Minutes

January/February

• Calvin wrote a simple data storing program in python for us to build off of.
• A specific job was given for each member to complete by the end of the week. These included:
  • Contacting Roland for additional information
  • Creating testing patterns (sine waves, steps) in mastercam
  • Converting these patterns into g-code
  • Setting up the continuous data saving file in python
  • Looking into using pydaq c as a middle step between the real time data, and the data being saved in python
  • Writing a new draft of our group's sponsorship letter

• We went over the space delimited g-code that will generate our sine wave.
• We decided that for a pretest, it is easier to manually move the load cell to meet the sample for now.
• We're in the process of fixing the three programs we need to test.
• We will meet next week Wednesday to possibly do pretesting and put our midterm presentation together.

• Calvin fixed the problem with continuous data saving in python and CNC movement in g-code
• The continuous saving program now collects data from the load cell and stores it in a csv file along side a time stamp (See an example excel file of acquired data on our programming page)
• Supplementary python programs (pyDAQ tools, etc.) were installed on two laptops in order to run the the program along side the CNC machine (Unfortunately these are not functional yet)
• The program for CNC movement was successful ( The video of the program in progress can be seen in the following video)
• The home screen for this new device is now being designed to allow sample parameters to be input and to provide a user friendly screen to control the tests.
• Each step was tested seperately and we plan to combine them in the near future.

• Our sponsorship letter was edited by the writing center and is now in the final stages of revision.
• It was discovered throughout the last two weeks that Python 3.0 and editions above do not read our data saving code properly. We are working on making a compatible one.
• Python 2.6 and all extra downloads needed were put on to a laptop so that we could test position, time, and load together.
• The data acquisition program was made more concise by minimizing the time lapse between collecting and dumping data.
• Tests were combined for the first time.
• The new signal conditioner we have has not been calibrated with the load cell yet, so our load data from these first tests are not accurate.

• The plans for the final step of the project were put in place
• Our final product will include a video tutorial, a list of programs needed to duplicate our mechanical test, and the written programs to run a user friendly test.
• The sponsorship letter was sent out to Roland.

• We had difficulty getting the GUI (home screen) to work.
• The GUI program will now be made in Python 2.6 rather than 3.1
• The calibration and testing python programs worked and will now be split so that the GUI program can call upon them seperately.
• A list of programs and downloads needed to run the tests were made, and it was discussed whether to list them online or provide them as a downloadable package.
• The final stages include putting together the calibration code, testing code, and g-code converter with the GUI, and creating a help menu to setup the physical connections.

• The GUI is now in the debugging stage
• The website template now has over 20 pages which are being edited. It includes tutorials, information, etc.
• The plan is to have a final video of everything working together made by monday. We will follow the finished tutorial to make sure that our instructions are concise and understandable.
• Roland responded to our letter saying that their engineers could not share with us the information we need for the Creep Test. A clarification letter is being formulated.

• The CNC Mechanical tester now works!
• Four minor changes were made to the GUI code
• The website is up and running!
• Written and video tutorials are running as well.
• The bulk of the project is complete, we will continue to troubleshoot as problems do arise. So far, data graphs for the packing peanut look promising.