ASUS EAX800XT: 127C / 260F !!

Taz · TweakNOOB Joined: 09 Nov 2004 Posts: 14 Location: Israel

I just bought a new computer (P4-3GHz LGA775, ASUS P5GDC, 2x512 Kingston DDR400, ASUS EAX800XT-TD).

It's all in a HEC CI-6A192F case with a 120mm intake fan (on HDs) & 120mm outtake fan, plus the CPU has

an air guide from the case door so that's taking air in too. When the air blowing out seems cool

(actually more like room temp).

The CPU fan (Intel Boxed) is running with no PWM - at it's max of 4500 RPM (/w air guide)
Both 120mm fans are running with no PWM as well - at their max of 1500 RPM
The GPU fan also has it's PWM switched off - running at max of 5000 RPM (/w ASUS's air duct)

Case sensor doesn't seem to work (stuck at 35C / 95F, but it's located at the bottom of the case).

Idle temps

CPU: 43C / 109F
GPU: 80C / 176F
V-RAM: 43C / 109F

CPU Load (two CPU Burner processes, each set to it's own virtual proc so both at 100%)

CPU: 52C / 125F
GPU: 80C / 125F
V-RAM: 43C / 109F

So far it seems fine, right?
So now with the X800XT...

GPU Load (after 20 secs of running ATI Radeon9800 Chimp Demo - lol it'll do)

CPU: 44C / 111F
GPU: 127C / 260F
V-RAM: 47C / 116F

Now to me it doesn't seem like the GPU could run at 127C, let alone without a hitch (i'm not gonna stress it for more than 30 secs... to afraid).

Could it be that the sensors are bad? The V-RAM sensor seems fine. I updated to the latest ASUS Special driver (/w Cat 4.10) and latest SmartDoctor.

The problem with verifing the temp by touching is the way this fan/heatsink combo is built. There's a large heatsink with a duct covered by a platic cover. The fan is about 6 cm away from the core, blowing air into the duct which flows over the heatsink (and the core area). There is no accessible part of the heatsink that is even close to the core to touch. The only way is to remove the plastic cover which will render the entire cooling assembly useless (and will prolly burn the GPU).

What do you think? is it safe to use? I don't feel like burning the GPU just so I can demand a replacment, i'd rather hand in a working one and say it overheats.

What should I do next? TIA!

Google · Sponsor

Taz · TweakNOOB Joined: 09 Nov 2004 Posts: 14 Location: Israel

Why can't I edit my post?

Anyway, after letting the fans run at max for a while, the new idle temps are:

CPU: 41C / 105F
GPU: 78C / 172F
V-RAM: 43C / 109F

[KoG]^weaZel · Posted: Tue Nov 09, 2004 4:04 pm Post subject:

can you feel the air coming out of the shroud? is it warm? If so then how warm? Because at that high of a temp, there would be alot of heat being pumped out. But if there is very little then there are two things that can be happening. 1 is that the reading is wrong. 2 the heatsink isnt properly mounted on the core.
I have had the second option happen on the two Asus 5900 U's that I have had. So I dont put it past Asus to strike again.

I would try getting that card exchanged before something burns up.
_________________
I tweaked and it tweaked back! So I Tweaked some more!

"Barney is like the Michael Jackson of PBS." - James Tybeerious

Taz · TweakNOOB Joined: 09 Nov 2004 Posts: 14 Location: Israel

No, the fan is only blowing out slightly warm air, and the heat sink is hardly warming up either. At that point I was thinking "wrong readings" but i proceeded to touch the exposed PCb under (or above actually) the GPU and it was very hot (I couldn't keep my hand on it). It seems that the heatsink isn't pressed against the core or some other thermal problem.

Opening my case seemed to help cool it. It's now idle at 76C and peaked just under 100C. Still not normal afaik.

What would you suggest I do? Demand a replacement claiming a defect or use it till it blows then I won't have to argue my case?

Josh · Posted: Tue Nov 09, 2004 7:18 pm Post subject:

Have you overclocked or physically changed your GPU? (sorry if you stated it, its late and didnt feel like finding it

)

If not, then i would check out a replacement.
_________________
»4 RS«»1Ж «»1 DENE«»1 FREAKIN LAME OH«»1 MONEY MONEY«
(-)

meobius · Posted: Tue Nov 09, 2004 7:25 pm Post subject:

it could also be a resitor that fell off... had it happen to one of my 9800pro's and my old geforce mx 440
but at that point you would feel it on the heatsink and the exaust....
so i would probably say that it is bad heatsink mounting.
_________________
If time is of the essence,
than why is it that we waste,
that which holds such great importance,
spent without a trace.

JayDubya · Posted: Tue Nov 09, 2004 7:34 pm Post subject:

Hmm, you say the gpu is too hot to touch? What about the gpu heatsink? is it too hot to touch?
_________________
JayDubya aka JW Jay JD ^> ﺵ

Xal · Posted: Wed Nov 10, 2004 3:41 am Post subject:

He said he cant get to it, it would seem that its a bad mount on the hsf.
RMA.......
_________________
Phenom II x4 955 @ Stock
Asus M3N78-EM
4gb Corsair XMS2 DDR2 667 @ 800
1gb Powercolor Radeon HD 5850 @ Stock
X-fi Extreme Audio PCI E
Nexus 600W Silent PSU
Nexus Fans
Custom case

Taz · TweakNOOB Joined: 09 Nov 2004 Posts: 14 Location: Israel

I brought it today to the computer shop where I bought it to demonstrate.

Same results (78C Idle, 127C Load).

The guy took out the video card and had a look. He played with heatsink assembly, and look down it the contact with the core. He could easly lift the heatsink off the core without disassembling anything. He said it was normal (is it?), but he found that one of the plastic pins that hold the assembly was loose. He pushed it till it was tight and put the card back in.

Now, on idle (with the case open), it was 74C. But it still peaked 127C on load (Chimp Demo).

He said that if I really insist, he can send the card back for repair (it's Retail so he can't just give me a replacment). That'll take a while.

He offered an alternative. It's got a 3 year warranty. He said that I should keep using it normally, and if it burns out, he'll be happy to give me a new one. He also said he'll find out what the nominal and maximum operating temps are and get back to me.

I chose the latter. The whole computer including all it's parts (except the case & PSU) have a 3 year warranty (case & PSU = 1 year), so if it blows up and burns everything in it, it would all be replaced under warrany. I just didn't feel like waiting 3 months for the video card to ship to Taiwan, get repaired then shiped back (been there, done that).

If after 3 years it will still work well, then I won't be worried about the warranty expiring, even at 127C.

Taz · TweakNOOB Joined: 09 Nov 2004 Posts: 14 Location: Israel

Oh and I didn't overclock the GPU (stock is at 500/740). I actually tried underclocking it to 450/740 and it still overheated (maybe 2C less).

Anyway thanks for you help! I'm gonna keep my fingures crossed....

[KoG]^weaZel · Posted: Wed Nov 10, 2004 12:14 pm Post subject:

I am sorry that guy is an idiot!! There should be no way of lifting the heatsink off of the gpu without some screws being removed. And as he should find that is not normal temps for that card. I have heard that they run hot but not that hot. And from what you have told us "barely warm air coming out but the back side PCB HOT" means that the heatsink isnt making good contact with the GPU. So yes it will burn out, but how long will it take I dont know.
_________________
I tweaked and it tweaked back! So I Tweaked some more!

"Barney is like the Michael Jackson of PBS." - James Tybeerious

agchenry · TweakNOOB Joined: 06 Nov 2004 Posts: 81 Location: FL

Your not kidding weazel 127C that is 27 degrees C over boiling that Heat sink should be Blazing if it is in proper contact w/GPU . But then again i guess if it had a good contact with the GPU it wouldnt be that hot

ToggleHead · TWEAKGURU Joined: 03 Mar 2004 Posts: 4360 Location: Jersey

agreed.....this is incorrect....get a new card......RMA....or return for reapirs either way.......that card WILL die sooner rather than later.....
_________________

JayDubya · Posted: Wed Nov 10, 2004 1:11 pm Post subject:

agchenry · TweakNOOB Joined: 06 Nov 2004 Posts: 81 Location: FL

I am going crazy because i am getting 84-86C after doom3 for about 2 hours 80C @ idle is plain Crazy! Tell that wish i had a brain sales man no need to find out if 27 degrees C over water boiling temp is normal cuz it it aint! lol

_________________
My Rig: P D 955 XE @ 4.35 (Stable), Abit AW8-Max mobo,1GB Patriot DDR2 1000 (2x512), EVGA 7800 GTX 512 74GB Raptor , 250 GB 16MB WD, 300 GB 16 MB WD

Hoedizzy · Posted: Wed Nov 10, 2004 1:52 pm Post subject:

You need to rid of that card asap!!

FreeTheRock [banned] · Posted: Wed Nov 10, 2004 6:57 pm Post subject:

why would he need too? the 6800U runs even hotter than that

agchenry · TweakNOOB Joined: 06 Nov 2004 Posts: 81 Location: FL

Have yet to see a 6800 run @ 127 C @ load has any one? Or even an 80C idle (anyone?)
_________________
My Rig: P D 955 XE @ 4.35 (Stable), Abit AW8-Max mobo,1GB Patriot DDR2 1000 (2x512), EVGA 7800 GTX 512 74GB Raptor , 250 GB 16MB WD, 300 GB 16 MB WD

Josh · Posted: Wed Nov 10, 2004 7:13 pm Post subject:

^^^Nope, anything over 100C sounds a bit dangerous to me
_________________
»4 RS«»1Ж «»1 DENE«»1 FREAKIN LAME OH«»1 MONEY MONEY«
(-)

2old2care · Posted: Wed Nov 10, 2004 9:22 pm Post subject:

Don't know how they are measuring these #'s...but those kinds of temps are serious on the laminations (over a bit of time) of the multi-layer PCB. Will soon dis-color the back of the board where the BGA chip has been attached.
If the back of the board (opposite side from the chip) is to hot to touch...and this monkey won't honor a retail card...I would be talking to ATI directly. (or whichever board manufacturer)
_________________
.
Liquid-Cooled Q9450 and an EeePC
.

Silicon Skum · Posted: Wed Nov 10, 2004 9:52 pm Post subject:

I just checked the max temp that GPU is designe to run at, 127°C is MAXIMUM temperature for the silicon running at REDUCED clock rate.

That GPU is gona fail, and it's gona fail SOON!

My guess is that there is no thermal pad / paste between the GPU and the heatsink, also that heatsink is NOT in contact with the GPU. Perhaps a warped heatink? Either that or that card has one BIG voltage problem.

That tech guy just blew you off, that temperature range is NOT acceptable for the card. I would RMA that card soon, no use trying to fix the problem, the damage has allready been done. Electromigration has probably allready began.

§
_________________
my sig disappeared from the image host (?)
But at least I have a Josh Award!

JayDubya · Posted: Wed Nov 10, 2004 11:30 pm Post subject:

Hoedizzy · Posted: Thu Nov 11, 2004 12:08 am Post subject:

agchenry · TweakNOOB Joined: 06 Nov 2004 Posts: 81 Location: FL

WTG Jay! How dare him talk bad about my VGA lol My 6800 gets to 84C after almost 2 hrs of playing Doom 3 and my card is OC'd to 422/1.14 lol

_________________
My Rig: P D 955 XE @ 4.35 (Stable), Abit AW8-Max mobo,1GB Patriot DDR2 1000 (2x512), EVGA 7800 GTX 512 74GB Raptor , 250 GB 16MB WD, 300 GB 16 MB WD

ToggleHead · TWEAKGURU Joined: 03 Mar 2004 Posts: 4360 Location: Jersey