PCIe lanes. A failed PCH lane or a dead slot does not mean the drive is bad.
If no second slot is available, test the drive in a different computer. You can also use an external M.2 NVMe USB enclosure, though some enclosures do not support all NVMe drives. If the drive appears in the second system's BIOS, the original motherboard slot is the problem, not the SSD.
If the drive fails to appear on any system, proceed to Step 5.
Common False Alarms Before Assuming Failure
M.2 SATA vs. M.2 NVMe Keying Mismatch
M.2 is a physical connector, not a protocol. An M.2 drive can be SATA or NVMe, and they use different pin configurations (keys). NVMe drives have an M-key notch (a single gap on the right side of the connector). SATA M.2 drives have a B+M key (gaps on both sides). Most modern motherboard M.2 slots support only NVMe. Plugging an M.2 SATA drive into an NVMe-only slot produces zero detection in BIOS; the drive is not broken, it is in the wrong slot.
Check your motherboard manual. Slots labeled "M.2 (PCIe)" or "M.2 (NVMe)" do not accept SATA drives. Slots labeled "M.2 (PCIe/SATA)" accept both. If you have a SATA M.2 drive in a PCIe-only slot, moving it to a compatible slot or using a SATA adapter will restore detection.
PCIe Lane Allocation and BIOS Settings
On many consumer motherboards, M.2 slots share PCIe lanes with SATA ports. Enabling a second M.2 NVMe drive disables SATA ports 5 and 6. Some boards disable an M.2 slot entirely when a SATA device is connected to a shared port. This is documented in the motherboard manual under "M.2 / SATA sharing" or "Storage Configuration."
If your NVMe drive was working and disappeared after adding a new SATA device (or vice versa), check BIOS storage configuration. Intel platforms route some M.2 lanes through the PCH; AMD platforms route the primary M.2 slot directly from the CPU. The CPU-direct slot is always available. PCH-routed slots depend on what else is connected to the chipset.
This Is a Controller Failure
If the NVMe SSD is absent from BIOS on multiple systems after reseating and verifying the standoff, the controller chip or a power regulation component on the SSD PCB has failed. The drive is not communicating on the PCIe bus at all.
No software, driver update, or BIOS setting will fix this. The controller is the processor that manages all communication between the NAND flash and the host system. When it fails, the drive becomes electrically invisible to the motherboard. Your data is still stored on the NAND chips, but nothing can read it through normal channels.
Professional recovery path: Tools like the PC-3000 Portable III can interface with the controller at the firmware level, work around the failed initialization routine, and extract data from the NAND. In cases where the controller is completely dead, chip-off NAND reading may be an option for older drives without hardware encryption.
Common NVMe controllers that fail this way include Samsung Phoenix (970 EVO/Pro), Elpis (980 Pro), and Pascal (990 Pro), as well as Phison E12 and E16 and Silicon Motion SM2262EN. Each requires a controller-specific recovery procedure.
Free evaluation. $200–$2,500. No data, no fee.
What Not to Do
Once you have confirmed the drive is invisible to the BIOS on multiple systems, stop troubleshooting. The following actions risk making recovery harder or impossible.
- ✗Do not flash BIOS updates hoping it will detect the drive. A BIOS update changes your motherboard firmware. It cannot fix a dead SSD controller.
- ✗Do not use a heat gun or "reflow" the SSD. Heating an NVMe drive damages the NAND flash chips. NAND is temperature-sensitive, and uncontrolled heat degrades the charge stored in the cells, reducing recovery odds.
- ✗Do not repeatedly power cycle the drive. Each power-on attempt forces the controller to attempt initialization. On a partially failed controller, this can cause further damage to the firmware area stored on NAND.
- ✗Do not open the SSD or desolder components. NVMe SSDs have no user-serviceable parts. Removing the controller or NAND packages without the right equipment and firmware knowledge destroys recovery options.
Why NVMe Drives Fail to Detect at the Hardware Level
NVMe detection requires a multi-stage initialization sequence: the PMIC delivers regulated power, the controller boots its firmware from NAND, & the PCIe link training state machine (LTSSM) negotiates a physical connection with the motherboard. Failure at any stage causes the drive to vanish from BIOS without warning. Four distinct hardware failure modes account for non-detection.
Controller ASIC Failure
The NVMe controller is a system-on-chip (SoC) that manages all communication between NAND flash & the PCIe host. It runs its own BootROM on power-up, loads firmware from a reserved NAND partition, initializes the ONFI/Toggle interfaces to each NAND die, & builds the Flash Translation Layer in DRAM. When the ASIC fails from thermal degradation, electrical overstress, or an internal logic lockup, it can't execute any of that.
The drive becomes electrically invisible. The PCIe endpoint never registers on the bus, so BIOS has nothing to enumerate. Recovery requires either reviving the original controller through board-level repair or, for older unencrypted drives, reading the NAND directly with PC-3000 Portable III.
Controllers that commonly fail this way include Samsung Phoenix (970 EVO/Pro), Elpis (980 Pro), and Pascal (990 Pro). Phison E12 and E16 controllers appear across Sabrent, Corsair, and Seagate NVMe drives. Silicon Motion SM2262EN and SM2263XT power ADATA and budget NVMe models.
PMIC Failure & Power Rail Collapse
Every NVMe drive has a power management IC (PMIC) that steps down the motherboard's 3.3V supply into multiple regulated rails. The controller core needs around 1.0V. DRAM runs on 1.2-1.8V. NAND flash operates at 2.5-3.3V. If any rail collapses, the component it feeds can't function.
Two failure signatures tell us what happened. A dead short draws over 1A on the 3.3V input line; the shorted component overheats & is visible through FLIR thermal imaging within seconds of power-on. An open circuit draws under 30mA, meaning the controller stays completely unpowered. Power surges, failing PSUs, & improper shutdowns are the primary causes of PMIC death.
Board-level repair for PMIC failure runs $600–$900 at our lab. We isolate the damaged component via FLIR thermal imaging, remove it with an Atten 862 hot air rework station, & replace it with a donor component using a Hakko FM-2032 microsoldering iron. The original controller stays intact, which preserves its firmware configuration and any encryption keys tied to its silicon.
Firmware & Flash Translation Layer Corruption
The FTL maps logical block addresses (what your operating system sees) to physical NAND pages (where the data is stored). During normal operation, the FTL lives in DRAM & gets flushed to NAND periodically. Power loss during a flush or during garbage collection truncates the write, corrupting the mapping table. On next boot, the controller can't load its FTL & either enters safe mode or fails to complete the PCIe handshake entirely.
DRAM-less NVMe drives are the most vulnerable. Drives using Host Memory Buffer (HMB) architecture, like those with Silicon Motion SM2263XT or Maxio MAP1602A controllers, cache their FTL in the host system's volatile RAM. A sudden power cut wipes the host RAM instantly, & the FTL checkpoint stored on NAND may be hundreds of milliseconds stale.
Symptoms range from 0MB reported capacity to an incorrect model string (garbled characters instead of "Samsung 970 EVO Plus") to complete enumeration failure. Phison E12 drives sometimes report 144PB capacity from FTL corruption. Recovery through PC-3000 Portable III starts at $900–$1,200 and involves entering the controller's diagnostic mode, uploading a loader to controller RAM, and rebuilding the logical-to-physical mapping from NAND spare area metadata.
NAND Die Failure & Controller Initialization Hang
The controller must read firmware, configuration tables, & translation maps from NAND during every boot. If the NAND blocks storing these critical structures suffer from charge leakage or high bit error rates, the onboard LDPC error correction can't decode them. The controller enters an infinite retry loop or locks into a BSY (busy) state.
In this state, the drive may reach PCIe link state L0 (active) but never responds to NVMe Admin commands like Identify Controller. BIOS sees an empty slot or hangs waiting for the unresponsive endpoint. TLC & QLC NAND are more susceptible to this failure than MLC because they store 3 or 4 bits per cell, with tighter voltage margins between programmed states. High program/erase cycle counts & elevated storage temperatures accelerate the charge leakage.
How PC-3000 Recovers Data from Non-Detecting NVMe Drives
PC-3000 Portable III works around the drive's normal boot sequence and communicates directly with the controller at the firmware level. The recovery workflow adapts based on whether the failure is electrical (PMIC/passive component), firmware-level (FTL corruption), or a combination of both. Board-level repair is a prerequisite when the controller won't power on at all.
- Bench-power the drive independently & measure current draw on the 3.3V rail. A healthy NVMe drive draws 0.3-0.8A during initialization. Over 1A indicates a short; under 30mA indicates an open circuit or dead PMIC.
- Scan the PCB with a FLIR thermal camera to identify hotspots that pinpoint shorted components. A shorted PMIC or decoupling capacitor will heat up within 2-3 seconds of power application.
- Perform board-level repair if needed. Remove the failed component with an Atten 862 hot air rework station & replace it using a Hakko FM-2032 microsoldering iron on an FM-203 base station. For BGA-packaged controllers, use a Zhuo Mao precision BGA rework station for controlled reflow.
- Enter the controller's diagnostic mode. Silicon Motion controllers use a Safe Mode entry triggered by pin shorting. Phison controllers enter Safe Mode through a similar hardware trigger. Each controller family has a different entry procedure.
- Upload a specialized loader into the controller's RAM. The loader replaces the corrupted firmware temporarily and gives PC-3000 Portable III direct access to the NAND contents without relying on the drive's broken boot sequence.
- Build a virtual translator to reconstruct the logical-to-physical NAND mapping. PC-3000 reads the NAND spare areas where the FTL metadata is stored and rebuilds the address map that the corrupted firmware couldn't load.
- Extract data through PC-3000 Data Extractor using the reconstructed mapping. The extracted data is written to a target drive for return to the customer.
Many NVMe SSDs implement AES-256 hardware encryption with keys tied to the controller silicon. If the controller is dead on an encrypted drive, removing the NAND chips yields only ciphertext. Board-level repair to revive the original controller is the only recovery path for encrypted drives. Even on drives without AES-256, proprietary LDPC error correction and data scrambling make chip-off recovery impractical without a functioning controller. For most failed NVMe SSDs, board repair IS data recovery.
NVMe Controller Failure Signatures & Recovery Approaches
| Controller Family | Common Drives | Failure Signature | Recovery Approach |
|---|---|---|---|
| Samsung Phoenix / Elpis / Pascal | 970 EVO/Pro, 980 Pro, 990 Pro | Controller lockup; drive may reach PCIe L0 but ignores Admin commands | Board-level repair to revive controller; limited PC-3000 Portable III support for data extraction |
| Phison E12 / E16 | Sabrent Rocket, Corsair MP600, Seagate FireCuda 520 | FTL corruption showing 144PB capacity or garbled model string | Safe Mode entry; L2P table reconstruction from NAND spare areas |
| Silicon Motion SM2262EN / SM2263XT | ADATA SX8200 Pro, HP EX950, various budget NVMe | Safe Mode boot; 0MB capacity; controller alive but firmware unloadable | Techno Mode pin short; loader upload; virtual translator rebuild |
| Maxio MAP1602A | Budget Gen4 NVMe (Teamgroup, Netac, various OEM) | HMB cache loss on power failure; FTL corruption with total non-detection | Firmware-level recovery only; AES-256 encryption makes chip-off unviable |
When Recovery Software Can't Help an Undetected NVMe Drive
Consumer recovery software like Disk Drill, EaseUS, PhotoRec, & R-Studio requires a functioning communication path between the operating system & the storage device. When an NVMe SSD doesn't appear in BIOS, that communication path doesn't exist. No software tool can create one. Software recovery addresses logical problems on physically healthy drives; it can't fix dead controllers or corrupted firmware.
These tools work when the SSD is physically healthy but has a logical problem: accidentally deleted files (before TRIM executes), a corrupted partition table, or a formatted volume. That changes the moment the controller is dead or the firmware is corrupted. Software can't talk to a drive that won't power on.
One more barrier applies to deleted file recovery on NVMe drives specifically. TRIM (called Deallocate in NVMe) is enabled by default on Windows 7+ and macOS 10.6.8+. When you delete a file, the operating system tells the controller which blocks are no longer needed. The controller unmaps those logical addresses and schedules garbage collection, which erases the underlying NAND pages. Once garbage collection runs, the data is gone. No software and no lab can reverse that. Recovery is only possible if TRIM didn't execute: the drive was pulled immediately, TRIM was disabled, or the file system doesn't support TRIM.
For a non-detecting NVMe drive, professional lab recovery with PC-3000 Portable III and board-level repair capability is the only path. NVMe firmware recovery starts at $900–$1,200. Circuit board repair for dead PMICs and shorted components runs $600–$900. +$100 rush fee to move to the front of the queue.
NVMe SSD Recovery in the Lab
SSD Recovery Pricing
NVMe controller failure recovery follows our standard SSD recovery pricing tiers. Free evaluation, firm quote before work begins. No data = no charge. Call (512) 212-9111.
Simple Copy
Low complexityYour drive works, you just need the data moved off it
$200
3-5 business days
Functional drive; data transfer to new media
Rush available: +$100
File System Recovery
Low complexityYour drive isn't showing up, but it's not physically damaged
From $250
2-4 weeks
File system corruption. Visible to recovery software but not to OS
Starting price; final depends on complexity
Circuit Board Repair
Medium complexityYour drive won't power on or has shorted components
$450–$600
3-6 weeks
PCB issues: failed voltage regulators, dead PMICs, shorted capacitors
May require a donor drive (additional cost)
Firmware Recovery
Medium complexityMost CommonYour drive is detected but shows the wrong name, wrong size, or no data
$600–$900
3-6 weeks
Firmware corruption: ROM, modules, or system files corrupted
Price depends on extent of bad areas in NAND
PCB / NAND Swap
High complexityYour drive's circuit board is severely damaged and requires NAND chip transplant to a donor PCB
$1,200–$1,500
4-8 weeks
NAND swap onto donor PCB. Precision microsoldering and BGA rework required
50% deposit required; donor drive cost additional
50% deposit required
Hardware Repair vs. Software Locks
Our "no data, no fee" policy applies to hardware recovery. We do not bill for unsuccessful physical repairs. If we replace a hard drive read/write head assembly or repair a liquid-damaged logic board to a bootable state, the hardware repair is complete and standard rates apply. If data remains inaccessible due to user-configured software locks, a forgotten passcode, or a remote wipe command, the physical repair is still billable. We cannot bypass user encryption or activation locks.
No data, no fee. Free evaluation and firm quote before any paid work. Full guarantee details. NAND swap requires a 50% deposit because donor parts are consumed in the attempt.
Rush fee: +$100 rush fee to move to the front of the queue.
Donor drives: A donor drive is a matching SSD used for its circuit board. Typical donor cost: $40–$100 for common models, $150–$300 for discontinued or rare controllers.
Target drive: The destination drive we copy recovered data onto. You can supply your own or we provide one at cost plus a small markup. All prices are plus applicable tax.
Frequently Asked Questions
Can data be recovered from an NVMe SSD that is not detected in BIOS?
Why did my NVMe SSD suddenly stop being detected?
Is an NVMe SSD that shows in BIOS but not in Windows recoverable?
What causes an NVMe SSD to suddenly stop being detected after a power outage?
What is the difference between an NVMe drive not detected in BIOS vs. detected but showing wrong capacity?
Related Recovery Services
NVMe drive missing from BIOS?
Free evaluation. $200–$2,500. No data, no fee. Ship your drive to our Austin lab.