P18229: Robotic Otter
/public/

Detailed Design

Table of Contents

Team Vision for Detailed Design Phase

During this phase, our team focused on furthering our design and adding more detail to the structure. Our main goal was to complete the structure design and gain more information on waterproofing the skin of the Robotter. Also, we further practiced the silicone making process to ensure that we are prepared to make the vertebrae spacers for the final project.

In addition to furthering our Robotter design, we began brainstorming ideas for our test rig and our Imagine RIT set-up.

Lastly, we began drafting our goals and timeline for MSD II. We wanted to make these goals challenging but realistic.

Progress Report

What we accomplished in Phase 4:

Decisions that have been made in Phase 4:

Questions for the customer and guide:

Prototyping, Engineering Analysis, Simulation

Tubing for Spine

As seen in Phase III, our team prototyped our basic design concept by making a spine out of 3D printed parts, tubing, and fishing line. The tubing used for this simulation was something we had on hand, which was 1/4 inch PVC tubing. After completing that prototype and doing some simulations, we knew that we wanted to have a larger PVC tube with thicker walls.

For our spine tubing simulation we tested two different PVC tube diameters, 3/8 inch and 1/2 inch. We attached a metal weight to the center of the tube to simulate the weight of the electronic box and electronics the spine will need to hold in the final design. The weight for this test was about 4 lbs. The results from this test can be seen in the photo below.

Spine Tube Prototyping

Spine Tube Prototyping

Both tubes showed significant improvement in strength from the 1/4 inch tubing used in our original prototype. Both tubes showed a better resistance to kinking, but, once kinked, the deformation in the 1/2 inch tube was more permanent than in the 3/8 inch tube. The 3/8 inch tube also seemed more resistant to kinking, especially from a torsional load, although this was not quantified.

When purchasing the material for these test, the 3/8 inch tube was cheaper per length than the 1/2 inch tube.

We calculated the weight per unit length for each tube diameter, and found that the 3/8 inch tube weighed 0.3189 g/cm, while the 1/2 inch tube weighed 0.4666 g/cm.

Despite the 3/8 inch tube having a larger sag, we decided as a team that the weight and cost of the Robotter was critical. This test was also run without the vertebrae or spacers on the tube, which would increase the stiffness. We realized that, if sagging becomes an issue, we could place the original 1/4 inch tube into the 3/8 inch tube, greatly increasing the stiffness. Thus we decided upon the 3/8 inch for our final design.

The spine sagging in between the legs will be controlled through active control.

Silicone Making Process

The initial plan for waterproofing the robot was to cover the entirety of the bot in a silicone "skin", with a single, sealed seam; To test our capacity to keep the skin sealed at this seam, and to get a better idea of the material properties of the resultant material, we cast a single sheet of silicone, which was to be cut down the center and then resealed (to mimic the seal on our otter); water was to be poured over this to check for ingress.
Silicone Making Process Set-up

Silicone Making Process Set-up

Silicone Making Process Set-up

Silicone Making Process Set-up

Silicone Curing in Vacuum Chamber

Silicone Curing in Vacuum Chamber

To this end, we carefully mixed equal parts A and B of the silicone ingredients, with a small amount (<10% total volume) of thinner added, to improve uniformity from mixing, and to allow air bubbles to escape more easily. The mixing was done directly in the container used for casting (this led to a small amount of unmixed, uncured silicone remaining just around the edge of the part, but did not affect the apparent overall material properties), and this was then placed in the vacuum chamber (as seen above) for 24h. The results of this silicone seem acceptable, with few and small bubbles appearing in the final part, but it nonetheless became apparent to us that it would prove prohibitively difficult a) to cast the final skin into the appropriate shape, and b) to create a waterproof, easily reusable seal on a silicone-silicone interface subject to various dynamic loadings. Based on this, we decided to further pursue a different route to create a waterproof sheath around our device.

Waterproofing Prototyping

Benchmarking off of the Pleurobot, we found that the Pleurobot's swimming suit was made from Lycra Nylon fabric laminated with 1mm layer of polyurethane. Our team wanted to prototype this waterproofing suit before spending our resources on expensive fabric for the robotter.

The two materials we tested were a 100% Nylon fabric (black material) and a 88% Nylon & 22% Spandex fabric (white material). The black material is going to be more accurate when comparing to the Pleurobot's swimming suit. The white material has more elasticity and can stretch with the robot as it turns, which is a desired characteristic our team would like to have.

We made three different waterproofing suit prototypes per material:
1. One layer of polyurethane on one side.
2. Two layers of polyurethane on one side.
3. One layer of polyurethane on both sides.

100% Nylon fabric (black material) and 88% Nylon & 22% Spandex fabric (white material) coated with polyurethane.

100% Nylon fabric (black material) and 88% Nylon & 22% Spandex fabric (white material) coated with polyurethane.

100% Nylon Fabric Waterproof Testing. Click to view.

100% Nylon Fabric Waterproof Testing. Click to view.

88% Nylon & 22% Spandex Fabric Waterproof Testing. Click to view.

88% Nylon & 22% Spandex Fabric Waterproof Testing. Click to view.

88% Nylon & 22% Spandex Fabric Waterproof Testing After Being Stretched. Click to view.

88% Nylon & 22% Spandex Fabric Waterproof Testing After Being Stretched. Click to view.


The 100% Nylon and the 88% Nylon & 22% Spandex demonstrated similar results during the waterproof test. Both materials were able to withstand the water pressure and did not leak through the material. After about 4 minutes, both materials started to feel damp.

Next, we stretched both materials to see how they would perform. This test showed different results. The 100% Nylon's polyurethane layer did not show any cracks. The results did not vary from the unstretched results and did not leak. However, the 88% Nylon & 22% Spandex proved unsuccessful after being stretched. The polyurethane layer cracked and was not able to withstand the water pressure, therefore proving unsuccessful.

Drawings, Schematics, Flow Charts, Simulations

3D Models - Full Assembly

Full Assembly

Full Assembly

Full Assembly - Front View

Full Assembly - Front View

Full Assembly - Top View

Full Assembly - Top View




3D Models - Ribs

Large Rib

Large Rib




3D Models - Battery Enclosure

Battery Enclosure

Battery Enclosure

 Battery Enclosure w/ Ribs

Battery Enclosure w/ Ribs




3D Models - Leg Design

Leg Design

Leg Design

Imagine RIT Goals & Plans

Our goal for Imagine RIT is to get the location on the lawn next to Kate Gleason College of Engineering building. The main reason is because there is a hose at that location that would be easily accessible to fill up our kiddie pool for the Robotter exhibit course.


Imagine RIT Set-up

Imagine RIT Set-up




Below is our first draft of our exhibit build for the Robotter walking and swimming course. This set-up requires a kiddie pool that is at least 12 inches deep (so the robot does not touch the ground while swimming and so there is enough room for the robot to dive) and have a diameter of at least 72 inches (so the robot has enough room to turn around to walk back up the ramp). The platform and ramp will be made out of plywood and wood posts. To ensure that the Robotter cannot walk off the platform, small rails/walls will be made.


Imagine RIT Set-up

Imagine RIT Set-up




Bill of Material (BOM)

Current Bill of Materials

Current Bill of Materials

As we are in the process of finalizing our design, we have yet to decide on the exact specifications for the majority of our components. As of now we have 28 items that need to be purchased, 5 items that the team is providing and 13 items that we are either 3D printing or molding out of silicone.

We are always aware of our budget, so we have yet to purchase any materials. The materials we used for the tests were all donated. The bill of materials also includes the items required for our anticipated Imagine RIT setup; they can be found in the Miscellaneous section at the bottom of the bill of materials.

Once we lock down our design, and then the electronic and leg components specifications, we will do another thorough cost analysis to determine our exact anticipated cost.

We do not anticipate long lead times on any items. The 3D printed parts are relatively small, and, if we place the order in the first week of the semester, as we anticipate, we should have them available by the second or third week of the semester. The purchased components are all standard items, so no custom orders are required. Most of the items can be picked up locally in Henrietta.

Our working bill of materials can be found here: Bill of materials

Motor Considerations

There are additional factors to consider with the selection of our motors (specifically, those for controlling the spine), as there are a large number of parameters that can have a significant effect on the power and torque capabilities of our motors. A change in the shape, scale, or material of the base spine, the vertebrae (including the positions and orientations of the motors thereon), and the spacers, amongst other things, can drastically change the base requirements for the motors; considering that the overall density of our assembly is critical, we want to ensure that we can fit our robot with the lightest motors possible, while simultaneously ensuring they fit our requirements (at least a F.S. of 5 as compared to the force required for static bending in the most bent configuration possible for the spine). Because the evolving design plays a monumental role in the selection of the motors, we have tentatively selected those shown in the BOM above, but fully expect that these requirements, and, as a result, our selection of motor, to change moving forward.

Test Plans

Test Rigs

Similar to the Imagine RIT set up, we will make a platform and slope out of wood. However, instead of using a kiddie pool for the swim tests, we will use Prof. Gomes' tank. This tank is 10 feet which is the swimming length engineering requirement for our Robotter. Our team will also set up a Go Pro camera to film the testing.


Test Rig

Test Rig

Test Plans (Working Copy)

Risk Assessment

Top 20 Risks

Top 20 Risks

A full list of our working risk assessment can be seen here: Risk Assessment

Parameters to calculate importance

Parameters to calculate importance

Plans for next phase

MSD II Goals:


Individual Goals

Team Member
Amanda
Corry
Curt
Ethan
Tiffany
Vaughn


Gate Review

Our team's self-critique after the gate review can be seen here: MSD I Self-Critique

Home | Planning & Execution | Imagine RIT

Problem Definition | Systems Design | Preliminary Detailed Design | Detailed Design

Build & Test Prep | Subsystem Build & Test | Integrated System Build & Test | Customer Handoff & Final Project Documentation