• 21.07.2025, 23:57
  • S’inscrire
  • Connexion

  • Vous n’êtes pas connecté.

 

Kryos NEXT

Sahnetorte

Full Member

mardi 12 juin 2018, 17:34

Mhhh. i don't think this is marketing.

Kaine

Junior Member

mercredi 13 juin 2018, 02:00

Citation de "Sahnetorte"

I understand what the pictures are showing. But what does this have to do with the venturi-effect?
I am asking for more optimizations to come. Now that vario technology is here, what is left to do?

Most are just trying to reduce the size of the fins while preserving a maximum flow, others continue to innovate like Alphacool with the Eisblock (which I see in many tests to beat the kryos next).

I'm talking about Venturi because it's another possibility I believe.


EDIT: No, no marketing, I just mention it because it's the finest fin technology. I am not sure whether the results are good (although some tests suggest that) but as everyone is going in this direction, it is interesting to analyze the results.

Maximum friction (hpdc) or maximum radiator / waterblock cycle (lpdc) were the most common questions some time ago. But we can now see block is between the two. Small optimizations do not make enough differences anymore, new ideas are needed.

Ce message a été modifié 3 fois. Dernière modification effectuée par "Kaine" (13 juin 2018, 02:20)

Kaine

Junior Member

mercredi 13 juin 2018, 03:22

This seems to me to be very effective for removing heat, maximum friction AND minimal flow loss :







Give me your opinion because this thought prevented me from sleeping several nights !!!

Ce message a été modifié 12 fois. Dernière modification effectuée par "Kaine" (13 juin 2018, 04:19)

Chris0706

God

mercredi 13 juin 2018, 10:09

My last thermodynamics lecture was a while ago, but I don't think the speed is the problem but the volume flow is (which dependency the venturi effect describes). You don't need speed you need volume for higher heat transport, I think. The even finer fins are used for generation of larger surfaces, so that more water can "access" the heat and absorb it. I doubt a much a higher velocity (at the same volume flow), will increase this much as water can absorb a huge amount of energy.

Kaine

Junior Member

mercredi 13 juin 2018, 23:37

For me the speed and the flow are the same thing, you meant flow and pressure?

My idea of the venturi effect used directly in the watercoling circuit is to be able to keep both. Keep the flow and use the pressure generated to reduce the boundary layer (the layer of water that forms at the bottom of the maze) to better "rub" the surface to cool.

The water already present on the maze brakes the jet of cold water trying to enter. For years now manufacturers have been bringing the injection nozzle of the maze closer to limit this.

Ce message a été modifié 1 fois. Dernière modification effectuée par "Kaine" (13 juin 2018, 23:42)

Sahnetorte

Full Member

jeudi 14 juin 2018, 17:18

can you draw a simple picture with ms paint to show what geometry you would change?

If you want to speed up and increase the velocity of your fluid at some point of your circuit (which would be the kyros next in this case) you will always loose static pressure. Otherwise this would be a perpetuum mobile. And this loss of static pressure means that the flow resistance of the whole circuit goes up and the flow rate goes down. With the same power of the pump.

You want to change the geometry of the jet (the part of the cooler, in which the velocity is raised, just before it reaches the fins)?

Like a bypass connected with the inlet of the cooler, so extra water ist sucked in because of the jet?

Chris0706

God

vendredi 15 juin 2018, 11:13

Citation de "Kaine"

For me the speed and the flow are the same thing, you meant flow and pressure?

My idea of the venturi effect used directly in the watercoling circuit is to be able to keep both. Keep the flow and use the pressure generated to reduce the boundary layer (the layer of water that forms at the bottom of the maze) to better "rub" the surface to cool.


Sorry, that's non-sense, I would highly suggest going deeper into the materia. Volume flow and Speed are not the same! Volume = Speed * Cross-Section! And a higher pressure doesnt do anything in a closed loop, what you mean is more likely the flow.. there are mainly two flow sorts, laminar flow, which is kinda staight forward flow and turbulent flow, which has a low of turbulences. In general science says a more turbulent flow can achieve better results in heat transport.

Kaine

Junior Member

vendredi 15 juin 2018, 13:28

I have to get my daughter to school but as soon as I have 5 minutes I answer and make diagrams. Regarding the thermal capacities of a turbulent flow, it is possible, but I would leave more on a flow controlled flow so that the volume of water injected be totally extracted at each cycle. In my opinion this is not the case when the injection nozzle is placed in the middle (because of the placement of the outlet nozzle that succes stronger on one side). it remains on one side of the waterblock a kind of dead zone where the liquid is evacuated more slowly.

Kaine

Junior Member

vendredi 15 juin 2018, 18:03

This is just a suggestion, what do you think of that?





90% of the cooling base can be used by the fins, the outlet flow being removed. The water present in the block is forced to follow its "tunnel" and is therefore pushed by cold water arriving, but also sucked as in a straw (no residual hot water, no air bubbles):



We can surely optimize the flow by installing flow guides installed upstream, where the diameter has not yet changed (optimized circulation for the outer sides):



I think we can preserve the effect:



Ce message a été modifié 18 fois. Dernière modification effectuée par "Kaine" (15 juin 2018, 22:11)

Kaine

Junior Member

vendredi 15 juin 2018, 22:36

Another possibility to keep the vario technology, same system but output divided by two (the diameter of the input):







The two outputs meet (y) then in an output of the same diameter as the input.

Ce message a été modifié 6 fois. Dernière modification effectuée par "Kaine" (17 juin 2018, 01:36)

Kaine

Junior Member

dimanche 17 juin 2018, 01:15

What do you think about this ?

Can someone from Aquacomputer comment, to know if it can be good?

sebastian

Administrator

dimanche 17 juin 2018, 08:35

Try it out or simulate it with a FEM tool. Then make a tooling an check/benchmark your results with your prototype and available coolers.
This topic make no sense.
When you think you have a better design - build and sell you cooler.

Kaine

Junior Member

dimanche 17 juin 2018, 11:18

I do not want to do it, otherwise I would have developed the idea in my corner and then filed a patent.
The goal was more to look for what could be improved, new ideas for the upcoming waterblock (I am not here to make a profit, only for pleasure).

I would like to have your opinion on the thing, as it is without testing (I'm going to send it to other builders as well to see what they think of that).

Ce message a été modifié 20 fois. Dernière modification effectuée par "Kaine" (17 juin 2018, 16:32)

gr3if

God

lundi 18 juin 2018, 23:00

That design looks like it would collide with the mainboard big time :)

But the idea looks interesting.

Chris0706

God

mardi 19 juin 2018, 07:27

I find it interesting, but its a very simplistic design. If that would be "the" thing a company like AC would already have built it. BUT the concept may have a chance, if the heat distribution on the IHS is quite balanced.

I think the state of the art design aims to focus on the center of the IHS, because thats normally where the DIE sits and produces the most heat. When that changes, so heat is distributed or generated more equally on the IHS, a larger cooling surface at the whole area, like you drew it, could become beneficital.

BUT keep in mind, with the fin and flow design you propose there is a HIGH risk of clogging and also it has a large flow resistance, you will need a very powerful pump.

In the end it depends, what the overall aim of your water cooling system would be, performance or silence. Most people will go in the middle and dont want loud pump.

Taubenhaucher

God

dimanche 24 juin 2018, 09:47

I have followed the discussion with interest, but I have a question that I cannot answer sensibly.
All water blocks of the last years move within a few degrees Celsius, the ice block XPC is 1° C better than the next best cooler, i.e. in the measuring error range.
This may also be the reason why competition is more likely to be based on other properties than on cooling capacity.
The question I ask myself is: Why haven't any of the engineers come up with this very simple and, if the data of 10°C temperature improvement would be correct, very efficient solution in the last 5 - 10 years?
Es gibt keinen Ausweg, den ein Mensch nicht beschreitet, um die tatsächliche Arbeit des Denkens zu vermeiden.
Thomas Alva Edison (1847-1931), amerik. Erfinder

Kaine

Junior Member

dimanche 24 juin 2018, 12:49

We do not know what kind of winnings we expect with a waterblock like this one.

I speculate to push people to think...

Reflection that seems to have virtually disappeared regarding the realization of new cooling engine. It must have been a good 10 years since EK innovated with a slot system and since everyone copied with minor improvements.

To copy is simple, to innovate, to create what the others will copy later, this is not at the reach of everyone.

Me I think we must completely see everything again, the waterblocks do not transfer enough heat to the coolant.

I am sure that many discoveries are to be made ....

The man manages to break atoms in two to draw an extraordinary source of energy, and we are satisfied with no real and really significant improvement in 10 years to cool a surface with a piece of metal and a little water ?

Ce message a été modifié 6 fois. Dernière modification effectuée par "Kaine" (24 juin 2018, 13:03)

gr3if

God

dimanche 24 juin 2018, 13:20

I mean it's not that one didn't tried, Aq's last cooler even tried to make up for the shape of the IHS. But in the end I think we've seen nearly anything.

Kaine

Junior Member

dimanche 24 juin 2018, 14:35

I'm talking about a lot, a lot more change.

For example a waterblock 100% active in the heat transfer, not only the bottom where is the maze. It is already necessary to optimize the intake of heat of the COMPLETE block empty (without water) to have big change.

Understand that if the base and the top of the waterblock are not joined by the thermal paste, it is not optimal. We need to review everything from A to Z to have a big change.

I want to clarify that this thread is followed by several manufacturers.

Ce message a été modifié 1 fois. Dernière modification effectuée par "Kaine" (24 juin 2018, 14:59)

gr3if

God

dimanche 24 juin 2018, 15:54

And why? CPUs suffer the most from using an ihs with tim or even soldered.

Yeah you can delid them, but this is the best point to attack

In terms of cooling wise we have not problem even cooling the biggest 250w tdp monster CPUs with the current products.

Discussions similaires