FR3073162A1 - MODULAR POROUS CORE FOR MOLDING ALVEOLAR STRUCTURE PARTS - Google Patents

Abstract

Modular porous core for molding honeycomb structural parts. This core is formed by several stacked plates, each plate being a layer of elementary volumes having the same general shape, preferentially spherical. When the plates are stacked, these elementary volumes are nested in compact stack, so that they define between them interstices in which the material to be molded infiltrates when it is in the fluid state. This core is characterized in that it comprises at least two plates of different porosities, so that the cellular structure of the part thus molded comprises at least two zones of densities (and, subsequently, masses and resistances) different. Method of manufacturing, by molding, a solid material having a cellular structure, using such a core.

Description

The present invention relates to a modular porous core for molding parts with a honeycomb structure.

Whether constructing buildings, vehicles or tools, it is generally desirable to reduce the mass of the parts produced.

Firstly because the said mass can constitute a handicap: the mass of a long-span beam can cause it to take a boom, the “unsprung” masses of an automobile harm its dynamic behavior, etc.

Then because "mass calls mass":

the mass of a slab requires, to support it, larger pillars therefore heavier which weigh more on foundations which must then be resized, etc.

the mass of a car body requires a larger engine, larger brakes, which increases the overall mass, etc.

But also because mass is an indicator of the amount of material used.

However, on the one hand, the material cost can pose a problem, especially when the material used is sophisticated (high-performance fiber concrete; light metal alloy; technical composite ...) and, on the other hand, industrial materials being often from non-renewable natural resources, it will be increasingly necessary to make frugal use of it.

Furthermore, the vocation of certain technical devices of the “mechanical fuse” type is to dissipate as gradually as possible the energy of a collision, an earthquake or any other type of shock.

These devices fulfill their mission by destroying themselves.

In order to degrade gradually, they can have several "strata" with different mechanical strengths.

To do this, they generally consist of different parts, the assembly of which forms a sophisticated, therefore expensive, and relatively bulky sub-assembly.

However, these devices are often positioned in reduced spaces and their security function does not exempt them from economic constraints.

Finally, the two issues mentioned above can encourage us to look for a way out in the association of materials with complementary characteristics.

Unfortunately, there is a third problem here: that of assembling materials of different natures.

If each material is able to fulfill its mission but the connection between them is fragile or unstable, the final part will obviously not give long satisfaction.

The present invention aims to mainly address these three problems (reducing the mass of structural parts without reducing their mechanical resistance; increasing their energy dissipation capacities; optimizing them by combining materials of various natures) by the same process of manufacturing parts with alveolar internal structure.

In the past, various researches have already been carried out to reconcile mechanical performances and economy of matter by using materials with alveolar structure: "cellular concrete" in the field of the building, "metallic foams" in the fields of the automobile and the , etc.

However, the foaming processes by injection or gas production generate materials whose irregular structure (the locations and dimensions of their cavities are random) is difficult to model and hardly conducive to obtaining optimal mechanical characteristics, especially for complex parts.

In addition, either because their chemical composition differs greatly from that of the reference material (case of cellular "concrete" whose basic components are very different from that of conventional concrete) or because it is altered by the process of foaming (in the case of metal foams which incorporate particles necessary for the stabilization of the foam but detrimental to its final performance), the mechanical characteristics of these foams are generally much lower than those of the material they are intended to replace.

It is to obtain the advantages of these foams without their weaknesses that several authors have sought to reproduce such foams by molding a regular core.

The application for a certificate of utility FR 2599664 of June 9, 1986 and the patent FR 2932705 of June 19, 2008 constitute two exemplary approaches to this research.

Even if one of the two seems more oriented towards concrete and the other more dedicated to metal alloys, these two patents are applicable to any material capable of being supplied in the fluid state and then returning to the solid state after molding.

In order to escape the random configuration of the foams previously mentioned, these two methods use a core composed of sheets (or plates), themselves made up of a plurality of elementary volumes.

Said elementary volumes being identical to each other and arranged together in a regular network.

The main difference between these two patents is the type of elementary volumes chosen:

That of 1986 uses, as elementary volumes, spheres of the same diameter naturally leaving between them interstices in which is poured the material to be molded;

That of 2008 prefers to use parallelohedrons (which have the advantage, and the main characteristic, of entirely paving the space), even if it means chamfering the edges of said parallelohedra to create the interstices necessary for the passage of the material to be molded.

In each of these two patents, the cavities obtained are identical to each other:

- For the first, the identity of the spheres between them is explained in the description text, in particular from line 33 on page 1 to line 3 on page 2 and the regular mesh configuration is the subject of the claim No. 9;

- For the second, the identity of the parallelohedrons between them is induced by the very principle of the tiling of space and, moreover, explained from line 26 to line 34 on page 4 of the description text;

In both cases, the accompanying drawings clearly show the identity of shape and dimensions of the elementary volumes between them within the same core.

Consequently, the result targeted by these two solutions (total homogeneity of the cellular network for perfect isotropy of the material obtained) is perfectly achieved.

Moreover, the 2008 patent is actually used industrially for this purpose.

The present invention therefore does not aim to improve on this point one and / or the other of these referenced patents which, in this regard, give full satisfaction.

The present invention pursues another object.

In a structural part, especially if it is complex, the intensity of the mechanical stresses differs from one zone to another of said part.

Optimal use of the material would therefore consist in finely "controlling" the quantity as a function of the level of mechanical stresses.

In addition, no part being indestructible, it is preferable to know, even to decide, at what level of constraints and in what way it will give way: it is the principle of the “mechanical fuse” which requires, there too, that we can modulate the resistance of a part according to different predetermined zones.

However, the solutions for producing materials with an existing cellular structure do not allow it: either because their density variations are random (case of gas incorporation processes), or because they aim, on the contrary, to obtain a perfect homogeneity of the honeycomb structure (case of the two patents cited in reference).

Finally, certain mechanical stresses (collision for a vehicle, earthquake for a building, detonation for a container, etc.) can generate large amounts of energy, and materials with a honeycomb structure then have interesting aptitudes for energy dissipation (c is why they are used for parts of the “crash-box” type such as the side members of automobiles for example).

However, it is not enough to dissipate this energy, it must be done gradually. Here again, materials with a homogeneous and isotropic cellular structure do not lend themselves to this progressiveness and therefore oblige designers to play on the dimensional variations of the parts themselves and on re-machining operations, which is not optimal neither in terms of pure efficiency, nor in terms of consequences (cost / mass / size).

Consequently, the present invention aims to modulate the density of the honeycomb structure (therefore, its resistance and also its mass) as a function of the intensity of the mechanical stresses which each of the different zones of the part to be produced must undergo.

To illustrate the mechanical principle of the present invention, it is necessary to observe the behavior of a structural part such as a beam for example:

The diagram of the beam which undergoes "buckling" under the action of a vertical force is a classic of the resistance courses of materials: its upper portion is considered to work in compression while its lower portion is deemed to work in tension and that its middle portion is supposed to participate in no way in the overall mechanical effort.

This is why we can hollow out the beam by removing part of its middle portion (which we call "neutral fiber") without affecting its mechanical performance.

In practice, the distribution of forces is not as binary and, to understand it, it suffices to observe what is happening between the middle portion and the surface of the beam.

If we subdivide this beam, for example into seven portions instead of three, we consider that:

the upper portion works a lot in compression and not at all in traction, the second portion works a little less than the first in compression and, like its sister, it does not work at all in traction, the third works a little bit in compression and not at all in traction,

- the middle portion works neither in compression nor in traction,

- the fifth works a little bit in traction and not at all in compression, etc.

Therefore, it can be deduced that the second portion does not need to be as resistant to compression as the first and that the third does not need to be as resistant to compression as the second, etc.

The principle is the same in the case of a part intended to dissipate the energy of an impact, only the arrangement of the "portions" differs: the area closest to the point of impact must be the most fragile, then the adjacent area should be a little less, then the next one even less, etc ... so that the energy dissipation is progressive.

However, the compressive strength of a part (or a portion of a part) depends, of course, on the nature of the material chosen but also on the amount of material used.

Of course, we can imagine a structural part whose core is made of a first material (light and inexpensive) covered with a second material a little more resistant (even if it is more expensive and / or heavier), itself sheathed by a third very resistant material ...

Similarly, regarding mechanical fuses, two structural parts can be coupled via a rupture sleeve made of a more fragile material ...

Such associations are common for simple parts, such as tubes, but are more difficult to industrialize for complex parts.

Another way is to select only one material, for example the most efficient mechanically, but not using the same quantity for all areas of the room. This second path is that which is mainly explored by the present invention.

At the same time, the process developed also provides an answer to the question of the association of complementary materials.

Indeed, when it is desired to combine materials of different natures, the mechanical assemblies are generally punctual (or, at best, multi-punctual) and the assemblies by adhesives are linear (or, at best, planar).

Therefore the tearing stresses are annoyingly concentrated, which generates shearing and / or peeling of said connecting means.

In order to disperse these constraints, the present invention therefore aims, secondarily, to allow “three-dimensional” assemblies of materials.

As in the two patents cited in reference (and many others, not cited here but based on the same principle of molding on a core), the method according to the invention consists in casting the material constituting the final part onto a porous core , lost or destructible, in order to obtain a part with a honeycomb structure.

The specific features of the invention are due to the fact that said core is here designed to be modular and to make it possible to differentiate its porosity (and therefore, inversely and proportionally, the density of the cellular structure obtained) according to predetermined zones.

As in the patents cited in reference, said core (10) consists of several stacked plates (20), themselves made up of elementary volumes (30) (see figures la, lb and le) And, like the certificate useful in 1986, the elementary volumes are preferably spheres (31) juxtaposed in “hexagonal arrangement” (see FIG. 2).

The elementary volumes of the present invention can be produced from other geometric shapes than the sphere, but this one has at least two advantages: it allows a very wide density variation spectrum, it generates completely “arched” walls of re-entering angles, therefore of incipient fractures.

This is why the rest of the presentation will assume, non-exclusively, however, that the elementary volumes are spherical or approximately spherical.

Thus, by stacking the plates so that the spheres of a plate are positioned in the hollows of the plate below and that of the top (see Figure 3), we obtain either a "cubic network with centered faces", or a "compact hexagonal network".

These networks being the two variants of the densest three-dimensional arrangement possible for spheres (see Kepler's conjecture).

And, as in the two patents cited in reference, the elementary volumes of the same plate have exactly the same dimensions (in the case of spheres, the same diameter).

There the similarity with existing patents ends because, in the present invention, the diameter of the spheres can vary from one plate to another.

Simultaneously, and so that the stacking in a compact three-dimensional network is respected whatever the diameters of the spheres of the different plates, the spacing between the geometric centers of the spheres is the same for all the plates of the same core, whatever the diameters of their respective spheres.

NB: the value of this spacing will be named "E" in the following description and in the claims (see Figure 4).

It is therefore only the diameter of the spheres of a given plate (depending on whether it is less than, equal to or greater than E) which determines the level of porosity of said plate.

And it is this variation in porosity of the modular core which, in turn, makes it possible to modulate the density of the cellular structure from one zone to the other of the molded part.

If this diameter is less than E, then said spheres (32) are kept at the right distance from each other by small cylindrical protrusions (40), cited but not claimed by the present application because already present in the reference of 1986 ( see figure 5).

In this case, some (33) of said spheres are an exception to the rule of equality between the spheres of the same plate by adopting a diameter equal to £ so that the distance between the median plane of the plate concerned and the median plane of the plate which is immediately juxtaposed to it is kept equal to E (see FIG. 6).

At least three of these spheres with a diameter equal to E are needed per plate (only one would constitute a spherical pivot and two would form a linear axis around which the plate could pitch). From an industrial point of view, the simplest is to provide a constant quantity for a given plate area.

In addition, without this being absolutely necessary, bringing them together in groups of three facilitates the positioning of the plates.

The stacking of these plates with "small spheres" generates an area in which the interstices are relatively large and, therefore, where the density of the honeycomb structure will be high.

If, on the contrary, the diameter of the spheres of a plate is greater than E, then said spheres (34) interpenetrate (Figures 7a and 8) and, in this case, two models of plates complement each other:

- A first model, called “thick plate” (see FIG. 7a), in which the spherical caps of the upper face and those of the lower face (50) of the plate are each faceted by three discs (60), each disc being defined by the virtual interpenetration plane between each of said caps and the three spheres of the adjacent plate with which said cap is in contact.

The stack of plates of this model generates an area in which the interstices (80) are relatively small and, therefore, where the density of the honeycomb structure will be low (see FIG. 8);

- A second model, called a “transition plate” (see Figure 7b), is identical to the first on one of its faces but differs from it on the other face, the half-spheres of which (70) have their diameter reduced to the value of E.

The name of this plate comes from the fact that it serves to ensure the transition between a thick plate and a plate whose diameter of the spheres is equal to E (or even a plate whose diameter of the spheres is less than E, as described above) without generating structural discontinuity, synonymous with local fragility.

A final type of plate (90), simply flat (except for bowls intended to wedge the spheres of the adjacent plates) and watertight (devoid of gaps) completes this system in order to allow the creation of "empty" zones (140), that is to say completely free of the material constituting the final part, within said molded part (see FIG. 9).

These empty zones, formed precisely where the molded part is not supposed to be subjected to mechanical stresses, lighten the molded part a little further and are conducive to the passage of electrical, aeraulic or other equipment.

Finally, when, on the contrary, it is desired to generate a sealed partition within the molded part, spacers (100) are previously produced in the material constituting said part. These spacers make it possible to hold two plates spaced from one another, thus generating a space (see FIG. 10) which, once filled with the molded material, will form a solid partition. Two partitions, spaced by a plate or a group of plates, define a watertight pocket which it will be possible, for example, to then fill with a liquid or a gas.

If necessary, it suffices to make grooves by cutting or grooving in the plate or group of plates) taken between these two partitions to compartmentalize this waterproof pocket.

The stages of the molding process are as follows:

porous plates are produced from elementary volumes, the plates of different types and porosities are selected as required,

- if necessary, said plates are cut according to the contours of the part to be molded,

- the plates are stacked including, where appropriate, any spacers,

the core thus formed is introduced into a mold,

the material constituting the part to be produced, poured (by simple gravitational casting or under vibration or by injection under pressure or vacuum) into said mold when it is in the fluid state, is introduced into the interstices of the core, once the material to be molded has solidified (by cooling, crosslinking, drying, etc. depending on the nature of said material), the mold is opened,

- possibly, the nucleus is destroyed (in whole or in part),

- just as optionally, the empty cells are filled with one or different materials with complementary properties.

For certain applications (for example prototyping), the part can be produced by additive manufacturing via a 3D printer.

In this case, the kernel is virtual: it only exists in digital form.

The core according to the invention is said to be “modular” because it is constituted, on a case-by-case basis, by the assembly of different industrially prefabricated plates.

The density of the honeycomb structure, identical to the porosity rate of the plates, will therefore be defined zone by zone (see Figure 11).

If for example, in a given area of the final part, the spheres have a diameter equal to E, then they will occupy 74% of the space available in said area (see Kepler's conjecture) and the interstices will only occupy the additional 26%.

Consequently, once the material constituting the final part has been poured, the density of the cellular structure in this zone will be 26%.

If, in a given area, the diameter of the spheres is less than E, the interstices will be large there and the density of the alveolar structure will be high (between 26% and 99% depending on the diameter of the spheres).

If in a given area, the diameter of the spheres is greater than E, the interstices will be reduced in size and the density of the alveolar structure will be low (between 1% and 26% depending on the diameter of the spheres).

Finally, if in a given area the core is made up of waterproof plates, the material making up the molded part will be completely absent (density of 0%).

We see here the advantage of the spheres compared to the parallelohedra: these, by only allowing to play on the importance, necessarily limited, leaves leave only too little amplitude as to the dimension of the interstices.

This is why, in practice, the parts resulting from the process according to the 2008 patent cited in reference are only capable of covering a narrow spectrum of densities.

Not only is their cell structure density uniform piece by piece but, moreover, it is limited to around 5% for the lightest of them against barely 15% for the heaviest.

This low amplitude is obviously not annoying in the context of the 2008 patent cited in reference, on the contrary even, since its purpose is precisely to obtain a perfectly isotropic cell structure and of low density.

But, even disregarding the problem of coexistence of different levels of porosity within the same core, this low amplitude would be too limiting to achieve the objectives of the present invention.

Conversely, the density of the structure of the parts produced according to the 1986 reference could be 99% for the heaviest (in practice, there is little point in exceeding 60%) against, at best , 26% for the lightest.

This low limit, geometrically induced by the process of said reference was not a problem, because a concrete 4 times lighter than the standard constituted, within the framework of the use envisaged in 1986, a largely sufficient performance.

But, there too, it would be too disabling for the objectives of the present invention.

Not to mention that this process, too, does not ensure a gradual transition between two different densities without generating structural discontinuity.

To summarize, if the spectrum of densities achievable with spheres is wider than that of the parallelohedrons, it can only cover densities below 26% if said spheres have the possibility of interpenetrating.

The 1986 process does not allow to descend under this barrier (which said utility certificate explicitly agrees from line 34 on page 1 to line 3 on page 2) precisely because it does not have this possibility interpenetration.

Concomitantly, said method determines the density of the part to be produced, unlike the present invention, not by varying the dimension of the spheres but the spacing between them (which is explained therein of line 36, page 2, line 12, page 3).

However, so that different densities can coexist within the same room without generating discontinuity in the honeycomb structure (which would generate a zone of fragility), it is necessary that the plates of different porosities remain compatible (that is to say that, when stacked, their spheres respect a compact three-dimensional stacking).

This last point in turn requires that, on the one hand, E be constant within the same core and that, on the other hand, the plates include the specificities described above.

NB: in practice it is likely that, for obvious reasons of standardization of the plates, the porosity is only varied from 5% to 5% and this in a spectrum probably between 5% and 60%.

Anyway, in the present invention, by positioning very porous plates in the zones intended to undergo high mechanical stresses and low porous plates in the less stressed zones, the density of the cellular structure is adapted to the intensity of mechanical stresses, zone by zone.

Thus, the mass of the molded part is significantly reduced while maintaining its overall cohesion and without affecting its mechanical performance.

On the contrary, even: because, in the case of a load-bearing element, its total load capacity remaining identical and its own mass being reduced, its payload capacity increases accordingly. The example of the beam is the most obvious (see Figure 12) but the method applies to any structural part, including complex shapes and / or extreme dimensions.

Similarly, the gain in mass authorized by the present invention corresponds, for suspended parts, to a proportional increase in range.

Let us illustrate this with the hypothesis of the high slab, for example with a thickness of 30 cm, of a performance hall whose number of pillars we would like to minimize (see Figure 13):

the skins of the upper and lower surface of said slab, fitted with fittings and undergoing the greatest stresses, have a density of 100% (110) over, for example, a thickness of 2.5 cm each,

- each of these skins is followed by a layer that is less mechanically stressed, for example 5 cm thick for a density of cellular structure of 40% (120),

- A last layer, for example with a density of cellular structure of 20% (130) for 15 cm of thickness, whose only role is to ensure cohesion between the upper face and the lower face of said slab.

Such a slab, while preserving its flexural strength intact and fully ensuring its mission of horizontal bracing of the building, will have its mass divided by 2.5.

Therefore, its range between two pillars would be multiplied by 2.5.

Of course, for the process to allow a significant reduction in the parts thus produced, the material constituting the spheres should be destructible and / or with a density much lower than that of the material constituting the molded part.

Furthermore, since the spheres can be produced from a wide range of materials (polystyrene, frozen water, agglomerated sand, foundry wax, elastomers, etc.), it is possible to combine layers (or portions of layers) of different natures if one wishes to destroy, after production of the molded part, certain layers (or portions of layers) but not others. This possibility can in particular prove useful for keeping within the molded part associations of materials with complementary characteristics: in the case of a residential wall for example, a material can bring its capacity of sound insulation while another brings its thermal insulation capacity.

Thus, it is possible to combine several materials in two different ways:

As we have just seen, by preserving all or part of the core within the molded part (for example a polystyrene core in a final concrete part);

- By replacing all or part of said core with another material (for example by pouring into a part made of metal, a second metal whose melting point would be lower than that of the first).

Coupling the two methods allows complex associations of multiple materials.

Many applications can take advantage of these assembly possibilities:

- A relatively high Shore hardness elastomer core kept within a molded piece of low Shore hardness elastomer, would allow gradual transitions between zones of differentiated flexibility (for example to produce a tire without an inner tube or a part intended to absorb vibrations);

- A transmission part, made of an alloy chosen for its high mechanical strength, could be added, replacing a part of its core, of an element cast in a material with very low coefficient of friction (provided that both have close expansion coefficients, of course).

This last example illustrates the advantage of replacing a conventional mechanical assembly or an adhesive assembly with a decreasing interpenetration according to the invention. In the case of rivets, studs, or other screws, the mechanical stresses will tend to shear these connecting elements and, in the case of bonding, they will generate a tearing or peeling. While in the method according to the invention, the assembly is three-dimensional.

Thus, mechanical stresses no longer have the possibility of concentrating on a single plane, they disperse over the entire volume where the two materials are mixed.

If the method according to the invention optimizes the mass / mechanical strength ratio of the molded parts and allows the association of materials of complementary natures, it concomitantly has another advantage: all the cellular structures have, having regard to their mass, great aptitudes energy dissipation, but the honeycomb structure according to the invention adds a programmable progressiveness.

Thus, by voluntarily creating a more fragile area within a room, we can determine:

the precise place where the part will break,

- the level of constraints which will trigger the rupture,

- the way in which the mechanical fuse thus created will break (an area of homogeneous fragility would lead to a frank and abrupt break, while an area made up of zones of different volume densities will generate a more gradual break).

We understand the interest of such a fuse in a machine but, on a larger scale, this process can also be applied to buildings: for example, a rupture zone provided in the beams (Figure 14) for use the “strong pole / weak beam” principle in the event of an earthquake. In all cases, the programmed rupture of the part will dissipate a quantity of energy previously calculated.

It is these same progressive energy dissipation capacities which, for an equal amount of material, allow the method according to the invention to optimize any part (vehicle crash-box, end of railway track stop or any other type of shield ) intended to absorb a collision.

In such applications, the plates of the modular core are positioned in a direction different from that which would be used for a beam: the plates are stacked, not parallel to the longitudinal axis of the part, but perpendicular to it .

The order also differs: the density zones are determined, not according to the intensity of the mechanical stresses, but according to their chronology.

For example, if the part produced according to the method of the invention is an automobile side member-stretcher (see FIGS. 15a and 15b), the density of the honeycomb structure will be low at the front of the part (first part stressed by collision) and higher and higher thereafter, so as to gradually reduce the violence of the shock.

In addition to these specific applications, the method according to the invention is adaptable to the usual applications of cellular materials (heat exchanges, vibration dampings, etc.), as soon as there is an advantage in having zones with differentiated volume densities.

For example, in the case of a prosthesis intended to be inserted within a broken bone, cellular structures are already used, the porosity of which allows "osteointegration": the organism manufactures bone which "colonizes »The prosthesis by filling in the empty spaces.

This phenomenon will be more effective with the method according to the invention since, the density of the alveolar structure being progressive, the nesting with the bone material will also be.

Whatever the application of the present invention, the nature of the material constituting the final part will largely determine that of the material constituting the modular core in the sense that a fluid material "at room temperature" (concrete, plaster, etc. .) gives more latitude in this respect than a material that can only be molded hot (thermoplastic polymer, etc.), or even at melting temperature (metal alloy, etc.)

Furthermore, the value of the spacing £ between the spheres will depend greatly on the particle size and the ductility of the material constituting the molded part: for example of the order of a few mm for a polymer or a metal alloy, it may, always as an example, be of the order of a few cm (or even a few dm) for a concrete.

For the record, it is recalled that in the method according to the invention, the density of the honeycomb structure does not depend on the size of the cells in the absolute, but on the ratio between this dimension and the spacing E between the centers of said cells .

The accompanying drawings illustrate the present invention:

- Figure la shows, in schematic perspective, any nucleus according to the invention.

- Figure lb shows, in schematic perspective, the plates that make up said core.

- The figure shows, in schematic perspective, the elementary volumes that make up said plates.

30- Figure 2 shows, in top view, the hexagonal arrangement of the constituent spheres of any plate of a core according to the invention.

- Figure 3 shows, in section, the stack of several plates according to Kepler's conjecture.

FIG. 4 represents, in side view, three plates constituting the same nucleus: one whose spheres have a diameter less than E, one whose spheres have a diameter equal to E and one whose spheres have a diameter greater than E.

- Figure 5 shows, seen from the front, a portion of a plate whose spheres have a diameter less than E.

- Figure 6 shows, in side view, a portion of plate whose spheres have a diameter less than E, except for some of them which have a diameter equal to E.

Figure 7a shows, in side view, a portion of plates called "transition".

- Figure 7b shows, in side view, a portion of plates called "thick".

10- Figure 8 shows, in front view, a portion of a plate whose spheres have a diameter greater than E.

FIG. 9 represents, in side view, a non-porous plate inserted within porous plates.

Figure 10 shows, in side view, spacers inserted between plates.

15- Figure 11 shows, in side view, a molding core according to the invention, consisting of several plates of different porosities stacked.

- Figure 12 shows, in perspective, the edge of a beam made by the method according to the invention.

FIG. 13 represents, in perspective, a building slab produced by the method according to the invention.

- Figure 14 shows, in longitudinal section, a beam incorporating mechanical fuses produced by the method according to the invention.

- Figure 15a shows, in section, a vehicle spar intended to dissipate the energy of a collision.

25- Figure 15b shows, in section, the same spar after the collision.

In summary, the porous core (10) according to the present invention comprises at least two plates (20) of different porosities.

Said Easter (20) consist of elementary volumes (30).

The value of the spacing E between the geometric centers of said elementary volumes (30), common to all the plates (20) of the same core (10), is independent of the cubing of said elementary volumes (30).

The cubing of the elementary volumes (30) of any plate (20) may differ significantly from the cubing of the elementary volumes (30) of other plates (20) of the same core (10).

On the other hand, the diameters of the spheres of two specific models of plates respectively called “thick plates” and “transition plates” are equal to each other and greater than £ so that said spheres interpenetrate virtually. Each sphere (34) of each thick plate is faceted by six discs (three on the upper face and three on the lower face), each of these discs (60) being defined by the virtual interpenetration plane between said sphere 34) and l one of the six spheres (34) of the adjacent thick plates (above and below) with which said sphere (34) is in contact.

Each hemisphere of the first face of each transition plate is faceted by three discs (60), each of these discs (60) being defined in the same way as the discs (60) of the thick plates, so that said first side adjusts with any thick plate,

The diameter of each hemisphere (70) of the second face of each transition plate is reduced to the value of E, so that each transition plate adjusts with any other plate (20) whose diameter spheres (31 or 30) would be less than or equal to E,

In order to avoid the formation of rupture angles on the honeycomb structure, the edge of the hemisphere of the first face is cut down by a convex fillet.

It should be noted that the plates (20) of the modular core (10) generally consist of destructible materials after casting and / or densities substantially lower than the density of the material of the molded part.

Likewise, the material constituting a plate (20) can be of another nature than the materials respectively constituting the other plates (20), so that it is possible, after casting, to destroy certain parts of said core. (10) while retaining others.

On the other hand, spacers (100) previously made in the material of the molded part are used to separate the core (10) from the edges of the mold and / or to separate certain plates (20) from each other, allowing the so that said material constituting said molded part is introduced in the fluid state into the empty spaces thus generated and to form watertight partitions around and / or within the honeycomb structure.

In order to form a mechanical fuse, the rupture of which occurs as soon as a predetermined level of mechanical stress is reached, the porosity of the core (10) according to the invention can be significantly lower in a particular zone than in the other zones of the structure. alveolar.

In other cases, the density of the honeycomb structure of each zone is chosen so as to be proportional to the intensity of the mechanical stresses which said zone will have to undergo.

The method of manufacturing a honeycomb structure, based on the porous core (10) according to the invention, comprises the following steps:

plates (20) of different types and porosities are selected, as described above,

- said porous plates (20) are cut according to the contours of the part to be produced, a core (10) according to the invention is formed, by stacking said porous plates (20) and, if necessary, inserting between some of spacers 100), positioning said porous core (10) in a mold, pouring or injecting said material, in the fluid state, into said mold so that said material infiltrates into the interstices (80) of said porous core (10), after solidification of said material, said mold is opened and, optionally, all or certain plates (20) of said core (10) are destroyed,

- again, optionally, filling the empty cells with one or more other material (s).

It should be noted that the porous core (10) according to the invention can be digital and virtual in order to allow the production of a part by additive manufacturing, by depositing successive layers of the material constituting said part by means of a device called a "three-dimensional printer".

Claims (10)

1) Porous core (10) for molding a solid material with a honeycomb structure consisting of a plurality of superimposed plates (20), each plate being a layer of elementary volumes (30) juxtaposed in hexagonal arrangement, characterized in that: the elementary volumes (30) are, preferably but not exclusively, of spherical or approximately spherical shape and are called “spheres” (31), the spacing £ between the geometric centers of said spheres (31) is invariable within each plate (20) and common to all the plates (20) of said core (10), the diameter of the spheres (31) is independent of the spacing E between their geometric centers and can therefore be less, equal or greater, the ratio between the spacing E and the diameter of the spheres of any plate (20) determines the dimensions of the interstices (80) left between said spheres (31), therefore the level of porosity of said plate (20) and, subsequently, the density of the molded honeycomb structure, - the diameter of the spheres (31) of any plate (20), and subsequently its level of porosity, can differ significantly from the diameter of the volume spheres (31) of other plates (20) of said core (10), said core (10) comprises at least two plates (20) of different porosities. 2) Porous core (10) according to claim 1 characterized in that: - The plates (20), the majority of the spheres (32) of which have a diameter less than E are equipped with at least three so-called "assembly" spheres (33) whose diameter is equal to E, so that the distance between the median plane of any one of said plates (20) and the median plane of the plate (20) which is adjacent to its lower or upper face is kept equal to E, with the exception of the so-called assembly spheres (33 ), all the spheres of the same plate (20) have the same diameter. 3) Porous core (10) according to claims 1 and 2 characterized in that: - The diameters of the spheres of two specific models of plates (20) respectively called "thick plates" and "transition plates" are greater than £ so that said spheres interpenetrate virtually, each sphere (34) of each thick plate is faceted by six discs (60), each of these discs (60) being defined by the virtual interpenetration plane between said sphere (34) and one of the six spheres (34) of the adjacent thick plates with which said sphere (34 ) is in contact, - Each hemisphere of the first face of each transition plate is faceted by three discs (60), each of these discs (60) being defined in the same way as the discs (60) of the thick plates, so that said first side adjusts with any thick plate, - the diameter of each hemisphere (70) of the second face of each transition plate is reduced to the value of E, so that each transition plate adjusts with any other plate (20) whose diameter of the spheres (31 or 30) would be less than or equal to E, 4) Porous core (10) according to the preceding claims, characterized in that: - The plates (20) of the modular core (10) are made of destructible materials after casting and / or of densities substantially lower than the density of the material constituting the honeycomb structure thus molded. 5) Porous core (10) according to the preceding claims characterized in that the material constituting a plate (20) is of another nature than the materials respectively constituting the other plates (20), so that it is possible , after casting, destroying certain parts of said core (10) while retaining others. 6) Porous core (10) according to the preceding claims characterized in that spacers (100) previously made in the material of the molded honeycomb structure are used to separate the core (10) from the edges of the mold and / or to remove certain plates (20) from one another, thus allowing said constituent material of said molded honeycomb structure to be introduced in the fluid state into the empty spaces thus generated and to form there watertight partitions around and / or within said alveolar structure. 7) Porous core (10) according to claim 1 characterized in that the porosity of said core (10) is substantially lower in a particular area than in the other areas of the molded honeycomb structure so that the low density of honeycomb structure thus obtained in said particular zone forms a mechanical fuse, the rupture of which occurs as soon as a predetermined level of mechanical stress is reached. 8) Porous core (10) according to claims 1 to 4 characterized in that the level of porosity of the plates (20) and, subsequently, the density of the honeycomb structure of each zone of the honeycomb structure to be molded is chosen so as to be proportional to the intensity of the mechanical stresses foreseen for said zone. 9) A method of molding a solid material with a cellular structure within a porous core (10) according to claims 1 to 5, characterized in that it comprises the following steps: porous plates (20) of different types and porosity rates are selected, said porous plates (20) are cut according to the contours of the cellular structure to be produced, the porous core (10) is formed, by stacking said porous plates (20 ) and, if necessary, by inserting spacers (100) between them, - said core (10) is positioned in a mold, said material is poured or injected, in the fluid state, into said mold so that said material infiltrates into the interstices (80) of said core (10), - After solidification of said material, said mold is opened and, optionally, all or some of the plates (20) of said core (10) are destroyed, again, optionally, filling the empty cells with one or more other (s) materials). 10) manufacturing process using a porous core (10) according to claims 1, 2, 3 and 8 characterized in that said core (10) is virtual and digital so as to allow the production of parts with a honeycomb structure by additive manufacturing.