The disclosure relates to a system and method for manufacturing laminated composites by selectively inhibiting lamination within a layer and between two adjacent layers.
Three-dimensional rapid prototyping and manufacturing allows for quick and accurate production of components at high accuracy. Machining steps may be reduced or eliminated using such techniques and certain components may be functionally equivalent to their regular production counterparts depending on the materials used for production.
The components produced may range in size from small to large parts. The manufacture of parts may be based on various technologies including photo-polymer hardening using light or laser curing methods. The present disclosure is directed to laminated object manufacturing (“LOM”), and in particular LOM using composite materials.
LOM uses sheet materials to make three dimensional objects, which allows for making parts out of pre-existing off-the-shelf sheet materials. LOM creates 3D parts by forming individual layers out of pre-existing sheet materials by cutting or etching them in patterns dictated by data representative of the three-dimensional object being built. The layers are adhesively bonded together. Each layer is cut or etched into object sections and waste sections. The object sections are those sections of the layer that define the desired finished object. The waste sections are those sections of the layer other than the object sections and are removed at the end of the object building process.
Current LOM systems and methods suffer from a number of drawbacks. Certain LOM methods cut object shapes out of the laminating material before adhering layers together. These methods require techniques for aligning or “registering” the various layers. Also, many methods require the inclusion of a process for depositing an adhesive onto the laminating material.
After LOM layers are formed to define the desired object cross-section, the layers include an object section that will remain part of the finished object and waste sections that will be removed. The object sections and waste sections define interfaces where the two sections meet. The waste sections are typically cut or etched to facilitate their removal from the finished object. However, within a given layer, the waste sections can sometimes adhere to the object sections such that removal of the waste sections damages the object sections. Also, adjacent layers may have regions where object sections in one layer abut waste sections in another layer. As a result, the removal of the waste sections in one of the layers can damage an adjacent object section in an adjacent layer.
Another drawback in many known LOM processes involves the use of a pressure roller to bond adjacent layers. In some cases, the pressure roller comes into contact with the adhesive at a temperature high enough for adhesion to the roller itself to occur. This is particularly a problem if the pressure roller itself is used to supply the heat necessary for adhesion and lamination to occur. In addition, many prior LOM processes failed to account for the compression of laminating materials when moving the build platform to laminate subsequent layers. Certain laminating materials such as those with a plastic binder component will undergo compression during a lamination operation, and if the build platform is moved by a distance corresponding to the uncompressed thickness of the layer, the mechanical properties of the resulting object may be compromised and/or delamination may occur.
Thus, a need has arisen for an apparatus and method for making laminated objects from composite materials which addresses the foregoing issues.
The disclosure will now be described, by way of example, with reference to the accompanying drawings, in which:
The Figures illustrate examples of a system and method for manufacturing. Based on the foregoing, it is to be generally understood that the nomenclature used herein is simply for convenience and the terms used to describe the invention should be given the broadest meaning by one of ordinary skill in the art. Unless otherwise specified, like numerals refer to like components herein.
The system and methods described herein are generally applicable to the laminated manufacturing of three-dimensional objects, such as components or parts (discussed herein generally as objects), but may be used beyond that scope for alternative applications. In such systems and methods, successive layers of an object forming material are adhered to one another and cut in a pattern that defines an object section and a waste section. The patterns are based on data representative of the three-dimensional object. The data may be provided, for example, as CAD/CAM data and sliced into a number of layers (using a data slicing technique) each representing a cross-section of the object. Any known three-dimensional object data construct may be used, including without limitation STL (Stereo Lithography) files or other CAD (Computer Aided Drafting) files commonly translated for rapid prototyping systems into formats such as SLC, CLI slice data files or voxelized data files which may include data formats such as BMP, PNG, and vector data that defines the object contours within each layer. In preferred examples, the three-dimensional object data is preferably converted to layer data and then to a format (such as G-Code data) useful for moving the cutting assembly 56 in the x-y plane, adjusting the z-axis position of the blade 140 tip and adjusting the rotational orientation of the blade 140 about its own longitudinal axis. In certain preferred examples, the patterns are based on G-Code data and used to guide the movement of a cutting blade that forms the contours of the object sections and cuts the waste sections into shapes that facilitate their separation from the object sections.
Preferred materials for making three-dimensional objects in accordance with the present disclosure include composite materials, and more specifically, reinforced fiber plastics. Composite materials are combinations of two or more chemically distinct and insoluble phases with a recognizable interface, in such a manner that its properties and structural performance are superior to those of the constituents acting independently. One class of composites of particular importance in this disclosure is reinforced fiber plastics, also known as polymer-matrix composites, and fiber-reinforced plastics. This class of composite materials includes fibers as a discontinuous or dispersed phase in a polymer matrix. The fibers tend to have high specific strength (strength to weight ratio) and specific stiffness (stiffness to weight ratio). The fibers have stiffness in the longitudinal direction but very little strength or stiffness in the direction transverse to the longitudinal direction. The reinforced fiber plastics comprise a polymer matrix of a thermoset or thermoplastic polymer with reinforcing fibers embedded therein. The percentage of fibers in reinforced plastics generally ranges from 10 percent to 60 percent (by volume). In some cases, the fibers are istotropic, and in other cases, the fibers are anisotropic. The thermoset or thermoplastic polymer acts a binder (adhesive) when it is heated above a lamination temperature and cooled to adhere adjacent object layers to one another.
In certain examples, the reinforced fiber plastics are provided as pre-peg (“pre-impregnated”) sheets comprising fibers in a polymeric matrix of a thermoplastic or thermosetting binder. The fibers are pre-impregnated with the binder, and the binder can be heated to act as an adhesive for laminating adjacent layers together. The advantage of using pre-pegs as laminating materials is that the apparatus for making the three-dimensional object does not have to include an apparatus for depositing a separate adhesive onto the laminating materials, which reduces equipment costs and processing times.
Pre-pegs may have continuous fibers or discontinuous fibers. Continuous fibers are those that have an uninterrupted length along some dimension of the sheet (e.g., length, width, or some angled orientation between the length and the width). Discontinuous or “chopped” fiber pre-pegs are short (relative to the sheet dimensions) and are randomized as to their placement and orientation. Pre-pegs are generally made in one of three ways: 1) laminating one side of the fibers with a binder, 2) laminating both sides of the fibers with a binder, and 3) powder coating the fibers with a binder.
Pre-pegs are typically classified as “isotropic” or “anisotropic.” An isotropic pre-peg has equal mechanical properties in any direction along any axis. An anisotropic pre-peg will have some directional variation in mechanical properties. Anisotropic pre-pegs may have fibers that are uni-directional, bi-directional, and multi-directional. There may be two or more axes along an anisotropic pre-peg along which the mechanical properties will be equal. However, there will also be one or more axes along which the properties will be unequal. Discontinuous, randomly-oriented fiber pre-pegs are isotropic. Continuous fiber pre-pegs are anisotropic. However, if they are multi-directional, as the number of fiber directions increases, the pre-pegs will generally approach anisotropy.
In some examples, the pre-pegs have anisotropic fibers which provide mechanical properties that vary with direction. In one example, the anisotropic fibers are parallel to one another throughout the entirety of the sheet. However, in other examples, groups of fibers with in the pre-pegs are oriented at different angles relative to one another in a way that provides for some variation in mechanical properties as between two different axes. For example, if two groups of fibers are provided and are oriented at ninety degrees relative to one another, the properties at forty-five degrees will vary from those at zero degrees and ninety degrees. As will be discussed further below, adjacent layers containing anisotropic fibers may be oriented with the fibers at different rotational orientations relative to one another to obtain a desired object strength or other desired mechanical properties.
Suitable thermoplastic matrix materials for use in making reinforced fiber plastics (including pre-pegs) in accordance with the present disclosure include HDPE (high density polyethylene), LDPE (low density polyethylene), polypropylene, Nylon 6, Nylon 66, polycarbonate polyetherketone ketone (PEKK), and polyetherether ketone (PEEK). Suitable thermoset materials include epoxies, polyesters, phenolics, fluorocarbons, polyethersulfone, silicone, and polyimides. Once thermoset materials cure, they cannot be reheated to flow. Therefore, when composite materials comprising thermosets are used, steps are preferably taken to prevent curing from occurring before lamination. In one example, a release liner with a release coating, such as a silicone coated paper, is placed over the composite material to protect the thermoset from atmospheric exposure. The release liner is removed as close in time to the layer being laminated to the three-dimensional object as possible. In addition, it is often necessary to keep composite materials comprising thermosets at a relatively cold temperature to avoid premature thermal curing.
Suitable reinforcing fibers include glass, carbon, graphite, boron, and aramid (Kevlar) fibers. Carbon nanotubes may also be used to form a composite instead of using carbon fibers. Carbon nanotubes are seamless, cylindrical hollow fibers comprised of a single sheet of pure graphite that typically have a diameter of 0.7 to 50 nanometers and lengths in the range of 10 s of microns.
In certain examples herein, a PEEK carbon fiber fabric is used. In one implementation thereof, the PEEK carbon fiber fabric has a glass transition temperature of about 289° F. (143° F.). PEEK carbon fiber fabrics are particularly well suited for aircraft applications due to their strength, light weight, and chemical and corrosion resistance to typical aircraft service fluids.
The ability to form three dimensional objects by laminating composite materials comprising reinforced fiber plastics provides superior mechanical properties (e.g., tensile strength, tensile modulus, flexural strength, flexural modulus, Izod impact, compression strength, compression modulus, and shear strength) as compared to current rapid prototyping technologies. In certain examples herein, the reinforced fiber plastics are anisotropic and alternate layers are arranged such that the length axes defined by their fibers are not parallel. This alternating of the fiber lengths further improves object strength.
Referring to
Apparatus 40 comprises a table lift 50 that holds a rotary table 98 (
A lamination assembly 70 is provided and is used to bind successive layers of laminating material 42 to one another. Material handling system 66 provides a free edge of laminating material 42 that can be gripped by a pair of parallel material advancement systems 124a and 124b (
A laminating material forming assembly 53 is also provided to form the laminating material 42 into shapes dictated by data representative of the three dimensional object being built. Laser assemblies and cutting assemblies may be used as forming assemblies. In the figures, the laminating material forming assembly 53 comprises cutter translation assembly 77 and a cutting assembly 56 (
In general, the cutting assembly 56 cuts sheets of laminating material 42 into object sections corresponding to the three-dimensional object being built and waste sections which comprise those portions of a sheet of laminating material 42 which are not object sections. The waste sections are typically cut into a pattern that facilitates their removal once the object building process is complete. In one example, the pattern is cubes. The object sections may be defined by vector data that is used to dictate the movement of the blade 140. One advantage of the LOM methods of the present disclosure relative to other methods such as stereolithography, is that rasterization need not be carried out to solidify the areas within the object sections. Instead, only the object contours need to be formed.
Referring again to
An exaggerated schematic view of the material handling assembly 66 is shown in
Apparatus 40 includes two material advancement assemblies 124a and 124b to pull a free edge of laminating material 42 over build platform 52 to laminate each layer. Material advancement assembly 124b is shown in
Material advancement assembly 124b (
Referring again to
Vertical rail 131b is connected to another linear bearing 127b via mounting plate 145b and bracket 161b. Linear bearing 127b slidingly engages a laminating assembly rail 108b (
Rack gear 128b is an elongated rail structure that includes teeth 129b which engage corresponding teeth on pinion gear 150b. Motor 126b selectively activates pinion gear 150b, causing pinion gear 150b to rotate about an axis of rotation through its center which is parallel to the build (z) axis. As pinion gear 150b rotates, the engagement of its teeth with rack gear teeth 129b causes linear bearing 127b to travel along the travel (x) axis. The material advancement assembly 124b is not visible in
Laminating assembly 70 (
Referring to
Pressure roller 86 is mounted on a shaft (not shown) through opening 101 that defines a longitudinal axis that is parallel to the y-axis and about which pressure roller 86 rotates as it translates along the travel (x) axis. Guide roller 102 is mounted on a shaft (not shown) through opening 107 that defines a longitudinal axis that is parallel to the y-axis and about which guide roller 102 rotates as it contacts current laminating material layer 108. The bracket 100b is attached to a linear bearing 116b which slidingly engages rail 108b. A corresponding bracket 100a and linear bearing 116a are provided on the opposite side of the build platform 52 along the y-axis. The linear bearings 116a and 116b are operatively connected to the timing belts 88a and 88b such that when the timing belts 88a and 88b (
Preheater assembly 105 is also provided as part of lamination assembly 70 to apply heat to the lower surface 112b of current laminating material layer 108 prior to the lower surface 112b of current laminating material layer 108 making contact with the upper surface 114a of previous laminating material layer 110. The preheater assembly 105 comprises at least one preheater, which in
The amount of heat supplied by preheaters 106a-106c is sufficient to cause the polymer matrix component of a reinforced fiber plastic to reach a lamination temperature, i.e., a temperature at which the plastic of the current and previous laminating material layers 108 and 110 is soft and hot enough to cause adhesive bonding between the lower surface 112a of current laminating material layer 108 and upper surface 114a of previous laminating material layer 110.
Pressure roller 86 comprises a heat conductive material, preferably a metal such as steel or aluminum. Pressure roller 85 houses a conductive heating coil used to selectively heat the outer surface of pressure roller 86. The outer surface of pressure roller 86 will generally supply less heat to upper surface 112a of current laminating material layer 108 than the preheaters 106a-104c will supply to lower surface 112b of current laminating material layer 108 so that the current laminating material layer 108 does not adhere to the outer surface of pressure roller 86. In certain known systems, all of the lamination heat is supplied by a pressure roller to the upper surface 112a, which causes the composite material to adhere to pressure roller 86. In preferred examples, the surface temperature of pressure roller 86 is maintained below the glass transition temperature of laminating material 42 so that the upper surface 112a of current layer 108 does not adhere to the pressure roller 86. In certain examples, the surface of pressure roller 86 is maintained at a controlled temperature by using a thermocouple or other temperature measuring device that measures the temperature of the external surface of the pressure roller 86 and adjusting the heat supplied by the conductive heating coil in pressure roller 86 (such as by adjusting the current or voltage supplied to the coil). Feed back control may also be used with preheaters 106a-106c by measuring their emission temperatures and adjusting the energy supplied to the preheaters 106a-106c (e.g., by adjusting the source voltage or current supplied to the preheaters 106a-196c. Alternatively, as indicated above, non-contact temperature sensors (such as IR temperature sensors) may provide a temperature of the lower surface 112b of current laminating material layer 108 and used to manipulate the heat supplied by preheaters 106a-106b. In certain examples, pressure roller 86 is used to apply a controlled pressure to upper surface 112a of current laminating material layer 108. A shaft (not shown) through opening 101 may be operatively connected at its ends to pressure sensors, and the position of build platform 52 along the build (z) axis may be manipulated to achieve a desired pressure. In such examples, the shaft is operatively connected to the side rails 108a and 108b so that pressure roller 86 has some play in the build (z) axis direction. Although not visible in
After each layer of laminating material 42 is adhered to the previous layer, a forming operation takes place. In some known LOM processes, the forming operation takes place before adhering the layers together. However, such processes generally require a means for aligning or registering the formed object sections. By adhering the layers before carrying out the forming operation, the forming operation may be carried out with respect to the same fixed frame of reference in the x-y plane without the need to make sure the layers are registered.
In general, it is desirable to maintain a constant layer thickness as between different layers when laminating layers of a particular composite material. Pre-peg composite materials will have a laminated (compressed) thickness that may be substantially thinner than the unlaminated pre-peg thickness. Variations in the laminated thickness throughout an object may cause undesirable mechanical property variations. In addition, the build platform 52 should ideally be moved downward by an amount Δz that is equal to the amount of the actual compressed thickness of the current layer not the uncompressed thickness, which is greater. If the build platform 52 is moved down by a distance greater than the actual compressed thickness of the current layer, the mechanical properties of the three-dimensional object may be compromised and/or delamination may occur.
A default compressed layer thickness (such as one provided by the supplier of laminating material 42) may initially be used to determine how far down along the build (z) axis to move the build platform 52 after each layer is formed. The compressed layer thickness is the thickness expected at a certain pressure and temperature applied to the layer. In the same or other examples, the actual compressed layer thickness is used after the default thickness is used for the first layer or some initial number of layers. In certain examples, the build platform 52 is moved downward by a distance equal to the measured thickness of the previous layer. In other examples, the build platform 52 is moved down by a distance equal to an average thickness of several previous layers. In further examples, the average is a sliding window average wherein a defined number of thickness measurements is used to determine the average thickness, and as new layers are formed, earlier measured thicknesses outside of the window are discarded in the averaging calculation. In certain examples, platform 52 is moved by a certain amount to affect a desired change in the downward (along the build (z) axis) pressure applied by pressure roller 86 to achieve a desired layer thickness.
A contact sensor (not shown) such as a button may be provided on pressure roller 86 and used to determine the compressed layer thickness of each layer following lamination. Following the completion of a layer, the build platform 52 is elevated until contact is made with the sensor. Instead of using a contact sensor, the pressure sensors used for pressure roller 86 may also be used by lowering the build platform 52 until the object is exerting no upward force on the pressure roller 86 and then elevating the platform 52 until a pressure change is detected. The contact with the sensor (or the change in pressure) defines a specific build (z) axis location of the upper surface of the most recently formed object layer relative to the earth and the stationary components of apparatus 40. When the next layer is built, the process is repeated. The build platform 52 also has a position detector (or the position may be determined by the operation of the actuating mechanism used to raise and lower the build platform 52) so that the build platform positions between the two sensor contact events (or two pressure change events) may be used to determine the thickness of the most recently formed layer. In certain examples, the currently detected layer thickness is used to manipulate certain build parameters (discussed below) to drive the layer thickness toward the nominal compressed layer thickness for the lamination material 42 (which may be provided by the material supplier). However, in other examples, a sliding window average of the layer thicknesses is determined by using a specified number of the most recent thickness measurements and discarding those measurements that preceded the specified number. The sliding average is then compared to the nominal or set point value of the compressed layer thickness and used to adjust build parameters to drive the operation toward the nominal value.
Several different variables may be manipulated to control the layer thickness and drive it toward a desired or set-point value. The variables may be manipulated separately or in combination, and in various sequences. One of the variables is the pressure applied by pressure roller 86. The pressure may be manipulated by changing the distance Δz by which build platform 52 drops after laminating a layer. The lower the value of Δz, the higher the pressure applied downwardly by pressure roll 86, and the greater the compression of the thermoplastic or thermosetting component of the composite material. In one example, if the measured compressed layer thickness of the last layer or the average of some number of the last several measurements of the layer thicknesses indicates that the layer is too thick relative to a desired value, Δz (the distance by which build platform 52 drops each layer) may be decreased to increase the pressure applied by pressure roller 86 to the next layer to be laminated. Conversely, if the measured layer thickness is too thin relative to the desired (nominal) value, Δz may be increased to decrease the pressure applied by the pressure roller 86.
In another example, if the layer thickness is too thick relative to the nominal thickness, the speed of movement of the pressure roller sub-assembly 71 along the travel (x) axis may be slowed down. The speed of movement of the pressure roller sub-assembly 71 influences the amount of heat that is applied to the lower surface 112b of the current layer 108 (
In yet another example, the heat supplied by preheater assembly 105 may be adjusted to adjust the layer thickness of each layer. As mentioned previously, a non-contact temperature sensor may be used to measure a temperature at a point along the lower surface 112b of current layer 105 (such as at the junction described previously). That temperature may be raised by supplying more current to the infrared preheaters 106a-106c (or to inductive heaters if those are provided instead) and/or by slowing down the speed of movement of the pressure roller sub-assembly 71 along the travel (x) axis. In certain examples, the pressure, speed of travel (x) axis of pressure roller sub-assembly 71 and/or the heat output of preheater assembly 105 are manipulated to control the layer thickness. The temperature of the lower surface 112b is limited by the propensity of the binder to burn and is preferably maintained at a temperature that is lower by some specified amount than one at which burning will occur. In one possible control scheme, the lower surface 112b temperature is maximized to a value that is within a specified tolerance of the binder burn temperature, and the speed of movement of the pressure roller sub-assembly 71 along travel (x) axis is adjusted to achieve the desired layer thickness. In another exemplary control scheme, the lower surface 112b temperature is maintained at a value that is within a specified tolerance from the binder burn temperature, the speed of movement of the pressure roller sub-assembly 71 is set to range between 1 inch/second (2.54 cm/sec) and 6 inches/second (15.24 cm/sec) and is adjusted within those limits to achieve the desired layer thickness. If the pressure roller sub-assembly 71 reaches a high or low travel (x) axis speed limit, the pressure applied by the pressure roller 86 to the composite material is adjusted by adjusting the distance Δz by which the build platform 52 is moved between layers. In general, the pressure roller sub-assembly 71 travel (x) axis speed is from about 1 inch/second (2.54 cm/sec) to about 6 inches/second (15.24 cm/sec), more preferably from about 1 inch/second (2.54 cm/sec) to about 4 inches per second (10.16 cm/sec), and still more preferably from about 1.5 inches/second (3.82 cm/sec) to about 2.5 inches/second (6.35 cm/sec).
Referring again to
Thermoset composite materials may be laminated using heat and pressure in the same way described for thermoplastic composite materials. Alternatively, thermoset composites may be laminated using pressure alone, formed into object and waste sections and separated from the spool 42 using cutting apparatus 56 (
As mentioned previously, the object data that defines the three-dimensional object being formed on apparatus 40 is preferably converted to layer data that describes he object and waste sections formed on each layer of laminating material 42 and the patterns cut into the waste sections to facilitate their removal from object sections. The sum of the thicknesses of each layer correspond to the build (z) axis height of the three-dimensional object. However, as mentioned above, the actual (compressed) post-lamination layer thickness may deviate from the nominal value provided at the beginning of the object building process and may also vary to some extent during the build. Therefore, in certain examples, the three-dimensional object data is dynamically sliced (also referred to as being sliced “on the fly”) during the object building process. In accordance with one technique, a first portion of the object (along the build (z) axis) is sliced using a first layer thickness (such as the nominal value)). After data is collected (such as using a sliding window technique), an average layer thickness for the first portion of the object is used to slice a second portion of the object based on the as-built height of the first object portion, the expected height of the second object portion, and the average layer thickness calculated for the first object portion. This process may be carried out with varying numbers of object portions of varying heights. Using more object portions requires more computational to carry out multiple data slicing operations but yields more accurate parts.
An object having a build (z) axis height of 10 cm is defined by three-dimensional object data, such as STL data. The predicted (compressed, post-lamination) layer thickness is 100 microns (100×10−6 m). To illustrate the technique, a simple case involving only two dynamic slicing operations is used. However, any number of dynamic slicing operations may be used. A first portion of the object height comprising fifty (50) percent of the height (5 cm) is sliced using the 100 micron predicted layer thickness, yielding 500 layers. The layer thickness Δz is controlled using the techniques described previously (manipulating the speed of the pressure roller sub-assembly 71 along the travel (x) axis, manipulating the heat output from the preheater assembly 105, and/or adjusting the movement of the build platform after each layer Δz to increase the pressure applied by pressure roller 86). At the end of the 500 layers, the actual object height is 4.5 cm, and the sliding average layer thickness is 95 microns. The remaining object height to be built is 5.5 cm. Using the current sliding average layer thickness, the number of slices in the second object section is 0.055/(95×10−6)=579 layers. The remaining 5.5 cm of the object is sliced into 579 layers, each having a predicted layer thickness of 95 microns. The predicted layer thickness of 95 microns is used as the desired (set point) of the layer thickness controller (which may be an algorithm embodied in software as opposed to a specific hardware controller), and the layer thickness manipulated variables described above are manipulated (alone or in any combination) to achieve it.
In preferred examples of the present disclosure, a cutting assembly 56 (
Referring to
A y-axis translation support 78 is movable along the cross-member 76 along the y-axis and is attached to mount plate 82. Ultrasonic cutter 137 (
The horizontal base plate 135 is attached to mounting plate 82 via bracket 141. Blade 140 projects downward along the build (z) axis beneath the horizontal base plate 135. A closed loop stepper motor 147 controls the rotation of blade 140 about its longitudinal axis. Timing belt pulleys 149a and 149b have a timing belt 144 mounted thereon such that rotation of motor shaft 151 rotates pulley 149a, causing the timing belt 144 to move and adjust the rotational position of ultrasonic cutter 137, blade housing 153, and blade 140 about the longitudinal axis of blade 140 and the ultrasonic cutter 137, which axis is parallel to the build (z) axis. Bearings 155 are disposed between an inner bearing shaft 156 and an outer bearing shaft 158 and support blade housing 153 while allowing it to rotate. Upper bearing retainers 152 and 154 and lower outer bearing retainer 160 retain the bearings 155, the inner bearing shaft 156, and the outer bearing shaft 158.
Thus, during an object section cutting operation, the cutter translation assembly 54 travels along the travel (x) axis as the y-axis translation support 78 travels along the y-axis while blade 140 reciprocates at ultrasonic frequencies along the build (z) axis and rotates about its longitudinal axis. The extent of blade 140 travel along the build (z) axis as it reciprocates is preferably such that the blade performs a “kiss cut” and only cuts the current layer without cutting the layer immediately beneath it. In certain preferred examples, various blades can be selectively attached to the blade housing 153. The blades may vary in overall length and in their profile, i.e., the variation in their width as a function as position along their length. As one example, a v-shaped blade may be used which narrows in width to a point as you move along the build (z) axis toward the build platform. Different blade lengths and different blade profiles may be tailored to the thickness or other properties of the laminating material 42. Suitable motors are provided to translate the cutting assembly 56 along the travel (x) axis and the y-axis translation support 78 along the y-axis. In preferred examples, the motors are connected to a controller that operates in response to data representative of the three-dimensional object (such as vector data or G-Code data defining the contours of the each layer's object section (s)) being built so that that the object section defined by the blade 140 corresponds to the object. In addition to defining object section 252, cutting assembly 56 cuts the travel (x) axis border 257 (
In certain examples, apparatus 40 includes a blade changing assembly (not shown). The blade changing assembly is similar to known CNC tool changers and provides a mechanism for automatically changing blade 140 by moving the cutting assembly 56 to a location where new blades are available and causing the cutting assembly to dispense the existing blade 140 and pick up a new one. The blade changing assembly allows the apparatus 40 to use varying blades of varying width profiles during a single object building operation if so desired. In preferred examples, the controller that manipulates cutting assembly 56 includes a program that positions the tips of new blades to a reference position along the build(z) axis and also positions the blade in a reference rotational orientation about the longitudinal axis of the blade, and in some cases, at a reference x, y position. From the reference position and orientation (xr, yr, zr, wr), the blade can be moved to the x, y location where cutting is desired and placed in the proper rotational orientation and at the proper build (z) axis height. In preferred examples, a button or other contact mechanism of fixed height relative to a portion of apparatus 40 that is fixed along the build (z) axis is used to align a new blade's tip along the build (z) axis. In one case, a button is placed outside the build envelope, and the tip of blade 140 is moved downward until contact is detected, at which point the tip of the blade 140 is at the reference build (z) axis position. As the button has a fixed x, y position, such contact fixes the location of the tip in all three coordinates: x, y, and z. Another sensor may be provided which detects the rotational orientation w of the blade 140 so that the blade can be rotated to a reference rotational orientation about its longitudinal axis. By using the rotational sensor and the button, a reference condition for all four degrees of freedom is defined and can then be related to a desired rotational orientation, blade position in the build envelope, and build (z) axis height so that the blade may be moved and rotated appropriately to arrive at a desired cutting location in the build envelope in the correct rotational orientation, the correct x, y, and z locations. In other words, every x, y, and z position and rotational orientation used to create an object section or to cut a waste section may be related to the reference condition xr, yr, zr, wr.
In certain preferred examples, ultrasonic cutter 137 includes a sensor that measures the actual frequency of reciprocation of blade 140 along the build (z) axis. A comparator circuit (which may be in software) compares the measured frequency of reciprocation and compares it to a nominal frequency or setpoint. The comparator signal is then output to a suitable controller and display unit. If the difference between the measured frequency and the nominal frequency is greater than a specified threshold, then the blade is presumed broken, and the controller manipulates the cutting assembly 56 to pick up another blade and discard the current blade.
Over time, particular blades 140 my wear out and become unsuitable for further use. In certain examples, an algorithm determines the amount of wear a blade 140 has endured based on a number of variables, including at least one of the number of hours of operation, the total lineal feet of cutting, and a materials parameter indicative of the difficulty of cutting the laminating material 42. A controller may be provided which comprises a processor and a non-transitory computer readable medium with computer executable instructions stored on it which, when executed by the processor, determine a level of blade wear. When the level of blade wear exceeds a certain threshold, the controller causes the cutting assembly to pick up another blade and discard the current blade.
An example of an object section is shown in
In certain preferred examples, an adhesion reducing material is applied along the interface 255 to better enable the removal of waste section 253 from object section 252 without damaging object section 252. The adhesion reducing material is preferably one that is selected based on the binder (polymeric matrix) of laminating material 42 to disrupt adhesion between object section 252 and waste section 253 such that following lamination, regions with the adhesion reducing material will not adhere or bond together. In one example, silicone oil is used as the adhesion reducing material. Other suitable adhesion reducing materials include natural or synthetic paraffin waxes. As discussed further below with respect to
In certain preferred examples, a printhead movable along the travel (x) and y axes is used to apply the adhesion reducing material to interface 255. When a printhead is used, the adhesion reducing material preferably has a viscosity that makes it jettable. The printhead also preferably includes an internal heater for controlling the viscosity of the adhesion reducing material. An exemplary printhead 162 is shown in
In certain examples, a controller is provided which comprises a processor and a non-transitory computer readable medium having instructions stored thereon, and when executed by a processor, the instructions cause printhead 162 to selectively open the printhead openings 172a-172v to dispense the adhesion reducing material. The pattern of the deposition is dictated by object data that defines the geometry of interface 255 as adjusted for the offset between the printhead 162 and the blade 140.
The cutter translation assembly 53 and y-axis translation support 78 respectively move the cutting assembly 56 in the x and y directions. With many three-dimensional objects, object sections will have curved object contours, and a technique is required to convert the discrete orthogonal movements of the cutter translation assembly 53 and the y-axis translation support 78 into curved paths. In one example, linear interpolation is used. With linear interpolation, a curved path is translated into a series of short linear paths. As the distance of each movement approaches zero, the path approaches a true curve.
In another example, circular interpolation is used in which the curved path is translated into a series of second degree polynomials. However, circular interpolation is limited to second degree curves. Thus, in a preferred example, curved paths are defined using spline interpolation, which allows for higher order interpolation. For example, cubic or higher order splines will more closely approximate cutting paths having an inflection point (where the second derivative y″(x) is zero) because the splines will themselves have an inflection point. In preferred examples herein, the object contours in each layer are approximated using splines that define the entire object contour as a single curve. Preferably, the speed of movement of the cutting assembly 56 is varied and optimized based on the degree of curvature at a given location, with the speed increasing at relatively straighter (high radius of curvature) regions than at relatively curved (low radius of curvature) regions.
Referring to
Table lift 50 provides a frame for mounting build platform 52 and allowing it to translate along the build (z) axis and rotate in the x-y plane. Rotary table 98 is mounted on threaded shafts 94a and 94b which engage corresponding threaded portions of rotary table 98. Rotary table 98 includes a rotating carriage 62 which is rotatable in the x-y plane relative to table lift 50 and the remainder of rotary table 98. Rotation of shafts 94a and 94b causes build platform 52, vacuum plate 58, and rotary table 98 to translate along the build (z) axis. A motor (not visible) rotates horizontal shafts 91a and 91b (not shown). Suitable gears translate the rotation of horizontal shafts 91a and 91b into the rotation of vertical shafts 94a and 94b to adjust the build (z) axis position of build platform 52. Rotary table 98 also includes linear bearings (not shown) which engage rails 95a, 95b (not shown), 97a, and 97b (not shown). Rails 95a and 97a are mounted on mounting plate 99a, and rails 95b and 97b are mounted on mounting plate 99b.
Rails 95a and 95b are spaced apart along the travel (x) axis and positioned at the same location along the y-axis. Rails 97a and 97b are spaced apart along the travel (x) axis and positioned at the same location along the y-axis. Rails 95a and 97a are spaced apart along the y-axis at the same travel (x) axis position. Rails 95b and 97b are spaced apart along the y-axis at the same travel (x) axis position.
Cutting assembly 56 forms the object section for current layer 250 and the waste section for current layer 250. As can be seen by comparing
As mentioned previously, the material advancement assemblies 124a and 124b have grippers 146a and 146b which can grip a free edge of laminating material 42 on spool 46. The grippers 146a and 146b preferably pull the free edge either to the edge of the build envelope or to the farthest position along the travel (x) axis at which the object will be present for the current layer. Note that the x and y axes remain fixed when the build platform 52 rotates and are defined by the direction of travel of the pressure roll 89 and the cutter translation assembly 54. However, the object data may be defined in a Cartesian coordinate system that is based on the build platform 52, thus requiring translation when the platform 52 rotates in the x, y plane.
In order for the grippers 146a and 146b to pull the free edge of the laminating material 42, they have to pass the pressure roller 89 and guide roller 102 of the laminating assembly 70. When the pressure roller 89 and guide roller 102 are in the orientation of
In certain examples, an adhesion reducing material of the type described previously is used to prevent adhesion between object sections and waste sections in immediately adjacent layers of laminating material 42. Referring to
Section 194 of layer n+1 is an object section. Adhesion reducing material 187 does not cover the entire downward facing surface of object section 194. Instead, two small edge portions or “spot welds” 199a and 199b are maintained. However, when the lamination steps are complete the entirety of layer n will be separated from the finished object, leaving behind the entirety of object section 194, including the edge portions 199a and 199b, as part of the finished object. By maintaining adhesion between layers n and n+1 at edge portions 199a and 199b, the waste sections and object sections remain an integral whole until the lamination of all object layers is complete. The waste sections thus act as supports for downward facing surfaces of the object and remain in place until all of the object sections have been formed.
Layer n+1 is then deposited on and laminated to layer n using the pressure roller sub-assembly 71 as described previously. Kiss cut 233 is made in layer n+1 to define object section 230 and waste section 234. Waste section 234 is cubed or otherwise cut appropriately by cutting assembly 56 to facilitate its removal from the finished object. Adhesion reducing material is preferably deposited in the object section/waste section interface defined by kiss cut 233 in the manner described previously.
Object section 230 includes an overhang 231 which is defined between kiss cut 229 in layer n and kiss cut 233 in layer n+1. Overhang 231 is a region that will be unsupported once the object is finished and waste section 236 of layer n is removed. Thus, a first portion of waste section 236 lies underneath and support overhang 231 of layer n+1 and another portion 237 of waste section 236 lies underneath waste section 234 of layer n+1. The adhesion reducing material 235 is applied only in a central region of overhang 231 leaving edge portions (or spot welds) 238a and 238b bound to waste section 236 of layer n. As with edge portions 199a and 199b in
One advantage of using the apparatuses and methods described herein as opposed to stereolithography or other photopolymer-liquid hardening 3D printing methods is that “nested” parts may more easily be produced. “Nested” parts are parts having the same relative geometry but different overall dimensions so that one part may fit inside the open space of another part, like a stack of successively smaller bowls stacked one inside the other. With techniques that harden a photopolymer, removable supports must be created between each successive object to keep them spaced apart. The creation of removable supports from the photopolymer adds additional processing time. With the techniques described herein, laminating materials 42 that would ordinarily become waste sections may be formed into a nested object and easily separated at the end of the build process (because the waste sections are formed into readily removable shapes like cubes). In LOM processes such as those described herein, minimization of waste sections is desirable because the waste sections cannot be reused to form three-dimensional objects and are discarded. In the case of a single bowl formed with LOM, the entire interior of the bowl would comprise waste. Thus, any additional nested bowls that can be made will necessarily reduce the amount of waste. This waste concern does not apply in photopolymer based systems because any material that is not used to form the three-dimensional object during a given object build process remains available to build subsequent objects and is not wasted. Thus, not only is nesting more important in LOM processes of the type described herein than in photopolymer based processes, but it can be done more quickly and efficiently because waste sections which act as supports between the nested objects in the methods of the present disclosure do not need to be solidified as is the case with photopolymer object supports.
Methods of making three-dimensional objects by laminating composite materials will now be described. Although not shown in the methods of
In step 1012 an adhesion reducing material is dispensed along the interface(s) between the object section and the waste section. In addition, if portions of the previous layer comprise an object section or waste section which will abut the other of an object section or a waste section in the next layer, all or some of those portions of the previous layer may have the adhesion reducing material applied to them. As explained previously, in certain examples, it is desirable to leave edge portions (spot welds) of abutting object and waste sections in adjacent layers untreated with an adhesion reducing material so that the waste sections provide object support until the object is complete and ready for removal.
In step 1014 a next layer of laminating material 42 is laminated onto previous layer of laminating material 42. In preferred examples, the surface of the current layer of laminating material 42 which faces a surface of the previous layer of laminating material 42 is heated (such as by using preheater assembly 105) to cause the binder (polymeric matrix) component of the current layer to adhere to the binder of the previous layer as pressure roller 86 rolls and translates on the upward (along the build (z) axis) facing surface of the current layer of laminating material 42. The current layer is then subjected to a forming operation (such as by using cutting assembly 56) to create an object section and a waste section in the current layer. The forming operation is carried out in accordance with data representative of the three-dimensional object. The waste section is cut in a pattern (e.g., cubes) that facilitates its removal from the finished object.
In step 1015 the adhesion reducing material is applied to the interface(s) between the object section and the waste section. In step 1016, a determination is made as to whether the last layer of the object has been completed by comparing the current layer index k to the maximum layer index kmax. If the last layer has been completed, the process ends. Otherwise, control transfers to step 1017 where the layer index k is incremented.
Another method of making a three-dimensional object in accordance with the present disclosure is described in
In step 1020 the build platform 52 is rotated (counterclockwise in the example of
In step 1022 a second layer of composite material is provided with its fiber length axis 254 oriented along the same axis (the travel (x) axis) as shown in
Referring to
The temperature of the pressure roller 86 surface is selected to be below the glass transition temperature Tg of the polymeric matrix component of the laminating material 42. In one example using PEEK carbon fiber fabric as laminating material 42, the junction temperature where lower surface 112b of current layer 108 meets upper surface of 114b of previous layer 110 is 735±15° F. (390±8.3° C.), and the pressure roller surface temperature is less r than the glass transition temperature 289° F. (143° C.). The pressure roller sub-assembly 71 is then translated along the x-axis with the guide roller leading the pressure roller as depicted in
In step 1034 a forming operation is carried out, such by using cutting assembly 56. The forming operation creates object contours that define an object section and a waste section on the layer which abut each other along an interface. The contours are defined by vector data. When cutting assembly 56 is used for the forming operation, the interface is where the cutting occurs. The waste section is then cut into removable shapes, such as cubes, to facilitate its eventual removal from the finished object.
In step 1036 an adhesion reducing material of the type described previously is applied to the intra-layer object section/waste section interface(s). In preferred examples, the adhesion reducing material is also applied to inter-layer object section/waste section interfaces, i.e., regions of waste sections which will abut object sections in the next layer and/or regions of object sections that will abut waste sections in the next layer (step 1040). In certain preferred examples, adhesion reducing material is not applied at the edges of such abutting regions, thereby creating spot welds that maintain the stability of the waste sections and object sections as the object is being built so that the waste regions do not prematurely separate from abutting object sections in an adjacent layer before all layers are laminated.
In step 1042 the current layer is separated from the composite material spool 46 such as by using cutting assembly 56. A determination is made as to whether the current layer is the last layer by comparing the value of the current layer index k to the value of the maximum layer index kmax. If the current layer index value equals the maximum layer index value, the process ends. Otherwise, control transfers to step 1045 and the layer index is incremented by one. In some examples wherein reinforced fiber plastics or pre-pegs with parallel fibers in each sheet are used as the laminating material 42, before step 1026 is repeated, the build platform may be rotated (depending on the strength requirements of the object being built) as described previously with respect to
In certain examples, wherein the x-y area of the object being built decreases as you move upward along the build (z) axis, “smart advancing” may be used wherein the free edge of the composite material (e.g., free edge 261 in
The present invention has been described with reference to certain exemplary embodiments thereof. However, it will be readily apparent to those skilled in the art that it is possible to embody the invention in specific forms other than those of the exemplary embodiments described above. This may be done without departing from the spirit of the invention. The exemplary embodiments are merely illustrative and should not be considered restrictive in any way. The scope of the invention is defined by the appended claims and their equivalents, rather than by the preceding description.