Final Design

Final Design Solution

Nerve Lengthening Device

Overview

The final nerve lengthening device design is a miniature motor and spool mechanism encased in a stainless steel enclosure. Individual nerve grippers are attached to the proximal and distal nerve stumps. Connecting them is a guide wire that runs through a metal backbone that will be surgically attached along a patient’s bone that also leads the guide wire into the nerve lengthening device. The device will also be surgically implanted within the patient's body along the same bone as the metal backbone and on the same end as the distal stump. The device utilizes the motor to rotate the spool, thereby pulling the guide wire and proximal nerve towards the distal nerve.

The device’s main components are the motor, drive shaft, enclosures, spool, and a one-way bearing mechanism. The motor, drive shaft, and bearing mechanism are encased by a two-part enclosure and spool. The spool is used as the component that secures the guide wire and as the guiding track for which the motor will use to reel the guide wire along. The enclosure is made up of the minor enclosure, which houses the one-way bearing mechanism, and the major enclosure, which houses the drive shaft and motor. The major enclosure also has mechanical stops that are designed to be utilized by the drive shaft to limit the distance actuated. The final design of the spool and enclosure utilizes stainless steel as the primary manufacturing material so that the device can follow medical device standards that allow the device to be implanted within a human body. However the prototype that is used in testing with rabbits utilized aluminum as the primary material.

Within the enclosure, the drive shaft is used to transmit the motor torque to the spool, and is placed between mechanical stops of the major enclosure. The mechanical stops act as the main mechanism that controls the step size of the distance actuated. The mechanical stop always stop the driveshaft from moving beyond 2.617mm, thus preventing the nerve from being overstretched during a lengthening procedure while also creating a set amount of distance actuated at every step. The bearing mechanism is two one-way bearings that each stop one component from rotating. The first bearing is press-fitted into the spool with the driveshaft inserted into it. This first bearing will allow the driveshaft to have free rotation when it is turned counterclockwise, but when the driveshaft turns clockwise, the bearing will lock and cause the spool to rotate with the driveshaft. The second bearing is placed within the minor enclosure with the spool inserted into it. This second will grant free rotation when the spool rotates counterclockwise, but will lock if the spool attempts to rotate clockwise. This combined bearing mechanism acts as a way to index the actuation distance and to maintain the tension of the nerve by locking the spool’s rotation when the motor is not powered. The nerve lengthening device is powered and controlled by a detachable external controller that is not implanted in the body.

        The external controller is an Arduino Mega 2560 powered by a 9V battery. This controller uses a program that will observe the current of the motor. When the device's motor is stopped by the mechanical stops of the enclosure, it is forced into it's stall torque and consequentially draws the maximum current it is able to. The program observes the current for this current spike and shut of the motor when it is measured. Then, the program resets the drive shaft by reversing the motor's rotation direction.

Figure 1: Final design of device within an implanted environment

Animation of CAD prototype performing the nerve lengthening procedure

Component Overview

Figure 2: Cross section overview of the nerve lengthening device components

Figure 3: External controller part overview

Motor

Figure 4: 700:1 Plastic Planetary Gearmotor 6Dx21L mm

In the Motor-Spool design, the primary purpose of this motor is to reel in a guidewire that will be attached from the distal nerve to the proximal nerve. Therefore powering the motor will result in the proximal nerve being pulled towards the distal nerve as the guidewire is reeled in. In order to satisfy the strict size requirements of the ideal nerve lengthening device while still retaining the ability to provide enough force to actuate a nerve, the optimal motor of the spool design should be less than 1cm in width, 4cm in length, and be able to output 12N, which is the maximum force required to pull a nerve. By pulling on the nerve at maximum required force of 12N around a spool with a radius of 5mm, the required torque to pull a nerve stump with a motor would be 0.06 N-m.

The final motor that was chosen for this design was the 700:1 Plastic Planetary Gearmotor 6Dx21L mm, manufactured and distributed by Pololu Robotics and Electronics displayed in figure 4 above. The chosen motor features a diameter of 6mm and a length of 21mm, thereby meeting size constraints of the ideal device while giving enough remaining space for other key components. The motor also specifies a stall torque of 12 oz-in (0.0847 N-m) at 6V. With the maximum required torque of 0.06 N-m, the motor sufficient was torque at 6V. However, the motor will be operated and controlled using an Arduino Mega 2560, which has maximum operating voltage of 5V. This will cause the motor to be undervolted when powered by the arduino board. Calculating the torque output at 5V to be 0.07 N-m, the motor would still be able to provide the necessary torque to pull the nerve, albeit with a lower factor of safety.

Enclosure

The key functional requirements of the major and minor motor enclosures, shown above in Figure 8, will be to ensure that the motor and other key elements can be safely implanted into a rabbit’s body. The one portion of the enclosure will have a built-in mechanical obstruction that limits rotary actuation step while the another portion will have an ISO interference fit opening for a one-way bearing. This enclosure will encase all device components excluding wires to power the motor. The enclosure will have to be made of biocompatible materials to ensure the device can be implanted within a body without any detrimental effects on the the implantee’s health. The primary standards to judge biocompatibility will be based on whether a material is non-corrosive and non-toxic. This limits choices of metals to most stainless steels, titanium, nickel-titanium alloys, and cobalt-chrome. However, plastics are also an option and most 3D printable plastics are biocompatible. Lastly, the enclosure will have to fit within the optimal boundary box of the device of 1cm diameter and 4cm length.

The final design of the enclosure was dictated by the OD of the motor and of the spool. Built in mechanical stops were included on the major enclosure in order to serve as index boundaries for the driveshaft, enabling a controlled step by step nerve actuation. These built in mechanical steps allow the nerve to be actuated at a linear distance of 2.617 mm each actuation. In addition to physically limiting the distance of motor travel, the stops also cause the motor to exhibit stall conditions when stopped. This in turn causes the current of the motor to spike because stalling a motor will cause it to draw the maximum amount of current in an attempt to overcome the resisting torque against it. This effect is the foundation of the controller design used to control when to stop the motor and is further elaborated in the controller section. Both the motor hole and the bearing hole were sized for an interference/press fit. In order to attach to the rabbit bone, a mounting interface of holes designed for size #2 screws was included. Finite element analysis was utilized in order to verify that the design could withstand the “worst case” scenario of the driveshaft transmitting maximum motor (i.e. stall) torque onto the mechanical stops in the major enclosure. Figures 6 and 7 below show the stress and displacement heatmaps, respectively, of the enclosure when the motor drive tooth is stalled against one of the hard stops. It can be seen that the highest stress within the model does not exceed the material yield strength.

Figure 5: FEA of enclosure's mechanical stops with motor stall torque.

One Way Bearing

Figure 6: Dual One-Way Bearing Mechanism Graphical Summary

For the final design, the method of using two one-way bearings in opposite directions was chosen as the mechanism to control and measure how far the nerve has been stretched. The primary reasons for choosing this method was because it would be relatively easy to incorporate to the spool design compared to the other two while also staying within the optimal boundary box of the device. A second design for the nerve lengthening device originally incorporated a ratchet-mechanism with a rack  as the primary method to control actuation. However, further prototyping and investigation on the design led to the decision that the ratchet-mechanism in the small scale workspace would be too difficult to utilize. Encoders were suggested as the best way to measure distance traveled by the motor, but an encoder that was small enough to attach to the motor was not found. With the double one-way bearing design, it was possible to utilize the one-way bearing’s backstop mechanism as a method to index actuation length by connecting two one-way bearings in opposite directions. The design revolves around having the first bearing press-fitted within the spool and the driveshaft inserted into the bearing. This bearing would mounted so that the drive shaft can transmit torque to the spool when rotating clockwise, while the driveshaft can spin freely without moving the spool when spinning counterclockwise. The second bearing is mounted inside the minor enclosure with the spool inserted into the bearing. This bearing is positioned in the opposite direction of the first bearing. This allows the spool to spin freely when moving clockwise attempting to spin counterclockwise will lock its position. This mechanism is designed to work in conjunction with the mechanical stops that were designed in to the enclosure so that the distance of nerve actuation can be indexed. When the motor rotates the driveshaft clockwise, the anti-reversing mechanism of the bearing transmits the torque to spin the spool clockwise. When the driveshaft reaches the mechanical stops, the program controlling the motor has the motor reverse its direction, thereby by spinning the driveshaft counterclockwise. The first bearing allows the driveshaft to have free rotation counterclockwise, but the second bearing prevents the spool from spinning counterclockwise as well and keeps it in place while the driveshaft’s position is reset. This design allows the driveshaft position to be continuously reset while the spool’s position is indexed. An illustration that summarizes device can be found in Figure 6 above.

Driveshaft

The connection between the motor and the first one way bearing must transmit the torque of the motor to the spool (through the bearing), and also contain a feature that contacts the two hard limits of the Major Enclosure. These hardstops define the rotary travel range of the driveshaft, and the matching tooth must be rigid such that the motor can be stalled for a sufficient time to read a spike in the current draw.

The final design for the driveshaft included a 3.175mm (⅛ in.) diameter sub-shaft sized for press fit into the spool bearing, a filleted trapezoidal extrusion/protrusion to index in between the major enclosure’s mechanical stops, a coaxial hole sized for an H7/h6 shaft-based locational/transition fit with the motor’s D-shaft, and a set screw hole perpendicular to the rotating axis sized for a #0-80 thread (to accommodate the set screw that holds the motor to the driveshaft). See Figure 10 for an illustration.

The design was verified with a simple torsional analysis, under the assumption that it was undergoing motor stall torque while the trapezoidal indexer pushed against the major enclosure’s mechanical stops; this simulated the highest amount of stress being applied to the driveshaft. A free body diagram can be found in Appendix: Calculations. The applicable torsions and forces were applied to a FEM and the model was meshed, run, and postprocessed. Figure 11 depicts the resulting stress heatmap; the maximum stress in the model was 57.97 MPa, which represented a safety factor of 4.74 over the 275 MPa yield stress for 6061-T6 aluminum.

Figure 6: FEA of driveshaft on mechanical stops

Prototype Performance

In addition to the aluminum prototype, an Arduino Mega 2560 along with an INA169 current sensing module was tested to verify reliable current feedback from the motor. Using the INA169 module, it was found that the current sensing resolution was much improved over simply utilizing a 0.1 sense resistor, since the INA169 module has a built-in amplifier. An additional major enclosure was machined to be used for code testing and debugging. The working code was tested, and verified to be able to detect the motor stall current, and subsequently shut off the motor.

    With this prototype the indexing mechanism was tested in order to see if it actuated in the prescribed step sizes. The test was conducted using a dial travel indicator, which was attached to a string to the device in order to observe how much it moved per step. The results of the test is shown above in Figure 21, with the slope of the lines representing the average displacement over each actuation. The expected step size was 1.706 mm per step, while the actual measured step size was 1.680 mm per step. This was only an error of 1.5%, so the indexing mechanism functioned exactly as planned.  The results of this experiment are displayed in Figure 7 below.

Figure 7: Experimental results of prototype actuation distance

This aluminum prototype was also test implanted within the dead rabbit, as shown in Figures 8 and 9. The placement of the device is consistent with the actual intended placement with respect to the backbone, which can be seen in the figures as the L-shaped tube. In this test, the nerve grippers were not installed since the rabbit nerve had already been removed for analysis previously. Instead, the pull cable was tied to a piece of adjacent tissue, which can be seen in Figure 9 below.

Figure 8: Second prototype partially implanted in a dead rabbit’s thigh

Figure 9: Side view of partially implanted prototype to show guidewire setup