Test Plan

This week, we worked on a test plan for the SCABMS.  We spent time working out a decent set of test cases to be run on the SCABMS when it is built and ready to be tested.  These test cases will examine each functional area of the project. In addition to the test cases listed, we will be conducting a large amount of user testing, or beta testing, in the evenings leading up to the Engineering Expo.  Our next task is to obtain the proper parts and build the system.  Then, we can calibrate it and begin running the test cases.

Warning System

This week, I worked on the warning system. We had to make a couple of changes to the database for it to work. The first was the addition of a bottle size element, so that we could know what size the bottles were that were in the system. The other is an amount remaining element, to let us know how much liquid is remaining in each bottle. For keeping track of the amounts, we implemented some code in the pouring screen that basically “pours” from the database at the same time that the liquids pour from the system. Each drink subtracts a certain amount from the amount remaining columns of the liquids used. We successfully implemented this portion Sunday night and a bit more Monday evening. I also worked with Derek on the search & sort algorithms this weekend. We successfully got that working Sunday night as well. Currently, we’re waiting on brackets for the system. Until then, we will be working on a test plan. We will also work on getting the warnings for when we are short on liquid to display upon loading the main page. This hasn’t been added quite yet, but should be fairly simple to add. Aside from waiting for parts and building the hardware, the remaining task that will consume the most time is calibrations. We will update more as we find out more. I’ll post another update when we complete the addition of the warning bar in the system.

Searchalicious

These past few weeks I’ve been quite busy with all sorts of SCABMS business… from development to logistics.

I finished up the search feature on the Drink Menu. This allows for the user, via the onScreen keyboard, to limit the drink list dynamically as they type to fit their search string. This limiting is based on both the name of the drink and the ingredients included therein. I made the code quite robust, so that it would be simple to add another search criteria, like ingredient type or drink genre.

I’ve also been in talks with several people about the manufacturing of the custom brackets for the refrigerator to hold the bottles inverted. Looks like CounterPart Fab here in town will be laser cutting them out of stainless steel for us. We hope to get them this week and to start final assembly and testing from there.

As always I helped both Dave and Mike with the PHP, HTML, CSS, JS, etc… aspects of their assigned modules… this always helps to keep me busy…

Next I will hopefully be posting pictures and results of the full assembly process.

Edit/Remove Drink

This round, we worked on and wrapped up the Edit Drink and Remove Drink modules. They both work exactly how you would expect them to work, properly removing the drink and all of it’s entries from the database (except the ingredients, or course) and editing the drink just allows you to change the name and the amounts of the ingredients, but not the actual ingredients right now. We decided to do this out of simplicity and time constraints and it will probably be implemented in the future sometime. Go try them out under the Maintain Drinks page.

Alternative Offerings

I’m writing this week about our seeming lack in progress in our hardware development. While we didn’t get all of the hardware configured as we’d hoped… we did make quite the impressive prototype of what the cap assembly will ultimately be as well as a rough prototype of the holding apparatus (now to find a suitable material to construct it out of).

I took it upon myself to knock out a few feature implementation on the coding side this weekend as to have something to show for the lack of physical progress. I added some robust features to the drink selection process that now allows the user to choose a glass size for a particular beverage (given that that beverage has more than one glass option).

I also created the code that generates the execute strings for both the prime and clean lines functions, as well as put some finishing touches on their corresponding pages. A user can now clean/prime all lines or a chosen selection of lines and the code generates the system calls.

I did this in an easily configurable manner (for when we finally calibrate the system) using an array in PHP to store ounce values for each line and generating the system calls using our existing serial driver (hooray for reuse). What this allows us to do is simply change a value (in two places, primeLines and cleanLines) for each line to increase or decrease the value when calibrating and fine tuning the system. This is a far better alternative than having to recompile a version of our C code each time we make an adjustment.

Currently our biggest issue is the lack of Linux support from the company that makes our touchsceen (Elo Touchsystems). They have linux drivers and tutorials on how to install them… but they are all a bit old… We have even tried switching to an older kernel, (like the one found in Fedora Core 6) but to no avail. We will keep you posted on this development, but by all means if you have a solution for getting an Elo 1545L (more commonly 2500U) USB Touchscreen to work in linux please let us know, either by leaving a comment or emailing me (fernholz -at- gmail -dot- com).

Project Pictures, Finally!

This is a rudimentary mock-up for the bottle bracket system that we’re going to use. We haven’t decided whether we’re going to stick with wood or get some metal brackets for when it is in the fridge. Derek will probably be able to elaborate a lot more on everything to do with the bracket system, since everything is laid out in his mind already on how it will finally look.

Otherwise, here is all the wonderful hardware and the prototype!

Img1

Img2

Img3

Img4

Img5

Img6

Img7

Img8

Img9

Why Not?

Nothing big to report on right now. Still building the hardware slowly but surely.

I will be supplying my camera so that we can put up some nice video when it gets running.

Two Weeks of Building

This was the first week of our two week build cycle.  Everything is still in the works, we’ll post a longer update, and perhaps some pictures, when we have more to talk about.  Next week’s post should be a bit more robust.

Happy Harry Hardware

This week is half of a two week regiment of building the physical aspects of the machine. Putting everything together, assembly, “gutting” the fridge, etc. There isn’t a whole lot to report as of the first week due to conflicting schedules and sickenesses. More to come soon!

Progress is a warm gun…

This week my task was to build a four valve prototype to test the pressure drop when multiple lines come into the picture. This task isn’t quite resolved for the following reasons.

  • I was quite ill most of last week
  • I was waiting on a part shipment (that came friday)
  • Said part shipment had a wrong part that I quickly fixed on friday (the replacement came today)
  • I wanted to wait for the rubber stoppers to arrive so I could make a more permanent cap assembly instead of rough prototypes from this point on.

Now that all (and I mean all) of the parts are here we can actually start construction (which is/was my task for the next 2 weeks). I am quite excited for this stage. I purchased a refrigerator to house the system (and keep all the ingredients fresh and cold) on friday as well!

I will be building a more extensive prototype with a variety of cap assemblies this week. As well as working with Dave and Mike to start “ripping” down the refrigerator to start final assembly.  I think this step is crucial so that the priming and cleaning functions can be completely calibrated.

More to come soon! Check back!