Check this homemade WB out

[JoeBar]

If i wanted a fan on my wb i would have stayed with aircooling...

[UrbanSmooth]

Very interesting concept, Nateman. Keep up the good work!

I have been following this thread (and Skinnee's GTX 480 block roundup thread) while in the peaceful embrace of vacation, and, I must say that this thread has kept me on the edge of my seat!

I would love to see more testing with a few other blocks and your very own with some consistent results. Also, why not send a few of your prototypes off to Skinnee, Vapor, and other testers here and on other forums that are willing to help you test?

Someone brought up a thought that the fin area would only be comparable to a 0.5x120mm radiator. Why not combine the idea of a full-sized air cooler (TRUE/Megahalems, etc) with a waterblock on the bottom? Just throwing that one out there, sir!

Stay strong, keep up the good work, and continue to improve this very interesting and POSSIBLY ground-breaking waterblock!

[Church]

UrbanSmooth: Big full-sized air coolers manage to drive heat far enough from cooled component by using smart gimmickry, heatpipes and stuff. Trying to make such from scratch or trying to solder ones from existing air cooler to this waterblock base should be very hard to do. Adding another 1x120 rad to loop is by two orders simplier. It's also simplier to plan out airflow if you have to take into account only rads, but not yet another thing you have to route air to. Let me repeat - increasing cooling by adding rads is 100x easier and what's LC about. While it's possible to make such hybrid solutions and with major PITA make air-cooler part be more effective, and while this is unusual idea and, at the end it's just that, plaything & waste of time on less effective subpar solutions.

[TJ TRICHEESE]

so when is the block going off to skinnee or vapour

[UrbanSmooth]

UrbanSmooth: Big full-sized air coolers manage to drive heat far enough from cooled component by using smart gimmickry, heatpipes and stuff. Trying to make such from scratch or trying to solder ones from existing air cooler to this waterblock base should be very hard to do. Adding another 1x120 rad to loop is by two orders simplier. It's also simplier to plan out airflow if you have to take into account only rads, but not yet another thing you have to route air to. Let me repeat - increasing cooling by adding rads is 100x easier and what's LC about. While it's possible to make such hybrid solutions and with major PITA make air-cooler part be more effective, and while this is unusual idea and, at the end it's just that, plaything & waste of time on less effective subpar solutions.

Cool response, bro. Let's see what Nateman thinks about this.

I'm thinking more of a hybrid fin/waterblock design, with the inlet/outlet holes going down the center, surrounded by the fins, if possible. Sure, it wouldn't be as big as a full-sized air-cooler, but, perhaps something bigger than his current setup (more fins).

[Charles Wirth]

Invalid data, looking at the thread on evga it seems ambiant temps are altered between the two setups to heavily favor his block is my diagnosis.

In July I will show you my new Intel chip (12 thread) that can OC to 4ghz on passive sink.

[Blacky]

Invalid data, looking at the thread on evga it seems ambiant temps are altered between the two setups to heavily favor his block is my diagnosis.

In July I will show you my new Intel chip (12 thread) that can OC to 4ghz on passive sink.

Now I look at mine and this makes me feel bad

[NaeKuh]

Now I look at mine and this makes me feel bad

its a sub 60W gulfy...

6c/12t.

Unless charles your gonna preview a new 12 core.

[shazza]

Hey guys - let nateman handle discussions with skinnee. This isn't a commercial product and isn't ready for prime time, so probably not something high on skinnee lab's agenda at this time.

@ Fugger - nateman_doo addressed the fact that he now sees something's wrong with his data and said earlier he shouln't have gone public with it yet. And - to be fair, he's not the one that posted it here.

[nateman_doo]

If i wanted a fan on my wb i would have stayed with aircooling...

different strokes for different folks

UrbanSmooth: Big full-sized air coolers manage to drive heat far enough from cooled component by using smart gimmickry, heatpipes and stuff. Trying to make such from scratch or trying to solder ones from existing air cooler to this waterblock base should be very hard to do. Adding another 1x120 rad to loop is by two orders simplier. It's also simplier to plan out airflow if you have to take into account only rads, but not yet another thing you have to route air to. Let me repeat - increasing cooling by adding rads is 100x easier and what's LC about. While it's possible to make such hybrid solutions and with major PITA make air-cooler part be more effective, and while this is unusual idea and, at the end it's just that, plaything & waste of time on less effective subpar solutions.

Wow.
...A subpar waste of time plaything...

I have no response for you. I really just don't have one for you.
maybe later I will, but now im not even going to bother.

[zeroibis]

For the greater good of the forum...

Naekuh has commited Seppuku...

Nooooooooooooooooooooo!!!!!!!!!!!!!111111111111111 11111

How about we get a thread kappa to monitor things and keep the wc in check?

[UrbanSmooth]

Hey, Nate, did you see my post?

[Church]

nateman_doo: BUT! You can use that 'unused' space above cpu waterblock for mounting some rad there aswell. Even if not 1x120, but there are smaller ones too, eg. HWLabs' single or dual 80 or 92 fan rads. And even those will have more fin area then your conceptual waterblock. And they are leaps and bounds easier to add to existing loop then making fins from piece of copper for 99.99% of average LC joes out there. Buy rad with two fittings, connect to loop, put fan on it. The End. Now with going your route it's a very tiny little bit more difficult, getting fitting for the job big piece of copper, quality tools, skills to use them .. and all those fins are extra/bonus, what sense in making good air cooled part, if main liquid cooling function of block performs worse then top blocks out there? Then all you get - hybrid block that performs better then stock water blocks only when they are in artificially handicapped situation (read - inadequate rad area/and little heat to dissipate at stock clock, resulting even little help from those aircooling fins helping seemingly so much).

[randomhro]

Nate - It would be nice for the water to not boil when a pump went out but I'd like to see some screen shots where the ambient temp is the same and where the blocks aren't rad limited to see what best case would be for both blocks in your testing. I can see why you wanted to test with such little rad surface area, it's what your block is optimized for. I would like to see similar testing where neither block has an optimized setup (i.e. not rad limited) for the max performance of the two blocks side by side. If I was going to buy something like this I wouldn't buy it based on worst case scenario since I don't care about worst case as long as it isn't going to mess my system up.

But that's just me...

Props on being creative with skillz I don't possess

[Vapor]

Hopefully this will calm the thread down some (though both FUGGER and I are guilty of adding fuel to the fire ): things are in motion for me to formally test this block

It'll go up against the HK3.0 LT, cuplex kryos XT, Supreme HF P1, Apogee GT (yes, the old one) and the results will be comparable to all the later results from my Roundup #3. I plan to test it with no fan, fan at half speed, and fan at full speed, and will be able to quantify just how much the block's radiator is dissipating. I'll also, hopefully, get some runs done on the TRUE Copper, if it can tame this chip at these settings.

[shazza]

Good to hear, Vapor.


I've deleted a couple more posts - we'll be looking at infraction points if there's any more generalization about other forums (God, I sound like my Mother - wait, I think I am my Mother...)

Keep your debate constructive - better yet, everybody can calm down and let's see what the next series of tests show.

[ridney]

good job nateman

[ohms]

Glad to see that things are calming down in this thread. I'm happy Auntie Shaz stepped in but it shouldn't have come to that. For the most part most of us here are adults and young adults and should know better then to act in this manner.

I'm looking forward to the results from Vapor and to see how they match up against the listed CPU blocks. I think Nate's design is innovative and I was always wondering why companies have never thought about adding fins to the blocks to help with heat dissipation before. I don't know much about deltas, and pressure drops and such but I can't see having fins and active air cooling on a CPU block being bad.

[Martinm210]

FWIW, I tried making a couple of custom waterblock with larger than your average cooling fin area. It seems to be a common thinking from a lot of folks that go from air cooling to water cooling that huge surface areas within the waterblock are key to improving current designs. Unfortunately it just doesn't work that way because of the huge difference between water and air thermally. First there is the 30X or so difference in thermal conductivity or ability to remove heat. Then there is the volumetric heat capacity differences which is on opposite ends of the spectrum.

It's these two water vs. air property differences that pretty much makes anything but the base of the waterblock completely irrelevant from a CPU>Water heat exchange.

I've tried making different tops for old waterblocks and have also tried making air cooler like waterblocks. At least for my failed ideas, nothing seemed to help outside of nozzles and fine microchannel/pin type matrixes right at the core heat exchange area. Thicker bases were often worse, and copper tops didn't do anything for thermal performance.

The only advantage I can see in an air cooled waterblock is the actual benefit of heatexchange on the coolant (Water>air delta). This same or greater benefit could be achived by an additional radiator of the same or greater size. Anyhow, I don't discount test results...never do until I've tested something for myself, I just see it's hard to compare since there is a difference in water>air heat exchange from the two testing scenarios. This is not all that different than testing one waterblock with a 220 rad, and another with a 320 sized radiator just that one is integrated into the block.

Going to be tough to make any sort of comparison here...

[shabby]

Interesting idea, have you thought about ditching the fins and having a flat surface where you can attach any heatsink on top of it?

[nateman_doo]

Hey, Nate, did you see my post?

Which one?

[Sideroxylon]

Martin!!!! It's about time you chimed in... Glad you did and I'm happy to see this block being sent off to some reputable testers.

I'm pretty sure what the outcome will be but I'll wait to see the official results. If, for some reason, this turns out to be a valid design then imagine the gain you would see by stacking a wb onto a wb in separate loops.

[nateman_doo]

What do you think the outcome will be?

[crazydiamond]

it would be interesting to see if you could make basically the same block except have no fins & no fan. then compare it to the one with the fins & fan to see if it can improve the block. comparing 2 different waterblocks (this & hk) will tell you what one of the 2 is better but it wont tell you how much of a gain you get from adding the fins & fan. get what im sayin? like the hk might do better because it has a flatter base & the microchannels etc. but that doesnt mean the fins & fan didnt improve the cooling capability of the block.

[Sideroxylon]

What do you think the outcome will be?

I think the outcome will be that we'll all be glad to see some independent test results done.