Overclock.net banner

1 - 20 of 33 Posts

·
Registered
Joined
·
19 Posts
Discussion Starter #1
So, I finished installing my water cooling loop yesterday and after carrying out the normal temp checks on how well the loop is performing, my GPU is performing really well, however, the trouble im seeing is on 100% load testing with prime 95, the CPU hits 89 degrees in a matter of seconds, this ins't my first loop, and from my experience the CPU should not really hit over 60 degrees, I can even confirm this with the using the Corsair H110iGTX and my CPU not getting above 60.<br><br>
I have drained the loop this morning and ensured the thermal paste is in contact, and it has definitely spread, the thumb screws to secure the block to the board are as tight as they will allow, and all washers etc are in place. Everything is as tight as can be.<br><br>
The block in question is <a href="http://www.watercoolinguk.co.uk/p/Water-Heatkiller-IV-Pro-Intel-Clean-acrylic_52903.html" target="_blank">http://www.watercoolinguk.co.uk/p/Water-Heatkiller-IV-Pro-Intel-Clean-acrylic_52903.html</a>, I did notice in the specs it is LGA115X compatible.<br><br>
I really dont know what to do with the block now, has anyone else had any issues with this specific block and LGA1151?<br><br><a class="H-lightbox-open" href="http://www.overclock.net/content/type/61/id/2961211/"><img alt="" class="lightbox-enabled" data-id="2961211" data-type="61" src="http://www.overclock.net/content/type/61/id/2961211/width/500/height/1000/flags/LL" style="; width: 500px; height: 375px"></a><br><br>
Is it just that this block does not work on 1151? I hope this isn't the case.<br><br>
Thanks for your help.
 

·
Registered
Joined
·
201 Posts
By any chance, did you forget to remove the bottom sticker of the waterblock? Also do you have the official heatkiller backplate for your waterblock?
 

·
Registered
Joined
·
19 Posts
Discussion Starter #3
Hi Duke,<br><br>
There is no sticker on the block, I am not using the official back plate, but its not compulsory, as far as I was aware, I kind of figured I didn't need it as the standard pillars suit 115X installs.<br><br>
Regards
 

·
Registered
Joined
·
201 Posts
Since you are not using the official ones from watercool, the backplate might not be providing the right tension that you need thus resulting of the high temp. You can try to loosen them a bit just incase you over tighten them.
 

·
Registered
Joined
·
19 Posts
Discussion Starter #5
I have just ordered the backplate, it should be here Tuesday, for now I wont be carrying out any CPU intensive work. I hope this is the answer! Thanks for your help, and I will keep you posted.
 

·
Just hangin' out
Joined
·
1,071 Posts
<div class="quote-container" data-huddler-embed="/t/1623095/awful-temps-with-heatkiller-iv-pro-on-skylake-i7#post_25838976" data-huddler-embed-placeholder="false"><span>Quote:</span>
<div class="quote-block">Originally Posted by <strong>RaiderUK</strong> <a href="/t/1623095/awful-temps-with-heatkiller-iv-pro-on-skylake-i7#post_25838976"><img alt="View Post" class="inlineimg" src="/img/forum/go_quote.gif"></a><br><br>
the thumb screws to secure the block to the board are as tight as they will allow, and all washers etc are in place. Everything is as tight as can be.<br><br>
.</div>
</div>
<br>
I am not sure about the HK IV, but my HK 3.0 installation instructions said to actually measure the length of the mounting screws protruding above the waterblock. I believe mine were to be set at 19mm.<br><br>
I used a digital caliper to measure them. They are definitely not as tight as they can be.
 

·
Registered
Joined
·
703 Posts
The mounting mechanism of the Heatkiller IV is (theoretically) foolproof. Just tighten the screws all the way till they hit the end of the thread. The thread-length is adjusted for the different sockets according to intels specifications.<br>
The Backplate for socket 115X is optional, you can absolutely install the Heatkiller without it. It helps in spreading the physical tensions over the socket area, but shouldn't affect the thermal outcomes at all.<br><br>
Actually, there IS a protective film on the bottom of the cooling plate (or, to be precise, normally there should be). So, if you didn't remove this, I'd say we have our culprit. Please disassemble the loop, clean all thermal compound from the cooling plate, and make a picture of it, so I can visually verify.
 

·
Registered
Joined
·
19 Posts
Discussion Starter #9
Hi Jakob,<br><br>
I have the back plate coming on Tuesday or Wednesday, but I can 100% guarantee that there is no big white sticker on the block that says "Please remove first". This isn't my first water loop, definitely not my first heat sink install. I did notice however that on one part of the thermal compound when I removed the CPU, there was very little in a certain place, this implies that it is tight at that point, but to avoid any uncertainty, I removed all thermal compound, placed some on the CPU and spread it this time evenly on top (normally I do the pea size process). When the back plate comes, I will have to disassemble the loop then so I will take photos when it is in bits. I will also take photos of the thermal compound when I remove the block.<br><br>
I have a feeling that there is contact, but it is insufficient, so fingers crossed, the back plate will spread the load on the pressure on the CPU to water block.<br><br>
Regards
 

·
Registered
Joined
·
703 Posts
We talked about this problem in the shop, and found two other possible reasons for your bad mount:<br><br>
- You mounted the water block in "goofy" orientation. It might be possible that on this specific mainboard, there are some components colliding with the Heatkiller in goofy orientation. We design our product according to Intels "KeepOut" Compliance Rules, but some motherboard layouts are not overly strict in respecting those rules. Please try the normal orientation (90° from your current setup), this works always.<br><br>
- Maybe one of your hardline runs is a tiny bit too long, thus creating a sideways force on the water block and ruining a perfect mount. Can you do a test setup with softtubes to rule this possibility out?<br><br>
- And I need to restate that ALL our waterblocks leave our shop with a big protective sticker on the baseplate. If you did not remove this sticker, chances are that you either bought an illegitimate product or forgot to remove it. Please double check that there is no sticker whatsoever.
 

·
Registered
Joined
·
19 Posts
Discussion Starter #11
Hi Jakob,<br><br>
Thank you for your further discussions with your team on this, i very much appreciate it. In answer to your points:<br><br>
- You mounted the water block in "goofy" orientation. It might be possible that on this specific mainboard, there are some components colliding with the Heatkiller in goofy orientation. We design our product according to Intels "KeepOut" Compliance Rules, but some motherboard layouts are not overly strict in respecting those rules. Please try the normal orientation (90° from your current setup), this works always.<br><span style="color:#FF0000;">The are around the CPU is clear, there are no components high enough to interfere with the water block.</span><br><br>
- Maybe one of your hardline runs is a tiny bit too long, thus creating a sideways force on the water block and ruining a perfect mount. Can you do a test setup with softtubes to rule this possibility out?<br><span style="color:#FF0000;">Im not going to rule this out, it may actually be pressure being applied from the fittings on the reservoir side of the CPU, I will be able to determine this when I dismantle the loop to implement the backplate, and check the thermal compound to where it has been applied most. Unfortunately, I dont have the resources or finances to buy more fittings and rubber tubing to test this, I am anticipating that the backplate will pull the whole area flat with each component that is connected.</span><br><br>
- And I need to restate that ALL our waterblocks leave our shop with a big protective sticker on the baseplate. If you did not remove this sticker, chances are that you either bought an illegitimate product or forgot to remove it. Please double check that there is no sticker whatsoever.<br><span style="color:#FF0000;">Definitely removed the sticker. <img alt="biggrin.gif" class="bbcode_smiley" src="http://files.overclock.net/images/smilies/biggrin.gif"></span><br><br>
The part is actually being delivered on Thursday so I will not know any outcome until late Thursday / Late Friday.<br><br>
Once again, I appreciate your investigation on this matter, and I will be sure to keep you updated.<br><br>
Many thanks
 

·
Registered
Joined
·
10,400 Posts
I've seen clear films on metal brackets and products and parts that was ALMOST indistinguishable except for the reflectiveness or it being hit by the right light. I guess that would be why they use big white stickers instead. I say that for OTHERS fiding this via google search.<br><br>
Another way to prove the block is seating correctly is the spread patter when using the rice grain method on the TIM. (also stating for others)<br><br>
<a href="https://www.youtube.com/watch?v=EyXLu1Ms-q4" target="_blank">https://www.youtube.com/watch?v=EyXLu1Ms-q4</a><br><br><br>
Hope you find out the problem as i'm curious what it is now.
 

·
Registered
Joined
·
2,024 Posts
Is it the SkyLake or the new KabyLake where intel went back to using cheap tim between die & ihs, resulting a freaking hot running cpu?<br>
I've seen some reports on the new cpu's running hot, but didn't really pay much attention, since i'm not in the market for an upgrade (unless Ryzen really is an intel killer)
 

·
Registered
Joined
·
2,023 Posts
To the OP: check if your motherboard is warped. Take pics of the TIM leftover in your block and cpu surface to check the spread pattern (and share!).<br><div class="quote-container" data-huddler-embed="/t/1623095/awful-temps-with-heatkiller-iv-pro-on-skylake-i7/10#post_25844651" data-huddler-embed-placeholder="false"><span>Quote:</span>
<div class="quote-block">Originally Posted by <strong>cdoublejj</strong> <a href="/t/1623095/awful-temps-with-heatkiller-iv-pro-on-skylake-i7/10#post_25844651"><img alt="View Post" class="inlineimg" src="/img/forum/go_quote.gif"></a><br><br>
I've seen clear films on metal brackets and products and parts that was ALMOST indistinguishable except for the reflectiveness or it being hit by the right light. I guess that would be why they use big white stickers instead. I say that for OTHERS fiding this via google search.<br><br>
Another way to prove the block is seating correctly is the spread patter when using the rice grain method on the TIM. (also stating for others)<br><br>
<a href="https://www.youtube.com/watch?v=EyXLu1Ms-q4" target="_blank">https://www.youtube.com/watch?v=EyXLu1Ms-q4</a><br><br><br>
Hope you find out the problem as i'm curious what it is now.</div>
</div>
I think seating correctness is mostly defined by the block's flow path and the CPU die shape.<br><br><div class="quote-container" data-huddler-embed="/t/1623095/awful-temps-with-heatkiller-iv-pro-on-skylake-i7/10#post_25845722" data-huddler-embed-placeholder="false"><span>Quote:</span>
<div class="quote-block">Originally Posted by <strong>RnRollie</strong> <a href="/t/1623095/awful-temps-with-heatkiller-iv-pro-on-skylake-i7/10#post_25845722"><img alt="View Post" class="inlineimg" src="/img/forum/go_quote.gif"></a><br><br>
Is it the SkyLake or the new KabyLake where intel went back to using cheap tim between die & ihs, resulting a freaking hot running cpu?<br>
I've seen some reports on the new cpu's running hot, but didn't really pay much attention, since i'm not in the market for an upgrade (unless Ryzen really is an intel killer)</div>
</div>
Intel stopped soldering heatspreaders with Ivy Bridge CPUs. Devil's Canyon CPUs (Haswell refresh) featured an improved TIM (or so they said), but they didn't keep on improving. See what you can improve by delidding (according to SiliconLottery):<br><div class="quote-container"><span>Quote:</span>
<div class="quote-block">Originally Posted by <strong><a href="https://siliconlottery.com/products/delid" target="_blank">https://siliconlottery.com/products/delid</a></strong> <br>
Ivy Bridge: 10°C to 25°C<br>
Haswell: 10°C to 25°C<br>
Devil's Canyon: 7°C to 15°C<br>
Broadwell: 8°C to 18°C<br>
Skylake: 8°C to 18°C<br>
Kaby Lake: 12° to 25°C</div>
</div>
 

·
Registered
Joined
·
10,400 Posts
I once had an arctic freezer 7 pro that wouldn't work. conducted heat form a soldering iron just fine but, the CPU was running way hot. bought another heat sink and it was running fine. my friend took the arctic 7 and it worked fine for him. never did figure out why it wouldn't work.
 

·
Registered
Joined
·
703 Posts
<div class="quote-container" data-huddler-embed="/t/1623095/awful-temps-with-heatkiller-iv-pro-on-skylake-i7/10#post_25844625" data-huddler-embed-placeholder="false"><span>Quote:</span>
<div class="quote-block">Originally Posted by <strong>RaiderUK</strong> <a href="/t/1623095/awful-temps-with-heatkiller-iv-pro-on-skylake-i7/10#post_25844625"><img alt="View Post" class="inlineimg" src="/img/forum/go_quote.gif"></a><br><br>
The part is actually being delivered on Thursday so I will not know any outcome until late Thursday / Late Friday.<br><br>
Once again, I appreciate your investigation on this matter, and I will be sure to keep you updated.<br><br>
Many thanks</div>
</div>
Hey Raider,<br>
I'm curious. Did the backplate help with your problem? Did you take some pics of the TIM spread? I'd really like to know what happenend there!
 

·
Registered
Joined
·
19 Posts
Discussion Starter #17
Hi Jakob,<br><br>
So I went back to the CPU and had a proper look around, and it turned out there were 4 small capacitors under the CPU block, schoolboy error on my part <img alt="whistle.gif" class="bbcode_smiley" src="http://files.overclock.net/images/smilies/whistle.gif">, attached is the image for what I found, however, the back plate is now also installed and I am happy with the 23 degrees idle temp / 50 degrees load temps . <img alt="biggrin.gif" class="bbcode_smiley" src="http://files.overclock.net/images/smilies/biggrin.gif"><br><br>
The CPU block is now at standard orientation, with the inlet in the centre and the out at the bottom.<br><br>
Thanks again for your efforts on finding a resolution. I cant believe ASUS let me down with capacitors in the way. <img alt="wink.gif" class="bbcode_smiley" src="http://files.overclock.net/images/smilies/wink.gif"><br><br><a class="H-lightbox-open" href="http://www.overclock.net/content/type/61/id/2966211/"><img alt="" class="lightbox-enabled" data-id="2966211" data-type="61" src="http://www.overclock.net/content/type/61/id/2966211/width/350/height/700/flags/LL" style="; width: 350px; height: 263px"></a>
 

·
Registered
Joined
·
703 Posts
Hey, thanks for clearing this up!

Yeah, it's a shame, because intel gives very very clear definitions on their "KeepOut" area (it's even officially called that!). And of course, we have to take those defintions as the baseline for our design process, trying to design a block that strictly stays within this area. As you now know, the "normal" orientation always works. But sometimes, motherboard manufacturers act sloppy on those definitions, and come extremely close to the socket area, so much that the "goofy" orientation of a bunch of waterblocks sometimes fails.

I am happy that we could solve your problem and you finally reah temperatures that make you happy! Keep up the good work!


Edit:
I need to correct my statements from 2017. The problem is NOT the motherboard manufacturers - it is, in fact, our mounting bracket. It does NOT follow the intel specification exactly.
The mounting bracket was designed in 2015, based on boarda available at the time, and it was perfectly fine to overstep the boundaries a little bit, as all motherboards had plenty of room. But later board generations moved components nearer and nearer to the keep-out area, so much so, that nowadays, our block DOES in fact, collide with them.


I apologize for my mistake in communicating this wrongly.
 

·
Registered
Joined
·
19 Posts
Discussion Starter #19
I think I can see what has happened here with ASUS designing and making the board, It seems as though the holes for the capacitors have been drilled almost perfectly in line with the edge of your water block design, most likely in line with Intel's "Keep Out" area, however, it looks as though they didn't compensate for the actual diameter of the capacitor. <img alt="thumb.gif" class="bbcode_smiley" src="http://files.overclock.net/images/smilies/thumb.gif">
 

·
Registered
Joined
·
1 Posts
I know I'm bumping an old thread but I wanted to add that this was also my issue. I have a Maximus XI Forumla and a Heatkiller IV Pro. Was getting absolutely god awful temps and the CPU wasn't usable. Reseated the waterblock like 10 times (always in the same orientation where the text was facing the correct way to read). Saw this picture with the capacitors and noticed that there were capacitors under the mounting bars of the waterblock, rotated the waterblock and it works perfectly.
 
1 - 20 of 33 Posts
Top