CPU changes temperature rapidly

[Curemylife]

Hey People!

 

(the processor in question 4790k @ stock - 1.2v)

 

I'm having a little trouble with my "loop", or rather my temperatures, I'm in the middle of building a scratch case so right now I'm using some random stuff in the meantime...

 

The loop as it stands right now consists of the following(in order of the loop): Reservoir, DDC pump, CPU-block(correct inlet/outlet), 480Radiator and then back to the res.

 

My issue is that my CPU right now is at a steady 25 degrees celsius during idle, but as soon as I stress it it goes straight up to 80... and really fast, same thing happens when turning the stresstest of.. My hypothesis is that this is due to a collaboration between, haswell's bad IHS seating and the small die, but when i used a smaller 200x200mm radiator (both CPU and GPU at the time) the CPU did not reach those temperatures. I also wanna point out that the radiator is also still very chilly, this is what leads me to believe that the heat transfer between the IHS -> water is not working like it should...

 

The things i have tested:

 

Making sure that there are no bubbles left.

 

Reseated the cpu block.

 

Making big changes in fan RPM.

 

And all of this to no avail..

 

Does any of this make any sense to you guys and does anyone have any tips and tricks?

[CharlieR1]

This sounds like some sort of throttling. I would suggest re-applying thermal paste.

[Necrodead]

If you are an advanced modder I would look into de-liding your cpu. It is possible you have horrible contact between the IHS and the die, since it seems you have checked all other issues off. Pump is working I take it?

[Curemylife]

If you are an advanced modder I would look into de-liding your cpu. It is possible you have horrible contact between the IHS and the die, since it seems you have checked all other issues off. Pump is working I take it?

Yeah, it almost seems so, but as I explained it seems wierd for it to be hotter now than with the smaller radiator and it's also been running cooler with some smaller air coolers, could it possibly be the pump? It's circulates water perfectly it seems but maybe it's not at a high enough rate? 

 

The only thing that i've actually changed between the two loops is cutting out the gpu and changing the pump top, maybe the pump top is screwed on too tightly?

[Necrodead]

Yeah, it almost seems so, but as I explained it seems wierd for it to be hotter now than with the smaller radiator and it's also been running cooler with some smaller air coolers, could it possibly be the pump? It's circulates water perfectly it seems but maybe it's not at a high enough rate? 

 

The only thing that i've actually changed between the two loops is cutting out the gpu and changing the pump top, maybe the pump top is screwed on too tightly?

The flow rate shouldn't affect your temps as long as water is moving. Your using the same cpu block? And your positive there is no air in the system? I'm gonna keep thinking on this.

[Ellis_D]

That honestly makes no sense for it to be the IHS if you weren't having this problem with a different radiator.  I'd suggest testing with the previous radiator to see if the temperatures go back to how they were and if so, there's probably a manufacture's defect in the 480 Rad that's preventing your coolant from flowing through it properly.

[Curemylife]

The flow rate shouldn't affect your temps as long as water is moving. Your using the same cpu block? And your positive there is no air in the system? I'm gonna keep thinking on this.

Yeah, I've been twisting and turning around every single part of the system just to make sure. Same block aswell..

 

That honestly makes no sense for it to be the IHS if you weren't having this problem with a different radiator.  I'd suggest testing with the previous radiator to see if the temperatures go back to how they were and if so, there's probably a manufacture's defect in the 480 Rad that's preventing your coolant from flowing through it properly.

You're so right! That's pretty much the same conclusion as I've reached only I've never actually heard about this kind of defect..

I've been trying to feel if theres some kind of heat difference between the lanes in the radiator but I have yet to find one.

 

Will try switching back to the former radiator when my new motherboard arrives, perhaps I'll get a chance to order a block for my gpu too.

 

Thanks for trying to help guys!