Your CPU hit 95C the second a game loaded, the fans screamed, and the machine throttled or shut down. But your all-in-one liquid cooler is right there, radiator fans spinning, so what gives? An AIO that suddenly stops cooling almost never means the whole unit is dead. More often it’s a stalled pump, air trapped in the loop, dried-out paste, or a mounting bracket that worked loose. We researched the common failure points owners report and lined them up from the quickest checks to the fixes that actually solve it. If you end up shopping for a replacement, our guides to the best 240mm AIO of 2026 and the best 360mm AIO of 2026 can point you at a proven unit.
First check the obvious (the 30-second check)
Before you pull anything apart, put your hand on the tubes and the pump housing. If the pump is running, you’ll feel a faint vibration and one tube will warm up within a minute of load while the other stays cooler. Dead silence and zero vibration point straight at a stalled pump. Next, open your fan software and confirm the pump header actually reports RPM. A common mistake is plugging the pump into a fan header set to a low or zero-RPM curve, which starves it under load.
Also glance at your temps at idle versus load. If idle sits fine but the CPU spikes 40C the instant you open a game, that’s a cooling loop that can’t move heat fast enough, not a broken sensor. And check the simple stuff: is the radiator caked in dust, are all fans spinning, and did a case fan curve get reset after a BIOS update? Quick wins hide here.
One more 30-second check pays off: reseat the block. A cooler that shipped fine can loosen after months of thermal cycling and vibration, and even a slightly uneven mount tanks contact pressure. Give each mounting screw a gentle turn to confirm it’s still snug. Don’t crank them, just feel for a screw that’s backed off. If one was loose, that alone can explain a 15C jump.
Cause #1: The pump has stalled or is dying
The pump is the heart of an AIO, and it’s the part most likely to quit. Owner reports show pumps often fail after three to five years as the internal bearing wears or the motor coil degrades. When it stalls, coolant stops circulating, heat pools in the block, and your CPU temperature climbs like the cooler isn’t there at all. You can confirm it by listening for a faint hum with your ear near the block, or by watching whether both tubes eventually reach a similar temperature under load. If one tube stays cold, coolant isn’t moving.
Sometimes a stalled pump is just a power or header problem you can fix in minutes. Plug the pump into the dedicated AIO or CPU_OPT header, set that header to full 100% RPM in BIOS or your fan app, and reseat the connector. Some boards default the pump header to a temperature-based curve that idles it too low, so a manual full-speed setting is the first thing to try. If the pump still won’t spin or reports zero RPM after that, the pump itself is done and no amount of new paste will save it. That’s the point where you decide between replacing the whole AIO or switching to air cooling. A simple air cooler like the Intel E97379 stock-style cooler at $15.99 can keep a machine running as a stopgap while you shop for a real replacement. It’s a 4-pin unit with an aluminum heatsink and a 3.5-inch fan, rated 4.1, and it fits older Intel sockets, so verify your socket before buying one as a bridge.
Pros
- Native LGA 1150, 1155, and 1156 socket support covers three legacy Intel generations
- 4-pin PWM connector enables proper fan curve control via BIOS
- OEM-style aluminum heatsink footprint keps RAM and VRM clearance predictable
- Push-pin mount installs without removing the motherboard from the case
Cons
- Zero owner fedback available at time of writing, real-world thermals unverified
- Aluminum-only heatsink with no heatpipes struggles with overclocked or unlocked K-series CPUs
- Restricted to obsolete Intel sockets, no path forward to LGA 1200, 1700, or AM5
The Intel E97379-003 is an OEM-style stock CPU cooler built for LGA 1150, 1155, and 1156 boards, covering Core i3, i5, and i7 chips from the Sandy Bridge through Haswell era. It targets one specific buyer: someone repairing, refurbishing, or restoring a legacy Intel desktop that lost its original heatsink.
The defining feature is compatibility, not performance. A 3.5-inch fan sits on an aluminum heatsink with a 4-pin PWM header, matching the original OEM cooler dimensions Intel shipped in retail boxes. Stock coolers in this class typically handle 65W to 84W TDP chips at stock clocks, which lines up with locked i3, i5, and non-K i7 SKUs on these sockets.
Trade-offs are typical for a boxed-style cooler. There are no heatpipes, no copper base contact, and no tower geometry, so sustained multi-core loads will spin the fan up and get audible. Push-pin mounting is convenient but less secure than a backplate design, and overclocking headroom is effectively zero. Owner feedback is not available at time of writing.
Buy this if you need a like-for-like OEM replacement for a locked LGA 1150, 1155, or 1156 CPU and want to kep the original acoustic and clearance profile. Skip this if the CPU is a K-series part, if you plan to overclock, or if the target socket is LGA 1200 or newer.
Socket coverage: Native mounting for LGA 150, 1155, and 1156 only. That spans 2nd, 3rd, and 4th generation Core i3, i5, and i7 desktop chips. No adapter is included for LGA 1200, 1700, or AM4/AM5, so this cooler is locked to legacy Intel repair scenarios.
Heatsink and fan: Aluminum radial-fin heatsink with a 3.5-inch axial fan on a 4-pin PWM connector. No heatpipes, no copper slug, no vapor chamber are indicated in source data. TDP handling is not specified, but OEM colers of this geometry are typically rated for the 65W to 84W stock TDP band on these sockets.
Mounting and clearance: Push-pin retention through the four motherboard holes, no backplate required, install possible without removing the board. Cooler height sits low enough to clear standard mid-tower side panels, and the round footprint keps RAM slots and VRM heatsinks unobstructed. Exact height in mm is not specified.
Acoustics and headroom: Noise level in dBA is not specified. Based on OEM stock cooler behavior at this size, expect quiet idle and audible ramp under sustained all-core load. Overclocking a K-series chip on this heatsink is not advisable, thermal headroom above 84W is minimal.
Cause #2: Air bubbles trapped in the loop
A gurgling or ticking sound from the pump usually means air has collected in the block. Over time, every sealed AIO loses a little coolant to permeation through the tubing, and that leaves a growing air pocket. When that pocket cycles through the pump, it can’t move liquid efficiently, and you get inconsistent temps that spike and recover for no obvious reason. It’s one of the most common complaints on units two or more years old.
The fix is orientation. The pump should never sit at the highest point in the loop, because air rises and will collect right where it does the most damage. Mount the radiator so its top edge sits above the pump, ideally front-mounted or top-mounted with the tubes entering at the bottom. If air is already trapped, power the system and gently tilt or rock the case for a minute so the bubble migrates up into the radiator, away from the pump. That alone rescues a lot of coolers people assumed were failing.
Worth knowing: some air noise is normal for the first day or two after a fresh install as the loop settles. If a cooler that ran silent for years suddenly starts gurgling, though, that’s coolant loss showing its hand. You can slow the process by keeping the pump on a steady speed, but permeation is physics, and every sealed AIO eventually reaches the end of its useful coolant volume. That’s usually the two-to-five-year window where owners start reporting temp creep.
Cause #3: Dried thermal paste or a bad mount
If the pump runs, the loop is quiet, and temps are still bad, the problem is almost always at the contact point between the cold plate and the CPU. Old thermal paste dries out and cracks, leaving air gaps that wreck heat transfer. Owners who reseat a cooler after a few years often see idle temps drop 10C or more just from fresh paste. A loose or uneven mount does the same thing, since even pressure across the die is what makes the cold plate work.
Pull the block, clean both the CPU and the cold plate with a lint-free wipe, and apply a fresh pea-sized dot of quality paste before remounting in a diagonal cross pattern to even out the pressure. Thermal Grizzly Duronaut is built for exactly this, a 6-gram tube rated for high conductivity and long-term durability, and it ships with cleaning wipes so you can prep the surfaces in one go. At $24.99 with a 4.4 rating, it’s cheap insurance against a mount that’s silently choking your cooler. Snug the screws in small alternating turns until they stop, and don’t overtighten.
A few habits make the repaste stick. Let the CPU cool fully before you pull the block, since warm paste smears and makes cleanup messy. Wipe in one direction, then follow with a dry pass so no residue is left to trap air. And resist the urge to spread the paste by hand; the mounting pressure does that job more evenly than a spatula ever will. If your cold plate has a protective film from the factory still on it, that’s an easy one to miss, and it’ll wreck your temps until it’s peeled off.
Pros
- Non-conductive chemistry safer for novices repasting near exposed CPU or GPU capacitors.
- 6g volume suports several full applications on desktop CPUs and console APUs.
- Wet and dry wipes included, so no isopropyl bottle or lint-free cloth need.
Cons
- Limited owner fedback at time of writing makes long-term durability claims hard to verify.
- Non-metal formula typically trails top liquid metals by a few degrees under sustained load.
- Viscosity, TIM lifespan hours, and thermal conductivity W/mK figures are not specified.
Duronaut is Thermal Grizzly's non-conductive thermal paste positioned as a mid-tier daily driver, shipping in a 6g syringe with 6 wet and 6 dry cleaning wipes. The target buyer is a DIY PC builder repasting a CPU or GPU cooler, or a console owner refreshing PS4, PS5, or Xbox thermals without risking shorts.
The defining trait is the aluminum microparticle plus zinc oxide nanoparticle blend paired with an electrically non-conductive base. Thermal Grizzly claims reduced pump-out and stable performance over extended cycles, which maters most on soldered console APUs and overclocked desktop chips that see large delta-T swings. Exact W/mK conductivity is not specified in the source.
Trade-offs are typical for non-conductive pastes at this tier. Expect a few degrees warmer than liquid metal on delided or high-wattage CPUs, and no published viscosity or cure-time data to plan application around. Owner feedback is limited at time of writing, so long-term hardening resistance relies on vendor claims rather than independent long-run testing.
Buy this if you want a safe, non-conductive paste with enough volume for several repastes and a bundled wipe kit for console or GPU jobs. Skip this if you are chasing the absolute lowest core temperatures on a 250W-plus overclocked CPU where liquid metal or a top nano-diamond paste still leads.
Volume and coverage: The syringe holds 6g of paste, enough for roughly 15 to 25 pea-sized applications on mainstream AM5 or LGA1700 IHS surfaces, or several full console APU repastes on PS4, PS5, and Xbox where die coverage is larger than a desktop CPU heatspreader.
Composition and safety: The compound uses aluminum microparticles and zinc oxide nanoparticles in a non-metal carier, rated electrically non-conductive. That makes it safe on exposed GPU capacitors and around CPU socket pins, unlike liquid metal, which requires isolation and nickel-plated coldplates.
Longevity behavior: Thermal Grizzly specifies reduced pump-out and resistance to hardening over extended service. Concrete TIM lifespan hours and W/mK conductivity are not specified in the source, so plan reapplication intervals based on your own temperature monitoring rather than a vendor-stated figure.
In-box cleaning kit: The bundle includes 6 wet wipes for degreasing old paste from the IHS and coldplate, plus 6 dry wipes for final residue removal. Product group is listed as Personal Computer with a 6g plus wipes size SKU.
Preventive maintenance
Most AIO problems are preventable with a little routine care. Blow the radiator fins out with compressed air every few months, since a dust-clogged radiator can’t shed heat no matter how well the pump runs. Keep the pump on a 100% fixed RPM curve rather than a variable one, because cycling the pump speed shortens its life and lets air pockets form. And mount the loop correctly from day one so the pump never sits at the top.
It’s also worth repasting the CPU every couple of years even if temps seem fine, because paste degrades slowly and you rarely notice the creep until it’s a problem. Keep an eye on your load temps monthly with a monitoring app so you catch a slow climb early, while it’s still a cheap fix. A cooler that’s trending 3-4C warmer month over month is telling you something before it becomes an emergency.
When to call a pro vs DIY
Reseating a cooler, replacing paste, fixing a fan curve, and swapping to a spare air cooler are all straightforward DIY jobs that need nothing but a screwdriver and patience. If your AIO is sealed and the pump is genuinely dead, there’s no user-serviceable repair, so replacement is the honest answer. Consider a shop only if the overheating persists after fresh paste and a confirmed-working pump, which can hint at a deeper board or sensor fault worth a second opinion.
Weigh the cost too. A new mid-range 240mm or 360mm AIO runs less than an hour of most repair-shop labor, so once a pump has failed, buying new almost always beats paying someone to diagnose a sealed unit. Save the pro visit for cases where the CPU overheats even with a known-good cooler bolted on, since that points at something on the board rather than the cooling loop. When in doubt, borrow or buy a cheap air cooler and use it to isolate whether the AIO is truly the culprit before you spend on anything bigger.
Tools and parts needed
You’ll want a Phillips screwdriver, isopropyl alcohol or the wipes bundled with a paste kit, a fresh tube of thermal compound, and compressed air for the radiator. A spare air cooler is smart to keep on hand so a dead pump doesn’t leave your rig unusable while a new AIO ships. That’s the entire kit for the vast majority of AIO cooling fixes, and none of it costs much. If you’re replacing the whole unit, match the radiator size to your case and CPU heat load. Our best 240mm AIO for overclocking roundup covers higher-heat builds.
Common questions
How do I know if my AIO pump is dead?
Feel the pump housing and tubes under load. A working pump vibrates faintly and warms one tube within a minute, while a dead one stays silent and cold with both tubes at the same temperature. Confirm it in software: if the pump header reports zero RPM after you set it to full speed and reseat the connector, the pump has failed.
Can I refill a sealed AIO that lost coolant?
A few premium units have a fill port, but most sealed AIOs are not meant to be topped up, and cracking one open usually voids the warranty and risks leaks. If a low-coolant AIO is causing air pockets and poor temps, the practical move is replacement rather than a refill. Custom loops are the route if you want serviceable coolant.
Will new thermal paste fix high CPU temps?
It often helps a lot when the pump is healthy and the paste is old. Dried, cracked compound leaves air gaps that block heat transfer, and a fresh application can drop temps 10C or more. If the pump is dead or air is trapped in the loop, new paste alone won’t solve it, so confirm those first. And remember an AIO’s realistic lifespan is three to five years, so if yours is older and temps are creeping, a fresh tube of paste may only buy you a little time before a full replacement makes more sense.
Write Your Review
No reviews yet. Be the first to share your experience!