DIY Soft Robotic Gripper: EcoFlex Mold & Casting | Purdue STEM

Electro-pneumatic control systems integrate electronic signal processing with pneumatic power transmission, where electronic controllers process input signals from sensors (such as proximity sensors, limit switches, and pressure sensors) to generate control commands that activate pneumatic actuators like cylinders and valves; these systems typically consist of three main sections: the signal input section (sensors and switches), the signal processing section (logic controllers and PLCs), and the power section (pneumatic cylinders and motors), with the interface between control and power sections requiring proper signal conditioning and amplification to ensure reliable operation.

Electro-pneumatic control systems integrate electrical and pneumatic components, where electrical signals control pneumatic actuators through solenoid-operated direction control valves. These systems offer advantages including reduced installation complexity, lower wear, smaller working space, and expandability with PLCs and sensors. Key electrical components include push buttons (momentary and maintained contact types), limit switches for position detection, pressure switches for pressure sensing, and proximity sensors (inductive for metallic objects, capacitive for non-metallic, and optical for reflective objects) that detect objects without physical contact.

Electro-pneumatic control systems use electrical signals to control air pressure for automation purposes, offering advantages such as higher reliability due to fewer mechanical parts, reduced planning and commissioning effort, faster installation with modern compact assemblies, and easy modification of control strategies using programmable logic controllers.

Electro-pneumatic control systems integrate electronic control with pneumatic power. The system consists of three main component types: input elements (push-buttons, limit switches, sensors) that provide initial signals, process elements (relays, contactors) that process and transmit signals, and control elements (solenoids, valves) that activate pneumatic components. The signal flow follows a sequential pattern where inputs send signals to relays, which then activate solenoids to control pneumatic valves. Pneumatic systems require a supply element (air compressor) that provides compressed air to working elements like cylinders. Common components include air compressors, push-buttons, proximity sensors, relays, contactors, solenoids, and cylinders. These systems combine electrical control with pneumatic power for automated mechanical operations.

Electro-pneumatic systems combine electrical control with pneumatic actuation. Unlike purely pneumatic systems, they use electrical signals for control while compressed air powers the actuators. Key advantages include lower costs (fewer valves needed), reduced system size, and faster response times since electrical signals travel at light speed. The system uses limit switches (finales de carrera) as sensors—S1 for retract position and S2 for advance position—connected as normally open contacts. A 5/2 valve controls air flow direction, with both pilot ports controlled electrically. The power supply connects to the relay coils, and a start button initiates the sequence.

Finite element methods (FEM) simulate complex materials by dividing them into small discrete elements (voxels in 3D) and defining interaction properties between neighboring elements. Each voxel connection behaves like a spring with stiffness and damping coefficients, plus additional parameters for torsion and shear resistance. By setting different coefficients for different voxels or connections, engineers can create materials with varying stiffness throughout—a technique essential for simulating soft robots composed of heterogeneous materials.

Traditional FEA suffers from multiple limitations when simulating soft robots: (1) Extremely slow computation times even for linear deformations; (2) Difficulty handling large deformations and rotations; (3) Challenges with hyperelastic material properties where Young's modulus curves are highly curved; (4) Actuation nonlinearity in pneumatic/hydraulic systems; (5) Repeatability issues due to material delays requiring long stabilization periods between measurements.

The Finite Element Method (FEM) discretizes soft bodies into small tetrahedral elements to simulate volumetric elasticity, using deformation gradient matrices to measure strain and energy. Unlike mass-spring systems, FEM provides volumetric constraints and mesh-independent behavior. Key strain measurement approaches include: (1) linear elasticity (F - I), which penalizes rotations; (2) small strain tensor (½(F + F^T - I)), which works for small rotations; (3) Green-Lagrange strain tensor (½(F^T F - I)), which is rotationally invariant but allows reflection; and (4) co-rotated linear elasticity using SVD, which separates rotation from scaling. Volume preservation is enforced through terms like (det(F) - 1)² or trace(Σ - I)², with the Neo-Hookean model providing a more sophisticated approach for large deformations.

Stress-strain graphs plot material response to loading, revealing elastic behavior (Hooke's Law), yield point (plastic deformation onset), ultimate tensile strength (maximum stress before necking), and fracture point. These curves, obtained experimentally, enable software to predict material behavior under various loads. FEA follows three stages: pre-processing (importing CAD models, verifying geometry, discretizing into elements, defining parameters), processing (software performs matrix calculations), and post-processing (visualizing results). Meshing uses triangular or quadrilateral elements based on application needs. Refinement improves accuracy through H-refinement (increasing element count) or P-refinement (increasing node count with higher-order shape functions). Grid dependency tests confirm mesh sufficiency. SolidWorks supports static analysis (constant loads), fatigue analysis (repeated loading prediction), frequency analysis (natural vibration determination), buckling analysis (compressive failure prediction), and drop test analysis (impact damage simulation).

Traditional hyperelastic material models (such as Neo-Hookean, Ogden, and modified decompositions) are unsuitable for modeling soft polymer foams because they produce strain-dependent Poisson's ratios that cannot be controlled; instead, hyperfoam models (implemented differently in Abaqus, COMSOL, ANSYS, and LS-DYNA) provide the key advantage of maintaining a constant Poisson's ratio through specific mathematical formulations involving power-law terms with parameters mu and alpha, allowing engineers to tailor the mechanical response between tension and compression by selecting appropriate parameter values.

Grippers maintain furniture stability under dynamic loads, demonstrated by a chair remaining stationary when someone jumps into it. The pads protect floors from scratches caused by sliding furniture. Real-world applications include preventing coffee tables from sliding when people put feet on them, keeping dining tables stationary when children put legs up during meals, and securing appliances like microwaves and knife blocks. The grippers work on various furniture types including sectionals, beds, dressers, buffets, sofas, love seats, and countertop appliances.

Advanced gripper technology can vastly expand the range of objects that can be grasped, but this expansion occurs in thousands of different directions. The challenge is identifying which applications represent the best fit and where there is the most need. Many existing gripper manufacturers already address existing needs well, so the opportunity lies in discovering applications that people never considered using robots for because they didn't know it was possible or affordable. The company conducted field research at farmers markets, bookstores, and hardware stores to test the hand and observe real-world applications. While mobile platforms and robotic arms are becoming more common, there is relatively less choice in what can be placed at the end of these arms because grippers are more complex to design. As mobile platforms and arms become more prevalent, gripper technology is expected to follow, with more companies developing new hand designs to complement the growing ecosystem of robotic systems.

Multi-joint grippers offer three key advantages over traditional grippers: compliance control for error compensation, directional force control for curved objects, and multi-modal grasping for diverse objects. Industrial applications span bin picking for manufacturing and logistics, kit assembly for component preparation, and handling challenging materials like sponges or thin films. The modular design enables easy maintenance and durability validation at the module level. Applications demonstrate the gripper's ability to reorient objects, perform tasks like screwing and connector assembly, and integrate with suction pads for objects difficult to grasp mechanically. The technology enables diverse object handling, object reorientation, and complex manipulation tasks that were previously manual, bridging the gap between human dexterity and industrial automation.

Material characterization is essential for accurate gripper simulation and involves: (1) Printing samples using appropriate parameters (e.g., 0.6 mm nozzle at 100 mm/s); (2) Performing tensile tests to obtain stress-strain curves; (3) Converting experimental results to material models for simulation comparison. Hyperelastic material models achieve approximately 90-95% accuracy for Finra grippers at strain rates around 15%. The three-stage grip simulation methodology includes: (1) Characterization stage - determining maximum contact force before gripper collapse; (2) Payload capability stage - measuring maximum payload by pulling object away after closure; (3) Optimization stage - iteratively adjusting parameters using optimization algorithms. This systematic approach enables automated pre-validation of gripper suitability for different product variants in fully automated production processes.

Form-fitting gripper jaws for lathes use pre-hardened tool steel jaws with interlocking key structures that bite into and deform material, creating a form-fitting connection that provides superior holding force compared to traditional soft jaws, allowing secure clamping of materials with minimal overhang and enabling machining operations on materials like Delrin, aluminum, and steel with high holding force at relatively low torque settings.