CPU-360 Midplate Testing

[Alexandr0s]

I was just using it as an example, nothing more. My point is that all we see are the same old designs being rehashed. It's like an Iphone, sure they are cool, but there really has been any "game changing" design to the Iphone since the second model, mostly because it is a pretty damn good phone and it just works. Same thing with CPU blocks, nothing new with the current designs. Until someone decides to do something radical CPU blocks for the most part will perform within a few degrees of each other.

We all want to see radical new blocks that chop of a few degrees (well, basically we just want to see the degrees getting chopped), however, since we're working with ambient cooling, it wont be a linear curve. At some point the progression ends. Maybe with being able to cool a i7 980x with a MCR 120 or something like that, but it will probably stop.

And Naekuh, leave the TEC's out of it!!

[NaeKuh]

OT but i've seen you type this a couple of times now and cannot hold back any longer .

http://dictionary.reference.com/browse/stagnant

bad spelling and excessive photo flashes are my trade mark on this forum.

Oh and zipties.

I am the ziptie god of this forum.

[Waterlogged]

Funny thing is, I've been reading that statement since P4 Northwood's and yet there are new CPU blocks coming out all the time. Look at the comparison between the Apogee GT and the Supreme HF, we have over 5C separation in performance... is that what we consider a few degrees? To me, 5C is quite a margin when you consider the medium we're using to cool our gear.

Here's an interesting question, How much of the temp (when measured in their respective era) difference for blocks like the original Apogee/GT to blocks from today (EK HF/HK3.0) is from the different TDP of the chips they're sitting on ? Chips have gotten a lot hotter over the years and we're still looking at relatively minor gains in temps.

[Piggy1918]

Just addressing the concept of a .5c decrease not making a difference, I think a few factors come into play.

WC'ing is an interesting medium because of how many choices and parts are involved. The block is the most important, but you need a good pump to get decent flow. And you need a strong radiator to handle the heat load. And you need good fans to cool the rad. If you use barbs with a large ID you can increase flow a tad.

All of these factors add up. .5, .3, .8, .7, .9c changes start to really add up. This is why it is so important to heavily consider the best of each component, because in the end, 4-5c total decrease can really help.

Eh, that's my view on it. So if I can eek out a little better from each component, I'm happy.

[Vapor]

Centipede and Millipede were filtered out after 1 mount. Looking like Niner will be too.

[eth0s]

Going back to the .5C decrease issue, I think we may have reached a point of diminishing returns as we approach the limits of water cooling. However, as we do approach that limit (which we will probably never actually reach), gains in performance will be smaller and smaller, and harder and harder to come by. So today, a .5C decrease might actually be something that should be celebrated, not shrugged off as "so what", or "it's not worth my time". This may be as good as it gets right now. And eventually we may have to celebrate .1C gains. Just because the numbers are not huge in terns of absolute values, does not mean that the gains are not significant or important. Now, I would love to see some kind of "game changer" as much as anybody else, but given that the laws of physics in our current universe cannot be repealed or changed, I don't know if that is possible. And using Titanium is not the answer (thermal conductivity is far worse than copper, or even brass. Although, solid Silver blocks may be an answer, but they would be prohibitively expensive). But that is not to say, that there is not some fabulous new technology out there waiting to be discovered that uses our current physics in some new and brilliant way. Unfortunately evap rads are not it.

And finally, Spel check? Who uses that? People who want to ruin spontaneity, that's who. And what's wrong with zip ties? Are they out of fashion now? I can't keep up with you trendy guys.

[Alexandr0s]

But that is not to say, that there is not some fabulous new technology out there waiting to be discovered that uses our current physics in some new and brilliant way. Unfortunately evap rads are not it.

If only graphene would work as a base for our waterblocks .

[NaeKuh]

Going back to the .5C decrease issue, I think we may have reached a point of diminishing returns as we approach the limits of water cooling.

actually no.

The only thing we are doing is increasing bulk efficiency on pickup side of the system.
Reason why we see such a big improvement is because we are working with more heat,
and also our eq and testing is a lot more accurate then it was 5yrs ago per say.

The dissipation side is something we have had for more then 100 yrs.

The biggest and noticiable improvements we can have is:

1. Evaporation = lets face it... its the safest way to get sub ambient, however bong coolers have there own sets of rules and problems. Biological being a big problem.

2. Compression = Sorry u cant compress water so that's completely out of the question.

In short, the biggest move is to reduce ambient, or reduce coolant temps.
The only way possible without putting work into the system is using evap.

[Alexandr0s]

The only way possible without putting work into the system is using evap.

Or TECs ?

[NaeKuh]

Or TECs ?

no TEC is putting work into the system.

Compression is also work into the system.

Putting your outlet @ a spray nozzle and having a fan blow at the droplets is NOT putting work into the system.

[Alexandr0s]

no TEC is putting work into the system.

Compression is also work into the system.

Putting your outlet @ a spray nozzle and having a fan blow at the droplets is NOT putting work into the system.

Aah, misread it . I'd love to try a bong cooler some time, though I'd definitely need a good way to get rid of humidity.

[Vapor]

Water isn't the issue, copper is.

In the grand scheme of things, copper isn't that good of a material for cooling. It's the natural choice based on availability, ability to be manipulated (machined), etc., but because of its thermal resistance it's really only effective within ~5mm of the thermal source (with the performance we're after at the thermal density/load we're at currently). With our current machining, we're approaching a limit of how much surface area/performance we can extract from that really small area. We're not there yet (likely years away, but it is within view), and in my opinion, we're not nearly approaching a limit with watercooling--just the limit of copper with our current ability to manipulate it.

There are better materials out there, not just silver and diamond and other forms of carbon, but non-carbon composites have some particularly interesting thermal properties that we may see start trickling into the market. Innovation and improvements will keep coming.

In the meantime, we keep pushing with what we have on the table.

[NaeKuh]

There are better materials out there, not just silver and diamond and other forms of carbon, but non-carbon composites have some particularly interesting thermal properties that we may see start trickling into the market. Innovation and improvements will keep coming.

In the meantime, we keep pushing with what we have on the table.

problem with carbon and carbon composites is that they are very hydrophobic.

LOL... im sure you remember Tony's nightmare on his graphite block.

Diamonds is just way too expensive.. even synthetic diamonds would cost more then 10x your system.

Then i heard people saying 1 way carbon nano tubes.
Which can transfer heat in 1 direction only... well once again.. 90% of carbon composites are hydrophobic.

Also vapor you still have the problem with the IHS being nickle / Copper, so you cant truely get rid of the copper.

[Vapor]

Updated with Niner. Just installed Dots, looks even worse, probably the worst so far.



Hard to be good when you only use roughly half the available channels (Niner and Dots are guilty of this).

EDIT: updated with Dots. Running Diamond now. Diamond is another dog....all these poor performing plates are really expediting the testing process

[Utnorris]

Ok, so since so many are missing my point on this, what I am talking about is efficiency, in other words, getting more out of water cooling with less, say rads. We can all agree that with water cooling you will be limited by your ambient temps, however getting close to ambient temps currently requires large rads and fans. A game changer to me would be a block that performed as good as the best current block, i.e. the Koolance 360, but did it with less rads, less flow, less block, whatever. My point is that there has to be a way to make blocks more efficient either through a different material, different design (i.e. dual inputs, single out) or something that allows them to transfer more heat to the water to be dispersed by the rads. Maybe it will require a different type of rad, I don't know, I am not engineer, never claimed to be one, but it seems we are stuck on the same ideas that have been kicked around for some time now.

As far as celebrating a .5c or even 1c degree difference, I never said we shouldn't be happy to see this or even cheer about it. What I said or what I meant was that I don't buy a block because it gets 1c better temps under load, but rather the way they look, mount or are something else that adds value beyond the performance.

With that being said, I applaud and appreciate efforts by Vapor, Martin, Skinnee and anyone else that takes the time to review and/or improve current and future blocks. This takes a lot of time and dedication, not to mention precision to get the results and reviews up for other folks to read and all this is done for the most part without compensation of any kind except for a pat on the back.

I am not attacking any efforts made by anyone willing to take the time to try and improve what we have, but rather the industry and the single minded thought process that is behind the current water cooling industry. I want to see companies take a chance and think outside the box for once. Stop being afraid to fail and learn that failure is actually success. I understand that this is a niche product and that most of the companies depend on thin profits to survive, but take a chance, many did when this hobby first started and while many did fail, a lot succeeded and have reaped the rewards.

[NaeKuh]

well norris as vapor stated, we cant do much with the materials we currently have.

We would also need to find a new revolutionary material which had better thermal properties then copper yet not cost the same price as a new honda civic.

The only way to get the improvement your asking for is to tweek the coolant medium then.
But coolant which has better thermal properties then water are also very poisonous. So your trying to kiss the devil.

Then the only solution left is to put work into the system to bring coolant down.
But that comes at the cost of power. Unless you want to run a bong chiller.

So your asking vendors to find a new way to tweak either coolant, which can be dangerous, or introduce a product which can put work into the system.
This is why ive been asking for a PWM controller for my TEC units, because i wanted to see exactly how efficient TEC's would be on PWMs.

[Utnorris]

That may be true Naekuh, but I also remember when they use to say we couldn't get more out of cpu's without adding more cores, but along came Intel with hypertheading, first time was a flop and the second time was the charm, and now we have the same dual/quad/hex cores doing almost 75% more work. And yes, a 775 and a 1366 chip are not the same, but the analogy still works. I might be asking too much, but if you don't ask for it, no one is ever going to give it to you.

[NaeKuh]

but if you don't ask for it, no one is ever going to give it to you.

LOL... and you know how much i try.

Remember my whole rant on we want silver? :P

[Vapor]

Mass-specific heat of water: 4.186 kJ/kg K
Mass-specific heat of air (20C, 50% RH, sea level): 1.005 kJ/kg K
_______________________________________________

Density of water: 998 kg/m^3
Density of air (same conditions): 1.2 kg/m^3
1 m^3/s = 15852 GPM
1 m^3/s = 2119 CFM
1 kg/s of water = 15852/998 = 15.86 GPM
1 kg/s of air = 2119/1.2 = 1766 CFM
_______________________________________________

Water has: (1/15.86) * (4.186) * 1 (K) = 264W heat capacity @ 1GPM per 1 degree K
Air has: (1/1766) * (1.005) * 1 (K) = .569W heat capacity @ 1CFM per 1 degree K

Said another way, you need ~464CFM of airflow to move the same amount of heat as 1GPM of water (265W @ 1 degree delta).

A perfect single 120mm rad with 0 airflow restriction (impossible) and an 80CFM (max flow rating) fan can only remove 265W at a 5.8C delta, that's the best possible scenario before physics kicks you right back in face--to get near-0 water-to-air deltas, a perfectly efficient radiator still requires way more airflow than our fans can do at a reasonable sound pressure levels. Because of this limit, we increase airflow by going parallel, i.e., multi-fan rads.

Based on skinnee's rad testing, a single Ultra Kaze 2000RPM fan (roughly 80CFM in open air) can maintain an air-to-water delta of ~12-13C at 277W. That's roughly 45-50% efficiency relative to perfect.

Radiators have two counteracting figures: air flow and air saturation. Increasing FPI and/or thickness increases saturation but decreases airflow (via restriction). We've had a few rads now that get really high air saturation numbers (high-90s in percentages), but they're restrictive to airflow. Surely rad manufacturers are working on lowering airflow restriction while maintaining saturation. But even if they get a perfect radiator, we'll only have radiators that are 2-2.5x as good as they are now, which isn't that impressive (IMO) considering that performance is already attainable by just increasing rad count or fan RPM

IMO, air-to-water deltas are impressive as is. Sure we can get more at similar noise levels with better fans and some radiator improvements--but I don't think there's much headroom there.

I still think the block is where it's at. You double block performance and you go from 76C CPU temps to 53C (at 30C water temps, 22C air temps), you double radiator performance and you go from 76C (same conditions) to 72C.

[Vapor]

Updated with Diamond plate. Starting tests on HKesque. Isn't going to make a second mount either....guess all hope is left with Jagged, Double, and Single.

[skinnee]

Double down on single.

[Vapor]

Hmmmm, should be noted I don't have a Bei Fei Single. I have a homemade, dremel'd Single. I'll probably forgo testing that (since it's not a repeatable plate for reproduction).

Maybe I'll take Double and block off one of the channels to emulate single.

I also have taken Centipede and Millipede and used 3M plastic tape to block off the face of the legs facing the base, leaving (essentially) Single with some structure for turbulence at the inlet of the final injection.

EDIT: pics of what I'm talking about: 1, 2

[Bei Fei]

Let me know what you need and I will make!

[Martinm210]

Votig for stock with a radius..
Great work!!

[Vapor]

Thanks!

Updated with HKesque. Going to install Double now.