GPU Fan Not Spinning on Boot Zero-RPM vs Failed Fan

Why are my GPU fans not spinning at boot?

When you press the power button and the PC starts but the graphics card fans stay completely still, your first instinct is panic — but on every modern NVIDIA RTX 20 series and newer, and on AMD RX 5000 series and newer, this is normal behavior. Modern cards ship with zero-RPM idle mode (NVIDIA calls it "0dB", AMD calls it "Zero RPM"), which keeps the fans completely stopped until the GPU temperature crosses a threshold — typically 50–60°C depending on the AIB partner (ASUS, MSI, Gigabyte, EVGA, PNY).

The real question is: are your fans dead, or are they just idle? This article gives you a 30-second diagnostic that answers it before you RMA a perfectly working card.

The 30-second zero-RPM test

1. Boot to desktop. Leave it idle 60 seconds — fans should be off.
2. Open any GPU monitor (HWiNFO64, MSI Afterburner, GPU-Z). Note idle temperature (typically 35–48°C).
3. Run FurMark for 60 seconds, or launch any modern game at native resolution.
4. Watch the temperature climb past 55–60°C.
5. Fans should kick in automatically between 50–65°C.

If the fans spin under load → it's working as designed, stop worrying. If temperature crosses 70°C and fans are still still → real failure, continue with the fixes below.

GPU load and expected fan behavior

| State | Wattage (RTX 4070 example) | Temperature | Expected fan RPM |

|-------|---------------------------|-------------|------------------|

| BIOS / boot screen | 15–25 W | 35–45°C | 0 RPM (normal) |

| Windows desktop idle | 15–30 W | 38–50°C | 0 RPM (normal) |

| YouTube 4K / dual monitor | 30–60 W | 45–55°C | 0–800 RPM (variable) |

| Light gaming (esports) | 100–150 W | 55–65°C | 1000–1500 RPM |

| AAA gaming / FurMark | 200–280 W | 65–80°C | 1800–3000 RPM |

If your card is pulling >150 W at >65°C and fans are still at 0 RPM, that's a real fault.

When does it occur?

• Brand new build — fans never spin even under load
• After moving the PC, transport, or working inside the case
• After a BIOS/driver update that reset the fan curve
• After cleaning the card with compressed air (debris jam or sticker stuck on blade)
• After 2–4 years of use (bearing wear, usually starts as click/grind first)
• Only one of the two/three fans spins (asymmetric failure)

Common causes

• Zero-RPM mode (NORMAL) — fans off below ~55°C by design on RTX 20+/RX 5000+
• Loose 4-pin fan power connector between the card's PCB and the fan shroud
• Sticker/protective film left on the fan hub from shipping
• Bearing failure after years of use (often preceded by clicking or grinding noise)
• Debris or cable physically blocking blade rotation
• Corrupted custom fan curve in MSI Afterburner, ASUS GPU Tweak, or AMD Radeon Software
• 8-pin PCIe power not fully seated, dropping the card into a power-limited safe mode where fan control disables
• Dead fan controller IC on the card (rarer, usually after a surge)

Step-by-step fixes

1. Run the 30-second zero-RPM test above first — confirm whether you actually have a problem. Idle fans on a cold card are working as designed on every card released since 2018.

1. Force fans to 100% via software — Install MSI Afterburner (works on NVIDIA and AMD). Click the fan-control icon, enable Manual, set fan slider to 100%, apply. If fans spin → mechanical and connector hardware is fine, the issue was just the fan curve. If fans still don't move → mechanical or power problem, continue.

1. Reseat the 8-pin/12VHPWR/16-pin PCIe power cable — Power down, unplug, hold power button 10 seconds to discharge. Unplug GPU power connector from BOTH ends (PSU and card), reseat firmly until you hear/feel the latch click. On RTX 4090/5090, ensure the 12VHPWR connector is fully home — partial seating triggers safety mode that disables fans.

1. Reseat the card in its PCIe slot — Remove the GPU, blow out the PCIe slot with compressed air, reseat firmly. A partially seated card may POST but enter a degraded mode where fan control fails.

1. Check for shipping film or sticker on fan blades — New cards from EVGA, PNY, Gigabyte often ship with a clear plastic film over each fan hub. Look under each fan with a flashlight. If present, peel off.

1. Reset the fan curve to default — In MSI Afterburner: Settings > Fan > Disable user-defined software automatic fan control, apply, restart. In AMD Adrenalin: Performance > Tuning > Reset to Default. In ASUS GPU Tweak III: click the reset arrow on the fan tab. A corrupted curve can leave fans pinned at 0% even when GPU temperature is 80°C.

1. Reinstall GPU drivers with DDU — Boot into Safe Mode, run [Display Driver Uninstaller](https://www.guru3d.com/files-details/display-driver-uninstaller-download.html) > Clean and restart. Reinstall the latest NVIDIA Game Ready / AMD Adrenalin driver. A broken driver-side fan control daemon will keep fans stuck even at 90°C.

1. Test the card in another PCIe slot or another PC — If you have a second PCIe x16 slot (or another machine), move the card. If fans still won't spin under load anywhere, the card itself is the fault — RMA it. Most manufacturers (ASUS, MSI, Gigabyte, EVGA, PNY, Zotac) cover fan failures under their standard 2–3 year warranty.

If it still doesn't work

If only ONE fan out of two or three is dead and the rest spin normally, you can often replace just that fan — most AIB partners sell replacement fans directly through their RMA portal, and on Gigabyte and MSI cards, the fans are individually screwed in and use a standardized 4-pin micro connector. For older cards out of warranty, a generic 92mm/100mm replacement fan from Noctua or Arctic can be wired into the GPU's fan header with a simple adapter. If the fan controller IC itself is dead (all fans dead, blades freely spin when you flick them with a finger, no resistance), the card needs board-level repair — Northridge Fix and similar specialists can resolder a failed PWM controller for far less than buying a new GPU. Before you proceed with any of that, confirm one last time with the test above — a card running cool with healthy zero-RPM is not broken.