Overheating e1705

Alright so I took apart my 1705 to paint it. After I put it back together, the video card (7900gs slightly overclocked) just keeps warming up until it gets to 88c and i got nervous and shut it off. I did this all 3 times i tried using the computer. So I took it back apart and checked to make sure its all there and clean and i Even swapped out the stock thermal pad for some arctic silver (did the cpu too). So now its still over heating and the cpu is running a little warm too. Im reloading the stock bios onto the vid card to see if that helps but it only brought the temp up 3c when i overclocked it so i doubt it will make much of a difference.

Any Ideas? Im leaving for college on Tuesday so i really need this taken care of.

 

Was it overheating before the painting? If no, then I would check all ur vents and whatnot because something you did is blocking the air from going in and out. Maybe you painted over the vents that should've remained clear.

If it was overheating before, then try cleaning out ur fans and vents of dust, and look into undervolting.

Also, I see you're OCed. Maybe you should not be doing that if the computer is overheating. I mean it's like slashing ur tires, and then complaining that ur car doesn't drive well and have no idea what's the problem...

 

The computer was not overheating before. The vents are clear and i blew out the fans. I just rolled back to normal clock speeds and it is still over heating.

 

What do you define as overheating? You say that it gets to 88* and then you shut of off. If the computer doesn't crash or artifact due to extreme heat, then it's not overheating. What you're doing is simply turing ur computer off when you think it's too hot (not when the computer thinks it's hot). That's not really overheating. The definition of overheating is when a component gets hotter than what the manufacturer feels is a safe operating temperature for that part. Not when the user thinks the part is too hot, but the maufacturer. And if it hasn't shut down or given you any warning, then it's only u that thinks the gpu is too hot, not the manufacturer. And it really doesn't matter what the user thinks, it's where the manufacturer draws the line. Whether or not the individual user agrees with this line doesn't matter.

For example, if I think that anything over 40*C for a gpu is bad and the manufacturer set the thermal limit to 98*C, does that make an 88* gpu overheated? No it doesn't. It makes it too hot for my liking yes, but not too hot for nvidia, or dell.

What were your temps before when it wasn't "overheating".

I think 88* for a pretty powerful card isn't that much. And if it's not crashing and artifacting, then I don't really think theres much to worry about.

 

before i would rarely hit 85* while gaming. now im am hitting 88 while the computer is just sitting there

 

Right, but as long as the system is stable in gaming and everyday computing, it's not really overheating nor should you worry about it.

If you say that the fans are clean and nothing is blocking the vents, then I would tear down ur computer again and check a couple things. First, make sure the AS5 is applied correctly and evenly. Too much of that stuff will actually hurt temps. Read the instructions on how much and where to put it. Also make sure that the gpu heatsink is seated correctly and all the screws are tight making good contact with the gpu and the AS5.

Both those things could have been done poorly when u put it back together and caused your temps to increase.

 

88 degrees isn't necessarily too hot. I have a friend who got an nVIDIA alert when his card hit the warning temperature - 139 degrees Celsius. Now that was a little too hot. It kept working for a few days, but after awhile of working at 140 degrees+ it blew up. You're still a good 50 degrees celsius below that point. You'll probably get an nVIDIA warning if you reach dangerously high temperatures.

Also make sure all the fans are connected and working. A non-working fan could be the culprit.

 

Fans are working, that was the first thing that i checked. The thing is that im worried when its goes from ambient temp to 80 in 5 minutes of just idling. And it was doing it in about 4 minutes before i did the as5 so i guess that helping a little bit. I guess i will just tear it down and clean everything again.

 

Are you sure the heatsink is still making contact? You said you replaced the thermal pad with AS5. Most of the time there's about a 1mm gap once your removed the thermal pad. I'd suggest you check it. Most of the time you need to replace the thermal pad because it could get some dirt on it just is unable to fill the gap after you break the bond.

My advice is to make sure the heatsink is making contact. If it isn't then order some Shin Etsu pads and replace it.

 

My 1705 will hit 76-80C + 8-degree offset using I8kfan, logging only, when idle - much faster when gaming

Once GPU fan kicks on, the temps should drop to 60C or so, stop, then creep back up again..and so on. Are your temps not dropping?

Note that this is my second GPU card, my first 7900gs did the same, my first 1705 with a 7800GO did it faster

 

with all fans on full the gpu rises from ambient to 90* in 7 minutes while the computer is doing nothing. I am getting a cooling pad and some new thermal pad's to see if they make a difference

 

did you get paint on/in the heatsink?

 

no, the internals were kept in a seperate room. im am not unexperienced in doing things like this and yet i cannot think of anything else that could be wrong with it.

 

You may be getting a bad thermal pad to GPU seating or heat pipe not right as in fluid/gas flow. Let us know how the artic silver works out..as Tebore said.