Jump to content


Photo
- - - - -

New Water Cooling Setup, Weird Core Temps


  • Please log in to reply
3 replies to this topic

#1 GaiusMaxwell

GaiusMaxwell

    Member

  • Members
  • PipPip
  • 244 posts
  • Gender:Male
  • Location:Newcastle, NSW, Australia

Posted 03 October 2008 - 02:12 AM

Hello all,

now i know this is blasphemy but i bought a thermaltake bigwater 735, now here is my reasoning behind it lol. it was on sale for $110 australian, a TRUE over here costs over $200 and even my previous maxorb cost $100, i heard that thermaltake WC kits work about as good as the best air coolers so i thought to give myself some experience with WC id pick it up

so i set it up outside the case, did leak tests for 4 hours and everything was good so i set it up and everything is alot lower than my previous maxorb, but now, after having it on for 3 days, my 1st and 3rd core (core 0 and core 2) on my Q6600 are at a constant 5c higher than the other 2, i tried repositioning the water block but they still remain 5c higher

My current loop is reservoir > pump > rad > water block > reservoir, i changed it to this today after previously having it as reservoir > pump > waterblock > rad > reservoir but nocited alot of condensation in the reservoir so i changed it and now its fine

so basically what im asking is, is the reason for the higher core temps on 2 of the cores a bad thermal compound application, positioning or just the way the chips were made?

p.s now that actually have experience putting a kit together i have already started ordering better parts, starting with a new water block - Dtek Fuzion and a black ice eXtreme II, is the pump powerful enough to add another waterblock (on NB)? its 400L/hr

#2 Jenova69

Jenova69

    Member

  • Members
  • PipPip
  • 410 posts
  • Gender:Male

Posted 04 October 2008 - 03:51 PM

all quads have that problem

4790k

Asus Maximus VII Hero
2x8gb G.skill Trident X

Pny gtx 980
XFX xtr 550w
Fractal Define S
Crucial MX100 128gb, 2xCrucial MX100 512gb

Acer XB270HU


#3 rud3bwoy

rud3bwoy

    Geek

  • Members
  • PipPipPip
  • 559 posts

Posted 05 October 2008 - 08:10 AM

same prob here on my quad 45 nm don't worry about it

#4 TheSternMystic

TheSternMystic

    Dell Apostle

  • Folding Member
  • 1053 posts
  • Location:Canadian

Posted 05 October 2008 - 11:25 AM

There's a couple things that could cause that. My first suspect would be the calibration of the thermal diode in the die itself. I seriously doubt that any chip manufacturer will spend a huge amount of time accurately calibrating the internal diodes to report the exact temp. +/- 5*C sounds like a decent margin of error to me. The lack of calibration doesn't mean the scale is off.

Second suspect would be load balancing. If you're running windows, background processes will only be able to use one core at a time. lsass.exe, svchost, etc... And I remember from back in the day of my Dual Pentium 2 rig, all BIOS events could only be run on processor 0. So one CPU would always run hotter.

Anyways, don't worry about it. Some dual core CPU's even experience the same thing, as well as AMD dual and quad core processor's.