Chapter 9 - The Documents

This was the final week before my set end date of March 28. As such, I had a few loose ends to wrap up, mostly with the documentation. The main thing I had to deal with this week was the pull data macro. This is the macro that opens each device’s spreadsheet and automatically copies and pastes the data into the master spreadsheet. After reviewing this problem in more detail, I found that the best way to handle it was to just have the user do that part. I was hoping to avoid having the user do this part since the user could incorrectly copy and paste the data into the wrong worksheet in the master spreadsheet. This would lead to major problems as you would then think a certain device registered certain points when in reality it was another device instead. Even with that though, I found that it was the best way to go, especially when this week, I ran into another problem with the pull data macro.

This week a staff member of Pam’s ran through the data downloading procedures and she informed me of the results. I was glad when she told me the documentation was easy to follow and that she was able to download the data. However, she did tell me something interesting. One of the devices registered no data points for some reason. So, she was only able to create five of the six spreadsheets. This is a problem for the pull data macro. The macro was recorded to open all six files in order. If one file does not exist, the whole macro cannot run at all. Also, if the name of one of the saved files from the data download was incorrect and did not match what is in the macro, the macro would again not run at all. So, the first thing I wanted to do was see what happened when a device did not have any data. I wanted to see why it was not possible to at least create an empty sixth spreadsheet so that the macro could still run.

To do this, I needed to redo the downloading procedures myself to see what happened. The HOBO shuttle keeps the data until a new set of data from a HOBO device is loaded off onto it. The shuttle will keep the previous months data until a new month is run. This is good in case something happens to the spreadsheets on the computer. All you would need to do to get the spreadsheets back would be to redo the procedures by simply plugging in the shuttle. Since the data was still on the HOBO shuttle, I was able to redo the procedures myself so that I could see what happened when the procedure came to the device that had no data points. After pulling the data off the shuttle as normal, I skipped straight ahead to the device that had no data (it was device L6 in Pam’s office). I tried to export the empty device’s data to a spreadsheet, but to no avail. BoxCar Pro informed me that it would not create the spreadsheet since the device had no data. So for every device that does not have data, the only way to get the file for the macro run would be to create it yourself. This was something I did not want the user to have to do as it would increase the probability of incorrectly naming the file, thus making the macro far more confusing for the user. Needless to say, this was pretty much the last straw with the pull data macro.

Although I wanted full automation, the pull data macro just created too many problems. If the files were incorrectly named or were not there, the macro would not run. If a device had no data, you would have to create the spreadsheet yourself. Also, if for some reason you needed to move all of the data to a new directory, you would have to recreate the macro. Due to all of this, I decided it would be better to have the user just copy the data out themselves and paste it into the master spreadsheet. From there the user could still run the rest of the master macro and most of the procedures would still be automated. With that decided, I had to start creating the documentation for this procedure. I also now had to edit the downloading procedures to let the user know what happens when a device does not have any data.

Before I could begin writing up the procedures, I still had to figure out why the sixth device did not collect any data points. First, I plugged the device directly into the computer and read it out using BoxCar. BoxCar picked up the device so it was indeed still functioning. While doing this, I noticed that BoxCar does not allow for apostrophes in the device name. Because of this, it was saving the spreadsheet file as Pam_s office instead of Pam’s. I went ahead and changed the name of the device to just Pams Office. While doing this, I remembered that I still needed to write the documentation on how to change the name of a device. So I started over, this time making sure to write down the necessary steps involved with changing the name of a device. Once completed, I wrote up the documentation on how to change both the name of the device and the interval (how often the device collects data points). With that documentation in place, the user could now change the interval and the name of the device should the need arise.

After completing that documentation, I still had to figure out what was wrong and why there was no data.  I was stumped. The device seemed to be working. I set the delay time on the device and put it back to its location. The delay time is merely a date and time you want the device to start collecting data points. I set the delay time for five minutes after I was finished. This would cause the device to hold off on collecting data points until that delay time hit. After putting the device back to its location, I still had no idea what the problem was so I decided to move on.

It was only until just now that I realized what may have happened. Pam had mentioned to me that her staff member told her that the shuttle at one point became full. At the time it did not hit me. I just assumed it was full because of the last months run. I thought I was just going to have to write a procedure to clear the shuttle each time. But then I realized that the shuttle is supposed to clear itself after a new HOBO device is loaded onto it. This means that once her staff member transferred the first device over, the shuttle should have been cleared and room should have been made for the new devices. This makes sense as when I ran it myself in February there were no remnants of January’s data.

I began by first looking at how to clear the shuttle. I found out that there was no manual way of clearing it. The only way to clear it is to transfer new data from a HOBO device to the shuttle. So I was still not sure what to do with this problem. Just now though, I think I realized what may have happened, and I do emphasize “may.” I have not talked to Pam or her staff member to confirm this. Her staff member may have transferred data from only five of the devices. If she went in order, Pam’s office would be the last device. When she got to the sixth device, the HOBO shuttle said it was full. From there she may have followed the procedures with just five of the devices. However, I am not sure what happened to the sixth device’s data. When I went into that device alone, there was no data. If the shuttle was full, I would hope it would just leave the data on the device. However, the shuttle did still recognize the sixth device as it had the devices name and even made the .dtf file for it when her staff member ran the procedure. So it did pick it up at some point, it just had no data points. Perhaps she tried to transfer the sixth and final device and the Shuttle was full, but not full enough to register that the sixth device was being transferred. Since maybe it was full, the device simply just lost its data. I hope this is not the case and that the shuttle does have some kind of a failsafe for this.

I will need to investigate this further.  I need to talk to the staff member and see what she did and what really happened with the full shuttle. I never expected the shuttle to be a problem. However, this was the first time we ran six full devices. So, I am going to have to take a look at this when they run the data set in April. I could have them run one now, but that won’t include a full month’s data set and as such may not fill up the shuttle.

Now, back to writing up the final documentation. I began writing up the procedure for creating the final results spreadsheet. I wrote up the documentation that made the user open all of the device’s spreadsheets. I made sure to note that there should be six files, but if a device had no data, there would be no spreadsheet. If a spreadsheet was missing, the user would just skip copying over anything to the corresponding sheet in the master spreadsheet. I then showed them how to copy and paste the data into the master spreadsheet. I was sure to make this part very descriptive and detailed so that there would be less chance for error. From there, the documentation told the user to just run the master macro. Once that is run, all they have to do is copy out the final results sheet to a new spreadsheet and then save that into the current month’s folder. For example, I went ahead and created the Mar08-results spreadsheet for the month. Every month, the user will run this procedure after the data download and they will create this new results spreadsheet. This spreadsheet shows each device and its out-of-range points all on one sheet. From there the user would then have to run the clear master macro which would clear the master spreadsheet of all data. With the master spreadsheet cleared, the user can copy and paste it into next month’s data folder when the user runs the procedure the following month.

I touched up and finished all documentation and saved them to the network drive in a documentation folder. I am confident in all the documentation I wrote up and I hope the new documentation for the results spreadsheet and how to change a device’s name are easy to follow. With the documents completed, I emailed them to Pam to let her know that I had completed all documentation.

As promised, I delivered all documentation and a working out-of-range algorithm by March 28. I do have to say that the out-of-range algorithm was not as easy as expected. I ran into a few more problems along the way then originally anticipated. I will not go into detail about what I learned from this project at this time as that is a specific assignment that is due two weeks from now. This assignment will be a final post that goes over what I learned from this project.

Even though I have completed the objectives of this project, I am still going to make sure that all is well when they run the data next month. I want to see what happens with the shuttle and what happens with the sixth device. Also, on April 17^th, I have a presentation to give to Pam and the course instructor Melissa Raulston where I will go over what was accomplished in the project among other things. After this presentation, Pam will review my performance and will fill out the sponsor evaluation form. This form is where Pam will have the opportunity to grade me on such things as professionalism, and how well she felt I met the objectives I set forth in the project plan. Once she submits the form, the project will be officially complete.

With that said, next week’s post should be really quiet. Next week, I will showcase the final project plan, which is due tomorrow for the class. I will also showcase some of the macros I created to show you how the master macro actually works. The master macro itself is nothing major as all it does is run a series of other macros. So I will be focusing more on the other macros. Until then, thanks for reading.