Method and apparatus for determining the condition of metal powder

Abstract

Determining the condition of a metal powder 33 in additive manufacturing by dark field illuminating a region of powder with electromagnetic radiation; detecting the intensity of radiation reflected ba

Description

Technical Field of the Invention The present invention relates to an apparatus and method for determining the condition of a metal powder. In particular, the present invention relates to an apparatus and method for determining if and the extent to which a metal powder for use in an additive manufacturing process has degraded. Background to the Invention In a known additive manufacturing process an additive manufacturing machine produces articles from a powdered metal or alloy. The machine deposits a layer of metal powder on a build platform and the metal powder is subsequently selectively fused with a laser or electron beam to form an article or articles. The process is repeated so that articles are formed layer by layer. On completion of a build, unfused powder may be re-used in another build. The composition and condition of metal powder used in a build process can have a significant effect of the integrity of an article formed by the process. During a build operation unfused metal powder is subject to degradation. A metal powder may gradually oxidise which alters its properties and thus those of an article produced from the powder. The tendency of a powder to oxidise typically increases with temperature, and exposure to temperature may also affect other powder properties. Consequently, the nearer unfused powder is to an article being built, or a heat zone, the more likely it is to suffer degradation. Also, when powder is fused the process may cause some heated particles of powder to be scattered on to the powder bed around the manufactured article, degrading the quality of the unfused powder around the article. To ensure adequate build quality of an article it is known to analyse used powder and stop recycling the powder when it has been degraded to a certain extent and / or to blend virgin powder with recycled powder so that the blended powder has an adequate bulk property for continued use. Powder condition is typically analysed by making a bulk oxygen content measurement. The measurement process involves testing a powder sample, which cannot then be re-used. As this test method is destructive the results never relate to powder that is re-used and so approximate the condition of re-used powder. Other aspects of powder composition and condition can also effect build quality. The presence of contaminant particles can have a similar effect to the presence of highly oxidised particles. Particle size and shape can also affect a build, as can the flowability, temperature and density of the powder. WO2019 / 097222 discloses a method of determining the condition of a metal powder for use in an additive manufacturing process involving processing an image of the powder to measure the colour of particles of the powder, which can be related to the degree of oxidation of the powder. In practice, though, obtaining a meaningful assessment of oxidation using the disclosed method is difficult or impossible. It is an object of embodiments of the present invention to address some or all of these problems. In particular, it is an object of embodiments of the invention to provide an improved method and apparatus for determining the condition of a metal powder in a non-destructive way and without exposing the metal powder to the external environment. It is a further object of embodiments of the present invention to determine the extent to which metal powder has oxidised. Summary of the Invention According to a first aspect of the invention there is provided a method of determining the condition of a metal powder for use in an additive manufacturing process, the method comprising the steps of: dark field illuminating a region of the powder with electromagnetic radiation; detecting the intensity of the electromagnetic radiation reflected back from the metal powder; and processing the detected intensity of the reflected electromagnetic radiation to determine the condition of the metal powder. According to a second aspect of the invention there is provided an apparatus for determining the condition of a metal powder for use in an additive manufacturing process, the apparatus comprising: an illumination device for illuminating a region of the powder with electromagnetic radiation; a detector for detecting electromagnetic radiation reflected from the powder; wherein the detector and the illumination device are arranged so that the detector is unable to detect electromagnetic radiation emitted directly from the illumination device; and a processor configured to determine the condition of the powder based on the intensity of the detected electromagnetic radiation. By dark field illuminating metal powder with electromagnetic radiation and detecting the intensity of the radiation reflected back from the metal powder, it is possible for a user to determine or infer the extent to which a metal powder has degraded in a non-contact and non-destructive manner. Advantageously, is has been found that the intensity of the reflected radiation is associated with the thickness of an oxide layer formed at the surface of the metal powder. Measuring the intensity of the reflected radiation thus allows a user to determine the thickness of the oxide layer, and so the extent to which the powder has oxidised. In some embodiments it also allows a user to the determine the condition of individual particles of powder and the extent to which they have degraded. The method may comprise the step of illuminating, and the illumination device may be configured to illuminate, the powder with radiation having a wavelength between 300 and 1500 nm. In some embodiments the powder is illuminated with radiation having a wavelength between 600 and 1300 nm. Suitably, the radiation may have a wavelength between 700 and 1100 nm. The method may comprise the step of illuminating, and the apparatus may illuminate, the powder with only a single wavelength of radiation or with wavelengths of radiation within a narrow band. Alternatively, or additionally, the method may comprise the step of detecting, and the detector may be configured to detect, only a selected wavelength or selected wavelengths in a narrow band. This may be achieved by filtering the illuminating or reflected electromagnetic radiation with a filter so that the detector receives only a selected wavelength of radiation wavelengths of radiation within a narrow band. Or a detector could be employed that is only, or predominately, sensitive to the wavelength or narrow band of wavelengths of interest. A single wavelength may be a nominal single nm wavelength. A narrow band may be up to lOnm, or 20 nm wide. The method may comprise the step of selecting the wavelength or wavelengths of electromagnetic radiation based on a maximum thickness of an oxide layer formed on the metal powder it is desired to detect. The selected wavelength may be at least four times the maximum oxide layer thickness. In some embodiments, the wavelength may be at least six times the maximum oxide layer thickness. The maximum oxide-layer thickness may be 80 nm or less. In some embodiments the wavelength or wavelengths of electromagnetic radiation may be selected based on the predicted refractive index of an oxide layer of interest and / or the predicted refractive index of the metal powder to be tested. In other embodiments, the wavelength or wavelengths of radiation selected may be based on the intended angle of illumination of the radiation on the metal powder to be tested. The powder may be illuminated, and the apparatus arranged to illuminate powder, with electromagnetic radiation at an acute illumination angle to the surface of the powder greater than 40°, so from 40° to 90°. In particular, the acute illumination angle may be between 40° and 75° or 40° and 60°. The powder may be illuminated by, and the illumination device may emit, unpolarised electromagnetic radiation. The powder may be illuminated with, and the illumination device arranged to emit, a substantially parallel beam of electromagnetic radiation. The method may comprise the step of determining, and the processor may be configured to determine, the thickness of an oxide layer formed on the surface of the metal powder based on, or only on, the detected intensity of the reflected electromagnetic radiation. In some embodiments the method and apparatus enable dark-field images of the powder to be obtained. This allows a user to determine or infer the extent to which individual metal powder particles have degraded with improved accuracy in a noncontact and non-destructive manner. The apparatus may comprise a shroud which in use prevents the detector from directly detecting electromagnetic radiation emitted from the illumination device. The detector and the illumination device may be housed within an enclosure. The enclosure may comprise a window for separating the powder from the detector and the illumination device. The window may be transparent or at least partially transparent to the wavelength or wavelengths emitted by the illumination device. In some embodiments the window may comprise glass, sapphire crystal or magnesium fluoride crystal. The apparatus may comprise at least two or three illumination devices. The illumination devices may be evenly spaced around the detector. The shroud, where present, may be disposed between the or each illumination device and the detector. The shroud may be annular. The illumination device or devices may be arranged to emit electromagnetic radiation towards the window. Suitably, the electromagnetic radiation is directed to illuminate metal powder adjacent the opposite side of the window to the illumination device or devices at a desired angle of illumination. 5 The or each illumination device may comprise one or more lamps such as LEDs. The window may comprise flat, substantially parallel sides, adjacent one of which, in use, powder to be analysed is disposed. The processor may be arranged to cause the apparatus to perform the method according to the first aspect of the invention. 10 The apparatus may be comprised in a powder transport container, conduit or an additive manufacturing machine. In some embodiments the apparatus is comprised in a pipeline. For example, the pipeline may lead from a powder transport container to an additive manufacturing machine. In some embodiments the pipeline may lead into or out from a sieve. The apparatus may comprise a sampling body. The apparatus may be attached to or integrated within a sampling body which is configured to limit the rate at which powder flows across the apparatus. The sampling body may be in the form of a conduit having a least one flat internal surface. Part of the flat surface may be formed by a window of the apparatus. Suitably, the sampling body may be in the form of a pipe. The sampling body may comprise an entrance, an exit and a passage extending between the entrance and the exit. The passage may be at an oblique angle relative to the longitudinal axis of the sampling body. At least a portion of the passage may have a cross-section that is smaller than the cross-section at the entrance or exit. The apparatus may be positioned to test powder at the angled portion of the passage, or portion of reduced cross-section. In this way, the flow of powder through the passage can be controlled as it passes across the apparatus, resulting in improved accuracy when determining the condition of the powder. The apparatus may be attached to or integrated within a powder transport container, additive manufacturing machine, or pipeline. The apparatus may be comprised in a device comprising one or more other apparatuses for determining the condition of and / or mechanical properties of metal powder. In this respect the apparatus may comprise apparatuses for determining flowability, temperature and / or density of metal powder. Advantageously, by determining the extent of degradation / oxidation, flowability, density and the temperature of metal powder a user can make a better-informed decision on whether the powder is suitable for use in an additive manufacturing process. The apparatus may be operatively connected to a communication network, for example a local area network. For instance, there may be a data connection between the apparatus and a local powder processing apparatus such as an additive manufacturing machine, blender, sieve or conveyance system. This may be achieved by using an industry standard protocol such as Modbus or I / O-Link. In this way, status information and point measurements can be communicated when polled. The apparatus may be operatively connected to a wide area network. For instance, the apparatus may be wirelessly connected to a cloud server so that the degradation measurements (and any other measurements that can be used to provide an indication of a powder’s health) can be logged. This also enables the apparatus’s firmware to be updated when needed and allows calibration information to be downloaded. The apparatus may comprise a power source, e.g. a rechargeable or non-rechargeable electric battery e.g. anon-rechargeable lithium primary cell which powers the apparatus during transit when connected to a powder storage and / or transport container. The apparatus may comprise a power connection to an external power source, e.g. a DC electrical connection to an external power source, such as a local powder processing apparatus, e.g. additive manufacturing machine, blender, sieve or conveyance system. Detailed Description of the Invention In order that the invention may be more clearly understood one or more embodiments thereof will now be described, by way of example only, with reference to the accompanying drawings, of which: 5 Figure 1 is a perspective view of apparatus for determining the condition of metal powder, with a cover shown separated from the apparatus; Figure 2 is a schematic block diagram showing components of the apparatus of figure 1 10 Figure3 is a longitudinal cross-section taken along the line A-A of figure 1; Figure 4 is an end view of the apparatus of figure 1; Figure 5 is a transverse cross-section taken along the line B-B of figure 3, with the cover removed, with an enlarged region; Figure 6 is an enlarged view of part of Figure 3; Figure 7 is a schematic cross-sectional view of the enlarged region of figure 5; Figure 8 is a graph showing reflectance against oxide thickness; Figure 9 is a schematic view of apparatus for determining the flowability of a metal powder comprised in the part shown in figure 3; Figure 10 is a schematic view of an embodiment of apparatus for determining the temperature of a metal powder comprised in the part shown in figure 3; Figure 11 is a schematic view of an embodiment of apparatus for determining the density of a metal powder comprised in the part shown in figure 3; Figure 12 is a schematic view of a conduit comprising the apparatus shown in Figure 1. Figure 13 is a partially transparent view of the apparatus shown in figure 1 as depicted in figure 12; Figures 14 to 16 are schematic views of a conduit arrangement comprising the apparatus shown in Figure 12 in different states; and Figures 17 is a side view of a metal powder container comprising apparatus for determining the condition of a metal powder expanded into an enlargement. Referring to the drawings, figures 1 to 11 show an apparatus 1 for determining the condition of a metal powder. The apparatus comprises a sampling body 2 which defines a conduit 2a for conveying metal powder in use. The conduit 2a comprises an inlet 3 and an outlet 4. Each defines a circular opening. The internal surfaces of the circular inlet out outlet openings transition smoothly to the internal surface of the conduit extending between the openings, which has a substantially D-shaped cross-section. That is to say. there is a flat, planar internal surface 6 to the conduit 2a, which transitions through smooth 90° curves at opposite longitudinal edges to form two opposed side walls which are straight and parallel to each other adjacent the planar surface and then transition into a generally semi-circular shape to join up. The conduit is formed by a machined metal sampling body 2 with a generally part-circular external cross-section with a single flat face 23 which extends parallel and adjacent to the internal flat surface 6 of the conduit 2a. In use the apparatus is typically arranged so that the flat planar internal surface 6 is lowermost, to form the bottom of the conduit 2a. The inlet 3 and outlet 4 comprise flanged connections enabling the apparatus to be connected in-line with another conduit or other apparatus for conveying metal powder. The part of the conduit 2a with a D-shaped cross-section has an overall cross-sectional area smaller than that of the inlet and outlet. A housing 7 with a lid 8 partially surrounds the sampling body 2 and, as best shown in figure 1, the housing 7 encloses a circuit board 9. The circuit board mounts components for analysing metal powder in the sampling body and associated electronics 10 including a processor 11 with memory. The circuit board also mounts an electric battery 12 and super capacitor 13 forming part of a power supply, global positioning system electronics (not shown), communication electronics 14, local communications connector 15, sim card 16, cellular modem 17, antenna 18, status LEDs 20 which are visible through corresponding openings 22 on the lid when fitted, and drivers and sensing circuit apparatus associated with various metal powder conditions sensing apparatus described further below. The circuit board extends parallel and adjacent to the flat external surface 23 of the sampling body 2. A first substantially circular aperture 24 is formed through flat side of the sample body at a position approximately mid-way along its length with a counter-bore 25 in the outside surface of the sampling body. A magnesium fluoride crystal window 26 is fitted in the counter-bore along with an O-ring 35 to form a seal between the window 26 and the sampling body. Other suitable materials can be used, eg. Saphire crystal, depending on the specific application of the apparatus. A spacer 27 is fitted over the flat external surface of the sampling body, configured to hold the window 26 and O-ring 35 in place. The circuit board 9 is fitted over and abuts the spacer 27. Coaxial circular apertures are formed through the printed circuit board 9 and spacer 27 centred at the centre of the window, with a diameter of about one twentieth of that of the window. A detector 29 is mounted on the side of the printed circuit board 9 facing away from the window over the aperture through the circuit board, so that it faces the window 26. The detector includes a focussing element 28, such as a lens, which is extends into the aperture, towards the window, and serves to focus received radiation onto the detector. In the illustrated embodiment the detector is a point sensor e.g. a photodiode. In other embodiments the detector is an image sensor such as a multi pixel sensor e.g. a CCD or CMOS sensor. An inner shroud 30 extends around the aperture on the side of the printed circuit board 9 facing the window 26, sitting in the corresponding aperture in the spacer 27, and extends from the circuit board to contact the window. The inner shroud 30 is annular in cross-section and has a substantially cylindrical inner surface. The wall of the illustrated inner shroud 30 tapers in thickness as it extends from the printed circuit board so that it defines a slightly diverging passage from the circuit board to the window with an inside diameter the same as that of the aperture through the circuit board where it meets the circuit board. The shroud has a frustro-conical external shape that tapers toward the window. This shape facilitates moulding the inner shroud 30 from plastics. Functionally the inner shroud 30 need only be a thin-walled cylindrical tube. The spacer 27 forms an outer shroud which extends around and concentrically with the inner shroud, extending parallel to but spaced just inside the outer edge of the window 26, and extending from the circuit board 9 to the window 26. The inside diameter of the outer shroud is about three times that of the maximum outside diameter of the inner shroud 30. There is therefore an annular region of circuit board 9 between the two shrouds. In this region three LEDs 31 are mounted on the circuit board, adjacent and equally spaced around the outer shroud. The LEDs 31 are arranged to direct light towards the centre of the opposite side of the window to the printed circuit board, and the apparatus is sized and configured so that light is directed by the LEDs is refracted to form a substantially parallel beam as it enters the window 26, the beam extending at an acute angle 32 to the surface of the window of about 41°. In the illustrated embodiment the LEDs have a beam spread of about 28° and are positioned so that the edge of the beam defines an outer edge of the area of the surface of the window adjacent to the LEDs, and the inner shroud 30 defines the inner edge of that region. This, in turn, ensures that a desired limited area of interest on the opposite side of the window is illuminated. In use, metal powder 33 to be analysed is introduced into the sampling body so that it contacts (and ideally is close packed against) the window 26. The powder may be static, or moving. The LEDs 31 are activated and emit a single wavelength (850 nm) of infrared radiation. The resulting beam illuminates particles of metal powder 33 lying adjacent the outside of the window 26 in the area of interest opposite and within the viewing area of the detector 29. Radiation reflected by the particles is received by the detector 29. The chosen wavelength and angle of illumination in this embodiment is suited to analysis of a titanium alloy. Other wavelengths and illumination angles may be chosen for analysis of other metal powder compositions, e.g. a nickel alloy is conveniently illuminated with a radiation having a wavelength of 850 nm at an illumination angle of about 56°. The inner shroud 30 ensures that light from the LEDs is not directly received by the sensor, and the outer shroud ensures that no ambient light is received by the sensor, thus creating dark field illumination of the metal powder. In use when the apparatus is connected in line with powder handling / processing apparatus no ambient light will enter the sampling body via the inlet or outlet. The sensor measures the intensity of the radiation it receives. The lens 28 associated with the sensor is configured to focus light reflected by the metal powder 33 in the area of interest on to the sensor 29. The radiation received by the sensor 29 is the sum of the radiation reflected by any oxide surface of the individual particles of metal powder and radiation reflected by the underlying metal / alloy surface of the individual particles of the metal powder. Where an oxide layer has a thickness of one quarter (or less) of the wavelength of the illuminating radiation, destructive interference will occur at the oxide surface between radiation reflected by the oxide surface and radiation which has passed through the oxide surface and reflected back from the underlying metal / alloy surface, leading to a sharp reduction in the intensity of reflected light received by the sensor. Where the thickness of the oxide layer is less than a quarter of the wavelength of illuminating radiation, as the oxide layer becomes thicker the intensity of reflected light decreases, and will reach a minimum as the oxide layer thickness approaches one quarter of the wavelength. So, by choosing a relevant wavelength for a particular metal powder type and oxide layer thickness, there is a strong correlation between the intensity of reflected radiation received by the sensor and the thickness of the oxide layer, and thus the degree of oxidation of the metal powder. Figure 8 shows reflectance on the y axis against oxide thickness on the x axis for various metal powders. In each case there is a generally linear, or at least a straightforward relationship, over the range of oxide thickness of interest allowing oxide thickness to be reliably determined by measuring the intensity of reflected radiation. As the detector 29 is a point sensor, the measured intensity effectively represents an average oxide layer thickness for all of the illuminated particles within the area of interest within the view of the sensor that reflect light back towards the sensor. Point sensors are suited to taking measurements when metal powder is both static and moving past the window. In the latter case the processor may cause the sensor to take periodic measurements which are averaged overtime, so, for example, an intensity measurement may be taken every 0.2 seconds and a mean of five values calculated every second. Mean values obtained over 15 seconds may then be used to calculate measurement statistics. Where a multi-pixel sensor is used this can record a dark field image of the metal powder being analysed. In an example a sensor having an array of 320 x 320 pixels is used to take an image every 0.2s. The image may then be analysed to determine the intensity of radiation reflected by individual particles enabling statistical information relating to oxide layer thickness to be calculated. The darkfield image also enables other characteristics of particles to be determined through image analysis, e.g. particle size and shape distributions. For both single point and image sensors illumination with and / or detection of a single or narrow band of radiation in a dark field arrangement enables oxide thickness to be determined by measuring only the intensity of reflected radiation. This requires relatively simple processing owing to a strong and in most cases approximately linear relationship between intensity and oxide layer thickness, e.g. oxide thickness may be determined from a look-up table. The apparatus also includes apparatus 40 for analysing the flow characteristics of a metal powder, positioned between two of the LEDs 31. This apparatus is shown schematically in figure 9 and comprises a vertical-cavity surface-emitting laser diode (VCSEL) 41 and a speckle image sensor such as a CCD or CMOS sensor 42 mounted alongside each other on the circuit board 9 behind the window 26. The sensor 42 is a “bare” sensor, in that it does not have any associated focussing optics. It will therefore capture a speckle pattern caused by light reflected by the metal powder, rather than an image of the powder. The spacer 27 forms a shroud extending around the laser diode 41 and sensor 42, and from the circuit board 9 to the window 26. The shroud forms a generally square cavity which accommodates the laser diode and sensor, the sides of which diverge away from the circuit board towards the window. In use the laser diode 41 emits a well collimated narrow beam of near infrared radiation 43 towards and through the window 26 to illuminate moving metal powder adjacent the opposite face of the window. The beam illuminates a spot of diameter about 2mm on the opposite face of the window. The radiation 43 is reflected and / or scattered back through the window 26 by the metal powder where it is detected by the speckle image sensor 42. The area of illumination and the relative position of the image sensor are chosen so that substantially the whole area of the image sensor is illuminated by radiation reflected back onto the sensor by the power. This improves the signal to noise ratio detected by the sensor. Speckle images are obtained as the powder 33 flows past the window 26. These are stored on the speckle image sensor where an internal state-machine computes the difference between the speckle patterns of the frames and thus the rate of movement of powder particles across the window 26 enabling flow speed of the powder to be calculated, e.g. in mm / s. In the described embodiment the processor 11 polls the sensor every second. The sensor takes five images per poll. These are stored on the sensor and the internal state machine computes the difference between sequential images to determine the rate of movement of powder particles, which is transmitted to the processor 11. If the measured rate of movement exceeds a threshold (e.g. lOmm / s but other thresholds could be chosen) the powder is determined to be moving the rate that images are captured is increased. The poll rate may be increased . The poll rate can be increased as desired but may typically be increased so that readings are taken from 4 to 25 times every second. And / or the rate of image capture on the sensor may be increased. For example, the sensor may capture an image at least every 10ms, and accumulated images are polled by the processor at least every 100ms. The rate of movement can then be calculated. If the calculated rate of movement falls below the threshold the powder is treated as no longer moving and image capture rate / poll rate is decreased. Based on the measured flow speed and the (known) cross-section of the sampling body where the sensor 42 is located, an indication of the metal powder’s flowability can be inferred, and a volumetric flow rate (e.g. in mm3 / s) of metal powder through the apparatus calculated. In other embodiments speckle images, or optical images of the powder, may be transmitted to the processor 11 for analysis instead. The laser diode 41 only illuminates the metal powder when not illuminated by the LEDs 31 used to measure oxide layer thickness, so that one light source does not adversely affect measurement made by using the other light source. By monitoring the flow rate of the metal powder 33 it is possible to detect inline blockages, or poorly flowing material. The latter is a key indicator of a metal powder’s ability to consistently spread on a metal powder bed of an additive manufacturing machine. This helps to minimise or prevent damage to an additive manufacturing machine or to a component of an additive manufacturing system. It also helps to ensure that additively manufactured products of sufficient quality are produced. Flow rate of the metal powder will be affected by the angle of slope of the apparatus. The angle of slope may be processed together with measured flow rate to provide a more accurate determination of the metal powder’s flowability. In some installations the slope of the apparatus will be fixed and known. For situations where the slope is not known an accelerometer 49, forming part of an inertial measurement unit, is provided on the circuit board 9. This enables the angle of the apparatus with respect to gravity to be measured. The apparatus also includes apparatus for determining the temperature of metal powder 33 within the sampling body, shown schematically in figure 10. The apparatus comprises a thermal IR sensor 44. This is mounted to the printed circuit board 9 behind the window 26 between two of the LEDs 31. The spacer 27 forms a shroud around the thermal IR sensor 44 defining a generally square cavity in which the thermal IR sensor 44 sits. The shroud extends from the circuit board 9 to the window 26 its side walls diverge from the circuit board to the window. The thermal IR sensor 44 faces the window 26 and is configured to detect thermal electromagnetic radiation 45 emitted by metal powder particles 33 adjacent to the window. Specifically, the thermal sensor is configured to detect radiation 45 in the mid-infrared region of the electromagnetic spectrum. The apparatus further comprises an internal temperature sensor (not shown) which is configured to directly measure the temperature within the housing. The output of the thermal sensor 44 is processed to compensate for leakage current of the IR sensor 44, owing to its ambient temperature. In use, the thermal sensor 44 detects the intensity of mid-infrared radiation emitted by the metal powder 33. This information is processed, together with the temperature measured inside the housing, to determine the temperature of the metal powder 33 adjacent the window 26, whether moving or static. The apparatus also includes apparatus for determining the apparent density of a metal powder, shown schematically in figure 11. The apparatus comprises an inductor 46 formed by a conductive coil positioned adjacent and behind a second window 47 fitted to a second aperture formed through the flat wall of the sampling body. The second window can be formed from any suitable non-electrically conductive and non-magnetic material such as a plastics material. The inductor is embedded in the spacer 27 behind the window. It could however be embedded in the window, or the circuit board 9 or provided as an encapsulated inductor connected to the printed circuit board. The coil 46 is electrically connected with a resonance capacitor 48 to form a resonant LC tank circuit. The coil 44 is provided in a folded line pattern that comprises two rectangular loops. To improve the sensitivity the line pattern can comprise more than two loops. The apparatus is operable to determine the apparent density of static and moving metal powder 33. In use, the inductor is driven with an alternating current by the LC circuit. This produces an alternating magnetic field which induces eddy currents 46 within the metal powder 33 proximal to the window 47 opposite the inductor. The concentration / density of the eddy currents induced into the metal powder 33 affects the amount of magnetic energy that the coil 44 can store, altering its inductance. Since the concentration of eddy currents 46 within the metal powder 33 is associated with / indicative of its apparent density, measuring the inductance of the coil 46 allows the apparent density of different types of metal powders to be determined or inferred in a non-contact manner. As the inductor is comprised in an LC circuit the frequency of oscillation of current in the circuit is dependent on its inductance, so a change in inductance caused by the presence of metal powder, and the density of that powder, alters the frequency of oscillation. Therefore, an indication of a metal powder’s density may be inferred or determined by measuring the frequency of oscillation of current in the circuit and, in particular, by determining the difference between the measured frequency of oscillation and the natural frequency of oscillation for the circuit when no powder is present. Apparent density is the density of the powder in the state it is in, in the sample body, in contrast to a bulk or “tap” density measurement is consolidated (e.g. by tapping a vessel containing the power) before density is measured. The apparatus may be calibrated so that a frequency or frequency difference from natural frequency of oscillation may be mapped to an apparent density value, for example by way of a look up table. Frequency and / or density values may also be mapped to one or more possible powder types so that a frequency measurement may be used to infer what type of powder is or is not present in the sample body. Multiple density measurements may be taken over time and stored by the processor 11, and a mean value calculated. In the described example measurements are taken every second for 15 seconds and a mean value calculated from the 15 measurements. Each of the apparatuses described above may be continuously sampled multiple times per second or polled at regular intervals as desired. The outputs from the apparatuses may be logged and stored by the processor 11 or an associated memory and / or on a separate computer. If one or more of the outputs fall outside of acceptable ranges for degradation, flowability, temperature and density, a user may be alerted. The apparatus may comprise one or more indicators and a user may be provided with a visual indication, an audible indication or an audio-visual indication if the outputs fall outside of one or more of the acceptable ranges. The visual indication could be a light source in the form of a light emitting diode (LED) 20. As an example, an LED may be configured to illuminate red if an output falls outside of an acceptable range. Alternatively, or additionally, a user may be provided with a visual indication on a display screen associated with a computer. This may be accompanied by an audible indication. In other embodiments some or all of the apparatus for measuring flow rate, angle of slope, temperature and density may be omitted, in any combination. Figure 12 shows the apparatus 1 connected in line to a conduit such as a pipeline 50 by way of the flanged connections of the inlet 3 and outlet 4. Such pipelines are used for conveying metal powder within an additive manufacturing machine and / or for conveying metal powder into and out of sieve and blending apparatuses, for example. Valves may be provided upstream and downstream of the apparatus 1 operable to selectively prevent or allow the flow of metal powder through the pipeline 50. The pipeline and apparatus are configured so that the flat internal surface of the conduit of the apparatus forms the bottom of the conduit and has a slope of about 45° to the horizontal along its long axis, and its horizontal along its short axis. Accordingly, when the upstream and downstream valves are both open metal powder is allowed to flow through the pipeline 50 it will flow along the flat surface of the conduit through the apparatus 1 and thus over the windows 26 and 47 in that surface. Information relating to the extent of degradation / oxidation, flowability, temperature and density of the moving metal powder can be obtained by the apparatus. This information can then be assessed independently or in combination to determine the condition or ‘health’ of the metal powder and whether it is suitable for use in additive manufacturing. It is also possible to determine the condition and / or mechanical properties of static metal powder in the pipeline 50. In this respect the valve located downstream of the apparatus 1 is closed which causes metal powder flowing through the pipeline 50 to accumulate in the sampling body of the apparatus 1 as shown in figure 12 whereupon measurements of the properties of the static metal powder can be made. Figures 14 to 16 show the apparatus 1 fitted into a pipeline, again with the flat surface of the conduit forming the bottom of the conduit with its long axis extending at an angle of about 45° to the horizontal. In this arrangement the pipeline above the conduit, connected to the inlet 3 of the apparatus, branches into two. One branch 52 connects to a source of powder, and the other 53 to a pressurised gas cleaning system. Each branch is fitted with a valve 54. Likewise, the pipeline connected to the outlet 4 of the apparatus also branches into two. A lower branch 55 connects to further metal powder processing apparatus and a higher branch 56 connects to the pressurised gas cleaning system. Again, each branch is fitted with a valve 54. In use, as powder flows through the apparatus it may contaminate the window 26 through which electromagnetic radiation is transmitted and received to measure properties of the metal powder. Contamination of the window restricts passage of radiation and the ability to make accurate measurements of properties of the metal powder. The processor is configured to determine when the window 26 has become contaminated during operation based on changes in the measured reflectance of the powder and / or measured reflectance when the apparatus is notionally empty of powder. Information from the apparatuses used to measure flowability, temperature and density characteristics of metal powder, can be used to supplement the reflectance data to improve the accuracy of detection. When the processor determines that the window 26 is contaminated a cleaning cycle can be initiated. Figure 14 shows the inline system in its normal operating state where metal powder flows through and open valve in branch 52, though the apparatus 1, and out through an open valve in branch 55. The other two valves are closed. When a cleaning cycle is initiated the valve in branch 52 is closed. When the apparatus detects that powder is no longer flowing (or after a pre-determined period of time) the valve in branch 55 is closed, as shown in figure 15. Then the valves in branches 53 and 56 connecting to the pressurised gas cleaning system are opened, allowing a pressurised gas (typically an inert gas such as argon) to flow into the apparatus 1 via branch 56 and blow any contaminants or trace amounts of metal powder from the apparatus and the surface of the window 26 out through branch 53. When the cleaning operation is complete measurement may be made by the sensors on the apparatus (with no powder present) to detect if the window has been decontaminated. If not, the cleaning process can be repeated. When the window is decontaminated, the valves are returned to their original state, shown in figure 14, and processing of metal powder resumed. Figure 17 shows a metal powder storage and / or transport container in the form of a hopper 60. The hopper 60 comprises container with a frustro-conical lower portion leading to an outlet 61 which may be fitted with a valve to control flow of metal powder out of the container. Apertures 62, 63 are formed in the frustro-conical wall of the container 60. One aperture 62 is fitted with an IR transparent window and the other 63 with a non-ferrous window. Apparatus for determining powder condition 64 is mounted over the windows. This apparatus is functionally the same as that shown and described in relation to figures 1 to 11, save that the windows through which measurements of powder properties are measured is / are into the container rather than into a conduit. This enables measurements of properties of powder in the container to be made. The powder may be static when the container outlet is closed or moving if the outlet is open to allow powder to flow out of the container. Measurements may be made whilst the container is in transit. In addition to measuring powder properties, the apparatus is also able to determine the location of the container by virtue of its global positioning system (where present) and therefore a user may determine the location of the container 60 during transit and confirm whether it has arrived at its intended destination. The accelerometer of the apparatus can be used to determine the tilt, vibration, shock, or free fall of the container 60, including during transit. Information regarding powder condition before, during and after transit, as well as movement and handling of the container 60 enables a user to determine the condition of metal powder in the container more accurately and whether it can be used in an additive manufacturing process. In addition, use of the accelerometer allows the apparatus 62, 63 to infer the consistency and thoroughness of a blending process when blending occurs by rotating the container 60. This allows a fully closed-loop blending process, with feedback, without any operator exposure to metal powder, and no metal powder exposure to the environment. The one or more embodiments are described above by way of example only. Many variations are possible without departing from the scope of protection afforded 5 by the appended claims.

Claims

1. A method of determining the condition of a metal powder for use in an additivemanufacturing process, the method comprising the steps of:dark field illuminating a region of the metal powder with electromagnetic radiation;detecting the intensity of the electromagnetic radiation reflected back from the metal powder; andprocessing the detected intensity of the reflected electromagnetic radiation to determine the condition of the metal powder.

2. A method according to claim 1, wherein the method comprises the step of illuminating the metal powder with radiation having a wavelength between 300 and 1500 nm.

3. A method according to any preceding claim, wherein the method comprises the step of illuminating the metal powder with radiation having a wavelength between 600 and 1300 nm.

4. A method according to any preceding claim, wherein the method comprises the step of illuminating the metal powder with radiation having a wavelength between 600 and 1100 nm.

5. A method according to any preceding claim, wherein the method comprises the step of illuminating the metal powder with a single wavelength of radiation or with wavelengths of radiation within a narrow band.

6. A method according to any of claims 1 to 4, wherein the method comprises the step of only detecting a single wavelength of reflected electromagnetic radiation, or wavelengths of reflected radiation within a narrow band.

7. A method according to either claim 5 or 6 where the step of illuminating ordetecting radiation involves filtering the radiation to restrict the illuminating or detected radiation to a single wavelength or narrow band.

8. A method according to any of claims 5 to 7 wherein the narrow band is up to 20nm wide.

9. A method according to any of claims 5 to 7 wherein the narrow band is up to lOnm wide.

10. A method according to any of claims 5 to 9, wherein the method comprises the step of selecting the wavelength(s) of electromagnetic radiation based on a maximum thickness of an oxide layer formed on the metal powder it is desired to detect.

11. A method according to claim 10, wherein the selected wavelength(s) is / are at least four times the maximum oxide layer thickness.

12. A method according to claim 10, wherein the selected wavelength(s) is / are at least six times the maximum oxide layer thickness.

13. A method according to any preceding claim, wherein the metal powder is illuminated with electromagnetic radiation at an illumination angle greater than40 014. A method according to claim 7, wherein the metal powder is illuminated with electromagnetic radiation at an illumination angle between 40 0 and 90°.

15. A method according to any preceding claim, wherein the method comprises the step of determining the thickness of an oxide layer formed on the surface of the metal powder based on the detected intensity of the reflected electromagnetic radiation.

16. Apparatus for determining the condition of a metal powder for use in an additive manufacturing process, the apparatus comprising:an illumination device arranged for illuminating a region of the metal powder with electromagnetic radiation;a detector for detecting electromagnetic radiation reflected from the metal powder;wherein the detector and illumination device are arranged so that the detector is unable to detect electromagnetic radiation emitted directly from the illumination device; anda processor configured to determine the condition of the metal powder based on the intensity of the reflected electromagnetic radiation.

17. Apparatus according to claim 16, wherein the illumination device is arranged to a emit electromagnetic radiation towards the powder at an illumination angle of at least 40 °.

18. Apparatus according to claim 16 or 17, wherein the illumination device is configured to emit electromagnetic radiation at a wavelength between 300 and 1500 nm.

19. Apparatus according to any of claims 16 to 18, wherein the illumination device is configured to emit only a single wavelength of radiation or wavelengths of radiation within narrow band.

20. Apparatus according to any of claims 16 to 18, wherein the detector is configured only to detect a single wavelength of radiation or radiation or reflected wavelengths of radiation within narrow band.

21. Apparatus according to either claim 19 or 20 wherein the narrow band is up to 20nm wide.

22. Apparatus according to either claim 19 or 20 wherein the narrow band is up to lOnm wide.

23. Apparatus according to any of claims 16 to 22, wherein the apparatus comprises a shroud which in use prevents the detector from directly detecting electromagnetic radiation emitted from the illumination device.

24. Apparatus according to any of claims 16 to 23, wherein the detector and the illumination device are housed within an enclosure.

25. Apparatus according to claim 24, wherein the enclosure comprises a window for separating the metal powder from the detector and the illumination device.

26. Apparatus according to claim 25, wherein the window comprises glass, sapphire crystal or magnesium fluoride crystal.

27. Apparatus according to any of claims 16 to 26 wherein the processor is arranged to cause the apparatus to perform the method of any of claims 1 to 15.

28. Apparatus according to any of claims 16 to 27, wherein the apparatus is comprised in a powder transport container, in an additive manufacturing machine or in a conduit such as a pipeline.

29. Apparatus according to any of claims 16 to 27 further comprising apparatus for measuring the rate of movement of particles of metal powder.

30. Apparatus according to any of claims 16 to 28 further comprising apparatus for measuring the temperature of particles of metal powder.

31. Apparatus according to claim 30 wherein the apparatus for measuring the temperature of metal powder comprises an IR temperature sensor.

32. Apparatus according to any of claims 16 to 30 further comprising apparatus for measuring the density of metal powder.

Images