Hot ageing laptop

[CtrlAltELITE]

Hey folks

 

Looking for some opinions on this. Have had my current laptop just over 3 years. At a good age where I could reasonably replace it without feeling like I'm throwing money away. Thing is though it serves me fine for my current purposes and the games I play.

 

However it has been running really hot, a few months after I bought it I replaced the stock thermal compound with Noctua NH1. This made a noticable difference to the level that it thermal throttles, even brought gaming CPU temperatures down into the 80s. However in recent months I have noticed temps going up again, and occasional BSODs have actually led to me reducing the degree of my CPU undervolt. (was at -63mv stable, now down to -55mv). Under gaming load CPU temps are sitting well in the 90s, GPU in the 80s. Laptop itself has 10500 power on hours. Laptop gets dusted every 6 months or so, although to be honest I have never seen a big problem in it with dust buildup.

 

So, I have a couple of questions

1. I know I am reaching the EoL of the Noctua thermal paste (3 years), I am assuming this is causing the increase in temps, is it reasonable to see decrease in cooling at this age?

2. Would high temperatures effect the components ability to be undervolt, or could this be a sign of fatigue from use?

3. I am tempted to experiment with a liquid metal paste, if I kill the laptop at the age it is at it would be less painful than a brand new PCB, but are you likely to see a increase in cooling efficiency in a slim laptop cooler? (a difference which looks like between 12.5W/mk and 73W/mk, between cermic and liquid metal compounds, looks dramatic on paper)

4. As per the last question, does anyone know if laptop coolers tend to be aluminium or copper? Don't want to take the lot a part to find I can't use the aforementioned coolant.

[Fasauceome]

Judging by the steps you took to cool it, I'll assume you dusted the fans off and cleaned the cooler. You should be able to undervolt regardless of temps, as undervolting would improve temps/

[CtrlAltELITE]

Yes, dust it every 6 months or so.

 

Sorry if I phrased what I said badly, my query was; when I first got the laptop I ran a -64mV undervolt stably. Now it will only take -55mV before instability sets in. Can high temps coupled with undervolt reduce stability? Or could this be a symptom of component fatigue.

[knightslugger]

6 minutes ago, CtrlAltELITE said:

Yes, dust it every 6 months or so.

 

Sorry if I phrased what I said badly, my query was; when I first got the laptop I ran a -64mV undervolt stably. Now it will only take -55mV before instability sets in. Can high temps coupled with undervolt reduce stability? Or could this be a symptom of component fatigue.

I suspect that the reason a -64mV undervolt was stable when you first got it and now no longer is, is due to the ever changing nature of use, addition of programs, new versions, and such. It is unlikely that what you use on that computer, and how it's used has remained the same. Therefore it is LIKELY that the updates to the computer has made it no longer able to sustain a -64mV undervolt and be stable.

 

CPUs don't degrade. The TIM might, the cooling solution might... but the silicon doesn't. It either works, or it doesn't.

[CtrlAltELITE]

Well, on an unrelated note. Might be new laptop time. Thought I'd just pop the back off to see what sized thermal pads I'd need, and to check if my heatsink was copper.

 

Aaand 2 of the heatsink screws are completely rounded off. Not quite sure how I go about this, all the screws are attached to the heatsink by a small screw, I am most definitely going to break something if I try to drill these out.

 

EDIT: Managed to get them out, used my smallest slot drill to take a bit of material off the head of the screw, then managed to ease it off with a flathead. Little bit of metal swarf never hurt anyone.