Gtx 1080 to strix getting too hot for water cooling?

[Dannylti95]

This is the block apart, see anything out of the ordinary?

[rasmuskrj]

Holy shit that is a crap ton of thermal paste  Not sure if its a problem though... There seem to be a spot on your GPU block that the thermal paste is not covering though. Could be an issue. Also looks like you have applied thermal paste on top of the thermal pads, which all seem unnecessary. But not sure it could cause your problems.

Perhaps if there's so much thermal paste that it creates a gap between the card and the block... Seems unlikely though. Exactly how did you apply thermal paste to your block?

 

Did you try to check the water temperatures?

[For Science!]

Hopefully this is the cause of your bad GPU temps.

 

 

Is the component in circled red with a thermal pad? it should not be. See instructions below.

 

 

If so, your mounting is suboptimal and would explain a long standing irk I had with your original photo.

 

 

See how the orange line is not parallel to the yellow and red line? It made me think that your block wasn't screwed on properly, but can be explained by that thermal pad if its in a place it shouldnt be. I personally would not have sliced the pads up like you did or used thermal paste, but I'm assuming that isnt the cause of the issue. Also double check that middle strip is the 1.5 mm variant, whereas all the rest is 1.0mm strip

[For Science!]

14 minutes ago, rasmuskrj said:

-snip-

While relatively speaking it is a lot of thermal paste since its a direct die cooling you will find that this amount of paste is not unusual. With regards to the contact its actually the otherway around, the middle is the only part that is making good contact with the block, and the rest of the die is insulated by the paste because the block is not screwed on tightly enough potentially because of the extra pad.

 

 

[rasmuskrj]

@For Science! yea I agree actually. It looks like the thermal paste hasn't been pressed hard enough against the block. There looks to be some mounting issue.

 

I will maintain that it is a lot of cooling paste, even though it's on a GPU . I used the "EK method"(which means gracious amounts of thermal paste) on my GPU and I think I used about half of that.

Don't think it's a problem though. Most tests I've seen that have worked with this seem to conclude that "too much paste" is a lot more than this.

[Dannylti95]

3 hours ago, For Science! said:

Hopefully this is the cause of your bad GPU temps.

 

 

Is the component in circled red with a thermal pad? it should not be. See instructions below.

 

 

If so, your mounting is suboptimal and would explain a long standing irk I had with your original photo.

 

 

See how the orange line is not parallel to the yellow and red line? It made me think that your block wasn't screwed on properly, but can be explained by that thermal pad if its in a place it shouldnt be. I personally would not have sliced the pads up like you did or used thermal paste, but I'm assuming that isnt the cause of the issue. Also double check that middle strip is the 1.5 mm variant, whereas all the rest is 1.0mm strip

Ok I'll remove the strip on that, but the strip wouldn't cause the warping look on the card because the strip wasnt coming into contact with any part of the block, there was a big gap between the top of the pad and the block. Also the card looks like that because of sag in my opinion but if it wasnt screwed on properly I'll make sure it's on right this time?

[Dannylti95]

3 hours ago, For Science! said:

While relatively speaking it is a lot of thermal paste since its a direct die cooling you will find that this amount of paste is not unusual. With regards to the contact its actually the otherway around, the middle is the only part that is making good contact with the block, and the rest of the die is insulated by the paste because the block is not screwed on tightly enough potentially because of the extra pad.

 

 

I did add a lot of paste lol I did the ek method but on steroids ? and I took the extra pad off, I gotta run to the store to get rubbing alcohol to remove the paste. And I'll see if I can get my hands on some new thermal pads. They're 1mm, 1.5 mm and 2.5 mm if I'm not mistaken?

[For Science!]

23 minutes ago, Dannylti95 said:

I did add a lot of paste lol I did the ek method but on steroids ? and I took the extra pad off, I gotta run to the store to get rubbing alcohol to remove the paste. And I'll see if I can get my hands on some new thermal pads. They're 1mm, 1.5 mm and 2.5 mm if I'm not mistaken?

Yeah thats right, the 2.5 mm ones are for the back. But be aware that the Bitspower thermal pads are the bees knees of thermal pads, unless if you get top of the range, the thermal conductivity is probably not going to be as good as the ones that came with the waterblock.

[Dannylti95]

52 minutes ago, For Science! said:

Yeah thats right, the 2.5 mm ones are for the back. But be aware that the Bitspower thermal pads are the bees knees of thermal pads, unless if you get top of the range, the thermal conductivity is probably not going to be as good as the ones that came with the waterblock.

Hmm ok well do you think I'm still ok with these thermal pads?

[For Science!]

1 minute ago, Dannylti95 said:

Hmm ok well do you think I'm still ok with these thermal pads?

just try the same for now.

[Dannylti95]

SUCCESS!!!!!!! So I reinstalled the block... BUT I don't think that was my problem at all, see I took the Evo block apart, I mean completely apart so I can see where the liquid flows, and I found a chunk of some material idk what it was it looked like hair cleaned it out and also I delidded the CPU. I know it's dangerous without a delidding kit, also I didn't have liquid metal. I used Arctic mx 4. I don't see a difference at all. The temps are a bit jumpy from 40-60 C° tops but that was the case before I delidded so no change. The flow greatly increased and didn't even take 10 mins to bleed the system the bubbles flew right out into the reservoir and that was it. These are my temps after 25 mins under full load on the gpu. I am overclocked on the CPU to 4.6 so that also might play a role in the temps but everything seems to be checking out fine

[For Science!]

1 minute ago, Dannylti95 said:

-snip-

Congratulations, too bad you changed 2 parameters at once, but I am 99% certain it was the thermal pad introducing a space between the acrylic and the board which subsequently made your nickel block hover over the GPU die. Doesn't matter though, congratulations, enjoy the low temps. and yes, you will only see big improvement on CPU temp for liquid metal.

[For Science!]

5 minutes ago, Dannylti95 said:

 I found a chunk of some material idk what it was it looked like hair cleaned it out and also I delidded the CPU. 

Could it be remains of the anti-cyclone foam? Its known to crumble so I wouldn't use it if I were you. But anyway, since you managed to fill the system last time, there clearly was flow. Anyhow, go game.

[Dannylti95]

4 minutes ago, For Science! said:

Could it be remains of the anti-cyclone foam? Its known to crumble so I wouldn't use it if I were you. But anyway, since you managed to fill the system last time, there clearly was flow. Anyhow, go game.

Yes I used that foam but took it out I guess some of it got dislodged and into the block but man I'm so happy! Thanks for your help!!!! Definitely a crazy and trouble some experience but I'm glad I came through. 

[rasmuskrj]

Congratulations man

 

I am also pretty sure your GPU block was mounted incorrectly. If you look at the thermal paste on the block you can see it hasn't been pressed all the way down. Anyways, enjoy your rig

[Dannylti95]

16 hours ago, rasmuskrj said:

Congratulations man

 

I am also pretty sure your GPU block was mounted incorrectly. If you look at the thermal paste on the block you can see it hasn't been pressed all the way down. Anyways, enjoy your rig

Thanks man, well both of those problems are resolved. Under ac at full speed fans for 1 hour playing GTA 5 these are my temps. I couldn't be happier!!!