Help with Diagnosis of Danger Symptoms: CPU or GPU—What to Do

[gremlinkurst]

I have a Win 10 x64 machine with an i5 6600K CPU, a Zotac GeForce GTX-970 / 4 GB DDR5 GPU, and a 650 W PSU, and lately I've been experiencing some alarming symptoms…and am having a bit of trouble isolating the cause of the issue. I can hear a fan speeding up significantly faster than its norm, and there is audible distress as well, sounding like fan bearing shaft lubrication failure, a very alarming whining and screeching, commingled with sounds that seem very much like the fan blades hitting things they're not supposed to be hitting.

My system was custom-built to my specifications, and while I have home-built several OEMs, I have "messed" with NOTHING inside the case during the warranty period—or after, for that matter—and the system has been functioning at peak proficiency in a non-smoking environment for three years. There is absolutely NO chance that cables, ETC, are interfering with cooler fan movement.

I've installed a lightweight CPU monitoring application that reports temperatures well below norm, but my Zotac Firestorm application is of VERY little help—I can't find a way to display temps in Fahrenheit; I have to use a conversion application to understand the Celsius temperature reports—which end up being below anything alarming.

Over time, I have isolated the occurrence of the alarming symptoms' appearance only while running GPU-intensive games, not while streaming HD video.

The CPU is "water-cooled," but the GPU is stock. I'm running 16 GM SDRAM (Skilljaws).

I'm trying to figure out where my liability is: Do I have to get a new CPU cooler, or is it the more expensive option: A new GPU?

 

[95Viper]

My system was custom-built to my specifications, and while I have home-built several OEMs, I have "messed" with NOTHING inside the case during the warranty period—or after, for that matter—and the system has been functioning at peak proficiency in a non-smoking environment for three years. There is absolutely NO chance that cables, ETC, are interfering with cooler fan movement.

Might need a good dusting, if you ain't been in it in three years or so.

I've been experiencing some alarming symptoms…and am having a bit of trouble isolating the cause of the issue. I can hear a fan speeding up significantly faster than its norm, and there is audible distress as well, sounding like fan bearing shaft lubrication failure, a very alarming whining and screeching, commingled with sounds that seem very much like the fan blades hitting things they're not supposed to be hitting.

Might be a something has sagged and touch a fan. Also, fan could have dust or other problems. Open the case up, get a paper towel roll core, and, be careful, listen around when it happens to figure out where the noise is coming from.

You might wanna check your temps, too, to see if any are abnormal.

If it is out of warranty... you, either, have to do it yourself or get someone else to do it.

 

[eidairaman1]

Might need a good dusting, if you ain't been in it in three years or so.




Might be a something has sagged and touch a fan. Also, fan could have dust or other problems. Open the case up, get a paper towel roll core, and, be careful, listen around when it happens to figure out where the noise is coming from.

You might wanna check your temps, too, to see if any are abnormal.

If it is out of warranty... you, either, have to do it yourself or get someone else to do it.

Great minds think alike. Now here's one thing for the OP, if it is in fact his graphics card fan he can either find them online through fleabay or contact the card manufacturer. If the card manufacturer will not help either he can do a ghetto mod and mount a different fan all together or he can find a aftermarket heatsink from like Arctic cooling or Thermalright, Thermalking, Gelid, etc.

 

[erocker]

Might need a good dusting

This. The entire machine needs some compressed air.

 

[Mussels]

you dont do any conversions from C to F, you deal with the temps in C - because thats how intel/nvidia/AMD etc list their max safe temperatures.

What are these temperatures you ran into? Was it on the GPU or CPU? Can you get accurate maximum readings from something like realtemp or GPU-Z, while stress test programs are running?
Odds are a fan is failing, or its simply clogged with dust. Without more info (or you investigating) we cant figure much more out.