CPU Water Block That Installs In 12 Seconds

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
You guys have to check out this water block design that installs in just 12 seconds without the use of any tools. This prototype water block actually uses Intel's CPU retention mechanism to lock it in place instead of the traditional method of using the bolt holes around the CPU socket. What do you think?
 

Armenius

Fully [H]
Joined
Jan 28, 2014
Messages
21,992
Pretty neat. Looks like a good solution to not worrying about if you'll tighten it down too much and crack the core on a delidded CPU. Now we just have to see the motherboard standing vertically like in a real-world scenario and if it stays in place :p.
 

Silentbob343

[H]ard|Gawd
Joined
Aug 2, 2004
Messages
1,782
What is the clamping force of the CPU retention bracket. How secure is it with a chunk of copper?
 
D

Deleted member 184142

Guest
First comes to mind is pressure on socket and hinge, how this does over time and if the little hinge can apply enough pressure to make good contact with the CPU core, this also seems to only be fore delided CPUs.
 

rgMekanic

[H]ard|News
Joined
May 13, 2013
Messages
4,546
Pretty cool but the cost would be astonishing since every one has to be cut to fit each specific die
 

Emission

Supreme [H]ardness
Joined
Dec 6, 2005
Messages
4,306
12 seconds not including the time it takes to delid the CPU.

That aside, I love how clean it is without retention hardware around the block.

Pretty cool but the cost would be astonishing since every one has to be cut to fit each specific die
Not necessarily, it just has to have enough clearance around the die. They might need to use some kind of thin thermal foil to make up for the different die heights, and if you look closely in the video it appears to do that already.
 

JDon

Weaksauce
Joined
Jun 24, 2015
Messages
123
I'm wondering how long until Intel "hops on board with delidders" and starts making a version without the lid that A: costs them less, B: they can charge more for, and C: isn't warrantied. :)
 

FrgMstr

Just Plain Mean
Staff member
Joined
May 18, 1997
Messages
49,897
I would not touch that with a ten foot poll. That CPU socket is in NO WAY designed to support the weight of that block much less the lateral forces that would be applied by tubing and installation of the system. I bet some folks are going to be pissed when they find the edges of their naked CPUs chipped off. Trust me, those naked chips do not like things rocking around on them.
 

JDanser

Limp Gawd
Joined
Feb 9, 2012
Messages
246
I could see this working if they came up with some clever way to put in a cross member that shares the load and just have it snap into the stock HSF holes in the motherboard.
 

JosiahBradley

[H]ard|Gawd
Joined
Mar 19, 2006
Messages
1,791
Really thought this used the push pins for quick mounting that is secured with force. This looks more dangerous. Then again I've strapped a kilo of copper to an old CPU die and it didn't break, but that was with proper mounting points.
 

rgMekanic

[H]ard|News
Joined
May 13, 2013
Messages
4,546
12 seconds not including the time it takes to delid the CPU.

That aside, I love how clean it is without retention hardware around the block.



Not necessarily, it just has to have enough clearance around the die. They might need to use some kind of thin thermal foil to make up for the different die heights, and if you look closely in the video it appears to do that already.
Looks machined to fit the die to me.

cooler.png
 

Jovian

Limp Gawd
Joined
Jun 8, 2004
Messages
362
This reminds me of the day when AMD had exposed cores and people were chipping the sides when placing CPU's heatsinks incorrectly on them. Never mind your warranty is voided when you remove the lid. This is too extreme of a solution to be viable.
 

Bandalo

2[H]4U
Joined
Dec 15, 2010
Messages
2,660
This seems like a bad idea. The MB and socket are designed to hold together by themselves just fine. The MB is designed to hold the weight of a cooler mounted to the mounting points and holding down on top of the CPU/socket stack. Putting all that weight on the retention mechanism means the whole weight of the cooler and any stress from the tubing is being held only by the solder points under the socket.

This would probably be OK for a horizontal mount and a case that never gets moved around, but I wouldn't do it for anything else.
 
D

Deleted member 93354

Guest
12 seconds not including the time it takes to delid the CPU.

That aside, I love how clean it is without retention hardware around the block.



Not necessarily, it just has to have enough clearance around the die. They might need to use some kind of thin thermal foil to make up for the different die heights, and if you look closely in the video it appears to do that already.
Every time you add another interface material it lowers the efficiency. Even a liquid metal compound would be questionable here. It does have to have some sort or TIM, but the more you use to accommodate more processors, the less efficient it becomes.
 

Setiri

n00b
Joined
Oct 29, 2005
Messages
51
Completely NOT glossing over all the good arguments here for why this, at a minimum, needs to be tested for not stressing the clam cover... I'm wondering why is the mention of install time an issue? From 3-4 minutes at most for a regular water block down to 12 seconds... and? You typically install a water block 1 maybe 2 times over the course of your CPU's life. If this was a daily installation, ok, that makes sense. Eh, maybe I'm missing something but it seems a weird selling point.
 

CrazyRob

[H]ard|Gawd
Joined
Aug 26, 2004
Messages
1,274
They lost me at angling the water block onto the naked cpu die, let alone all the other good reasons on here that this is a bad idea. I do appreciate that he's trying something different, though. Maybe a water block that replaces the cpu retention hardware itself?
 

AK0tA

Gawd
Joined
Mar 4, 2012
Messages
813
I think I just witnessed a small miracle. Well probably not but I like the idea, would be a novel concept if MB manufactures gave an option to buy a Mobo with an easy clip on solution simular to this. I also have an Hyper 212 EVO that only takes minutes to install however I read so many forums where people need help because they can not comprehend the directions or process many water or air blocks come with. An easier solution would be great but for those who cant figure out the directions they may not be [H] enough and should go back to taking selfies.
 
D

Deleted member 126051

Guest
I mean, if you're going to delid your CPU ALREADY?

Yeah okay. Maybe.

But, as others have said, I'm not altogether confident in the retention mechanism's ability to resist the shearing forces that all the tubing will bring into play.
 

Emission

Supreme [H]ardness
Joined
Dec 6, 2005
Messages
4,306
The speed at which I can attach a cpu cooler has never been a factor in determining which cooler I purchase.

I just mounted a Hyper 212 EVO that easily took 3500 seconds.
That's almost an hour! :ROFLMAO:
 
Joined
Jun 4, 2008
Messages
2,647
Would be nice if some one did make a water block for a delid CPU . Since that seams to be the best way to get your temps down.. this though is not it..
 

lcpiper

[H]F Junkie
Joined
Jul 16, 2008
Messages
10,611
No they are not, the clam cover for the CPU is in no way used to hold the stock HSF on, thats what the holes in the mobo are for.
Also agreed, all the weight and lateral forces for tubing and such are essentially being held in place by a structure that is solely designed to hold the CPU in place and form a back-stop for HSF mounting forces.
 

lcpiper

[H]F Junkie
Joined
Jul 16, 2008
Messages
10,611
Completely NOT glossing over all the good arguments here for why this, at a minimum, needs to be tested for not stressing the clam cover... I'm wondering why is the mention of install time an issue? From 3-4 minutes at most for a regular water block down to 12 seconds... and? You typically install a water block 1 maybe 2 times over the course of your CPU's life. If this was a daily installation, ok, that makes sense. Eh, maybe I'm missing something but it seems a weird selling point.
Agreed, simplicity and ease of mounting yes, time not so much. I hate those Intel push-pin type mounting mechanisms, I can never get all four to mount proper. I'll take nuts and bolts and springs and thumbscrews any time over the push-pin mounts.

Take this and blend it with an All-in-One cooler pump assembly and I think you have a winner. Just replace that Lucite (or whatever it is) top block with a pump assembly, use the clamp to mount that so it's held in place, then slip a ring over the top that has arms that extend out over the mount points with holes for mounting hardware and you are there.
 

{NG}Fidel

Supreme [H]ardness
Joined
Jan 17, 2005
Messages
6,286
The speed at which I can attach a cpu cooler has never been a factor in determining which cooler I purchase.

I just mounted a Hyper 212 EVO that easily took 3500 seconds.
Exactly, you only install the thing once (not including reapplying thermal paste later). I want the best performance and noise levels. Not the quickest install.
 

Ur_Mom

I'm Not Serious
Joined
May 15, 2006
Messages
20,223
How does it perform and how reliable is it? I don't care about the time it takes to install. I want the job done right, not right now. Of course, most of us also take the time to route our cables nicely and make it look good and have good airflow, etc..

Time to install is not a selling point for me. It's neat, but it's not the feature that I'm interested in when buying a new cooler.
 

ecmaster76

[H]ard|Gawd
Joined
Feb 6, 2007
Messages
1,150
Heh, remember the T-Bird days and all the aftermarket shims? Those were the days :D
Maybe they should make and sell shims for the delidded chips as well. That would help. Kyle's point about the weight on the mechanism is still a problem.

I'm not really sure who the market would be. Competitve overclockers swapping out CPUs on a horizontal bench until they find the golden sample?
Anyone planning on keeping a CPU more than a few minutes shouldn't mind spending the time on putting something sturdy in.
 

rgMekanic

[H]ard|News
Joined
May 13, 2013
Messages
4,546
Maybe they should make and sell shims for the delidded chips as well. That would help. Kyle's point about the weight on the mechanism is still a problem.

I'm not really sure who the market would be. Competitve overclockers swapping out CPUs on a horizontal bench until they find the golden sample?
Anyone planning on keeping a CPU more than a few minutes shouldn't mind spending the time on putting something sturdy in.
Anyone hunting a golden chip wouldn't delid it
 

pendragon1

Fully [H]
Joined
Oct 7, 2000
Messages
18,655
what if he just added an x-bracket with push pins like the oem heatsink? that might support it well enough. attach the block pop on the x-bracket. still under 20 seconds :rolleyes:
 

lcpiper

[H]F Junkie
Joined
Jul 16, 2008
Messages
10,611
Maybe they should make and sell shims for the delidded chips as well. That would help. Kyle's point about the weight on the mechanism is still a problem.

I'm not really sure who the market would be. Competitve overclockers swapping out CPUs on a horizontal bench until they find the golden sample?
Anyone planning on keeping a CPU more than a few minutes shouldn't mind spending the time on putting something sturdy in.
Well, in this case you wouldn't need a shim cause although the CPU is delidded the base of the waterblock is essentially a replicated IHS underneath.
 

FrgMstr

Just Plain Mean
Staff member
Joined
May 18, 1997
Messages
49,897
Not disagreeing, but the base of the waterblock is essentially a replacement IHS "lid". You can see this clearly for a brief moment when he is removing the block at about 2:30 in the video.
Yeah, held onto that mating surface by a metal spring loaded flap designed to hold the CPU in place till the cooler is in place. Not that little metal flap is holding in the CPU as well as the weight of all that copper, the hose, and the water, not to mention any other force applied to the tubing. That design will never see the inside of my case.
 

pfc_m_drake

[H]ard|Gawd
Joined
Jan 7, 2004
Messages
1,354
Heh, remember the T-Bird days and all the aftermarket shims? Those were the days :D
I personally cracked TWO of those Athlon T-Bird cores during different builds.
Nothing is more disheartening than the whiff of ozone and seeing the magic smoke come out :(
The kicker is that I was actually using a shim the second time.
 

Ranulfo

[H]ard|Gawd
Joined
Feb 9, 2006
Messages
1,895
The speed at which I can attach a cpu cooler has never been a factor in determining which cooler I purchase.

I just mounted a Hyper 212 EVO that easily took 3500 seconds.
That top mounting bracket threw me for a loop the first time I installed a 212 evo.

Heh, remember the T-Bird days and all the aftermarket shims? Those were the days :D
Those were the days that caused me to stick with stock cooling for 3-4 years after my T-bird system.
 

lcpiper

[H]F Junkie
Joined
Jul 16, 2008
Messages
10,611
The way that the block goes inside of the CPU hold down bracket, doesn't it take the place of the lid with the same offset and depth of the lid. He basically replaced the cpu lid with a copper water block. To me it looks like the block is supported by both the top and bottom of the cpu hold down bracket.

Am I looking at that correct?
Yes, you are looking at it correctly. The problem is, the entire socket and cpu retention bracket is not designed for the weight it would have to bare, more over, the weight is going to have additional stress on it from the tubing which tends to pull in one direction or another relative to the socket.

With a traditional CPU mount for a waterblock, the mounting method takes all of this weight and stress and controls the weight that is placed on the CPU socket by directing the weight evenly against the IHS, (lid).

But in this case, you remove the lid and replace it with a new lid that is several times heavier and can have water tubing pulling on it in one direction or another. It's simply more stress than the CPU package and the socket are designed to accept.

It looks good, and it seems reasonable, but it's a flawed idea because it violates the basic engineering requirements and tolerances of the CPU sockets. If he adds in traditional mounting bracket support it's probably a go, but it's going to fail without it and the people who buy into it will pay with damaged equipment.

Many years ago I received a test sample of a water block, as far as I know, it was the first design to use a single inlet hose barb with two outlet hose barbs. The concept was to lower the outlet pressure increasing the inlet force against the baseplate allowing the coolant to push in a little closer to the baseplate and cool a little better. It was a good concept, engineering wise, it was sound, and other manufacturers copied the duel outlet design successfully, but this sample I received failed. It failed because the inner chamber of the waterblock was too deep, the water was moving through it fast enough but it was only flowing through the upper part of the block and so it wasn't really moving the water down near the baseplate. To make their initial design function they had to use something like a pipe to force the inlet down closer to the baseplate and later they redesigned the block so that the water chamber was much shorter and the inlets and outlets much closer to the baseplate so it would actually move that water around down there.

This design is like that older waterblock design, it has a nice new idea but it is being executed poorly. They are forgetting that the CPU socket just isn't made for this.
 
Top