User:Anthony Salvagno/Notebook/Research/2010/10/26/DOG Signal Analysis and Protocol

I know if I don't write down what I am doing and how to do it I won't remember in the future, and also someone else won't be able to do it without Koch or my own help so this is for them too.

Protocol
I'll have to write up the full protocol on the KochLab protocols page one day, but for now bits and pieces will be here.

Ok so I've already done the experiments of scanning stuck beads with the trap and have saved the data.

Displaying Data as HTML
This is a great method to compare all the data at once.
 * First open LabView 8.5 and open Auto Report 98. It must be run in LabView 8.5 otherwise it won't run.
 * When you click the run program arrow a series of pop-ups will um... pop-up and you must go through the set.
 * First select the directory where the data is.
 * Next are a bunch of data parameters and in the first parameters pop-up make sure you have Quick Conversion selected. It is towards the bottom under Conversion Type.
 * For the purposes of this procedure you can just click ok on the remaining pop-up menus.
 * Next make sure the axis are the exact data that you want to display. It has to be exact otherwise it won't display properly or at all.
 * Make sure autoscale is turned on or off depending on your preference or need.
 * When everything is set, just click Go and the program will start converting all the .dat and .ini files to .jpeg and html and will be uploaded to the server.

Exporting to Excel
In order to analyze the data and fit the linear portion of the DOG we need to get it into excel
 * 1) Open Secret Peeking Software
 * 2) open the data set you want to convert and then go to the segment that you wish to analyze
 * 3) there is a section called Save converted data to text file, make sure everything looks ok then press the button under the header.
 * 4) Open excel and navigate to the folder and select the file that ends in -dfstweak.dat and open it.
 * 5) select the data range that you wish to select and do your business.

Analysis
Since Friday's data is bad due to the funky bead, I will only look at Monday's data. The files I'm going to analyze are:
 * This is where the html report saved:
 * Big and small bead report from Mon 10/25/10
 * Misshapen bead report from Fri 10/22/10
 * 0279.dat
 * segment 0
 * segment 29 (for comparison because it is the worst of the group)
 * 0280.dat
 * segment 0
 * segment 33
 * 0281.dat
 * segment 0
 * segment 1 (I wanna see the differences in the two graphs)
 * segment 41
 * 0282.dat
 * segment 0
 * segment 33
 * 0283.dat
 * segment 0
 * segment 27