Yes I meant to say this in my last post it could be a bad mount.Stefan all you can do is a process of elimination to rule everything out, unfortunately takes time
Huh, oddly enough, I did not do that... how sometimes the simplest thing does not come to my mind, I'm always a bit complicated I guess... :3eyes: Oh, yes it did, the bottom rad is turned around now, I'll see to take a picture and post it here. edit: Here's a shot of the current rig, as you can see, the bottom rad is turned around, and a little less tubing length is used, thanks to Vet's tips of how to change a bit of it. As I was running Prime95 again, the CPU almost instantly went for 80°C temperatures on half it's cores, but then again, the Block did not become warm to the touch at all... so... should I try to tighten the screws of the block I bit more? I remember trying to not overdue it to not wanting to tear out the threads, but more so, avoid infamous error bd...
I don't see any problems with the layout of the loop so I think we can pretty much rule that out as being an issue, as the guys have said, it could well be that the blocks not tight enough. How are the GPU temps? Are they worrying you as well or is it just the CPU that's the issue? It does look like you have a fairly small reservoir but with the size of the radiators I can't see water volume being a problem. EDIT: It might even be worth removing the block and checking that the thermal paste has spread nice and evenly across the CPU, with the cores jumping in temperature that quickly it would suggest to me that it's not making great contact.
Actually I haven't tested the GPU much, as I haven't found a benchmark problem for them so far (except cinebench which is too short), and of course, it would influence the CPU also I thought, with it having little to no headroom for actual use at all. Yeah, I think it will come down to removing the block, with all the toruble coming with it, and maybe checking on the top rad and how much water actually is in there. I've just picked up two big syringes of GC extreme again, better safe then sorry, and tomorrow I couldn't get any. I'm just worried about the thermal pads that came with the block and were supposed to be placed on the VRMs... I don't really have any spare ones, and I don't kow if I could reuse the ones on there? VRM temperature didn't scare me so far, but I can hardly clean them off any thermal paste on there (I did that because it was adviced in the EKWB manual for the block).
It might be worth running the heaven benchmark as that's fairly light on the CPU but really stresses the GPU, just make sure you are monitoring temps carefully while running the bench. You shouldn't have any problems with reusing the thermal pads as long as they survive the removal of the block without any tearing.
On my side (ek wb also) I tighten the screws as much as I can with my fingers. But if you are as strong as a green bodybuilder then you may need to pay attention
Well I monitor temps with everything I do on this computer, as I'm aware that anything above a single core full loaded means that tempertures will rise. Might give Heaven a try, I would need it sooner or later anyway. hehe certainly not, I tried to tighten them properly without giving too much, as I had to use a screwdriver for the montage. I'm still not sure if I should try to tighten the screws a little, or take it off completely. Because the way I see it, if I need to reseat the monoblock, I will have to take my rig apart again.
Well, as I've been busy during the last days, I didn't have the time to deconstruct my rig. I've done so now, and here are some pics of the TIM's spread on the CPU and mainboard chip: Looks pretty good on the CPU, maybe a bit too much on the mainboard. But certainly, nothing that would make me worry, tbh. BUT.... here it comes. I will enter any entry for epic fail contest with this one, it pretty much explains everything. I'll bring it to you without any further notice, and get to remount the monoblock, and put everything together once again. Well, laugh at me, because I feel I deserve it:
LoL you left the label on, well i have never seen that one before! I was actually going to mention that in an earlier post but i though "Nar he would have taken that off as nobody would leave that on" LoL
Yeah, I know it's weirdly dumb But to be honest, I try to see the good side in it and have a simple answer what could be the problem. Simply to correct, that is... no default hardware problem except the assembler in front of the rig, kind of relieving I'm currently bleeding out air and seemingly making good progress, faster than the first time. Turning that thing upside down even to help get air out of the top rad, maybe I'll be hooking it up tomorrow, Monday the latest, and have a look at stock temps again. Can't wait to see the power beneath :banana:
Well, been bleeding air out of the rig, turning it upside down, tilting, starting / stopping the pump. I guess I'll let it run for a few hours, and tomorrow I will hook it up and check on stock temperatures
Ah hahaha maannn... you left that thing on!! It's cool, at least you got a simple solution to your temp problems lol
Only very, very little bubbling sound as I turn on the pump, and as it sounds i't close to the reservoir again (either in the fan controller, now below the res, or somewhere there in the tubing in between the fan controller and the reservoir). I'm quite confident I will see an instant improvement in temps, unless I've managed to mess up something else of course ... :3eyes: Can't wait to see the temperatures
Well, I've plugged everything up yesterday, and idling at 1.2GHz the CPU reaches about 25-27°C, which is about 5°C above ambient (before: ~34°C, ~10°C above ambient). Doesn't look too bad so far during light load (usually one or two cores at 3.7GHz clock), it climbs to 30-33°C (was getting to 48-55°C). The hottest I saw yesterday was when giving cinebench a run, was getting to about 38°C, never exceding 40°C (was getting into the 70's before...). All in all, I think it shows an improvement above the horrible temperatures EKWB's sticker made me have I will bench a bit today. The only thing that was kind of odd, at stock (with 2133MHz ram), I couldn't beat a 3930K's result in cinebench... is it because of stock clocks, or is there something else to consider?