Overview

The team decided to utilize fused deposition modeling (FDM) based 3D-Printers to construct the majority of the components for the rope gripping mechanism. The main advantage of this technique was that it allowed for rapid and inexpensive prototyping. Most components of Design II were redesigned, printed, and tested at least two times. The ability to almost effortlessly generate functioning prototypes allowed the team to rapidly improve the device and incorporate changes that would have otherwise been too costly in terms of time and money using other prototyping methods. The various iterations of redesign for the base plate of the gripping device are shown in Figure 1. The second redesign for the baseplate incorporated a mount for the servomotor and the third redesign allowed for a cover to enclose the linkage arms and servo.

Figure 1: Iterations of Design for Prototype II Baseplate

  The primary challenges of using FDM printers  approach were quality and performance of the prototype parts. Strategies for improving FDM print quality are addressed in the 'Printer Calibration' section below. Performance of the product was maximized via part redesign and rigorous testing. From experience the team learned that parts fail easily when in tension in the axis intersecting the printed layers, and future parts were printed with an orientation that minimized this mode of failure. Additionally, harsh tests of the prototype were utilized to find weak links in the mechanism.

FDM Printer Calibration

The most common quality problem encountered during the project was consistently undersized holes. Undersized holes were compensated for with the aid of a calibration sample. The sample consisted of 95 holes with five holes for each expected diameter between 1.0 and 10 mm in 0.5mm increments. Hole diameters were measured and recorded using digital calipers at room temperature. The average error for each diameter was plotted and a quadratic regression was fit to the data. The plot as well as the polynomial regression terms are shown in Figure 2. The regression terms were used to add an offset to each hole diameter on parts to be printed. The bushings press-fit into the plastic in Figure 3 are representative of typical results using the calibrated hole sizes.

Figure 2: Calibration for Undersized Holes

Pinned Joints

The smooth motion of the prototype is largely attributable to the use of oil impregnated sintered bronze bushings as bearings to support radial loads. The bushings shown in Figure 3 were press-fit into seats in the links. The seats and the slight interference of the fit prevented the bearings from slipping out of the links. Precision shafts were cut to size and used as pins to constrain the links. The pins were locked in place using retaining rings. Linkages were designed so that the distance between the outer edge of the link and the outer diameter of the bushing was greater than or equal to the diameter of the bushing. This geometric constraint proved sufficient for preventing tear-out in the linkage arms. The team plans to continue developing the assistive gripping device with the ultimate goal of fully integrating all mechatronic components into the base of the mechanism.

Figure 3: Bushings Press-Fit into 3D Printed PLA housings

Opportunities for Integration

One of the advantages of additive manufacturing is the ability to easily enclose components such as sensors and wires. Mechatronic elements such as the servomotor and hall-effect sensor were integrated into the base of the gripper prototype and all wires were routed through channels incorporated into the base-plate. Integrating the hall-effect sensor into the base in an enclosure similar to the one shown in Figure 4 makes the prototype more visually pleasing and more importantly protects the sensor from abrasion and debris. Wire channels have similar advantages with the additional benefit of preventing the wires from being snagged as the device is in use.

Figure 4: Hall-Effect Sensor Encased in Plastic

 

Next Page (Lessons Learned and Conclusion) ->

  • No labels