Balance and Mobility Module [Duke University]

Brian Bigler, Anita Raheja, Margaret Widmyer, and Matthew Davis

ABSTRACT

Our client has difficulty stabilizing herself on her bike due to former cancer related treatments and surgeries.  The overall goal of this project is to create a device to provide stability during intervals of starting and stopping throughout a bike ride.  The focus of this phase, Retractable Wheel Attachment was to design an attachment for the bike, similar to adult training wheels, which as a part of phase two can be rotated off the ground once the rider is rolling and stable. This attachment design includes two balance wheels, a wheel rotation system, a wheel t-support system, and a stopper and spring system to hold the wheels down. With the completion of the second phase, Rider Actuation, the Balance and Mobility Module will give riders the stability of balance wheels when they want them, and the freedom of riding on two wheels as they choose.  The final prototype provided stability for the client and has the ability to rotate, but the control system is still in development.

BACKGROUND

Our client is a two-time cancer survivor and former tri-athlete.  Since childhood, she has been very passionate about riding her bike.  She has completed duathlons, triathlons, and biathlons. Numerous surgeries and treatments have left her with a weakness in her left hip and leg. Therefore, she has difficulty stabilizing herself and her bike during starting and stopping on an outdoor bike ride.  Although she can cycle on a stationary bike, she has not ridden a bike outside in five years.

We performed extensive background research to establish which commercial devices and patents might meet the client’s needs.  A gyroscopic system (1) uses the gyroscopic effect of a spinning wheel, powered by a motor, on either side of the back wheel to add stability to the bicycle. This approach did not meet our functional specifications since it did not limit the client from falling in event of motor failure. The motor also was cumbersome and added substantial weight to the bicycle. Adult training wheels were found to meet the requirement of adding stability to the bicycle (2) (3). However, there was no ability to remove the wheels from the ground for the duration of the ride. We also examined the option of a three-wheeled bicycle, but our client desired the feel and maneuverability of a two wheel bicycle.  Finally, a system with wheels that rotate out of the way has been designed (4), but it required a second person to change the position of the wheels from contact to non contact. In conclusion, commercial devices could not adequately meet the client’s needs.

PROBLEM STATEMENT

Our goal is to create a device to provide stability during intervals of starting and stopping throughout a bike ride.  The device should be operable independently by the rider and allow for normal two-wheel bike motion when desired.  Safety is a main concern; therefore, we want to allow the rider to start and stop without falling.

DESIGN AND DEVELOPMENT

We initially met with the client and discussed her capabilities and needs.  We then developed a list of functional specifications.  It was determined that the device needed to safely allow the client to independently ride her bike, provide stability during periods of starting and stopping, and allow for a normal ride during the rest of the ride.  After several prototypes, we created a device that mechanically lifted two stabilizing wheels off the ground, and which could be raised and lowered at the rider’s discretion.

This device has four major components: a bike rack, a triangular rack attachment, a rotating wheel attachment, and wheels.  The wheels rotate around a fixed rod that is mounted to a standard bike rack over the back wheel.  An aluminum stopper plate prevents forward rotation in this upward position.  A spring, attached to the rotation system helps return the wheels to the lowered position.

Figure 1: Spring and Stopper

The device addresses problems observed during the early stages of the design process and has: 1) low friction bearings, 2) a rigid support structure and 3) low friction wheels.   Stainless steel ball bearings replaced a previous brass sleeve bearing as the means for rotating the wheels off of the ground.  The ball bearings slide onto the ¾” aluminum rod directly and are press-fit into both the wheel support plate and the aluminum stopper.  The bearings fit into a ½” aluminum section instead of a ¼” section, reducing the risk of bending at the point where it was previously observed.  Moreover, rotation occurs within the bearing rather than relative to the aluminum rod so the bearings will not wear against the rod like the brass sleeve did.  The bearings are locked in place by lock collars attached on the inside of each one.

Figure 2: Wheel Rotation System

The second major design consideration improved bending stiffness of the current device based on issues discovered during prototype testing.  First, we connected the bottom of the stopper and the top of the wheel support with two 4.75”x 2”x ¼” aluminum plates which link the motion of the two pieces together while providing torsion strength and support against bending at the attachment point of the wheel support.  Next, we strengthened the wheel support plate by bracing it with two 2”x ¼” thick aluminum sections that attach around the stopper bearing and thrust downward at 45°.  These plates prevent strain at the center of the wheel supports by turning the bending load experienced there into a compressive load.  Finally, the bottom half of the wheel support was strengthened by the addition of a 3/8”thick aluminum bar.  This bar effectively converted the plate of the wheel support into a T-section, increasing the bending stiffness by more than a factor of 5.

Figure 3: Wheel Support System

Figure 4: Bending Prevention T-Section

After testing a number of different types and sizes of wheels, we selected a pair with a diameter of 12.5” and a pneumatic tube rather than hard rubber.  The pneumatic tubing absorbs some of the impulse experienced by the supports when the wheels first touch ground or they pass over rough terrain.  The wheels are placed 2.5” further away from the rear wheel of the bike at the same height from the ground which helps to increase stability for the rider by reducing the amount of tipping she needs to experience before the wheels begin to support her.  The wheels are attached by ball bearings to an axle which in turn screws into one side through the T-section of the wheel support.  Held in place by nuts on each side of the wheel support, the 8mm axle holds the wheel firmly in position while the wheel rotates freely on the bearings.

Figure 5: Completed Device

Figure 6: Top View of Completed Device

COST

EVALUATION

To evaluate the completed device, the load was estimated by placing a scale underneath one additional wheel. Then, we had a 135 lb. rider lean on the wheel with her full weight several times. From this test, we measured that the wheel would be loaded with a maximum of 50 pounds and no possible failure locations were observed.

After testing without motion, a test during a bicycle ride was performed with a rider of approximately 130 lb leaning his full force on one wheel support. For three 20 minute rides, the device withstood the forces of the bicycle ride.  However, a future failure location will be the bicycle rack’s attachment location to the bicycle as well as to the triangular plate.  Because the tubular aluminum rack is loaded in bending rather than in compression, as designed, the aluminum shows signs of fatigue. Although the client was able to ride the bike safely during testing, the device will not be delivered until the issues have been adequately addressed.

DISCUSSION AND CONCLUSIONS

The device provides a short term solution to the problem, but we are not comfortable with the device as a long term solution without further testing due to uncertainty in weather, road condition, long-term fatigue, and rider safety compliance.  One benefit of the current design is that it does not add too much hindrance to a bicycle ride. An advantage of the device is that it is completely mechanical. There are no electronics on the device for the safety of the rider. The device is made out of aluminum to prevent rust.

One limitation of the device is that it cannot yet be moved into and out of position.  A mechanism was tested involving a thumb-actuated gear shift that was mounted to the handlebars.  When the gear shift was pushed, it pulled a cable that ran from the gear shift to a custom gear box.  Inside the gear box, the mechanical advantage of the pulleys converted the 0.9 inches of cable movement into the 2 inches of movement required to lift the stabilizing wheels. However, it proved too difficult to use this mechanism while riding the bicycle. Another issue with the current design is that it rotates rearward when lifted, which means that an object on the ground could push on the wheel while it is in the downward position, which could result in the device rotating unexpectedly. This could be dangerous to the client and is likely to occur under normal riding conditions.  This risk could be mitigated by redesigning the device to rotate in the direction of the bicycle’s motion.  The tubular aluminum in the rack shows signs of yielding when loaded in bending as it currently is by the device. The likelihood of this potential failure could be reduced by increasing the number of attachment points or by reinforcing the aluminum tubing with cross sections.

All parts are diagrammed in Solidworks to allow for easy replication and manufacturability. There is currently a large need for a device of this nature in the biking community. Thus, the market potential for this device is large if the price was reduced and the durability was increased.

REFERENCES

1. Goldstein, Linda K. Bicycle Balance Training Apparatus. USA: Patent 6676150. 13 January 2004.

2. “Welcome.” Stabilizer Wheels. Bike USA, inc., 2008. Web. 6 Sep 2010. <http://www.stabilizerwheels.com/>.

3. “Products.” FatWheels. Mechanical Innovations, Inc., 8/10/2010. Web. 6 Sep 2010. <http://www.fatwheels.com/products.html>.

4. Lytle, Kimberly M. Bicycle Training Aid with Dynamically Deployable Balancing Features. US: Patent 0029994A1. 7 February 2008.

ACKNOWLEDGMENTS

This project is supported by the National Science Foundation under Grant No. BES-0610534. We would like to thank our professor, Dr. Laurence Bohs, for his exceptional help throughout the process, Steve Earp for assistance in design and machining, and our client for supporting us throughout the design process.  Additionally, we would like to thank Veronica Rotemberg, Dr. Mark Palmeri, Dr. Roger Nightingale, and the people at the All-Star Bike Shop in Raleigh, NC for their helpful suggestions and support throughout the process.

Brian Bigler: 3311 Cardigan Court, West Lafayette, IN 47906;

Matthew Davis: 320 North St, Saco, ME 04072;

Anita Raheja: 1609 Arch Bay Dr, Newport Beach, CA 92660;

Molly Widmyer: 1434 Cattail Run Rd, Charles Town, WV 25414;