4790k EKblock high temps

Swapped out my H100i for a custom loop hoping to get lower temps. With the H100i I was in the idle range of 30degrees celsius. However once constructing my loop my idle temps are significantly higher =( hanging around 40degree celsius.
( I am aware idle 40c is not dangerous as high load temps are really the issue, my goal is to cool my 4790k as close to ambient as possible (no OC applied)
specs are as follows

EK supremacy water block and compression fittings
xscp ~300ml tube res
xspc bay res X20 750 (with pump)
HX1020 or something anyways its a 360 rad with af120 and radflow fans in push/pull directing air (to the outside of the ceiling of case)

How is it that the H100i outperforms this setup. I am running the cpu block in a horizontal configuration and I believe i have the correct 1150 jet plates installed. (will look again)
this setup also has 1liter of aurora mayhem in it as well... something isnt adding up here more water more cooling right? So the bottleneck is the cooling plate attached to CPU?

reapplied thermal paste > 6 times (still crappy temps) *pea sized method light circular pressing pattern.
going to put a water sensor in later today
checking mounting position of cpu block ( applied crisscross tighten pattern)
*read a bunch of different forum and posts, seems like im on a tangent here chasing this ghost.

(ignore the vertical cpu water block orientation ( i am running horizontal or ccw atm)


I have read that pulling cold air through the rad is better on these forums I will attempt that later its difficult as the case is small corsair graphite 760t and I just cut my knuckle in half after a camping trip. =(

Idle temps mean nothing. Please post H100 load temps and new setup load temps.

I see you used some Mayhem dyes. Please refer to this post. Just updated.

http://www.overclockers.com/forums/...Need-help-with-new-loop?p=7850322#post7850322

Cold air is better. It would be maybe 1C better if you were to change fans around. It's best to leave it as it is. It's the best setup for your case.

Just read that article a few minutes ago.. very insightful how dyes can cause discoloration. Can that translate to worse temps, i dont know, i recently added 1oz of distilled water, and will switch the fans, maybe it was the jump from z87 to z97 that caused it. I think the z97 board runs way hotter than my msi g45 gaming. Its just kind om embarrassing as my roomates who run corsair units laugh and say my loop is garbage ::sigh::

this is what it looks like for the custom loop at almost full load, temps are OK. (sentry 2 probes say ambient in case is 28c)

Dyes are a no-no in my book as well.

OP, some of your questions and your loop configuration makes me think you didn't fully read up before taking up on this expensive fun filled hobby. You don't need two reservoirs and to answer one of your questions, to get better delta-Temps you'll need more heat surface. The more the merrier until you hit diminishing returns but again, that all depends on what you're going to cool.

EK blocks and their jet plates can get tricky or at least confusing for me at times since I've never owned a EK CPU block but I'd advise on giving them a call and see what they say.

It is possible and would make sense why you've lost some thermal performance from your loop once the dyes were introduced. Distilled water has the highest thermal properties than anything of the dye fluids out there, AFAIK. I've been using distilled water and a silver kill coil with a few added drops of dead water for many years now. If you were worried about the eye candy, all you had to do is purchase some colored tubing from PrimoChill Advanced LRT. Can I ask what tubing is being used in your loop atm? If not the right tubing is purchased, you could set yourself off with more headaches as there are some tubing that like to leech into the loop. Its basically like Slimmer from Ghostbusters, your stuff will get slimmed and than harden up over time which will result in less flow and a gradual heat build up on your components. Lets not forget some of the stains that dyes will leave behind with leeched tubing or w/out.

Hope you read up a bit more and get all of this settled. Before I go, as Con said, its not about idle temps but its all about Delta-Temps and load temps as well as ambient temps and if you have the added luxury of knowing, water temps.

Ya it was an impulse purchase the mayhem fluid, I was under the impression dye as an additive was a nono, however using premade fluids from manufacturers was A ok?

tubing is LRT, I basically purchased alot of the best seller items from performance-pcs.com, saw what other people were using and added the extra res since I thought I could beat the heat using a higher volume of water. I could add an extra rad in front, and am willing to try it but like most folks funds are limited. as far as the watercooling beginners guide, ive read it, and ready to get my hands dirty. at this point im thinking flow rate could be an issue, also pump heat transfered into the water from the x20 xspc unit.

- - - Updated - - -

here are the load temps, for some reason I am having trouble posting atm, sorry for delayed posts. so any manufactured coolant that has coller is an additive? I was under the assumption that premade mixes were ok? Will be switching to distilled water to try that out!

Can you run the XTU benchmark instead of just the stress test? That way we can get a max temp.
That said, 53°C is good.

Higher water volume won't mean anything since after 30 mins of minimum run time, the fluid will reach its equilibrium.

The tube reservoir was a waste of money that could have been spent towards either a much better pump like the XSPC D5 or got a front radiator if you were going to be adding the GPU to the loop.

Its ok, everyone learns and you'll always keep learning as you go. You might have rushed this a bit as we usually get folks to provide us with a run down list as we just give them the thumbs up if everything looks good. We would have caught you in time but unfortunately you came to us after. Again, its all for fun and a learning experience.

here is the idle temps from hwmonitor z97 seems to run hotter than my 4790k in my msi z87 setup. could be the additional rails since this is techinically an "OC motherboard" frustrated for now but i wont be giving up any time soon!

I'll say what was said before:
Idle temps mean nothing, give us load temps.

Ya its tons of fun and I agree completely, I should have gotten a better pump, but I think i will be migrating to a larger corsair case as this one has cut me a bunch. for the xtu benchmark the picture was actually from the benchmark for score, did you want me to run an extended cpu stress test from the menu item above benchmarking. Thanks for the help guys, Im pretty stubborn but do have the desire to learn so be patient with me ^.^

- - - Updated - - -

ATMINSIDE said:

I'll say what was said before:
Idle temps mean nothing, give us load temps.

Click to expand...

is this it? Hoping i get it right this time lol =)

We want you to run something like Prime 95 small ffts. Use Realtemp to monitor temps. Use CPU-Z for status on clockspeed etc.

You let your PC idle for an hour or so, just monitor realtemp. Take a screenshot of realtemp and CPU-Z. Write down the room temps. That's called ambient temp. Post the screenshot.

Then, you run P95 for at least 30 minutes and take another screenshot, and write down the room temps.

So your screen should have all three programs up when you take the screenshot. Post the screenshot.

Only then can we tell.

gottcha! going to set that up asap, pretty sure I got the software, just need the time to run the tests, will try tonight afterwork, or after class tomorrow thanks again!

pr0jectm said:

Ya its tons of fun and I agree completely, I should have gotten a better pump, but I think i will be migrating to a larger corsair case as this one has cut me a bunch. for the xtu benchmark the picture was actually from the benchmark for score, did you want me to run an extended cpu stress test from the menu item above benchmarking. Thanks for the help guys, Im pretty stubborn but do have the desire to learn so be patient with me ^.^

- - - Updated - - -



View attachment 167788 is this it? Hoping i get it right this time lol =)

Click to expand...

Yep, that's what I was looking for, temps are great!

that's a sweet looking rig, love the coolant, to bad it has to come right out before it gums stuff up.
don't let em dog you to much on the pump/res, I use the same, works fine and looks good doing it.
for my ek block on my 4790K I use the plate with the widest slot, it's now on my 8350, guess what.... same plate.

those damn plates are hard to distinguish which is widest not to mention no label as far as I can tell, can you tell me what kind of temps your getting caddi on ur 4790k.. also your rad configuration would be awesome. got class in the am, hoping to run some tests and experiment tomorrow..

I just built my first loop too. Don't worry, I'll be tearing everything down since the dye I put in it is apparently breaking down already. I been researching for months, and still have made mistakes, it's just one of those things you have to learn as it happens. Experience is the best teacher. I did learn beforehand that the aurora dye is only meant for show PCs, and will lose it's "aurora" effect pretty fast.

It would be nice to know what kind of flow rate you're getting. I'm running two Alphacool UT60 360 rads, same CPU, same CPU block, and two MSI 980s with nickel blocks (only option.) Idle temp on my CPU right now is 30 C. Usually gets up to like 44 while playing Witcher 3 maxed out at 1440p. Same with BF4.

Comparing your voltages to mine, yours are significantly higher. Might be why you are seeing higher temps, combined with the aurora dye, which will raise temps some. I'm just going to post a screenshot for you to compare.

I'll also add that these CPUs are known to have varying heat issues due to intel's TIM. But seeing as how you had lower temps with the H100i, I wouldn't assume that's a problem. I just put a pea sized dot on my CPU, and then set the block on it, I don't move it around, it will spread out on it's own.

Third time editing post. I believe it's the second thickest jet thing that you need to use. But even still, it's only like a degree or two difference. If you left it the way it was, it's not that big of a deal. I guess also make sure everything else with it is installed correctly if you took it apart.

4th edit: Take a further out pic so we can see the whole loop. I just don't see having two res on one pump as being wise. Seems like it would create a lot more resistance. Honestly (this is without seeing the rest of the loop) it looks like you are running your CPU block in reverse. As in, it looks like your fluid is entering what should be the exit of the block.

dude you rock thanks!! cali is super hot, as places are literally catching on fire atm. thanks for the info, im definitly going to switch cases, this handme down sucks fan controller broken no space for res and push pull had to take a dremel to smooth some edges ect.. great to know your temps thanks seriously! Ill probably emulate your setup, and the aurora liquid has already broken down, its sad man I kind of want to write a letter to mayhem why their products are inadequte yet so awesome looking! can i see a pic of your pc? =3 I am very grateful for all the help on here, I definitly want to bench and get screenshots hopefully later today, for now I will start reducing my voltages so I can see where i bsod..