Additive manufacturing is now leveraged to digitally fabricate complex and customized sand molds for castings. Mathematical equations can be employed in design to gracefully vary strut sizings volumetrically within lattices and the corresponding sand molds can be realized with binder jetting additive manufacturing. Consequently, geometrically tailored periodic structures are now castable—not possible previously with traditional sand casting. However, inspecting these 3D-printed molds and cores prior to pouring metal can be dramatically more challenging given the required verification of both (a) the intended dimensions of internal cavity features and (b) the absence of unbound sand within difficult-to-access internal passages (proper cleaning). This work evaluates tapered cylindrical sand cores (100 mm in diameter at the top and 200 mm in height) and each core includes a complex internal cavity that captures the negative of one of four spatially varying lattice architectures. The four lattices were designed with software from nTopology and included strut dimensions that linearly decreased from the top to the bottom of the cylinder. For each of the four lattices, a pair was fabricated (eight cores total) using an ExOne SMAX printer. Each pair included: (a) one core cleaned based on best practices and (b) one core not cleaned leaving unbound sand in all internal passages (as harvested from the binder jetting print box). Cleaning included air blowing and brushing to remove internal unbound sand without damaging the delicately bound structure. Subsequently, X-ray computed tomography (CT scanning) was used to evaluate the cores prior to casting in order to: (a) evaluate the effectiveness of the standard cleaning techniques at removing all unbound sand and (b) verify compliance of inaccessible internal features in comparison with targeted geometries. Unbound sand was clearly identified which if present would indicate the need for additional cleaning. The bound sand was geometrically evaluated and dimensional deviation from the CAD-intended geometries was less than 200 microns in each case.

增材制造现在被用于数字化制造复杂和定制的铸件砂模。可以在设计中使用数学方程来优雅地改变晶格内的支柱尺寸，并且可以通过粘合剂喷射增材制造来实现相应的砂模。因此，几何定制的周期性结构现在可以铸造——以前用传统的砂型铸造是不可能的。然而，在浇注金属之前检查这些 3D 打印的模具和型芯可能更具挑战性，因为需要验证 (a) 内部型腔特征的预期尺寸和 (b) 在难以接近的区域内不存在未结合的沙子内部通道（适当清洁）。这项工作评估了锥形圆柱形砂芯（顶部直径为 100 毫米，高度为 200 毫米），每个砂芯都包含一个复杂的内部空腔，可捕获四种空间变化的晶格结构之一的底片。这四个晶格是用 nTopology 的软件设计的，包括从圆柱体的顶部到底部线性减小的支柱尺寸。对于四个格子中的每一个，使用 ExOne SMAX 打印机制造一对（总共八个核心）。每一对包括：(a) 一个根据最佳实践清洁的核心和 (b) 一个未清洁的核心，在所有内部通道中留下未结合的沙子（从粘合剂喷射打印盒中收获）。清洁包括吹气和刷洗以去除内部未结合的沙子，而不会损坏精细结合的结构。随后，X 射线计算机断层扫描（CT 扫描）用于在铸造前评估型芯，以便：(a) 评估标准清洁技术在去除所有未结合的砂方面的有效性，以及 (b) 比较难以接近的内部特征的合规性具有目标几何形状。未结合的沙子被清楚地识别出来，如果存在则表明需要额外清洁。在每种情况下，对结合砂进行几何评估，与 CAD 预期几何形状的尺寸偏差均小于 200 微米。未结合的沙子被清楚地识别出来，如果存在则表明需要额外清洁。在每种情况下，对结合砂进行几何评估，与 CAD 预期几何形状的尺寸偏差均小于 200 微米。未结合的沙子被清楚地识别出来，如果存在则表明需要额外清洁。在每种情况下，对结合砂进行几何评估，与 CAD 预期几何形状的尺寸偏差均小于 200 微米。