chip temps

[elBogg]

i know this is polly an old issue, but i couldn't find any posts that said what i needed. what kinda temp should be the max that i run my chip at? i have a 1.4 athlon on a shuttle ak31 mobo, and i wanna start overclockin it . . . MBM5 says that the chip at max load (F@H for a few hours straight) is around 50-51C. the crappy thing is that i've got like 8 case fans total and i still can't get it much cooler . . . anyway, if anybody's got a guideline for the temp for me to go on that'd be cool . . . also, when you're o/c'ing, is there any way to tell the difference between your puter crashing cuz it's trying to go too fast, the chip voltage ain't set high enough, the memory speed is maxed out or the mem voltage needs to be upped? thanks, guys . . .

[.. AE 86 ..]

try reapplying ASII to lower the temp

u will know if a computer is unstable

programs crash and all
sigh

[elBogg]

hehe, yah, i know about arctic silver, it's on there . . . i've got a delta heatsink & fan, too . . . i really just wanted to know what kind of temps are good and what is like the borderline for being "hot" . . .

[RED Hot Machine]

You want temps around 40/45c load for 1.4 b4 overclocking it. When overclocked you really don't wan't it to go above 50c.

What case cooling have you got?

[elBogg]

sorry, ain't got a sig yet, but i have an antec sx830 case w/an enermax 431w ps, i have a vantec cck-6035d heatsink/fan combo(delta), and the stepping on my chip is eyhja. i have three 80 mm fans blowing in the front, and two blowing out the back. i have 1 120mm fan blowing in through the case door (i cut a hole) fairly much on top of the cpu, and 1 92mm fan blowing out the top, and also a ~42cfm pci slot fan ducted to the video card blowing out. i really thought i'd be getting better temps than this and i don't know why. i have the chip o/c'd to 1468 (133*11), and i've got arctic silver on it. dvd, cd-r, floppy, radeon le, soundblaster, 10/100 ethernet card, and that's about it. oh, yah, and it's a shuttle ak31 rev2. i realy wanna o/c it more, but i don't wanna risk it if the temp's already 51-ish under load.

-edit- yah, like i said, that temp is measured with mbm5. the case temp's like 30-35, depending on the outside temp. thanks a lot, guys . . .

[hunter00]

my thoughts -->

i think i see a problem:

" i have three 80 mm fans blowing in the front, and two blowing out the back. i have 1 120mm fan blowing in through the case door (i cut a hole) fairly much on top of the cpu, and 1 92mm fan blowing out the top,"

so:

3 80mm intake
1 120mm intake

2 80mm exhaust
1 92mm exhaust

i'm no expert, but as far as i understand it, and from my own experiences playing around with different arrangements, you want to have at least as much exhaust as you do intake, which kinda makes sense if you think about it, its more important that you get all the hot air out of your case than to keep blowing more air in.

i would probably hack in another exhaust or two in the back, and then either upgrade your 92mm blowhole to a 120mm (make your hole bigger) or rheostat your side 120mm to make it slower.

my setup is actually a lot like yours,

3 80mm intake
1 120mm intake

3 80mm exhaust
1 120mm exhaust.

thermoengine heatsink

i'm running 1.53 at 35 degrees right now.

[elBogg]

hunter-
i know what you mean about the fans, but when i installed all of the fans, i went by the cfm ratings, which right now are almost exactly equal at ~170 cfm (in and out). even so, i'd like to try what you mentioned about "rheostat-ing" the 120mm if anyone can tell me how to do that . . .

[hunter00]

rheostats are these really cool things that can control the fan speeds, i'm not really an expert on them or anything, but a quick google search would probably turn up some good sites.

yeah i didn't think of that, as long as you're equal on both sides you should be okay . . .


hmmm . . . i really can't think of the problem, sorry. if i think of anything though ill be sure to let you know.

[elBogg]

coolness, i'll keep working on the problem . . . anyway, thanks for all the help, people . . .