Most PC gamers never think about their NVMe SSD until performance falls off a cliff mid-session. The classic symptom pattern looks like this: everything runs fine for the first 20–40 minutes of gaming, then load times balloon, texture streaming stutters, and frame times get choppy in a way that GPU or CPU monitoring doesn’t explain. Open a tool like CrystalDiskInfo or HWiNFO64 and you’ll see the drive sitting at 75–85°C — sometimes pegged near its thermal shutdown point. Most consumer NVMe drives enter thermal throttling somewhere between 70°C and 80°C, dropping sequential throughput by 30–50% or more until temperatures recover.
This isn’t usually an emergency, but it’s not something to ignore either. Sustained heat above roughly 85°C accelerates NAND wear and can shorten the drive’s usable life, and sudden thermal shutdowns can corrupt in-flight writes. The good news: in the vast majority of cases, this is a 15-minute fix that costs nothing. This guide walks through diagnosis and repairs in order of difficulty, from free software tweaks to a five-dollar heatsink. By the end of 2026, most current-gen motherboards ship with adequate heatsink coverage, but plenty of older or budget boards still leave M.2 slots bare.
Likely Causes
| Cause | How to confirm | Fix difficulty |
|---|---|---|
| No heatsink on the drive | Open the case and visually inspect: a bare M.2 stick with no aluminum plate over it | Easy |
| Drive mounted under the GPU | Check motherboard manual for M.2 slot positions; the slot between the CPU and PCIe x16 slot is almost always the hottest | Easy–Moderate |
| Restricted case airflow | Drive temps climb over a long session even at low workload; dust on intake filters; GPU dump fans blowing hot air into a dead zone | Moderate |
| Factory thermal pad missing or misaligned | Remove the motherboard’s M.2 heatsink and check that a thermal pad contacts the controller chip, not just the NAND | Easy |
| ASPM / power states disabled | Run powercfg /a in Command Prompt; check drive idle temperature in HWiNFO — an idle NVMe at 60°C+ is often never entering low-power states |
Easy |
| Heavy background write workload | Open Task Manager > Performance > Disk and look for sustained activity with no apps running (Windows Search indexing, OneDrive sync, telemetry) | Moderate |
Fix 1: Enable ASPM So the Drive Idles Cool
What causes it
Active State Power Management (ASPM) lets PCIe devices, including NVMe drives, drop into low-power L1 states when idle. If it’s disabled in the BIOS or overridden by the Windows power plan, your SSD stays partially powered at all times, adding a constant heat load even when nothing is reading from it.
How to check
Boot into the BIOS and look under PCIe or advanced power settings for “ASPM” or “Native ASPM” — if it’s set to Disabled, that’s your problem. In Windows, open Control Panel > Power Options > Change plan settings > Change advanced power settings > PCI Express > Link State Power Management and see whether it’s set to Off.
What to do
Set Windows Link State Power Management to “Moderate power savings” (or “Maximum power savings” if the drive is not your OS boot drive — on boot drives, Moderate avoids rare wake latency). In the BIOS, set ASPM to L1 if the option exists, and enable “Native ASPM” if your board offers it. Reboot and let the machine idle 10 minutes; idle temps should drop noticeably.
How to undo it
Return the Windows setting to Off and the BIOS option to Disabled. No lasting changes are made.
Fix 2: Reapply or Install the Motherboard Heatsink Pad
What causes it
Many motherboards include an M.2 heatsink, but the factory thermal pad is often thin, misaligned during installation, or stuck to the heatsink rather than the drive when someone swaps SSDs. A pad that bridges the wrong chips does almost nothing for the controller, which is where the heat is generated.
How to check
Power down, flip the PSU switch, and hold the case power button for 5 seconds to drain residual charge. Remove the M.2 heatsink (usually one small Phillips screw with a torqued-down standoff). Look at the pad: it should be positioned directly over the black controller chip in the center of the drive, not just the rectangular NAND packages. If the pad is torn, compressed flat, or left stuck to the heatsink, it needs replacing.
What to do
Peel off the old pad residue with isopropyl alcohol (90%+) and a lint-free cloth. Install a new thermal pad 1mm–1.5mm thick over the controller (many drives also benefit from a second pad over the NAND on the front side). Reattach the heatsink and confirm the drive doesn’t rock or sit proud of the standoff — a pad that’s too thick prevents proper contact.
How to undo it
There’s nothing to revert — but if you used an overly thick pad and it caused contact issues, swap back to 1mm.
Fix 3: Move the Drive to a Cooler M.2 Slot
What causes it
On most ATX boards, the primary M.2 slot sits directly under the graphics card’s fans and heatsink. The GPU dumps 200–400W of heat straight into that region, and the SSD absorbs it. Secondary slots below the GPU or above the PCIe x16 slot typically run 10–15°C cooler.
How to check
Consult your motherboard manual’s layout diagram to identify all M.2 slots and which share lanes with SATA ports or PCIe slots. Note which slot your drive currently occupies and whether it’s above or below the GPU.
What to do
If a lower slot shares bandwidth with nothing you’re using, move the drive there. On AMD boards, be aware that the second slot often drops to PCIe x2 or shares lanes with SATA — check the manual before assuming it’s a downgrade-free move. Your OS drive can be moved without reinstalling Windows; the board will simply enumerate it from a different slot.
How to undo it
Move the drive back to the original slot. No software changes needed.
Fix 4: Improve Case Airflow Around the M.2 Area
What causes it
Even a drive with a good heatsink needs moving air to dump heat into. Negative-pressure cases, clogged dust filters, or GPU dump fans blowing into a stagnant corner can create a hot pocket around the M.2 slot that no amount of heatsinking fully solves.
How to check
With the side panel off, note whether drive temps drop by 5°C or more during a gaming session — that’s a strong sign of case-level airflow restriction. Also inspect front intake filters for dust and confirm your front fans actually spin under load.
What to do
Clean intake filters and fan blades. Aim for balanced or slightly positive pressure: at least two front intakes and one rear/top exhaust. If your case supports a bottom intake fan under the GPU, adding one does double duty, pulling cool air across both the GPU and a lower M.2 slot. As a quick test, run one session with the side panel off and compare drive temps in HWiNFO.
How to undo it
Return fans to their original configuration if you relocated them. Cleaning and filters require no reversal.
Fix 5: Install an Aftermarket M.2 Heatsink
What causes it
Bare drives — especially PCIe 4.0 and 5.0 models with high-end controllers like Phison E18 or E26 — have no way to shed heat on their own. Controller chips on fast Gen4/Gen5 drives can hit throttle points in minutes without a mass of metal to absorb and spread it.
How to check
If the drive shipped without a heatsink and your motherboard’s M.2 slot has no factory heatsink either (confirm in the manual or by visual inspection), this applies to you. Drives rated above 7,000 MB/s are especially heat-sensitive.
What to do
Buy a low-profile aluminum heatsink with a thermal pad — most are under $10 and fit any standard M.2 2280 drive. Options include simple finned aluminum plates, or copper shims sandwiched between aluminum. Clean the controller with isopropyl, apply the pad, and clamp the heatsink with the included clip or screws. Verify clearance with your GPU — low-profile designs are 5–8mm tall and fit under almost every card. Note: if your drive is a Gen5 model, a passive low-profile sink is usually adequate; a full motherboard-integrated heatsink with a fan is better but requires removing the stock label and pads.
How to undo it
Unscrew or unclip the heatsink and peel off the pad. Residual adhesive comes off with isopropyl alcohol.
Fix 6: Kill Background Write Workloads
What causes it
Windows Search indexing, OneDrive/Dropbox sync, telemetry uploads, Steam’s client bootstrapper, and shader compilation can keep a drive at 20–40% activity for extended periods. Sustained low-level writes generate steady controller heat even though nothing “big” appears to be happening.
How to check
Open Task Manager > Performance > Disk (or Resource Monitor > Disk tab) and sort by total bytes/sec with no games or apps running. Any process sustaining writes for more than a few minutes deserves investigation.
What to do
Exclude your game libraries from Windows Search: Settings > Privacy & security > Searching Windows > add folders to exclusions. Pause OneDrive sync during gaming hours from its system-tray icon. In Steam: Settings > Downloads and disable “Allow downloads during gameplay.” Move your pagefile’s heavy load off the SSD only if you have another drive — otherwise leave it; a missing pagefile causes worse problems than heat.
How to undo it
Remove search exclusions and re-enable downloads during gameplay from the same menus.
Fix 7: Update Firmware and the Storage Controller Driver
What causes it
SSD firmware governs how aggressively the controller manages power and thermal headroom. Vendors have shipped firmware updates that improve idle power states and throttle behavior, and Windows’ generic StorAHCI/NVMe driver sometimes lacks power-management features the vendor driver supports.
How to check
Identify your exact drive model in Device Manager > Disk drives. Check the manufacturer’s support page (Samsung Magician, WD Dashboard, Crucial Storage Executive, Kingston SSD Manager, Corsair SSD Toolbox — most vendors offer a utility that reports current firmware). Also check Windows Update > Optional updates for any storage driver offerings.
What to do
Back up important data first — firmware updates carry a small risk. Run the vendor utility and apply any pending firmware update. Keep the process uninterrupted: don’t reboot or sleep the PC mid-flash. Afterward, confirm in the utility that the new firmware version is active, then re-check drive temps over a normal session.
How to undo it
Firmware downgrades are generally not supported and shouldn’t be attempted. If behavior worsens, check for a follow-up hotfix from the vendor.
Fix 8: Last Resort — Add Active Cooling to the M.2 Slot
What causes it
Gen5 drives and some high-end Gen4 drives in poorly ventilated small-form-factor cases can throttle even with a good passive heatsink, because the surrounding air itself is too hot for passive transfer to keep up.
How to check
If drive temps still hit 80°C+ with a verified heatsink, good airflow, correct power settings, and updated firmware, passive cooling has hit its ceiling in your case.
What to do
Two options: replace the motherboard’s stock M.2 heatsink with an aftermarket version that includes a small 20–30mm fan (these plug into a motherboard fan header or SATA power via adapter), or add a small ducted fan blowing across the M.2 area. Set the fan to run off a motherboard header controlled by BIOS fan curves — target ramping up when a temp sensor near the PCH exceeds 50°C.
How to undo it
Disconnect the fan, revert to the stock heatsink, and delete any custom BIOS fan curve.
If Nothing Works
If the drive still throttles after heatsinking, airflow fixes, power-management changes, and firmware updates, suspect a deeper issue. First, test the drive in another system or a USB NVMe enclosure — if it stays cool there, your motherboard’s slot or chipset region may simply run hot, and the slot-swap in Fix 3 becomes mandatory rather than optional. Second, run the vendor’s diagnostic tool (Samsung Magician’s Diagnostic Scan, WD Dashboard’s SMART check, or CrystalDiskInfo’s SMART values) and watch for reallocated sectors, rising “Controller Temperature” alerts, or “Percentage Used” climbing fast — a failing controller can overheat as a symptom of internal faults. Third, consider whether the drive is simply undersized for your workload: a small-capacity drive (500GB or less) thrashes more during large game installs and shader compilation, generating heat that a 2TB model in the same slot would not. If SMART data shows deterioration, replace the drive — thermal issues plus failing SMART is not a combination worth gambling your library on. Back up first, always.
FAQ
What temperature should my NVMe drive run at while gaming?
Under sustained gaming load, 50–70°C is normal and healthy. Throttling typically begins between 70°C and 80°C depending on the controller, and most drives force a protective shutdown near 85–90°C. Sustained temps above 80°C should be addressed even if you haven’t noticed stutter yet.
Does thermal throttling damage my SSD?
Throttling itself is a protective mechanism, not damage. However, the sustained heat that causes throttling does accelerate NAND and controller wear over time. Occasional throttle events won’t kill your drive; daily sessions spent at 85°C+ will shorten its lifespan.
Do all NVMe drives need a heatsink?
Not strictly — but PCIe 4.0 drives above 7,000 MB/s and all PCIe 5.0 drives benefit enormously from one, and some Gen5 drives refuse to sustain rated speeds at all without a motherboard-integrated heatsink. Gen3 drives are usually fine bare unless they sit directly under a hot GPU.
Will a heatsink interfere with my GPU or motherboard warranty?
No. Adding an aftermarket heatsink doesn’t modify your motherboard or GPU, and applying a thermal pad to your SSD is likewise non-destructive. Just avoid removing any label or warranty sticker from the SSD itself, and check GPU clearance (5–8mm typical) before buying a tall finned cooler.
