Possible overheating problems

I am running a Alienware M17x10 with an I7 processor @ 1.6GHz and 8 GB RAM. In the past week I have had two occasions with my laptop instantly shutting off randomly. I feel my back left side and its hot the the touch. After some trial and error before the PC turns off my CPU is running at about 72 C. Looking over other posts and threads I see this is hot but normal. I have cleaned out my laptops middle cover multiple times with no avail. Would it be wise to remove the 4 screw ( 2 on each side) to remove the outside left and right panel for excessive cleaning. Or could this possible be an issue why my intake fan giving out. I really just want to hear some thoughts before I put words into action. My warranty has expired, if I can fix the problem without sending it back to dell that would be great.

 

The CPU temperature you are seeing is normal and is not likely to cause a shutdown. Check the GPU temperature instead.

 

Well I downloaded a GPU temp gadget from addgadget.com

This is the feedback I received from the gadget. So either its not able to give me the fan information or its not even on. If you have a recommended gadget that gives me GPU temps from another site I'm glad to try them out.

 

Try GPU-Z. That should work.

 

The attachment below is what happened while playing LoL (League of Legends). This game was easily handled by my computer when I first purchased it but as you can see I'm hitting over 103 C. At the very end I just turned on my own fan which was blowing on the left side of my laptop to cool it down. Also you can see my Fans RPM are all over the place and way low while the GPU is heating up.

 

Sounds like dried up thermal paste. Time to open 'er up, clean GPU and heatsink with alcohol, and apply some Arctic Silver. That, or you have a bad fan. Does the fan make any suspicious noise, like grinding?

 

How does a fans RPM have to do with thermal paste? I'm not arguing your statement I'm just not that tech savy. My tech skills only reach passed the basics. Also do you know of any tips or videos explaining the process of applying new arctic silver specifically to a m17x?

 

Short version: The thermal sensor may be on the fan or heatsink and not on the GPU. Sure, there is an internal sensor in the GPU, and it reports high temperatures. The fan should run at high speed. However, some manufacturers decide to go on a different route.

Back in the days of P4 i had this stock cooler for my 3GHz Prescott that sounded like a jet engine when the CPU would heat up. The thing went to over 5000rpm. I could not understand how that would happen, since when connected directly to a 12v supply it would run at a "normal" 2500ish RPM. I was also unable to control this fan from the motherboard, which did support fan speed control. Later i found that the fan had a temperature sensor and step-up converter built in. As soon as the temperature would exceed a certain value, jet mode enabled. So if the thermal transfer between GPU and heatsink is poor, then the fan may not run at maximum speed. While a failed fan is a more plausible scenario given your fluctuating RPM readings, you'll still have to take it apart to find out.

Applying Arctic Silver is simple. They even got instructions for every CPU model (yes, even for laptop CPUs). For the GPU, always use "middle dot", it spreads by itself. Spreading it around manually will inevitably end up introducing dust into the paste. If you want to double check, remove the heatsink again after you have reinstalled it, the paste should have spread evenly and left no corners bare.

 

Thanks you very much for all the information and tips. Hopefully I can get my M17x back into working order by the end of the week.

 

Btw, if the fan is responsible and feels like there is significant resistance when you try to rotate it by hand, then you have a large chance of getting it back in working order by applying grease to the hub and bearings. If you can get to them that is. If not, you'll have to get creative (like use a toothpick or something to spread the stuff inside the fan).

Only if the fan makes a "clickety clack" sound does it mean that the bearings have failed, and then you really need a new fan.

 

Well I opened my computer up to find my graphics card and graphics card fan. I found out my GPU headsink was completely clogged with dust. Its sitting at about 70-75 C instead of reaching over 103 C

 

Your temps are too high for running LoL. My nephews play this game on my R2 and it's nowhere near that. It's not really a demanding game. 8800GTS and 4870 handles it well because that's what I gave my nephews with my two old desktops.

Try the suggestions above and let us know.

Edit: that's a good screwdriver you are using btw. I use a similar one

 

Good job. Same thing happened to a friend's 4870 i cleaned a couple days ago - it was the type with the dual slot cooler with squirrel cage fan (desktop card, laptop style cooler). It was running at 100ish deg C IDLE and would pretty much crash in any game after 5-10 minutes.

Dust had built up in the heatsink right next to the fan. So the poor fan did push cold air in, but the air never had a chance to pass over the heatsink fins. After cleaning and applying new paste, it starts at 60C, and stabilizes at 71. After 15 minutes of Crysis 2 - still 71C and the fan didn't go past 40% (ie silent). The fan on that card is programmed to run it in the low 70s - if the card temperature is 70C or lower the fan will slow down. At the desktop it's barely turning at all now.