Published papers

Femtosecond laser-induced microstructural visualization of stainless steel surfaces was investigated as a simple approach for revealing crystallographic heterogeneity in polycrystalline metals. Pocket machining experiments were performed on ferritic and austenitic stainless steels to compare their laser-induced surface responses. For ferritic stainless steel (SUS430), the resulting surface morphologies were further correlated with crystallographic orientation using electron backscatter diffraction. Under identical irradiation conditions, localized smooth and roughened regions appeared within the same machined pocket. In ferritic stainless steel (SUS430), grains oriented near the (111) plane showed an earlier transition to roughened regions, whereas other orientations remained relatively smooth. Three-dimensional surface measurements showed that the observed contrast mainly originated from differences in surface roughness rather than differences in material removal depth. The roughened regions were more consistent with chaotic laser-induced pattern (CLP)-type morphology than with wavelength-scale periodic structures. The clearest microstructural visualization was obtained at 9.35 J/cm2, indicating that an appropriate fluence window exists for this approach. These results provide a fundamental basis for a femtosecond laser-based auxiliary method for metallographic observation of stainless steel surfaces.

Additive manufacturing (AM) has garnered significant attention because of its numerous benefits and transformative potential across industries, particularly in internet of things manufacturing systems. The application of AMs with gravitational acceleration has considerable scope for exploration. Material extrusion (MEX) is a versatile technology, to evaluate the effects of high gravitational forces on MEX, an experimental approach involving controlled gravitational acceleration up to 5G was implemented using a specialized high-gravitational material extrusion system. This system, leveraging centrifugal acceleration, facilitated a comprehensive assessment of material flow dynamics and resultant surface quality under varying gravitational conditions. The results clearly indicate that applying high gravitational acceleration, reaching up to 5G, notably improves the material extrusion quality and appearance. Gravity affects the flow of materials during the extrusion process, high gravity can help achieve better material deposition. The experimental assessments clearly demonstrate that a high gravitational field enhances the surface quality and appearance of fabricated parts.

Many studies have focused on the stabilization of additive manufacturing (AM) in microgravity for its use in various space projects. Nevertheless, this paper presents a vital clue for innovating metal AM technologies from the perspective of high gravity. High-gravitational powder bed fusion has an excellent potential to address various challenges in AM, such as density enhancement, spatter suppression, and precise fabrication. This study summarizes an analogy among phenomena in different gravitational fields and establishes a combined machine for centrifuge and powder bed fusion. The results confirm the spatter suppression and fine-powder availability in high gravity, both theoretically and experimentally.

Although metal additive manufacturing (AM) gathers attention from various industries, the process stability in metal AM still remains in an important issue for ensuring the production quality. In particular, powder-based directed energy deposition (DED) is difficult to stabilize because large amount of metal is contentiously molten with a high-power heat source. Furthermore, the material powder flow needs to be precisely converged on the meltpool; otherwise the considerable amount of materials would be wasted. Against these challenges, this study aims to analyze the turbulent flow in the fabrication space for converging the powder trajectories on the meltpool stably and accurately. By constructing a gas-solid multiphase-flow simulation for the coaxial nozzle for DED, the influence of carrier and shield gases on the powder flow is numerically calculated. According to the estimated particle distribution, the particle concentration on the top surface of deposit decrea ses in the higher-layer deposition. Moreover, the particle-flow convergence is enhanced with higher carrier gas-flow rate. These analyses highly agree with the experimental results on powder distribution measured with a laser light sheet system. As a conclusion, it is clarified that the carrier gas supply should be higher at higher layer deposition to keep the powder convergence stable in DED.

Directed energy deposition (DED), an additive manufacturing process, is a suitable approach for freeform production with metallic materials. By generating a melt pool with a high-power laser beam, injected material powder is contentiously laminated on a baseplate by melting and solidifying. Although all supplied powder should be molten and solidified in order to reduce the material waste, powder distribution is difficult to converge owing to turbulence around the melt pool. Furthermore, an inappropriate powder supply easily leads to sputter generation, which also increases the material waste. In this study, the gas flow under a powder nozzle was analyzed by a computational fluid dynamics (CFD) simulation in order to achieve a high convergence in the powder supply. By measuring the powder distribution with a laser light sheet system and conducting deposition tests, the powder distribution and supply efficiency were experimentally evaluated. According to a gas-solid multiphase-flow simulation, the convergence distance of the powder flow should be shorter than the laser beam focus distance to improve the powder convergence with a lower gas-flow rate. Moreover, the powder nozzles were redesigned by taking the simulation results into consideration, and deposition tests were conducted to evaluate the powder supply efficiency, porosity rate, and penetration rate. The experimental result of the designed nozzle shows a clear improvement in the powder supply efficiency from 50.2 to 66.0%.

Directed Energy Deposition (DED), which is one of the Additive Manufacturing (AM) processes, can deposit metallic production efficiently by irradiating a high-power laser beam on a baseplate and supplying material powder simultaneously. DED has a problem in low mechanical strength due to residual pore inside of the production. However, the residual pore is also available to fabricate low density production as an opponent idea. If the residual pore could be increased intentionally and drastically, DED is applicable to foam metal fabrication. Foam metal could be produced continuously with high degree of freedom in shaping, if melt-pool can be foamed in DED. Therefore, this paper proposes a simple fabrication method for foam stainless alloy with DED by applying titanium hydride (TiH2) as a foaming agent for melt pool. TiH2 releases hydrogen in a high temperature environment as more than 500℃. Considering that high viscosity and low surface tension are required to stabilize pore generation, material powder also should contain a thickening agent (CaO: calcium oxide) and a surface active agent (Te: tellurium). According to the experimental results, an effect of CaO was hardly observed because the liquid stainless may already have high viscosity enough. On the other hand, Te certainly reduced the critical film thickness and prevented the pore collapse in stainless foam fabrication, according to the porosity rate calculation, the observation with a high speed camera, and the elemental distribution analysis. Furthermore, the proposed method successfully produced a foam structure in stainless steel with high porosity rate as 61% at maximum.

The joining of dissimilar materials is an important process to produce a large production. In other words, the reliability of such a production is determined by the joining technique because the joint interface often becomes the weakest point against stress. In case of metals, welding and riveting are popular approaches for joining dissimilar materials. However, these techniques generally involve manual and complex operations; therefore, the production quality cannot be maintained, because the accuracy and efficiency of these operations strongly depend on the worker’s skill. From this viewpoint, additive manufacturing (AM) can be useful to produce parts using a combination of dissimilar metals. Metal AM has attracted considerable attention from aerospace and automobile industries recently because of its flexibility and applicability in the production of various complex-shaped parts. Directed energy deposition (DED), one such metal AM method, forms a deposit of powder material and simultaneously irradiates a laser beam on the baseplate. DED can be applied to cladding and repairs as it can be conducted on the surface of the part. In particular, a combined part of dissimilar metals can be easily and directly produced from scratch by changing the powder material of the process. A graded material can also be produced by blending different powders and changing their ratios appropriately. In order to realize such applications of DED, the mechanical properties of the produced part must be evaluated in detail. In this study, a part combining a nickel-based superalloy (Inconel 625) and a stainless steel alloy (SUS316L) is produced using DED; the produced part is evaluated through a tensile strength test, Vickers hardness measurement, metal structure observation, and element distribution analysis. In addition, a graded material is also produced to evaluate the basic characteristics of the joint produced using DED. The experimental results show that the produced joint is sufficiently stiff against tensile stress and its hardness is increased because of the solid solution of niobium in the stainless steel area. The results of the elemental distribution analysis and the Vickers hardness test indicate that a graded joint of Inconel 625 and SUS316L can certainly be produced using DED.

Directed Energy Deposition (DED) of metal materials has been widely studied with Inconel 718 using in aerospace industry. Void generation in a deposited part causes deterioration of the mechanical strength. Hence, void should be avoided in the DED process. In this study, a mechanism of void generation was investigated through the dynamic observation by high-speed camera, the thermal observation and the evaluation of the porosity. The experimental result shows that solidification time has an influence on the void generation.

Direct energy deposition (DED) is a metal additive manufacturing process that involves the application of a large amount of heat energy, which generates a thermal gradient across the deposited part. This usually results in unexpected anisotropy and weakening of the deposited part due to residual pours. These problems need to be solved to ensure satisfactory results when using DED. This study led to the proposal of a simple heat treatment procedure for DED-produced parts in the form of a remelting process, which controls the metal crystal orientation and eliminates residual pours. The use of a high-power laser for remelting certainly obtained high-density metal structures with high hardness.

Although the applicability of additive manufacturing (AM) to the production of complex shapes has attracted attention from the automobile and aerospace industries, companies hesitate to introduce AM processes because of their low reliability, which is due to pores inside the produced parts. Consequently, many researchers have experimentally evaluated the relation between the pore evolution and production conditions in AM processes. On the other hand, several studies have focused on finishing processes in order to enhance the quality of AM production, considering that production quality cannot be improved enough only by modifying the production conditions in AM processes. To reduce pores in a metal product, hot isostatic pressing (HIP), which applies high pressure and heat energy to metal AM products and enhances production density, has proven to be an efficient approach. However, special equipment is required to produce a high-temperature and high-pressure environment, leading to high cost and low productivity. From the view point of practicability, a simple finishing process would be a fundamental solution to make metal AM processes highly reliable. This paper therefore proposes a method of reducing pores through a remelting process in the direct energy deposition of Inconel 625. Furthermore, a method of doing a graphical analysis to evaluate the bias of pore distribution in the deposited object is proposed. The pore reduction effect in remelting is experimentally evaluated by irradiating the low density area with a laser beam, and a graphical evaluation clarifies that the concentration of residual pores occurs in the top layer of a deposited object. As a result, residual pores are eliminated with certainty through the remelting process. The density of the deposit can be enhanced easily and without any complicated finishing systems with just the laser system originally introduced in a DED machine.

Direct energy deposition attracts attention from automobile and aerospace industries because of its applicability to complex shape production. Although the meltpool temperature has to be controlled to enhance shape accuracy, industries remain hesitant over introducing a complex process monitoring system because of high cost and necessity of frequent maintenance. In order to keep the meltpool temperature constant, this paper proposes optimal laser power command generation by creating a finite-element thermal conductivity model for a gradient descent calculation. The experimental results clearly indicate that the obtained laser power command enhances the shape accuracy of the deposited objects.

In recent years, the Directed Energy Deposition (DED) method using Inconel 718, in which supplied metal powder is melted by high energy laser and metal layers are deposited layer by layer, has been actively studied and attracted in the aerospace industry. Voids generated in the deposited part cause deterioration of the mechanical strength. Hence, avoiding voids is one of the most important issues in the DED process. In this study, the mechanism in which voids are generated was investigated by observing the porosity rate, analyzing gas included in voids and observing thermal distribution during the deposition process. The result indicated that the deposition with a laser power of 2000 W and a feed rate of 1000 mm/min has the lowest porosity, and constant laser energy density is effective to keep the constant porosity. Furthermore, Argon gas for shield and most atmospheric components were detected in the voids of the deposited part by analyzing gas in the voids.

This paper presents a sensor-less on-line chatter detection method for a turning process by introducing a mechanical energy factor (MEF) and a mechanical power factor (MPF). The MEF and MPF serve as indexes for self-excited chatter and forced chatter, respectively. The indexes are based on the power-factor theory, which generally represents the electrical-power efficiency as having a correlation with the phase difference between the current and the voltage. By applying this theory to a mechanical system, the MEF and MPF can be employed to monitor the phase difference between the cutting force and the displacement/velocity of the tool system, respectively. By monitoring the phase difference, chatter vibration can be detected in time domain with a high response and small number of computations. The MEF and MPF can be calculated without using additional external sensors employing the sensor-less cutting-force estimation technique based on the disturbance observer. The monitoring performance of the proposed method was evaluated through several outside turning tests with a prototype precision lathe. The results showed that both the self-excited and forced chatters were successfully detected with unique thresholds, which did not depend on the cutting condition or the workpiece material.

In recent years, many composite materials have been used in industry. Among such materials, the demand for carbon fiber-reinforced plastic (CFRP) is increasing. Although CRFP is used in various fields such as the aerospace industry, automotive industry, and sports equipment because of its light weight and high strength, it has poor production efficiency. Thus, carbon fiber-reinforced thermoplastic (CFRTP), with characteristics similar to CFRP but higher in production efficiency, has attracted attention in areas such as the automotive industry. Because CFRTP is used as a structural element, it is usually drilled to allow connection to other parts. However, it is a difficult-to-cut material requiring the development of a high-accuracy, high-efficiency processing method. This study proposes high-feed-speed processing as a method that achieves high-quality drilling at low cost. The cutting temperature was estimated to verify the effect of the high-feed-speed processing method, and processing conditions that prevented delamination and burring were identified.

This paper proposes a sensorless approach for realtime tool fracture detection in milling by means of servo information. Cutting force and torque can be estimated in a wide-frequency range by applying disturbance observer to the ballscrew-driven stages and the spindle controllers. By integrating the estimated information in each axis, a fracture-induced variation in cutting force and torque can be accurately captured with parallel sliding Fourier transform which is an analytical approach of low computation load in time–frequency domain. Validation of the proposed method is presented through milling tests of various milling conditions with several fractured endmills.

Electro Adhesive Gel (EAG) has the unique characteristic of changing its surface adhesive property with the intensity of the electrical field applied. This property makes EAG useful in applications to fixing devices and mechanical brakes. Although its adhesion performance depends on the distribution of the electro-rheological particles in the EAG, it is difficult to arrange the particle distribution uniformly in a wide area from the perspective of production process. In this study, a novel functional elastomer that has the same function as EAG is developed, Electro Adhesive Surface (EAS). In EAS, micro photolithography is used to fabricate strut pyramids distributed uniformly on a substrate, and then silicone gel is poured into the structure. When an electrical field is applied, the silicone gel rises to the tops of the pyramids formed by the struts, and adhesion occurs to an object on EAS. To determine a micro structure design for EAS, the fixing force was measured with various struts diameter and gaps. Experimental result shows that the larger struts diameter and the narrower gaps enhance the fixing force of EAS.

Chatter vibration must be avoided to ensure the cutting accuracy in milling. Several studies have proposed analytical prediction methods for chatter stability based on a frequency-domain milling model and a modal analysis. However, the prediction accuracy is difficult to enhance because the modal parameters vary according to the spindle rotation and heat generation. In addition, necessity of special measurement devices for the modal analysis leads to high cost and complicated operations overwhelming the operator. In this paper, we propose a novel analytical method for identifying stable spindle rotations against chatter vibration in milling, which only requires once milling test without any measurement devices. Chatter frequency gradually gets larger with higher spindle rotation and drastically shifts at the stable spindle rotation, at which the locally largest depth of cut can be applied. Based on this characteristic of chatter frequency, the stable spindle rotations can be captured only from the servo information by decreasing spindle rotation continuously and observing frequency shifts with disturbance observer in the spindle control system. The disturbance observer is a technique to estimate the load torque only from the servo information, which is utilizable to monitor a fluctuation in cutting torque due to chatter. By applying time-frequency analysis to the estimated cutting torque, the drastic shift of chatter frequency can be captured. The experimental results show that the chatter frequency shift can be clearly captured at stable spindle rotations by applying a time-frequency domain analysis on the estimated disturbance torque.

Chatter vibration deteriorates the machining accuracy and shortens the tool life. Thus, In-process monitoring methods have been proposed by using additional sensors such as dynamometers and acceleration sensors. However, introducing additional sensors leads to high cost and reduction of machine-tool stiffness. To solve these problems, a sensor-less monitoring method applying the disturbance observer theory was proposed in our previous research which has experimentally shown that chatter vibration can be detected only from the servo information in milling. For further demand, because chatter vibration can be classified into self-excited vibration and forced vibration, an assorted detection method is required. In this study, we propose a novel frequency analysis method combining moving variance and moving Fourier transform algorithms, which can obtain the power spectrum density of each chatter vibration type separately with small computational load. The validity of the proposed method is evaluated through milling tests with different rotational speeds.

In current practice, the buffing process required to finish the surface of mechanical parts is performed manually by a technical expert as it requires a delicate adjustment of the buffing force. The automation of this process is desirable in an effort to shorten the process time and reduce labor cost. To automate the buffing process, a beneficial process monitoring technique that supervises the buffing tool conditions in real time must be developed. From a practical perspective, an observer technique that does not require additional sensors would be most suitable for monitoring the tool operating condition. The authors propose a technique that estimates the buffing tool stiffness based on a disturbance observer. The validity of the proposed method as a buffing tool condition monitoring technique is verified through numerical simulations and experiments.