Too tecnical for my english too, but let my try... (I'm italian, bla bla bla..., you already know itOriginally Posted by stoner
)
Ybris ACS
The history of ACS begins in the first half of 2005, during the phase of development of another waterblock, the K11 Leviathan.
In that period the most of WB builders were developing the fins technologies but, at the same time, we began to also look for possible alternatives, well knowing that very soon the limits would be reached and that it was not possible to subsequently develop some complex fin system, able to improve the performances of the waterblock.
The first prototype of the waterblock had been realized building a rectangular matrix of needles directly elicited through a series of close perforations.
The structure that results had an interesting characteristic: the fund was not flat but it was formed from so many small half-spheres and in the points of union the needles departed working as the cooling element.
***Image 1, 2 & 3***
The practical realization however wasn’t simple: too much time spent in CNC cutter but above all a notable quantity of dribbles to be eliminated after the work.
Here we see the only prototype immediately realized after cleaning with sand.
***Image 4***
The waterblock received the water through a shower diffuser realized in delrin and, beyond the difficulties in the realization, it was very effective during the tests.
***Image 5 & 6***
The principal problem of this structure however was represented by the excessive thinness of the plate in copper of the base.
If the performances were wanted it needed to be near to the core but in this case the simple pressure of fixing of the grapevines was sufficient to deform the plan of contact and to immediately worsen the temperatures of job. Increasing the thickness, the matrix began to lose in performances for which we needed to look for the correct compromise.
The matrix, anyway, had the tendency to assemble too much dissipation to the center; it was fantastic on the processors with naked core but it lost a lot of its characteristics when it needed to cool the new cpu that were arriving in the meantime on the market: the new Dual Core Intel.
So the project was radically modified.
The initial idea (the structure of the fund made with half-spheres), were maintained eliminating the needles in relief and, this time, we studied a different distribution of the water.
It was necessary to get an elevated speed of slide, great strength of impact on the copper but, above all, it needed to extend the effect on a very ample surface, equal at least to that a square from 29 mms of side.
So we thought the idea to realize a double circular matrix of contrasted half-spheres: the core of Ybris A.C.S was born.
***Image 7***
In the image the superior half-spheres have been cut to let you see the underlying part but imagine them completely closed.
***Image 8***
As you can see, departing from any point of the cell, every time that a fluid fillet enters an inferior half-spheres (water goes down) it finds at least three possible streets of exit (water climbs) exploiting the passage created by the bewilderment with the superior half-spheres.
In this way water was forced to do a series of casual rebounds before finding the street of exit.
We built a hypothetical external form: the matrix was closed in a wrap and we started a series of simulations to study the fluid-dynamic behavior inside.
In the image: one of the so many simulations
***Image 9***
That was the moment to try a live the structure: it came so a first prototype and was tested in the laboratory.
***Image 10, 11 & 12***
The resultant effect of this structure was impressive: the turbulence was inclined to a levels that the waterblock start "to blow" but the performances allowed to intend that the way was correct.
Found defects: general courses too limited (therefore necessity of powerful pumps) and a mechanical shape not so good looking at it.
It was the case to slightly reduce the effect "puff" trying to maintain unchanged the performances.
So the matrix was modified again.
In the actual version every cell is composed of 3 half-spheres, every half-spheres is surrounded by 6 spheres (therefore the cells melt him among them because they have some common spheres) and water is sent in a point that breaks the flow in three parts.
***Image 13***
From that moment, we were able to exploit the rebounds among the "domes" and the "cave" and water looked for a casual passage crossing the matrix up to reach the peripheral zone of harvest.
This is what happens in ONE elementary cell. Now expand the same concept to the whole matrix: every cave will receive a part of a direct throw and at least two parts of throw coming from the adjacent caves.
We increase therefore the number of the throws and we pass from the initial idea to unique central collector to a series of holes predisposed in some specific points of the structure:
This is a hypothesis of the inside distribution calculated on a limited number of fillets and a distribution of 7 holes (they are the red zones of acceleration)
***Image 14***
To be continued…
Reply With Quote
Bookmarks