Cookies op Tweakers

Tweakers is onderdeel van DPG Media en maakt gebruik van cookies, JavaScript en vergelijkbare technologie om je onder andere een optimale gebruikerservaring te bieden. Ook kan Tweakers hierdoor het gedrag van bezoekers vastleggen en analyseren. Door gebruik te maken van deze website, of door op 'Cookies accepteren' te klikken, geef je toestemming voor het gebruik van cookies. Wil je meer informatie over cookies en hoe ze worden gebruikt? Bekijk dan ons cookiebeleid.

Meer informatie

  • Helvetios
  • Registratie: augustus 2019
  • Laatst online: 22-07 14:03
Hi everyone,
First of all, sorry for typing in English but Dutch is not my first language and it is very difficult for me to get the terminology below right. Thank you for the understanding!

Last night I took my liquid cooling for regular 6th month cleaning cycle. To my surprise I found galvanic corrosion in the EK FC Vega block that is nickel plated copper. The loop is made entirely of EK components, therefore I cannot understand why there is corrosion. There are no aluminium parts as far as I can see:
- Fittings - Brass nickel plated
- CPU block - Acetal and copper nickel plated
- GPU block - Acrylic and copper nickel plated
- Radiators - Brass chambers and copper flow channels/fins
- Tubing - EPDM rubber

As coolant I always used the same EK Cryofuel Premix without adding more additives. Interestingly, the place which is discolored does not have running coolant by design. It's a sandwiched area between the block and the acrylic cover. I checked the rest of the the components and I do not see any discoloration in the micro fin stacks of the blocks.

The block was purchased new 1.5 years ago and still has the EK Quality Control sticker intact. Do you think that such damage should be covered by warranty? I do not see if I am doing something wrong since all components are EK and this is probably 5-6 time I am building a loop so maintenance and contamination control were done accordingly.









On a separate note their stacked tolerances calculation is all over the place. Having designed and machined several water blocks in the past, they should not have a failure mode where gap pads are not compressed. I think the problem boils down the the standardized standoffs that they use in every single EK block. That way the milling waste is less compared to making everything from a single block of metal. There is tolerance of the SMD component, tolerance of the reflow process, tolerance of the standoff, tolerance of the standoff contact surface, tolerance of the thermal contact surface, tolerance of the gap pad. These are all very tiny on their own but on a 1mm thin gap pad that compression ratio variation can be huge! If the complete block was one piece including standoffs it will be more expensive to manufacture, the standoff tolerance and the plinth reference are eliminated completely. Anyway mechanical engineering OCD kicks in...



Your input will be greatly appreciated,
Ivan!


Apple iPad Pro (2021) 11" Wi-Fi, 8GB ram Microsoft Xbox Series X LG CX Google Pixel 5a 5G Sony XH90 / XH92 Samsung Galaxy S21 5G Sony PlayStation 5 Nintendo Switch Lite

Tweakers vormt samen met Hardware Info, AutoTrack, Gaspedaal.nl, Nationale Vacaturebank, Intermediair en Independer DPG Online Services B.V.
Alle rechten voorbehouden © 1998 - 2021 Hosting door True