What should I do if my water-cooled graphics card is crashing or freezing up?
The following is a list of typical symptoms and causes of malfunctioning graphics cards.
Overheating of the GPU and/or memory IC due to inadequate or insufficient thermal interface between the water block (cooler) and IC: We highly recommend users to follow the installation manual, unique to each EK water block. Remove the card and the water block from the computer. Please check for thermal imprints to see whether all surfaces, covered by either thermal grease (TIM) or thermal pad, are making contact with the cooler. In case of bad contact, please check the height of the standoffs (FC water blocks) and whether they are installed correctly. In the case of generally good contacts, please consider options 2 and 3.
You can see on the picture that there is no contact on the thermal pads and standoffs from the water block are stuck on the PCB.
Water block (cooler) short-circuiting electrical components: Remove the graphics card/water block assembly from the computer and visually inspect the graphics card for any sign of suspicious warp, mechanical damage, or contact between unwanted parts of the graphics card and the cooler. Make sure to check whether the TIM you are using (in case you are using 3rd party TIM) is not an electrically conductive one - in that case, remove the remains of thermal grease from all surfaces using denatured alcohol.
- Damaged GPU and/or memory IC due to physical damage: Remove the graphics card/water block assembly from the computer and remove the water block. Check the edges of the GPU die and memory IC for any sign of mechanical damage. GPU die edge can easily break when using excessive force.
We recommend checking for stability using Furmark or a similar GPU torture test utility.
Upon exhausting all of the possibilities mention above, feel free to contact our Technical Support.
Article is closed for comments.