You
Hardware

A card stopped working: check this before calling it dead

Last updated: 5 November 2026

Swap parts in an order that leaves the fault one place to hide, learn what common symptoms mean, and handle a return if the card really is dead.

Is the card really the problem?

Usually not. A card that has vanished from the rig has a short list of suspects. The card itself is near the bottom of it.

Above it sit five other parts: the riser, the riser's power cable, the slot, the card's own power cable and the software. All of them fail more often. All of them cost less to replace.

So the method is elimination in a fixed order. Change one thing at a time, cheapest and most likely first. Keep going until the fault has one place left to hide.

The order to eliminate faults in: riser, riser power cable, slot, the card's own power cable, software, and only then the card itself in another machine
Most cards returned as dead were one of the first four.

In what order should you swap parts?

  • Swap the riser for a known-good one. A riser is the small board and cable that connects a card to the motherboard. Risers fail more than anything else in a rig, and a spare on the shelf answers this in two minutes.
  • Swap the riser's power cable, and look at both connectors as you unplug them. Browning or a burnt smell is the answer. It means the cable and whatever it was plugged into.
  • Move the card to a slot that is known to work. Move a known-good card into the failed card's slot too. Doing both tells you which of the two is at fault, instead of only suggesting it.
  • Change the card's own power cable, and use a different socket on the power supply.
  • Reset the software: stock clocks, current driver, one card in the rig. An undervolt that was stable in winter can stop being stable in summer. That looks exactly like a hardware fault.
  • Only now, test the card alone in another machine. If it fails there too, it is the card.

What do the common symptoms mean?

  • Missing from the list of cards entirely: the riser or slot, far more often than the card.
  • Present but producing nothing: usually the driver or the clocks. Sometimes it is memory that has stopped being stable.
  • Fine for twenty minutes, then gone: heat. Check temperatures before anything else. If they are high, check the thermal paste and pads.
  • Errors rather than a crash: memory, especially on a card that has been running hot. Bring memory clocks back to stock and see whether the errors stop.
  • The whole rig reboots: the power supply or a cable, not the card. A failing card does not usually take the machine down with it.
  • It works alone but not in the rig: power. Check the supply, the cabling, then the circuit, in that order.

What should you do before sending a card back?

  • Put the original settings back. That means stock clocks, stock power limit, stock fan curve, and the original BIOS if it was ever changed. A card that arrives with a modified BIOS will come back to you.
  • Clean it. This is not cosmetic. A returned card thick with dust invites questions about how it was treated.
  • Photograph it before it goes into the box. Shoot every side, the connector, the serial number, and the packaging once sealed. This costs nothing and settles arguments.
  • Find the receipt and the serial number. Coverage is usually tied to one or the other.
  • Get the return authorisation before shipping. Almost every manufacturer requires it. A card that arrives without one can be refused or lost internally.

What should you know about warranty coverage?

Graphics card warranties are often non-transferable. So a card bought used often has none, whatever its age. Some manufacturers do transfer them, so check for the specific brand rather than assuming.

Several manufacturers also state that mining on a consumer card voids the warranty. In practice, treat any coverage as a bonus, not a plan.

If the rig's earnings matter to you, keep a spare card or two. A rig that cannot lose a card without stopping has a single point of failure. And cards fail.

What can you do with a card that is really dead?

A card out of coverage is not always finished. Fans, thermal paste and thermal pads can all be replaced. They cause a large share of what actually stops a mining card working, and the tools are ordinary.

A worn, cracked fan is taken off a graphics card and marked with a red cross. A new fan with a green tick is lowered into the empty slot, next to a tube of thermal paste and two thermal pads, while a robot holds a wrench.
Fans, thermal paste and pads are cheap to replace and often bring a card back.

Beyond that, some repair shops work at the component level. Get a quote before scrapping a card, especially one that was expensive when new.

A card that cannot be repaired is still worth something as parts. Its cooler, bracket and fans all fit other cards of the same family.

Ready to put this into practice?
The How to Mine guide picks your closest server and builds your exact miner command.
Start Mining Qubitcoin