Jared Suttles' Lab Notebook

Week 08

February 26, 2006 (4 hours):
Today we all met and worked on the Design Review presentation. We made updates to the schematic and PCB, and we created all of the slides. We also delegated speaking roles, and we are well prepared for the design review. Some work will need to be done tomorrow, but not much. Josh wasnt available all day, so we need to finish his part tomorrow.

February 27, 2006 (2 hours):
We made some small changes to the presentation, which included taking updated screenshots of the PCB and schematic. We also took time today to print the 11x17 plots of the schematics. Finally, tonight we printed a bunch of PCB handouts... This was tough since the 1:1 printing wasnt working when using Adobe PDF as the printer. So I had to photoshop and scale it. I will try to figure this out later.

February 28, 2006 (4 hours):
After getting THOROUGHLY grilled by the panel at todays Design Review, we spent some significant time reviewing the schematic and PCB and attempting to implement most of the changes that the panel suggested. The major one we wanted to tackle was to purchase all of our parts so that we can make meaningful changes to the PCB. I ordered pretty much all of our parts from DigiKey, and we hope to receive them before the end of the week. This has lifted a great weight off of our shoulders, and I think I'm going to take a few days off to study for some other classes.

February 28, 2006 (1 hours):
I spent an hour checking all of our data sheets to make sure that the voltage swings on logic pins for our peripherals was large enough to accept the 3.3V output from our microcontroller. This had been a concern during the Design Review. I was pleased to find that all of our parts were 3.3V tolerant. Thus, we dont need to buy and level translators.

March 2, 2006 (2 hours):
I found today that I was actually wrong about the input tolerance of all of our peripherals. I hadn't thought that the RS-232 level translator wouldnt be correct, but it in fact worked only with 5V inputs. I spent a short while surfing Maxim's website and I found the MAX3233E Level Translator that uses internal capacitors, and is made for 3.3V inputs. I am very happy that this didnt take long to find, and that it is the same footprint and basic part as the previous. It does have a slightly different pinout, so I spent some time redoing the schematic to reflect this change.

March 4, 2006 (5 hours):
Josh works Sunday nights, so he couldnt attend, but Jonathan, Jennifer and I spent a long time in lab working on a lot of things. I focused most of my time on generating custom footprints for the PCB layout of our specific discrete components. This took some time, since we hadn't even thought of what passive components we would be using during our previous iterations. I also helped Jonathan load some startup code onto the evaluation board for the first time. Josh had worked on it before, but we had never tried, so it took a little while to figure out. At first, Code Warrior was saying that we were missing a NE64.pe file. We found that file by running a NE64 patch. We then realized that we wouldnt actually need to implement the Processor Expert tool, so that .pe file wasnt necessary. I learned a lot tonight and set up a good framework for further work on the PCB.

WEEK 08 SUMMARY
Accomplishments: Prepared and presented the Design Review. Got ripped apart during the review, so took a lot of the advice and implemented the suggestions. Bought all of our parts, and made sure that they all worked. Began the process of completely redoing the PCB and schematic.
Weekly Work Total: 18 hours
Project Work Total: 88.25 hours

Week 09

March 7, 2006 (2.5 hours):
I went to office hours today with Jen. I wanted to ask Brian about some final concerns involving part selection and interfacing. First, there was a 1500 pF capacitor in the circuit for the switching regulator, and I accidently thought that was 1.5 uF, when it was infact a 1.5 nF capacitor. He told us we could use a non polar capacitor, which was great news, since I could find that in the cabinet. I was also concerned about how the printer we received had a male DB9 port on the back, and I had anticipated a female. Brian said that I can just buy a gender changer for our purposes.

We also got Brian to do our part layout and he said that everything was good, which was awesome news. All that is left is to re-route the whole PCB.

March 7, 2006 (5 hours):
I just had a small break for 495f lecture, but Jonathan and I spent the rest of the evening working on the PCB layout. Specifically, I routed all of the ethernet circuit, which was a major concern during the design review. It was somewhat challenging since there was a lot of hardware to pack into a 1 inch area. It took some time, but but I finished routing that section, and moved on the the rest of the micro pins. Jonathan took over at this point and managed to finish the entire board by the next morning. Meanwhile, I worked on the circuits for the power supply. This was the first time that I had actually wired up a prototype circuit and checked the output voltage. Luckily, the outputs were perfect. I actually cant prototype the 12 V switching regulator circuit since I dont have any through-hole inductors. It is probably OK though, since the RFID reader can run on a wide range of voltages, and the most important thing is that the signal is clean and regulated, which will hopefuly be the case. When I left the lab, Jennifer and I made a few small changes to the schematic in order to update the pin orientation of the LCD header and make a couple more small changes. The final version of the schematic can be found on the documents page.

March 8, 2006 (3.5 hours):
Since Jonathan did a great job routing the entire board last night, I spent most of the morning taking small time chunks to change things. I made the vias a little smaller and then changed some routes to be more optimally routed and spaced. I also fixed a couple of small things mentioned during the Progress Briefing today.

March 9, 2006 (2.5 hours):
I spent some time updating more small changes to make the design more optimal today. Later, Jen and I redlined the entire PCB and schematic and corrected a couple of traces that should have been routed more directly, rather than through multiple components. After making sure the whole PCB was correct, I ran the post processing, and submitted the files it generated to the freedfm website. It mentioned quite a few errors that it could automatically fix, but no "show stoppers", which is good news. So I zipped everything up and emailed it to Barry. It felt GREAT to be done with this a day early. The final top and bottom coppers can be found on the documents page.

WEEK 09 SUMMARY
Accomplishments: We finalized all of the parts and footprints for our design. We also placed and routed the entire board from scratch again. When it was all said and done, we verified our layout, and we submitted it to the website and it came back good. A huge part of this project is now complete!
Weekly Work Total: 13.5 hours
Project Work Total: 101.75 hours

Week 10

March 19, 2006 (6 hours):
Because I was on RA duty this weekend, I decided to spend the time working on the project, and Jen and Jonathan were also here, so we met in lab to work on software. Josh had gotten a BDM from Chuck that works for our micro (the first one was 5V only). So we began writing code and loading it onto the chip to test functions. The serial code is going to be the hardest at this point, because we want to keep the serial monitor code loaded on the chip in case the BDM fails. However, we decided to work on the Keypad code. I thought it would be a good idea to hook up a 7 segment LED and test the DEMO kit we received from Freescale. We are using Port G to drive the leds, and we are using Port S to control the keypad polling. The code basically uses a modulo counter to iterate through the 4 output signals we use to choose columns, and then we check the row signals to see if one is high. It isnt working well right now, as no matter what we try, the Port S[0] pin is driving a 3.3V high, though it should be an input... I plan on attending office hours later to get this working.

WEEK 10 SUMMARY
Accomplishments: Worked on software, mainly the keypad algorithm. We are still having problems, but I think we actually did decent work for one night (especially since its Spring Break still).
Weekly Work Total: 6 hours
Project Work Total: 107.75 hours

Week 11

March 21, 2006 (1.25 hours):
After lectures today, Jonathan, Jen, and I went to lab to work on the keypad software again. Jonathan had to leave shortly thereafter for a research group meeting, but Jen and I kept working on the code. We decided that for some reason the Port S[0] pin just wont work. We then moved the row reading pins to Port J, and things finally started to work correctly. We first got the 7-segment LED to output a different hex character for each of the 16 keys. We then thought it would be a good idea to test out the internal pull devices of the chip. We found it was really easy, and we could remove our external pull down resistors. Jen and I considered stopping here, but our sudden success made us want to test more, so we added code to receive and check a PIN against a hardcoded "correct" pin value. We anticipated that the code would be working much too fast to allow for 4 key presses without software support (one press would be read as a lot, due to the clockrate). When we confirmed this fear, we added a simple loop that holds the code until the button is released. This worked well, and we succeeded in accepting and checking the PIN. We still dont know why Port S[0] wouldnt act as an input, but it will be used for serial communication anyway, so that shouldnt be a problem.

Keypad + PIN Check Code Listing

March 22, 2006 (3 hours):
The software is really starting to come along. Jonathan, Jen and I went to lab and began working on new modules today. I thought it best to start working on the evaluation board since it had some 2 free serial ports, and serial communication was our first goal. We first hooked up the RFID reader to one of the serial ports, and we started writing serial initialization code. We ran into our first problem when we had to figure out what Baud divisor to use to get the 9600 baud we needed. We needed to know our internal Bus Clock, and we found a seemingly contradictive statement in the datasheet. It stated that the Bus Clock was half of the Core Clock, and the external crystal we are using is 25 MHz. Hence, we assumed (without reading very much) that the Core Clock was 25 MHz and hence that the Bus Clock was 12.5 MHz. The baud divisor we first tried didnt work, so we then remembered that in order to have the ethernet run at full speed, the Bus Clock had to be 25 MHz, so we tried a new Baud divisor and it worked perfectly. We knew that it was working, because we used the other serial port to send the just-read data to the computer and we viewed it on hyperterminal. We knew that there might be a race condition where the RFID reader would broadcast the RFID serial number multiple times, and we confimed that it was in fact the case. This can be easily handled later on by keeping track of serial numbers that have been previously scanned.

We then began working with the printer, and quickly succeeded in getting it configured to be used with Serial communication, 9600 baud, no flow control, and the US character set. We then attached it and found that the serial printing standard is awesome, and it simply accepts the ASCII value for most characters and it prints them. We easily made a simple printing program, and we then coupled it the RFID reading, and output the tag ID read and some text. We found that it at times wrote a random character before the RFID value. We think its a product of transferring the data directly from one serial port's read to the other serial port to be written, which wont happen in our final product, so we dont anticipate this being a problem.

We are excited that things are going so well in software, but Josh is still trying to figure out the ethernet code. Hopefully that goes as smoothly as the rest of our code.

RS232 Communication Test Code Listing
RFID + Printer Test Code Listing

March 23, 2006 (3 hours):
I came over to Jonathan's room tonight to help out with the LCD code. I ended up taking a long time to convert the monochrome bitmap we had into a binary array that we could use in our project. I first had to download a program called Batch Converter (though I'm sure other programs would work as well) that converted the BMP file to a PBM (portable bit map) format. This PBM format is basically a very small header, followed by the byte by byte values of the bitmap. This worked brilliantly for what we needed. I then just needed to get the actual binary out of the file, which wasnt very easy since notepad encodes it to ascii. However, I found a website that provided a solution:

perl -wne'printf "%08b ", ord $_ for split //' foobar.bmp (I added a "> logo.bin" to the end to capture it in a file.)

After that, I imported it into excel to line up the bytes in a nice order and do some string manipulations. I then copied it all into word to format it right, and I finally had an array. That took a while, but it was worth it. Here is the binary before being manipulated: binary file. And here it is in ASM format: asm text file.

March 24, 2006 (4 hours):
Today, I worked on 2 parts of the project. Josh and I worked on the ethernet code for a while, and Jennifer worked on ohming out the board and checking it under a microscope for mistakes. It turned out that the board looks fine, but our ethernet code was not going so well. A major problem was that we couldnt use the BDM to program any of the ethernet sample code onto our evaluation board. We had to use the serial monitor, which isnt an option in the future since we need both serial ports for our project. When we couldnt really get anything to work, we decided to shift our focus the power supply. Josh went to research ethernet in his room, and Jen and I because to work on the board.

We soldered on the two voltage regulators without much issue. I did solder a polarized capacitor on backwards, but with the solder wick, I was able to easily remove it and resolder it correctly. We checked these two power signals and they were both accurate and clean. We then worked on the switching supply, where we had a couple of problems. First, we soldered a resistor in the wrong spot. Second we realized that we had alotted a small capacitor footprint for a 100uF capacitor, which was much larger. We found a slightly smaller capacitor, and we soldered it on as best we could. It made a good connection and it seems to work just fine. We were unsure as to the actual value of this voltage, since we were unable to test it with the surface mount inductors that we had bought, but it turned out that we had a perfect 12.00 volt signal. It was also very clean. We left the power supply running over night so as to "burn" it in.

March 25, 2006 (2 hours):
Today, Jen decided to solder on all of the bypass capacitors. I thought it would be a good idea to ask around about the BDM problems we were having, and James Doty showed me how to use it without any of the settings I was trying to configure. This is huge, because we have struggled for a long time on how to get it working.

Josh also came to lab, saying that he had gotten the BDM to work, which was funny since I had also just figured it out. So we began to work on the ethernet code again, and we had a huge breakthrough. We found that we could spoof the MAC address and all of the other ethernet settings on our board to mirror our provided ECN computer. When we did so, and we plugged the CAT5 cable from the computer into the board, we could ping the board via a PAL internet connection. This meant that the board was accepting and sending TCP/IP packets over the internet. Basically, we need only find out how to send specific packets now to make it so that we can access our external database and send emails.

March 25, 2006 (2.5 hours):
I went to Jonathan's room again tonight to work on the LCD code. He had a test circuit up and running. He was trying to figure out how to get his old assembly code from 362 to run on the demo kit. After struggling with it for a while, we decided to just convert it to C code. We used his assembly as the base, and we wrote C to do the same thing. We never got it to work before calling it a night, and we realized right before we did that it was because we didnt attach the LCD logic Vss and Vdd signals. That would make it hard to turn on... We plan on working on that tomorrow again.

March 26, 2006 (5 hours):
The LCD was the main focus of the night. Jonathan and Jennifer came to my dorm room because I was on RA duty, and we worked there. Last night, we had decided it would be a better idea to just use C code, since neither of us really understood the Assembly all that well. We might try and go back to assembly just for efficiency sake, but we'll decide that later. We worked for a long time, and we realized that we were having trouble because the 5V supply that provided the backlight and the logic supply had a different ground from the 3.3V signal coming from the demo kit. We didnt know if it was a great idea, but we tied the two together and everything started working! We made a few test screens, and we ran into an error about allocating memory... It might be because of the size of the bitmap we are using for the logo is too big, but then its only 7k, so that shouldnt be a problem. Anyway, we chopped off the last few lines, and it worked. So we decided to minimize the drawing by taking all of the whitespace off the top and bottom and hard coding it into the code. This saved us about 2-3k of memory. Now, everything in our project has at least some code that works, and that is awesome.

<- wicked cool logo, props to Jonathan with his MS Paint Skills

LCD Logo Display Test Code Listing

WEEK 11 SUMMARY
Accomplishments: Coded RFID / Printer Serial Test Code. Ability to ping the eval board over general internet connection. BDM now works to flash any code. Coded keypad polling test code. Created logo binary file. Coded working LCD initialization and Logo Display code.
Weekly Work Total: 22.75 hours
Project Work Total: 130.25 hours

Week 12

March 27, 2006 (4 hours):
Today, I spent probably too much time on the cool RFID xpr3ss Code Viewer webpage... Afterwards I determined that shay really should have PHP installed on it, as using javascript as a hack to do what I needed is really kind of shady. I also found a really strange bug where if we put a index.something file in the Codes directory, it will break the autoloading link list on the right side of the page. This is because of the hack way I did it in javascript, so we just will make sure to not use a code file named index.

As far as progressing the project, I did assist Jonathan in further testing the LCD. He had simplified our code to use repeatable functions, but the test code has ceased to work. I looked over the code, and we quickly got it working. His reset logic was inverted, so the LCD was in constant reset mode. He got the Character mode working tonight, so everything is going very well.

LCD Test Code v2

March 28, 2006 (2 hours):
After lecture today, we went to lab and worked on a lot of things. I went ahead and soldered our microcontroller onto our pcb, Jennifer developed a code heirarchical diagram and pseudocode for our design, Josh continued to work on ethernet code, and Jonathan continued to update his software flowchart to match our design. Soldering the microcontroller was stressful, since I wanted to make it perfect, and there were inherently some pins that did not want to solder nicely. I ended up finishing the job under the microscope, which was actually a really cool experience. It was neat to watch the solder heat up and then flow with the flux, and it was easy to know when to stop heating the pins. We all had to leave for other classes, but we accomplished a lot in this short period.

March 28, 2006 (3 hours):
We worked on the Software Narrative TCSP tonight. We basically had to decide on a lot of different aspects of the design from a software perspective. We discussed many options of program flow and implementation before settling on a hybrid system we are calling Interrupt Flag Driven, Polling State Machine. It is based upon a main polling loop which first checks the state of the system, and then checks certain flags depending upon which state we are in, and then handles the flags as needed. This gives the most control to the main loop, which adds flexibility and ease of development to the program flow. You can read more about it, as well as the actual TCSP we came up with by visiting docs page or the direct link below.

Software Narrative TCSP

March 29, 2006 (2 hours):
I spent a couple of hours in the lab today populating more of our board. I added the Reset circuit and began the oscillator circuit. With these on, the board is actually near completion, which is exciting.

March 30, 2006 (4 hours):
Tonight, Jennifer and I helped Jonathan and Josh with revisions to their papers. Though we have waited a long time to finish these assignments, the night went well. I helped Jonathan the most with his paper (Software Narrative) and Jennifer helped with both papers. Inherently the Software paper was slightly more complicated, and thus it required quite a bit of discussion and revision. Both papers can be found on our docs page or in the direct links below.

Software Design Considerations
Patent Liability Analysis

WEEK 12 SUMMARY
Accomplishments: Added the microcontroller, reset, and most of the oscilattor circuit to the PCB. Developed the Code Viewier webpage. Finished the Software TCSP, Software Narrative, and Patent Liability Analysis.
Weekly Work Total: 15 hours
Project Work Total: 145.25 hours

Week 13

April 4, 2006 (5.5 hours):
Today was a very exciting day! We made a lot of progress populating the PCB and we were also able to load some successful code onto the board. Jennifer populated the oscillator, RS232, and EPHY circuits, and I worked on a couple of pieces of code to test the boards functionality. I wrote a heartbeat function that simply flashed an LED on and off. Once the board was populated enough to try it out, we loaded it up and it worked beautifully. This was a great thing to see since we were all nervous about the board. I then ported over the RS232 code from the code written for the demo kits to code that would work on our board. We tested it out, and it worked again! So I added some code that would do some simple manipulation and would look up a name and price based on the serial number. It took some debugging, but it worked! We dont have any code to test the ethernet code, but we are anxious to test that out too.

April 5, 2006 (4 hours):
Our progress briefing in lab today was excellent. We could actually say that we had a lot of things working on the actual board and that we anticipated having the project completed ahead of schedule. We mentioned that our voltage regulator was running a little hot, and Chuck offered us a piece of copper to dissipate the heat.

We also had one of the worse experiences of the project so far today... Jennifer and I realized that we had made a major error in the design of the ethernet circuitry. Basically, an RJ45 connection has 4 lines that are of major concern: TxP, TxN, RxP, and RxN. We had swapped the Tx and Rx pairs, and had also swapped the N and P in each pair. This was mostly due to an inconsistency between the datasheet for our RJ45 jack and the suggested ethernet schematic for our microcontroller. We literally had the soldering iron hot and the solderwick in hand before I thought of a better idea. We could just make a custom CAT5 cable that would work for our application. After searching long and far for a kit to make a cable, we found a friend of Josh's that made us a cable. Unforunately, one of the connections was bad on that attempt, so we had to find Chuck, who gave us a kit to add a new end to the cable. This time it worked perfectly. The Startup Demo code allowed our board to establish a link with the desktop computer and we were able to ping the board's IP address. This means that every major peripheral works on our board!

April 6, 2006 (3 hours):
Jennifer and I spent a long time tonight working on the main state machine code for the project. We started with the pseudocode that Jen wrote before we had done much decision making, so many changes were needed, but it did provide an excellent framework for the eventual final product. Although much is left to be done, this is most definitely going to be a work in progress for the remainder of development. A copy of our various versions of main code can be found on the codeviewer page.

April 7, 2006 (1.5 hours):
Jennifer and I again worked on the main code today. We finished the state machine and began attempting to integrate the code modules that we had finished in previous testing. We identified many of the different modules that have yet to be written and need to be. We made some changes to our original ideas, such as storing the PIN as an array of characters rather than an integer. This is best, as it adds ease of use to the code, as well as ease of computation. The latest version of our code is here.

April 8, 2006 (2 hours):
I went into lab today with Jen and I designed a large test circuit meant to utilize all of the different peripherals at once. I hooked up the keypad, RFID reader, printer, and the EVB board all in one large circuit. I never got around to trying it out, but I am sure I will do that tomorrow, as we intend to get to lab early. Jennifer designed just about all of our message screens for the LCD based upon the various states. We had determined that it was easiest to have no real arguments to the UpdateLCD method, and rather to just have it switch off of the state variable and determine what to display based upon that. I began to compile the main code before we left, but there are still many errors to be ironed out.

April 9, 2006 (6 hours):
The whole team got up bright and early and went to lab today. We worked on a lot of things and got a lot done. My first task was to get the RFID interrupt code working. I ran into my first problem when the EVB board wouldnt even acknowledge that an RFID had been scanned... I hooked the reader up to our real board, and the code worked nicely, although interrupts still dont work... I developed a non-interrupt solution that should work since our code is running so fast, but it would be nice to have that functionality as well.

The other things we did today were integrate Jonathans LCD code into the main function, as well as Jen's printer code. Jonathan's code had a few bugs to work out, as we realized that we needed to write a simple string copy method, which we made quickly, and things fell right into place. See below for and image of the shopping cart user interface. The printer code also didnt want to work at first. We wanted to run the code on our board, but currently our board uses the printer's power supply, so we tried to use the triple output power supply to do the trick, but it had a bad connection. We used the DEMO kit, and it now works awesomely. A picture of the receipt format can also be found below. The test code, with the LCD and Printer methods in it can be found here.

WEEK 13 SUMMARY
Accomplishments: Completely populated the board. Worked extensively on the main function code. Saved the ethernet circuit from certain doom by creating a custom CAT5 cable. Created fully functional versions of the RFID handler, LCD update, and Receipt Printer codes.
Weekly Work Total: 22 hours
Project Work Total: 167.25 hours

Week 14

April 10, 2006 (2 hours):
Today was meant to help us figure out the problem with the interrupts... We have tried many things. We have put a simple piece of code into the dummy ISR that the entire vector table points to, and the code never reaches that code. Anyway, I just thought that the CCR's I bit was meant to be set when interrupts were enabled. Well, I was apparently wrong, because adding a simple "asm cli;" to the code definitely fixed everything. Now the RFID interrupt works, as well as the timer interrupt.

April 11, 2006 (5 hours):
With the interrupts now working, I added the RFID interrupt code to the main program file that we have been modifying. Jonathan also began using the crimper and heat gun to make some good wires to connect the LCD and the keypad to the PCB. I finally got a seemingly perfect timer interrupt working that runs every 1/25th of a second. I quickly whipped up some code to increment the needed variables and create a clock on our LCD. In the few minutes that I spent watching it tick, I think it was pretty perfectly timed. I am going to leave it on all night and see what time it shows tomorrow.

April 11, 2006 (3 hours):
We also worked on the TCSP presentation for tomorrow. It involved blocking out our schematic even further than before, and creating FMECA tables about the safety and reliability of the project. Jenny and I worked on calculations and the like, while Jonathan and Josh created the powerpoint for our tables to go into. It wad time consuming, but the beauty of it is that Jen's paper will be a lot easier to write now.

April 12, 2006 (3 hours):
First, I checked the clock, and it was an hour off plus 5 seconds... The hour was due to the fact that when I calculated a 12 hour change, I set the hour value to 0 rather than 1 oclock. The 5 seconds is an acceptable loss over a 12 hour period for our application, especially since the system can synch with the server clock whenever rebooted.

Second, we worked a long time on the keypad today... We are getting some very odd results... All of the keys work with the exception of the 3 and "Enter" keys (they are next to each other in the same row). It doesnt make any sense at all that all of the other keys work just fine. We made a lot of changes to see if we could get it to work, including polling on an RTI interrupt, Timer interrupt, and attempting to wire the ORd value of the rows to the IRQ pin... Some things worked, but never all the keys at once. Basically, my hypothesis was that it was simply a race condition that was brought because the column strobes took too long to reach the keypad before we checked the rows. So, I placed the column strobing in the timer interrupt and the row polling in the main code. Now everything works great, as this increases the time that the columns are asserted and the rows are monitored. You can check out the code here.

April 13, 2006 (6 hours):
I wrote the Ethics and Environmental concerns paper tonight, with help from Jen in terms of editing. It took a long time, but in the end I feel that I wrote some good info about the concerns our project should face. Jen also completed her paper, and I helped her edit it.

HW11 PDF
HW12 PDF

April 15, 2006 (3 hours):
Today in lab, I worked on integrating Josh's ethernet code into our main design and Jennifer and Jonathan wrote some support functions for the UDP server that would help send emails. Jenny and I then set up the total test circuit using the crimped wire that could span the headers correctly. Before we test the main code on our test circuit, I have to make it work with correct pins, since some arent available on the demo kit/ EVB kit and likewise. We didnt get it to work tonite, but tomorrow is prommising.

April 16, 2006 (4 hours):
We went to lab today and began debugging the overall circuit. First, the LCD wont display or initialize, so that makes it really hard to know what is going on, but we quickly rectified that problem. It, along with the keypad was the victim of shady connections from the wires we were using for testing. After a lot of debugging, we had almost all of the functionality working in one way or another, except for the printer which is acting really funny. I dont know whats wrong, but it wont do anything at all. We will fix that problem tomorrow hopefully. My first inclination is that the tx and rx lines are switched, which might mean we have to hack up a cable, but we'll determine that tomorrow.

Regardless of that, the main loop pretty much works, and the UDP server application works great. Check them out below.

Main.c
UDP_Client_app.c
RFIDudpServer1Thread.java

WEEK 14 SUMMARY
Accomplishments: Figure out interrupts. Fixed the keypad. Built the overall test circuit. Got ethernet working. Began debugging the overall test circuit. Found out about the printer problem. Weekly Work Total: 26 hours
Project Work Total: 193.25 hours

Week 15

April 18, 2006 (3.5 hours):
Today was both troubling and rewarding... First, Jennifer and I continued to debug the complete system, and we got almost everything to work. On the other hand, the printer did not feel like working, and we couldnt figure out why. Jen and Jonathan ended up going to class, and I opted to hang out and try and fix the problem. Bryan and I were using an oscilloscope to see the serial data coming off of the port on the micro, but it disappeared on the other side of the level translator. Bryan then looked at the datasheet and found that we had tied the FORCEOFF signal (active low) to ground, rather than the Vcc it should have been attached to. I chopped the pin off the board and soldered it to its neighbor (Vcc) and it started to work! This was the last functional piece that we had to get working on the board.

April 19, 2006 (3 hours):
Today was the final progress briefing, and it was awesome to be able to say that everything works perfectly. We got a copper flag to solder to the heat dissipation pad of our 5V regulator from Chuck. Jen and I worked on a few bugs in the design, including one where you could scan a user keyfob as an item, and garbage would be printed to the LCD. We fixed it by adding comment lines between the database sections, as well as making the lookup code look for them. We also finally realized that our printer ALWAYS printed a 1 in the cents of the total thanks to a messed up looping index. Jennifer then added comments to our code, and I worked a little on the packaging with Josh.

Later, we assembled the items all together in order to film our PSSCs. All of the sudden the keypad was acting erraticly and the LCD was REALLY messed up. We believe that it was interference with the RFID reader and the rest of the base unit. This wont be a problem in the future since the RFID reader is separated from the base unit by an aluminum sheet and plastic packaging.

April 20, 2006 (5 hours):
We finally began taping our PSSCs after another bout with random operation. It seems that each time we touch the system and reorganize it, something else goes wrong. It must be either massive contact errors on our wires, or interference with the base unit. We really hope(believe) that the problem will be rectified when we get it into its packaging. After taping the PSSCs and on the way to class, Chuck gave us our alumninum sheet, and the hole and the "X" were perfect.

Jen and I spent a brief amount of time drilling holes for the LCD mounting screws, then we made sure they worked before we left to study for our exam later tonight.

When we returned from the exam, Josh had brought the packaging, spray painted black, and had mounted the whole design. Unfortunately... it didnt work. The keypad was acting erraticly. After much discussion, we had Jonathan tighten the wire crimps, and we soldered a new ground header to the PCB in the PLD expansion slot. We then attached the a grounding wire to the header and a screw attached to the aluminum face of the packaging. After grounding the chassis, it worked perfectly. Images of the finished product (minus a few desired cosmetic dress-ups) can be seen below.

April 21, 2006 (4 hours):
We re-taped our PSSC video in one continuous session today. The packaging is finally complete, so it looks a lot better in the video. It turned out to be about 3 minutes long, and it will be pretty easy to edit into specific sections.

I came back later with a piece of acrylic that I bought at Home Depot. Josh and I cut the acrylic to fit the entire case, and we drilled the appropriate holes in it. We then mounted it on the case, and it looked awesome. Josh accidentally screwed one of the large screws too tight, and a small spider crack appeared near the upper right of the front. It's not a big deal, so we'll be alright.

April 22, 2006 (2 hours):
Jennifer and I went to Radio Shack today and bought some blue LEDs. We then returned to the lab, only to learn that LEDs we bought weren't very bright. We spent a long time trying to think of a good way to rig them up to the chassis, and we couldnt really think of anything. What we really need is a brighter set of LEDs. We'll go back to Radio Shack later and buy the brighter ones.

WEEK 15 SUMMARY
Accomplishments: Fixed the RS232 Transmission problem. Finished the packaging. Filmed the PSSC Videos. Attempted to put cool blue light into our cool case.
Weekly Work Total: 17.5 hours
Project Work Total: 210.75 hours

Week 16

April 24, 2006 (3 hours):
I spent the night editing the PSSC video for class tomorrow. I only had Microsoft Movie Maker at my disposal, but I guess it does a decent job... You can watch it here if you'd like...

April 27, 2006 (4 hours):
Jen and I finally made it back to Radio Shack to pick up the mighty blue LEDs. We tried a bunch of different configurations, but nothing was really easy. In the end, we decided to just solder a resistor to one lead of the LED, and then we crimped connectors onto the resistor and the other lead and simply wired them to 5 V and Ground directly. Now the "X" cut out of the case glows a funky blue. It does however shine directly in the eyes of a person standing in front of it... maybe we can bend them to a less obtrusive angle later... We also printed out some labels onto decal paper and affixed them to the front of the case.

April 28, 2006 (2 hours):
Jen and I finished the User Manual tonight. She had already written most of it but I helped her finish it and edit the final version. Check it out below.

RFID Xpr3ss - User Manual

April 29, 2006 (5 hours):
I worked a long time on fixing up my papers for the final report. I also began compiling everything together and started to linearize the references. This is going to be a lot of work...

April 30, 2006 (12 hours):
I spent literally my entire Sunday working on this mammoth paper. I pity the teams who waited until today to start... Anyway, the editing took forever, and the references needed to be completely reworked, the software didnt format nicely... So basically it wasnt fun, but it is now finished and the final product is beautiful. Check it out below...

RFID Xpr3ss - Final Report

WEEK 16 SUMMARY
Accomplishments: Finished almost all of the papers due Monday. Still need to finish the poster...
Weekly Work Total: 26 hours
Project Work Total: 236.75 hours