Evan Coombs Individual Wiki Report
| Date: November 29, 2006 | Actuator Concept |
 |  |
| Author : Evan Coombs | Team : Robot Hand Mechanics |
Problem Definition
My part of this project is to reduce the friction in last year’s transmission. As you can see from the link below, last year’s transmission is bulky, poorly constructed, and has lots of friction. You can find the initial calculations from last years Team Robotic Hand’s Wiki Site. For reference, the home page of last years Team Robotic Hand 2005–2006 can be found below.
Requirements and Specifications
I am required to work with the same Shape Memory Alloy (SMA) Actuation Array that was developed by Team Robotic Hand 2005. This actuation array controls the application of hot and cold water to the SMA wires. Hot water is needed to change the phase of the SMA wire (contraction), and cold water is needed to cool the SMA wire back to its initial phase (extension).
In addition, I am to meet the requirements and specifications that were determined at the beginning of this project. The main constraints that I must meet are to maintain an actuator force of 10 Newtons, and not exceed the physical size of the previous SMA wire array.
Transmission Concept Generation Sketches
Below you will find drawings of the initial concept generation.
 |  |  |
| Transmission Concept | Transmission Actuator Concept | 3D rendering of Transmission Concept |
Through the process of benchmarking and research, I found a Shape Memory Alloy (SMA) Spring that altered all of my further concepts. By using a SMA spring to actuate each position of the finger joint, as compared to transferring the SMA wire force through a transmission to create the position, all frictional losses can be completely eliminated. In addition, using a SMA spring allowed for up to 30mm of displacement where as the SMA wire can only muster 5mm of displacement.
Increasing the displacement does have a tradeoff. By increasing the displacement the force that can be applied by the spring goes down. Currently, the existing SMA wire can exert 10 Newtons of force. The SMA spring can only exert 3.5 Newtons of force. In order to equal the force generated by the SMA wire, 3 SMA springs will have to be utilized.
Initial Downselect & Design Refinement
After you glance at my Initial Down Selection Matrix, you can see that the spring SMA Actuator design is easily superior to the previous transmission design. A refined picture of the SMA Spring Actuator Concept can be found here, as well as a drawing with the dimensions can be found here.
 |  |  |  |
| Initial Down Selection Matrix | 3-D Drawing of Initial Actuator Concept Design | Refined 3-D SMA Spring Actuator Concept | Schematic of Refined SMA Spring Actuator Concept |
Final Concept Selection
|
| SMA Spring Actuator |
After evaluating all of my conceptual designs, and evaluating 2 down selection matrixes, I decided to build an SMA spring actuator.
The final design consists of 3 components, the end plugs, the silicone boot, and the SMA springs. Below you will find the conceptual drawings, as well as the schematics for the end plugs and the silicone boot. Because the SMA Springs were purchased, there will only be a short explanation of the physical attributes of the spring, and the website where they were purchased.
End Plugs
The end plugs serve two functions, attachment points to the SMA springs and the tendons and access ports for the water.
The End Plugs are not COTS parts, but are easily machined on a manual lathe. The total time needed to machine the end plugs was 90 minutes. This included all machine set-up time as well as drilling and taping. With the speed and flexibility of this design, it can be easily modified in the future to help address any design changes.
 |  |  |
| 3-D Model of End Plug | Drawing Schematic for End Plug | Machined Endplug |
Silicone Boot
The most important part of this actuator is the Silicone boot. Its function is to incase the SMA springs, to flex with the SMA springs, and keep the Hot/Cold water inside the actuator. After an extensive search, no suitable commercially available part was available. A mold and core pin was designed and manufactured to generate a suitable silicone boot. Below you will see the conceptual pictures, drawing schematics and the actual manufactured mold.
Silicone Boot Mold
- Centering Block
 |  |
| 3-D Model of Cenetering Block | Drawing Schematic for Centering Block |
- Core Pin
 |  |
| 3-D Model of the Core Pin | Drawing Schematic of the Core Pin |
- Outer Mold
 |  |
| 3-D Model of the Outer Mold | Drawing Schematic of the Outer Mold |
- Manufactured Mold Pictures
 |  |  |
| 3-D Model of Mold, Exploded View | Open Mold for Silcone Boot | Assembled Mold for Silicone Boot |
Results of the Silicone Boot Mold
In the next picture you can see the result of the first silicone boot. The idea was to place the mold in a vacuum chamber, load the top of the mold with silicone, then vacuumed the air out. As you can see, there was not enough silicone entering the mold. This process has been refined by first placing the silicone in a vacuum chamber to de-gas the mixture, then injected into the mold to fill the mold cavity.
 |  |  |
| First Result at Manufacturing the Silicone Boot | Silicone Boot Seperated From Mold Core Pin | Actuator Assembly with Incomplete Boot |
SMA Springs
The SMA Springs were purchased from www.robotstore.com at a cost of $10 a piece. The specific type of SMA Springs purchased are called Nitinol. After additional calculations, that were completed by my teammates, it was realized that 10 Newtons of force was not sufficient to meet our requirements. The new requirement was that of 40 Newtons of force. The current SMA Springs are no longer adequate to meet 40 Newtons of force.
In order to meet this new force requirement, and keep the current actuator design, it was decided to purchase untrained Nitinol SMA wire. This wire can be purchased in larger diameter that the current SMA Springs, placed in a spring shape of our teams design, then cured to retain the desired shape. This process will be re-evaluated by our advisor and possibly completed in the spring semester after adequate calculations have been made.
 |
| SMA Spring Photo |
Pictures of Conceptual and Manufactured Parts
- Solid Models
- Machined Parts
Lessons Learned
By completing a CFP (Critical Function Prototype) I learned the following things about the following components:
- End Plugs
For the time required to manufacture this item (90 minutes), it was an effective design that can be easily altered to meet any future needs. Possible alterations would include:
- Increase or Decrease in number of SMA Springs attached to the end plug
- Increase or Decrease in the Actuator Diameter
- Increase in water flow port diameter
- Silicone Boot
The process to manufacture the silicone boot was extensive, but considering, it only cost time. Materials and machining time were donated. The total time to design and manufacture the mold was 40 hours.
I learned the following from this process:
- If possible, alter your design to allow using COTS Parts
- Schedule extra time for unanticipated problems such as
- Machining equipment failures
- NC code debugging
- Redesign time to accommodate available tooling
Even with the extensive amount of time required to design and make the mold, I feel that it was a valuable experience, and contributed to our team’s efforts. By making a part specifically for the designed actuator, our team vision could be realized. By accomplishing the exact vision of our design, an evaluation of our design process could be completed to determine the effectiveness of our designs.
Conclusions
By eliminating the previous design teams transmission, I fulfilled all requirements of my problem statement. Actuating the finger by SMA Springs allows for multiple degrees of freedom for each joint in the finger while eliminating the friction inherent to the previous design teams transmission.
Back To Top