Over winter break, Nicole and I got together three separate times. The first time, we went to Home Depot and PETCO to look at tubing to build air muscles. At PETCO, we purchased tubing that we thought we could use for the connection between the air muscles, compressor, and solenoids. We also bought a plastic gang valve at PETCO to see if we could use it as a manual solenoid for the time being. However, found out later that this gang valve would not work due to the high pressure pumped through the valve. We also bought silicone tubing and barbs at 3/8 to purchase to make air muscles. We also purchased a clothesline pulley that we decided we could use for our pulley on our elbow joint as well as PVC pipe to represent our bones. Two weeks later, we purchase ordered mesh to make the air muscles. We spent the remainder of that day making a hinge for the elbow on a milling machine. We got together the following week and finished the hinge on the milling machine. At the start of school, I spent some time thinking about the microcontroller and solenoids we would be using. I found some websites that mentioned different ways they created air muscles. This included using zip ties or clips to tighten the tubing that connects to the barb on the air muscle. That Wednesday, Nicole and I had our first meeting at Castleman. We discussed running tests on the lymphatic tubing. This test would be that we would have three cups. One cup would have the inlet to the lymphatic tubing we were testing. The inlet cup would be full of water to the same height. The tubing would then enter horizontally into the second cup by making incisions into the sides of the cup. This tubing would have perforations and would be exposed to fluid of a certain height. The tubing would exit the second cup through the second incision and then output to a third cup where volume obtained would be measured. We also discussed the idea I had of using a wetsuit as the skin layer. This would allow a flexible, more elastic holding similar to skin. We also spent some time discussing my findings with the microcontroller we may use. We also talked about using a sponge as the foam layer for the lymph fluid. This would help in creating a fat layer for the arm. We also spent some time discussing that we should use a gang valve if we can find one that is inexpensive because the solenoids can be so expensive. On Thursday, I had an idea that we should use a vacuum sealer for our lymphatic layer. This is similar to the idea in cooking where one sous vides food. When this is done, a piece of food is placed in a bag with certain liquids and spices and then all the air is removed the bag. This causes the bag to seal in anything that was inside. Then the food is placed in a warm bath to cook. In our case, we would avoid the cooking of our lymphatic layer, but could use this concept to seal in our lymphatic layer. A chicken being sous vide can be seen in Fig. 1 below.
Fig 1. : Chicken being sous vide. http://www.instructables.com/image/f0e9ds8fyiyo7an/what-is-sous-vide.jpg In our case, instead of chicken, we may have a sponge filled with fluid in it s place. Inserted into the sponge would be a lymphatic vessel in the form of a tube with perforations. The tubing will stick out of the bag when it is vacuumed sealed. To be certain a seal is created, of course it is important to block the opening of this tube. Once the seal is created, on would have a sealed bag of fluid. I spent some time looking at pricing and found that some can be as cheap as $23 while others can cost as much as $1000. One vacuum sealer, the Rival vacuum seemed to get good reviews. If we were to use this, it may be a simple way to get lymphatic packets. I also had a thought that for these bags, we could punch holes in the area that is completely sealed and connect multiple bags with string. If we connect two or three, we could have a cylindrical lymphatic packet set that fits around the arm. Another thought I had that night while searching for vacuums was the idea of how tight a wetsuit may be. We could adjust this if we cut it down and then resew it. Or, alternatively, we could get a large size and then cut it down to an appropriate size. That Friday was Nicole and my first class meeting at Castleman. I told her about my findings then we performed the tests we had established on Wednesday with mixed results. We found that the smaller tubing didn t have much difference to the larger tubing. In addition to the apparatus discussed, we used a TCS micropump that we found may be useful in setting up flow similar to that seen in the lymphatic system. However, we found that the smaller and larger tubing output about the same amount. Especially when massaging the perforations to open them and close them. Furthermore, we found the pump tended to slow after about seven minutes of testing. In the end, we emailed the company to ask about any specifications that could cause the micropump to slow in such a fasion. The following week I spent some time working on what muscles should be activated in which position. I found a website that discussed the ways that muscles are activated as well as the
sizing of the muscles(ie when the muscle stops and the tendon begins). Using this data, I set up a table that displayed what solenoids will be open and depending on the state. The muscles that I defined were (a) the brachioradialis, (b) the extensors, (c) the flexor carpi ulnaris, Palmaris longus, and pronator teres, (d) the biceps, and (e) the triceps. State 1: arm is silent, no activated State 2: arm is extending State 3: arm is bending, not gripping A B C D E Solenoid Solenoid Solenoid Solenoid Solenoid Solenoid Solenoid open Solenoid Solenoid Solenoid open Solenoid open Solenoid Solenoid open Solenoid open Solenoid While not entirely apparent at first, it should be realized that columns A, C, and D were found to have the exact same solenoid positioning for each state of the arm. This was discovered by Andrew Bligh. I discussed with him the possibility of using one solenoid for muscles a, c, and D and another solenoid for b and e. This would decrease the number of solenoids that should be used. Additionally, we discussed using the microcontroller to turn the compressor on and off. This way, a 3/2 solenoid could be used. A 3/2 solenoid has 3 positions/ports and 2 states. The two states are either that the air muscle is being inflated by the compressor or a state in which the air muscle is being deflated, or air is being sent to the exhaust port. A third state can be created if the compressor is shut off while the air muscle is being inflated. In this way, the air muscle is not deflating but it is not being inflated either. I worked on the air muscles lengths. Using a lower arm size of 24 centimeters, I deduced the following information: Upper Tendon Muscle Length Lower Tendon (A) Brachioradialis 2 centimeters 17.6 centimeters 19 centimeters (B) Extensors 2 centimeters 18 centimeters 6 centimeters (C) Flexors 2 centimeters 8 centimeters 16 centimeters (D) Biceps* 2.54 centimeters 21.9 centimeters 5.715 centimeters (E) Triceps* 2.54 centimeters 20.27 centimeters 2.54 centimeters The following day, I spent some time learning about microcontrollers and the setup for solenoid valves. I spent a lot of time on message board looking at projects involving specifically, Arduinos and solenoids. The first thing I learned is that solenoids and relays are both inductors. This means that if they are used with a microcontroller, they must be in parallel with a diode. This diode prevents kickback to a microcontroller caused when a microcontroller shuts off the current supplied to the inductor. Furthermore, the way that a solenoid is turned on depends on the strength of the system. If this solenoid is high powered, a relay is typically required.
However, if a solenoid is a lower voltage, a transistor can just be used without a relay. Both relays and transistors are switches but one is used for higher voltage solenoid. Some other things I learned about solenoids is that typically solenoids require a great deal of current--at least a great deal more than a microcontroller typically outputs. For this reason, a transistor is used to amplify the current between the microcontroller and the solenoid. A Darlington transistor has to amplifiers in it. In electrical terms, this would mean that the gain is multiplied by an additional gain causing a far greater current to flow through the transistor. The last thing I need to look into but I believe is true is that a transistor is required for a relay to operate. Also if a relay amplifies like a transistor. If it doesn t, a transistor will be necessary to amplify the current the correct magnitude. While I learned a bit about how a transistor works(its essentially a switch that will allow current to either pass or not pass to turn on a solenoid or some other load), I also learned that a transistor can sometimes heat up when supplied with too much current. For this reason, transistors are sometimes used with heat sinks to prevent overheating of the transistor. They essentially fit around the transistor and cost a few extra dollars more. Having put all of this together, I was able to come up with a sketch in my journal of how our setup may look. This can be seen in Fig. 2. Fig 2. Arduino and solenoid apparatus.
After talking to Drew about the compressor switching on and off, we may need to incorporate a relay into the compressor because the compressor is hooked up to a wall. This will produce a large voltage which needs to be managed by a relay. However, I m not unsure if we require relays for all our solenoids which is something I will have to look into. On Friday during class, Nicole and I created air muscles with the lengths of air muscle we had measured. One concern we did have was whether to create the air muscle using the measurement with the air muscle stretched(like it would be on our model) or to measure with it relaxed. In the end, we tried to approximate the amount of stretching required and measured using the length of the muscle/tubing when stretched. We also have some concerns about making bigger air muscles. Air muscles require some tension to inflate. Will there be enough tension for these longer muscles, like the biceps and triceps, to inflate and create volumetric expansion? I spent time during the class looking up Styrofoam for our stagnant muscle layer The idea we have for the stagnant muscle is that we will by a rod of Styrofoam, about 6 inches in diameter. We will cut a rod in the center of this rod that will be equal to the diameter of the bone and place the bone inside this Styrofoam rod. From there, we will glue the Styrofoam in place. Then we will cut into the Styrofoam and shape the the Styrofoam into the shape of the muscles that we are not including as air muscles. This will allow the arm to be dimensionally accurate with respect to the air muscles. Nicole and I discussed how we would connect the bone. But we will need to reconstruct this because the Styrofoam will be in the way of the bone.