This week I finally received all the parts I ordered and was able to build and test the sensor. Below is a picture of the photodiode circuit.

It works quite well. Under blue illumination the photodiode creates a slight voltage. However, under red illumination the increase in voltage is 10 times more then under blue illumination. The use of a red filter may not be necessary because the red LEDs are not sensitive to blue light.

            Next I need to start testing the response of the LEDs to dim red light, like that produced by fluorescence. This will tell me how many red LEDs I should use in parallel (I am thinking four), and how much resistance to put in the above circuit.

            Lastly, I’ll be taking a trip to home depot soon to look for a PVC housing. I need something that can be hold the Arduino and the LEDs. Acrylic can be glued to the PVC to provide a clear window for light transfer. Another benefit of using PVC is that heating it in the oven makes it very soft. So I can reshape PVC tubes if need be.

I also received the data logger and SD card this week. It took a lot of time to figure out how the data logger works, but I now have it writing values to a text file. 

Capstone 1

I begin looking into temperature dependence of LEDs, and found that temperature does have an effect on wavelength. However, if I can identify the range of temperatures my sensor will be exposed to, it may not have a significant effect.

According to one paper by Chhajed et al. (2005), the peak wavelength for UV LEDs only changed by about 5 nm over a 100 degree C temperature range. The temperature range of my sensor will be far less then that. However, the sensor will likely be in temperatures below 20 degrees. I will look for someone who has investigated this relationship down to lower temperatures, but I’d expect the same relationship to continue beyond 20 degrees. Red LEDs appear more sensitive to temperature. Although, I am not sure if the same relationship will hold true for when LEDs are used as photodiodes.

I also looked into data loggers and found one that should work well.  http://iteadstudio.com/produce/arduino-sd-card-shield-and-filelogger-library/

Here is another link to instructions for building an amplified color sensor: http://www.robotroom.com/ReversedLED.html

Chhajed, S., Xi, Y., Gessman, Y., Gessman, L., Schubert, E. F. (2005). Influence of

junction temperature on chromaticity and color-rendering properties of

trichromatic white-light sources based on light-emitting diodes. Journal of Applied Physics. 97: 054506.

Week 14

            This week we finished up our ROV and tested it in the pool. The ROV worked, but not quite as well as we had expected. It had a few problems with buoyancy once we put it in the pool. The write up on our ROV will be along shortly.

            As far as feedback on the course, I can’t say enough good things. This class was excellent and was like no other class I have taken before. The freedom to experiment on your own, make mistakes, and problem solve made the class fun and educational. However, I do have some recommendations for future classes. I saw this past semester that many students including myself wanted to (or needed to) come in during hours other than class time. It would be good if students knew specific times when you or Jim would be around. It wasn’t really a problem for me, but I know other students wanted more time in the lab. I worked with Colin in the lab a few times, which I didn’t mind doing, but I won’t be there in future semesters. So you should be sure let students know when the lab is open or find another student you trust to open and close the lab.

            I also think that using Arduino microcontrollers would add to the class. From what I have seen it is much better then the ones we used in class that were programmed in BASIC. It will also give students the choice to use the Lego brick or an Arduino to do data logging on their AUV/ROV. The Arduino could also be used in place of a Lego brick to control an AUV, if a student wished to take on that task. I also like the programming environment much better then the Lego software. The Lego software was useable but I found myself frustrated with it multiple times. I think that Arduino code is much more realistic as well. Learning how to write code using the Arduino (or BASIC for that matter) will prepare students to use other languages, whereas the Lego picture code does not. However, the Legos provide an excellent way to construct robots. So even though the programming environment is not great, it was still one of the most rewarding parts of the class.

            As I mentioned before the format of the class was excellent. I can see a number of different courses taught in a similar format. I wish there were biology courses taught using this hands on format. It might be possible for someone to graduate from the marine biology program here at UMaine without ever handling marine life (hopefully not). My point is that the majority of our courses are lecture based. The marine science department is lucky enough to have relatively small class sizes. This is perfect for having more lab-based courses. If students were able to study live animals and perform dissections, no one would be falling asleep in class! I think this is definitely true of physics courses as well. However, the physics department does do a pretty good job of this. The labs reflect what is done in lecture and are usually engaging.

            Lastly, I enjoyed the TED talks and definitely think you should keep them in the curriculum. It’s definitely a good way to inspire students and in many cases remind them to think out of the box.

            Finally just as a side note, I will get the diver visibility data to WHOI at some point next week. There is one-year day of Tara data that I have found the TSG file for but its in four different files. I will try and process this data when I get time. I will also get the ROV write up to you as soon as I can. I have got a million different things going on, so I apologize that my progress has been slow lately.

Week 13

            I really liked the TED videos this week, especially the toaster one. We really take for granted all the things that are provided for us. I heard a comedian make fun of this idea before, “People think they are smart. But if I were to leave you in the woods, how long would it be before you could give me a phone call.”

            In the lab this week, Dana and I got very close to finishing our ROV. We have our sensors built, the frame built, and the control box completed. Really all we need to do now is fix the buoyancy of our ROV. It sits pretty awkward in the water, so there is some work to be done there. Our control box is very compact, and contains all the batteries needed by the ROV (AA batteries for the motors, C battery for the hydrophone, 9 volt for the speaker). The speaker for the hydrophone is mounted on the control box, so the pilot should be able to hear it. Dana’s recorder will also be located in the control box. In contrast my sensor will be fully contained on the ROV with no connection to the surface.

            If all goes well, we will be done next week. This is good timing because I will have to leave early on Thursday for the marine science recognition reception at 4:00pm.

Week 12

This week Dana and I made good progress on our ROV and sensors. Dana’s hydrophone is completed and will have a tether to the surface, where sound can be recorded and heard in real time. My light sensor should be completed next week. This week I mostly worked on the programming aspects of it. It will now log data from the light sensor every second. There are 512 bytes of data in the eeprom memory of Arduino, so this will provide me with about 8.5 minutes of data. The analog output is from 0 to 1023, however the eeprom memory only stores values between 0 and 255. So I will have to divide the output of the sensor by four before saving it to eeprom. I’ll put the sensor inside something to block white light (maybe a piece of PVC) and place a red filter over the top of that. I’m also interested in building a depth sensor, because it would be nice to have the depth the light measurement was taken at.

            Last night I went up to the lab with Colin for a bit and discovered that I would be able to use conductivity to detect when the ROV is submerged. When I put a wire at 5v in water with a wire connected an analog port, the output jumps right up to 1023. Next step is to test it with pool water. I was hoping I could use this same method to detect when the ROV leaves the water. However, I am worried that having 5v constantly sent to a wire in the water will drain the battery. If this were the case, I can program the Arduino to cut power to that pin after it detects 1023. In any case, I already wrote the code to wait for 1023 on analog 1 then start logging light data on analog 0. In addition I hope it will start logging depth on analog 2, but I guess I need to look into how difficult a depth sensor is.

            Overall, our ROV is coming together nicely. My next step is to solder a 5v voltage regulator and a light sensor to wires that will connect it to the Arduino board. I was worried at first, but at the rate Dana and I are going we should have plenty of time to build our ROV.

Week 11

            This weeks TED talks were really good! I didn’t know that “inventor” was a real profession. Woody Norris’s invention to direct sound was awesome. This idea was very original and has many applications in our modern society. I also like how Dean Kamen showed another side of invention that is often forgotten.

            In other news, I am exited that our robot challenges are completed and we can now move on to a new project. I feel a little rushed for time to finished the final project. I guess that’s just how it has to be; there is only so much time in the semester. However, our hydrophone should be done soon. The instructions to build the hydrophone are very easy. That should be done next week, no problem.

            I want to build the frame for the ROV and mount the waterproof box by the end of next week. I think I am going to make a turbidity sensor that can go on the ROV in addition to the hydrophone. I am looking forward to getting back in the lab to continue building our sensors and ROV. 

Week 10

            Dana and I are slowing getting through our robot challenges. Now that we have completed line runner, we have two left. I am happy to see new challenges added to the previous ones. It’s just as fun to watch other people complete challenges as it is to do them yourself. I learned this week that even though programming of the Lego robots can be frustrating, it’s totally worth it. Finally getting the code right for line runner, after working on it for a few hours was satisfying. I can only imagine the satisfaction of completing more complex tasks, like depthx (the Bill Stone TED talk). Depthx has 96 sensors! That is an enormous amount. Being able to get all these sensors to work together is quite feet. I can see the allure of working as an engineer on projects like depthx.

            The talk by Richard Feynman was awesome as well. He provided at great reminder of why science is so interesting. It is fun to look at everyday things more in depth. I think this is why I like physics and chemistry courses much more then my peers. Most of my friends find the material in these courses irrelevant. I not sure why but I find these topics very interesting. I suppose I am more curious about fundamental things.

Next week Dana and I should have our robot ready for can grab. I have some good ideas for this robot. My ideas may be a little complex, but I’ll know next week whither my ideas are feasible. I am also planning on making the robot capable of picking up a full can. I should also say that I would like to build an ROV that has onboard video. I am not sure if this requires any parts we don’t already have, but I thought I would give you a heads up.

Week 9

            This week Dana and I finished our tablebot and it was a great success! It is very rewarding to have your robot and your program perform without error. When building the tablebot I was frustrated with the software at first, but once I got the hang of it I was having a lot of fun. We are now working on line-runner for the next challenge. On Thursday we started writing the program and by the end of class we got a pretty good start. We first attempted to have the robot measure the time it took to get to the first line, then go back to the beginning and suppress the light sensor for the amount of time recorded in the previous run. This did not work very will so we are now changing our code so the robot counts the number of line encountered.

            I used to believe that even simple robotics, like what we are doing now, was extremely complicated. I learned this week that it is not that complicated and can be done by anyone. I am having a lot of fun using the Lego robots and as a soon as a finish one task, I am excited to start the next one. I am especially looking forward to building the ROV and/or AUV. This is another task that is much less complicated then I had imagined.

            In the future, I think there should be more robot tasks to choose from. This way not everyone is doing the same tasks. This will make for a diverse set of robots and will allow students to watch robots complete more tasks then we currently have

Week 8

            This week Dana and I worked our way through the Lego robot tutorials and then started building the tablebot. The tutorials were very strait forward and didn’t take much time to get through. We started our tablebot with only one light sensor on the front. Our plan was to have the robot drive forward until it detected the edge, back up and turn a random direction and drive forward again. However, we decided it would work much better to have two sensors, one in the front and one in the back. This would make sure we didn’t back off the edge and it would allow us to record twice as many contacts with edge of the table.

            We attempted to write the program today and we were both a bit frustrated with the programming environment. We are having a hard time translating our ideas to code. I am considering adding a third light sensor to the robot as well. This would solve one of the problems we are having. Our robot has three wheels, one in the front and two the back, and the back wheels tend to go off the side of the table. I would position the three sensors so they were just in front or behind each wheel. However, the programming software seems to be the limiting factor. This is because I not sure I will be able to have the robot respond to cues from two sensors simultaneously. In any case we should have the tablebot ready for next week. 

Week 7

            I just finished watching Sylvia Earle’s TED talk, which was very well done. She covered many of the problems facing our oceans today. Talks like these definitely help to give me motivation to make a difference in the field of marine science. The first point she brings up is that we used to think the oceans were inexhaustible, but of course now we know better. This reminded me of my grandmother who always tells me that the oceans are so vast humans could never damage them. It seems completely unreasonable to think that today, but I suppose this was the consensus year ago.

            The second sensor project didn’t turn out quite like Dana and I had planned. We ended up changing our sensor to a temperature sensor, because we could not get a stable reading using our fluorometer set up. We integrated our fluorometer with the microchip, but when looking at the output, the counts were jumping over a very large rang.  Then when we changed the sample to a more concentrated dilution, the change was to small to be significant. We were running low on time, so we changed to a temperature sensor. Calibrating the temperature sensor was much easier then calibrating the fluorometer. We also had much less variation in the counts then we did with the fluorometer. We averaged six readings in order to reduce the variability in the counts even more.

            We ran into a couple problems when fitting a line through our measured temperature points. The calibration curve was not linear, so we found the best fit by using a fourth degree polynomial regression line, which fit almost perfect (R^2=0.9998). However, when we took the equation excel gave us for the line and entered the count values, it consistently gave us the wrong temperature value. I did the regression analysis again in matlab and got the same equation, however this time I got the equation to a few more decimal places. Using the equation matlab generated worked much better.

            I then spent more time then I had wished trying to insert the equation into the code. I wanted to the output of the microchip to be already in degrees Celsius, not in counts. No matter how I entered the equation, it would not work. It would output numbers that were far too large (>1000). Perhaps the microchip cannot perform more then a few operations in a row. However, I expect that the chip can perform the math and I am just entering something wrong. We ended up just entering the counts into the equation in excel to find the temperature. Our sensor performed well yesterday when we had a mystery sample; it found the temperature within one degree Celsius.

            We also ended up watching the space shuttle launch at the end of last class, which was pretty exciting. I am terribly disappointed that our space program is being downsized. It may be difficult to justify spending large amounts of money on a space program, but I believe it’s a mistake to cut funding to NASA and our space program. I believe discovery unites people, and there is nothing like discovery beyond our planet to unite people around the globe. 

Week 6

            I am looking forward to intergrading our fluorometer with the microchip on the breadboard. I think we will get much better results this way, instead of using the LED readout on the RadioShack kit. I also plan on putting our entire fluorometer set up in a box next week. This way we can keep out a lot of the ambient light. You will open the top of the box to put the sample in and then close it to read the sample.

            Last night I read a few chapters in the physical computing book, which I wish I had read before last class. Last class I was a bit confused about how the microchip works and what pins correspond to what. However, after reading the book I think I have a good understanding of how it works. I also read a bit about the BASIC programming language, it seems pretty strait forward and should not be hard to work with.

            My understanding of circuits and how electronics work is also coming together. For example I did not understand how components in a circuit could be controlled by a microchip. After doing the simple set up with the microchip and the LED last class, I realize the microchip can control the current sent to the LED and thus regulate the flashes. The example that we did last class tells me that the microchip also has some way of measuring time, which was alluded to in one of the chapter in the book.

            In the future I need to make sure I know the function of every pin on every component, this will greatly enhance my understanding of the circuits I build. One of my goals in this class is to be able to build circuits without instructions. I want to be able to build circuits just based on my knowledge of the components. 

Week 5

            I have chosen to go ahead with the fluorometer for my sensor. I am working with Dana on the sensor and we have made good progress this week. We started out by building a light detector using the photo resistor provided in the RadioShack kit. The circuit design was very simple and is connected to a potentiometer so we can adjust the sensitivity. The light intensity is displayed on the led readout on the RadioShack kit. That was the easy part.

            Next we set up a system of clamps so we can arrange a place for our light source, photo resistor, and our sample. We used a small test tube to hold our sample and a blue led light from the lab as the light source. We narrowed the beam of the light by pointing it at a small aperture. Our sample is located behind the aperture where the restricted beam of light will hit it. Perpendicular to the blue beam of light is our photo resistor. It is placed about a centimeter away from the sample. Over the photo resistor we have placed five layers of thin red transparent plastic. This stops the majority of scattered blue light from entering the photo resistor.

            The third step was obtaining a sample of known concentration to calibrate the fluorometer. This did not go so well, because we could not find any fluorometer to measure a sample. Neither Jim nor Bob Kirk (head of the chemistry laboratories) knew of a fluorometer we could use. So we decided to make a stock solution of unknown chlorophyll concentration and make our calibration in reference to this stock solution. We did this by using a quick and dirty chlorophyll extraction procedure. We boiled a few spinach leaves then mashed them into a paste and allowed the paste to sit overnight in rubbing alcohol. The next day we sieved the remaining chunks of spinach out of the mixture. It worked very well, the solution lights up bright red under blue light.

            We then started making dilutions of this stock solution in rubbing alcohol. We tried MANY different dilutions but did not find any correlation between concentration and the led readout. The next day we rearranged our set up and made sure every piece of the layout was very steady. Most importantly we made a new holder for the samples, so the samples would be in exactly the same position every time. We finally found good results by making dilutions using only drops from the stock solutions in six milliliters of rubbing alcohol. We blanked our fluorometer system by putting in a sample of only rubbing alcohol and adjusting the potentiometer so the led readout was zero. We increased the concentration by five-drop intervals until we reached 55 drops and the led readout was 10. We made a calibration curve from these measurements. Our curve is linear from 0 to 20 drops, and then follows another linear trend with a lesser slope from 20 to 55 drops. I don’t quite understand this, but it seems to work.

            I hope everything works for us on Tuesday although I am a little worried. We really struggled with making a calibration curve for our fluorometer because our results were not consistent. Also I am not sure how much our stock solution of chlorophyll will degrade over the weekend. Perhaps on Tuesday our stock solution will have lost some of its chlorophyll content. As of now it is in a jar in my fridge and I have my fingers crossed. Overall, I had fun making it and even more fun watching different solutions fluoresce under blue light. The picture below is our fluorometer while a sample is being read.

Week 4

            The talk by Bonnie Bassler was very thought provoking. What I liked the best was the connection she made between intercellular communication and multicellular organisms. It makes perfect sense that this would lead to multicellular organisms, because the cells inside multicellular organisms still have to communicate with each other. I also liked the introduction where she talks about the number of bacteria on a single human. The number of bacterial cells on a human is ten times that of human cells, I find that to be an impressive amount. This is a symbiotic relationship that seems to get little attention.

            I had a lot of fun building simple circuits on Thursday. The little bit of circuit building I did left me wanting to do more. I am excited to start creating my sensor, however I still have not chosen a specific sensor yet. I am considering a light sensor that will measure light intensity. Next class I want to try building a circuit only from the circuit diagram, not the step-by-step instructions in the book. One thing I am confused about is the need for so many resistors even in simple circuits. Why doesn’t a single resistor proved ample resistance? I am sure it will all make sense as I become more immersed in the subject. In any case, I can’t wait till next class. 

Week 3

            As presented in John Delaney’s talk, the future of oceanography is promising. The idea of setting up an autonomous underwater base is very cool. It is definitely advantageous to use automated and remotely controlled instruments when working in the ocean, especially on the floor of the ocean. These instruments can remain on site and can be controlled day and night from almost any location. The convenience of this is amazing. The speed and amount of data collected can be increased significantly by using remotely operated stations, such as the ones in Delaney’s talk. I was amazed when he said that we have the capability of doing in-situ DNA analysis. I can’t wait to see remotely operated stations on the floors of all the world oceans.

            As far as my scratch project is concerned, I am getting a bit frustrated with my simulation. The bacterium is having a hard time find the food source. The issue with particles moving diffusively (Brownian motion is what I’m trying to simulate) is that no information can be gained by sensing the direction of the moving particle. If a particle touches the left side of the bacterium, this does not indicate that the food source is to the left of the bacterium. I do have the bacterium moving further in the same direction when it encounters more particles, however it can still be moving up the concentration gradient diagonally and miss the food source. It’s a hard problem to solve. However, I have some time this weekend to figure it out. My next step is to figure out what to have the bacterium do when it senses it is going down gradient.  In this case it is definitely moving away from the food source, so turn 180 degrees? Add or subtract 90 degrees? I am not sure what to do. This is a thought problem not a programming problem, I will not find the answer by look at my scratch program. What would I do if were blind and contacting randomly moving particles? How would I find the source? Lastly I may have to make more particles in my program in order to increase the concentration gradient, this should make it somewhat easier.

Week 2

            I enjoyed watch Ken Robinsons talk, however I don’t think the education system is a flawed as he thinks it is. He believes that our education system is impeding our progress as a civilization and destroying young peoples creativity. No matter what I agree there will always be room for improvement in the education system, however in my experience I have found the current system to be satisfactory. The major flaw in education as I see it, which he did no mention, is the cost of education. People do not have the opportunity or choose not to attend a university because the costs are so high.

            He also mentioned that there is an emphasis on math and science and not art or dance in our schools. I may be somewhat bias because I am interested in the science, but I believe that science is much more important then art or dance. That’s not say that people interested in the arts should not peruse it. There are institutions specifically for the arts, and that’s how it should be. However in elementary school and high school, students should be taught a significant amount of math and science (with some freedom of choice to take more or less of a specific subject). It is my opinion that it is unacceptable for people to be ignorant when it comes to the sciences. In my eyes, understand the world around us is currently the most important and pressing task of all humans.

            It is also people who are uneducated in the sciences that are impeding the progress of discovery. I watched Randy Olson’s film Flock of Dodos a couple months ago, and it brought up a good point, which is what I am trying to get at. There are people who believe the earth was formed 6,000 years ago. These people do not understand modern science and are interfering with the teaching and progression of science. However, now I have gotten onto an even larger subject of debate, so I will leave it at that.

            My second scratch project is already going very well. I have made most of it last night because I couldn’t stop tweaking things in the program. My simulation is going to be a flagellated bacterium that has to find a food source that is emitting small molecules. At first I had the molecules leaving the food source on a strait trajectory, however I found this unrealistic. So now I have the molecules moving diffusively, the x and y positions of the molecules changes by a random number every 0.5 seconds or so. In the final product the user will be able to move the source to any point in the stage (top left, bottom right, center, ect.).

            I am now in the process of designing the bacterium “brain”. As of now I have colored each of the four sides of the oval shaped bacterium with a different color. The different colors act as sensors, so the bacterium will know what side the molecule touches. However, this design was originally for the strait trajectory molecules, so I may redo the code for the bacterium. I may take a sum of the number of molecules that touch the bacterium over a certain time period and if the sum increases over two successive time periods it will continue in that direction. If not then it will turn and move another direction. This will make the bacterium move up concentration gradients. I look forward to showing this simulation in class.

Week 1

After a couple months of absence, I am back to writing weekly updates here on Trials and Tribulations of a Lab Tech. The weekly updates will be on my progress in and thoughts concerning my class this semester (Engineering Literacy For the 21^st Century).

            I am excited for this class and expect it to be much more engaging then most of my other lecture-based classes. So far I am enjoying the first scratch project. I have a pretty neat idea about how I can make an interactive advisement for Sustainable Fisheries Foundation (SFF). In essence it will be a game that demonstrates increased fishing intensity over time, and will be ultimately impossible to beat. The message being that everyone losses when fisheries are not managed.

            So far I have found scratch easy to work with and I have been able to do most of animations I had in mind. It is becoming a bit more complex then I had planned, but I’m still having fun with it. There is a timeline included in my advertisement that will have to run a little fast. I suppose I could make it run slower, but I would have to make a lot more sprites, you’ll see what I mean next week. Overall, so far so good!

            Randy Pausch’s last lecture was very inspiring. One of the points he made was that, brick walls are there to stop people who are not willing to try again. Also, on one of his slides was, “The best gold is at the bottom of a barrel of crap.” I found this funny and true. I think that these ideas would definitely give me motivation when I feel overwhelmed or defeated. I liked the entire lecture and his overall optimism; I would suggest this lecture to just about anyone.