i7-4790k reaching 100 degrees after GTX 1080 install

[numberz]

Computer Specs:

i7-4790k @ 4.00 GHz (Turbo Enabled)

CPU Cooled by a H80iGT Corsair Water Cooler

GTX 1080 FE

32 GB DDR3 of G.Skill Ram @ 2400 MHz

Corsair CX 750 Power Supply

 

I have been using the i7-4790k since December of 2015. I have never overclocked it once (just have had the Turbo Boost option enabled). I recently upgraded my GPU from an AMD HD 6700 series to a GTX 1080. When I made this upgrade, I thought it would be a good time to overclock the 4790k a bit especially since I do have a water cooler. I decided to try overclocked to 4.6 GHz with the voltage set at 1.260. This is where I ran into issues.

 

I was able to play BF4 on Ultra at a constant 200 FPS which was exciting however, when I closed out of the program and launched Warzone, the computer instantly crashed and gave me a bluescreen. I also stress tested the CPU prior to launching BF4 to make sure the OC was stable, and it was as the CPU maintained 80 degrees at full load. The computer then restarted however this time it didn't even allow me to login to windows as it again crashed giving me a blue screen. So, I went back into the bios and removed the OC settings. I was able to re-login to windows however, after stress testing it again the CPU managed to reach temps of 90+ eventually getting up to the 100 mark and then crashing.

 

So, I’m not sure why my CPU is now crashing at the settings it was previously at (no OC just Turbo) when I have been running it like that for years. I decided to take out the GTX 1080 and re-install the AMD HD 6700 (I also deleted all NIVIDA drivers using DDU prior to putting the AMD card back in). Temps looked to be back to normal, so I thought that the GTX 1080 caused the issue because the CPU now needs to run a lot faster for the GTX 1080. So, to make sure that that was the case, I re-installed the GTX 1080 expected the CPU to reach 100-degree temps. However, the CPU ran like normal, reaching an 80-degree high after a 15 min benchmark. So now I was really confused because it wasn't the GTX 1080.

 

I decided to then give OC another go but this time I decided to increase it to 4.4 GHz with a 1.22 voltage and this crashed the computer again reaching 100 degrees. At this point, I knew the issue had to be my cooler or my thermal paste was old. Since I haven't changed my thermal paste since I built the computer back in 2015, I decided to remove the AIO and replace the thermal paste. I also noticed the radiator was absolutely clogged with dust, so I thought that had to be a contributing factor. Once I re-installed the AIO back into the PC and then ran some benchmarks (with non-OC) it seemed to work (sort of) only getting again 80-degree highs which was the same as before. But I thought since the cooler has been cleaned and I applied new thermal paste that I could once again try OC the CPU.

 

I tried to do the 4.6 GHz at 1.260 voltage again however it again crashed. I then tried the 4.4 GHz with 1.22 voltage and yet again it crashed. So, at this point I reverted the BIOS back to where there was no overclock and once, I booted back into windows and ran a benchmark, it again crashed. This was extremely frustrating as it seemed as if overclocking the CPU at all causes a never-ending loop of bluescreens and 100-degree highs under 100% loads even if I revert back to non-OC settings.

 

I managed to get it back to operating normal (with the GTX 1080 installed) and I decided to not even mess with OC since apparently my CPU has major issues with it (even though its an i7-4790k). I decided to test some games (GTA 5, Warzone, BT4) and then all ran really well (except GTA 5 which had a stuttering problem, but I fixed it by changed the Direct X version to use 10 instead of 11 in the game settings). This is where everything went wrong again!

 

After doing that I decided to launch Minecraft, a game that is much easier to run than GTA 5 and when launching into the game, the CPU temps skyrocketed to 100 degrees and crashed. Keep in mind, I didn't even make it to the main menu screen. The computer then restarted and I tested it again and the same thing happened. So, at this point I was super upset since I don't know what is wrong with my PC.

 

When applying the thermal paste, I did notice that it seemed a bit chunky/chalky (but I'm not sure since this was the first time, I applied thermal paste). My dad who gave it to me mentioned he got it a while ago (maybe 1-2 years ago). So that might be the problem, which is why I am going to replace the thermal paste again with something newer.

 

However, I don't know if that will fix it. I feel like there is no way my CPU is suddenly broken because of the first overclock that I did to it. I also understand that the GPU upgrade results in the CPU working harder however, I think that the i7-4790k is more than capable of handling the GTX 1080 as the CPU came out 2 years before the 1080 was released. Plus, I have seen many people on YouTube running the same CPU and GPU as mine.

 

I don't know what to do at this point and I need some help desperately. Hopefully you guys can help me out. Thanks

(While writing this, my CPU temps are at 60 degrees with no overclock using Brave Browser)

[HQuan]

I once using same CPU as your (200mm AIO), and never ever reached 50oC when using browser (just check my profile, didn't update it)

Please check your thermal solution.

[crxefx]

I'm more of a mechanic than a computer guru. Are you sure that your cooling pump was running when you were reaching 100? I wonder if it is possible that the pump is going out? Or is aerated? I have a 4790k in my old rig, I never bothered trying to overclock it though. Good luck, I'm just browsing, learning stuff!

[Somerandomtechyboi]

What case do you have?

Fan configuration?

 

Id reccomend upgrading the cooler and delidding the cpu to change the tim for liquid metal cause the stock tim is horrible

 

aim for 4.8g, 5g+ prefferably, volts wise 22nm spec is 1.55v so just ignore any volt danger since temps are gonna be the limiter, you may also be able to run higher speeds at a given volt if you play around with bclk strap (125-130bclk)

[191x7]

Are you sure the AIO pump is working fine? If it's a cooler from 2015, it probably isn't fine, the average lifespan of an AIO is 3 to 5 years.

 

Simply try replacing the cooler.

[numberz]

8 hours ago, crxefx said:

I'm more of a mechanic than a computer guru. Are you sure that your cooling pump was running when you were reaching 100? I wonder if it is possible that the pump is going out? Or is aerated? I have a 4790k in my old rig, I never bothered trying to overclock it though. Good luck, I'm just browsing, learning stuff!

I have Corsair Link 4 installed on my computer and it is showing the pumps RPM to be around 3000. I also noticed when re-installing the cooler after I re-applied the thermal paste that I heard the water going through but all of sudden (after about 30 sec) nothing was there to be heard. During benchmarks, I was able to feel that the water was warm in one tube than the other which to me indicated that the cooler was working as intended.

[numberz]

8 hours ago, Somerandomtechyboi said:

What case do you have?

Fan configuration?

 

Id reccomend upgrading the cooler and delidding the cpu to change the tim for liquid metal cause the stock tim is horrible

 

aim for 4.8g, 5g+ prefferably, volts wise 22nm spec is 1.55v so just ignore any volt danger since temps are gonna be the limiter, you may also be able to run higher speeds at a given volt if you play around with bclk strap (125-130bclk)

I don't know if it is a safe idea to de-lid the CPU considering its an older CPU, plus I don't want something to break. The fans used to be configured pulling air in from the back (these fans are the ones attached to the radiator) and pulling from the front and pushing from the top. But now I re-configured it to pull air in from the front and push out the back.

 

My case is a CoolerMaster HAF 912

[numberz]

7 hours ago, 191x7 said:

Are you sure the AIO pump is working fine? If it's a cooler from 2015, it probably isn't fine, the average lifespan of an AIO is 3 to 5 years.

 

Simply try replacing the cooler.

I think you are right about the cooler. I have had this cooler since 2015 and I think there is some block preventing the water from circulating through.

[Somerandomtechyboi]

17 minutes ago, numberz said:

I don't know if it is a safe idea to de-lid the CPU considering its an older CPU, plus I don't want something to break

Its literally just buy a chinese tool, use it to delid the cpu (basically just pushing othe ihs till it pops off), apply liquid metal, then reseal cpu

 

There are alot of delidding tutorials on yt so have a look at those

[191x7]

1 hour ago, numberz said:

I think you are right about the cooler. I have had this cooler since 2015 and I think there is some block preventing the water from circulating through.

Just get another cooler. Don't use the system like this.

[WoodenMarker]

Where is the cooler mounted?

[numberz]

I fixed the issue. It was the AIO. The pump was not circulating the water properly. After having it for more than 6 years I guess it was destined to fail at some point. I replaced it with a Noctua D15 air cooler and it is way better. 67 degree high on full load. Thanks guys