Mitchell Lum

EE499 - Spring 2000

6/8/00

The Force Reflective Endoscopic Grasper (FREG) is a tele-connected, force-feedback endoscopic surgical grasper. The FREG offers real time force feed back between the grasper handles and the tool tip jaws. The both the grasper handles and the tool tip jaws have rotary encoders which sense their orientation. Force feed back is related to the operator via voice coil actuators when there is a difference in handle/tool tip orientation.

One of the drawbacks of the FREG is that it does not allow a large amount of force to be fed back to the operator. This is due largely in part to the limitations of the voice coil actuator. With the current set-up the voice coil actuators are using the strongest magnets available to produce a maximum amount of force. However, the level of force desired is much greater than the voice coil actuator can produce.

Purpose:

During the Winter 2000 quarter, Jeff Longnion began investigating the feasibility of using a geared motor system in place of the voice coil actuator. A geared motor can provide much higher torque output than a voice coil actuator. However, rather than building a new FREG with a geared motor system in place, we first wanted to see the effect that the geared motor would have on haptic perception. Along with greater torque output, the geared motors mechanical design introduces into the system inertia, friction and backlash, which could effect perception.

In order to find a starting point, Jeff designed a flywheel system, which would have exactly the same inertial properties as the geared motor. The flywheel system would provide a baseline for how the inertia of the geared motor system would effect haptic perception.

The "Haptic Pliers" were designed to be very modular. We could take off the Geared Motor system, and put on the Flywheel Inertia Simulator very quickly. In the future if we wanted to investigate the frictional and backlash characteristics of the geared motor, separate "add-ons" could be developed to easily integrate with the "Haptic Pliers"

We were going to evaluate subjects on the "haptic pliers" (1) without the geared motor attached, (2) with the geared motor attached, (3) with a flywheel system that has the same inertia as the motor, but not the friction or backlash of the motor. Each trial consisted of two silicon samples of different stiffness. The test subject would be asked to determine which sample was stiffer. There were eight distinct silicon samples, and each subject would be tested on each pairing, resulting in 16 trials per subject.

My role in this experiment was to aid Jeff in building the flywheel device.

Procedures:

During Winter 2000, Jeff developed the basic design of the "Geared Motor Haptic Pliers". Jeff also did research into which geared motor would be the most effective in terms of cost, weight and output torque.

At the start of the quarter, Jeff already had the device built and configured for the geared motor. The initial parts were fabricated by a professional machine shop. My initial involvement was to aid in the construction of the Flywheel Inertia Simulator. In order to start setting up for the new "add-on," the geared motor system was going to be removed. However, upon attempting to remove the setscrew, which held the motor shaft, from the Plier's Pivot area, the setscrew sheared in the hole. In a failed attempt to remove the collar, which sat between the Plier's Pivot and the motor shaft, the part was damaged. I ended up putting the Plier's Pivot with the damaged collar and broken setscrew in a mill to drill out the setscrew, then used a lathe with an end-mill to bore out the damaged collar. I attempted to clean up the Plier's Pivot as much as possible.

The old collar (which was destroyed) had obvious drawbacks. It was purely cylindrical. If it ever got stuck (which it did) there was no way of removing it without destroying the part. I designed a new collar that was nearly identical, except I added a flange on the end, so that if the collar ever got stuck, one could slide a rounded needle nose pliers between the Pliers Pivot and the collar flange in order to wedge it out.

After a failed attempt, I found that the lathes in the Electrical Engineering shop had too much play in the lathe beds to be able to accurately machine this part. I used the Mechanical Engineering shops lathes to turn the new collar with a high degree of accuracy.

Now that the Haptic Pliers was in usable condition once again, I began to design the base in which the flywheel shaft would rotate. Jeff had a bronze bushing mounted in a cast metal base. We felt this would make a low friction pivot base for the flywheel. Using the existing design for the geared motors base I set forth to create a similar base upon which the bronze bushing would be attached. When the first Flywheel Pivot Base was nearly complete, Dr. Jacob Rosen observed that the bushing system might allow too much lateral movement of the flywheel shaft. It was his opinion that we needed to create a pivot base, which had much less play.

Based upon Dr. Rosen's suggestion I redesigned the pivot base to house a ball bearing rather than having the bushing attached to it. This new design required much higher tolerances in order to produce an accurate fit.

I attempted to mill the base on the EE shop's mill, but after producing a very rough inaccurate block, I decided that I should use the ME shop which has much more accurate mills. The block was machined on a manually operated Bridgeport mill. The bearing hole needed to be counter-bored to within two thousands of an inch (0.002") to produce an accurate fit. It was the recommendation of Mr. Noe to use their newer (and highly accurate) lathes with the Pivot Base mounted in a four-jawed chuck. He showed me how to properly align the piece in the four-jaw chuck, then discussed how the turning operation would take place. Upon doing the actual turning, I inadvertently over-bored the hole in diameter. When the bearing was placed in the hole, I used a piece of paper (0.005") to act as a shim, in order to achieve a very tight fit.

Jeff designed and did the initial machining of the flywheel shaft. He did his machining on the EE shops lathe, but left it slightly over sized so that I could finish cut it on the ME shops more accurate lathes. I used a collet lathe to turn the Flywheel shaft. Upon turning the piece it was discovered that the shaft was not even round, and needed significant work. After the piece was turned to its specific radii, I filed two flat spots onto the sides of the shaft at the top. This was done so that a 9/16" open ended wrench or an adjustable Crescent Wrench could be used to tighten or unscrew the shaft from the threaded rod.

The flywheel assembly fit onto the Haptic Pliers very well. However, then we used the pliers to pinch various silicon samples, we were very surprised. My initial impression was that the flywheel made it much more difficult to differentiate between two samples, compared to the geared motor. Because the flywheel only simulated the motors inertial characteristics and not its frictional or backlash characteristics the flywheel felt much different than the motor (as might be expected). We discussed the possibility of designing a frictional simulator a backlash simulator as part of possible future testing.

Jeff conducted the subject testing. The data and results from this part of the project can be found in his write up.

Data/Results:

The result that my contribution produced was a functioning Flywheel Assembly, which Jeff used to test three situations. The effects of the Geared Motor vs. Flywheel Assembly vs. No Load on the operators haptic perception can be found in Jeff's write up.

Conclusions:

While this project did not involve anything directly related to the study of Electrical Engineering, it gave me insight into the research process. My involvement in this project primarily involved fabricating aluminum parts for the Flywheel Inertial Simulator "add-on" for the Haptic Pliers. While it would have been very easy (but possibly expensive) for the designs to have been sent out to a professional machine shop, the experience I gained will undoubtedly be beneficial to me as a future engineer.

During this project I gained a deeper of understanding of the machining process. I learned how quickly inaccurate tooling can ruin hours of work. I am thankful to Mr. Noe from Mechanical Engineering for sharing with me his time and his shop. I learned how to use a lathe to perform a precise counter-boring operation. I also learned to think from the machinist's point of view when producing designs. Unless the machinist can plug a set of CAD drawings directly into a CNC system, the dimensions given must be useful. It is often helpful for all the dimensions to be referenced off a common origin in order to make the machining process more efficient. It was a good experience to be on both sides of the development of a device, both seeing the design process and carrying out the manufacturing of the parts.

This project has shown to me that engineers cannot simply sit behind a desk or read a textbook and learn all they need to know. Having done some of the machining myself, I can understand the design process from concept to tangible results. In the future I may or may not be machining the parts I design. If I am not the one to do the machining, it will be especially important to design from the machinist's point of view. After all, the machinist is the one who we depend on to an idea and a block of raw material and make it reality.

This being my first EE499 research project I feel it went very well. I gained insight into the research process and learned how involved and complex it can be. While my contribution to the overall project was a very small one, I hope that what was discovered, as a result of my project will benefit the team as a whole.

Back Home