External Links Bar:
Home
Helix Light Controller

Standalone, Wireless, Networkable Light Controller



Left Hand Side Navigation and Information Bar:
Helix PCBs
Helix PCBs are now available. Please send an email to Orders to request information.
What's New
The New Helix Main Board v2 is available. This site will be updated soon!


Main page content starts here

A Standalone Controller


Since the Prop 128 v1 only used six of the eight cogs and had 12 free I/O pins I felt there was too much power and capability left on the table. Some people say "if it isn't broke, don't fix it", most engineers say "if it isn't broke, it probably doesn't have enough features." So in early 2008 I started thinking about improvements I could make to the Prop 128. One of my first thoughts was to try to make it a standalone system. I needed to figure out a way to store the sequences locally on the main board and be able to play MP3 songs. After more research and a couple of false starts I found a couple of breakout boards that I thought might work. The first was an SD card socket breakout board. This would allow me to store the sequences and MP3s on a SD card for easy access by the main board. The second was an MP3 player breakout board. Since I had brought 10 of the 12 unused I/O pins out to headers on the Prop 128 v1 board I was able to start testing the breakout boards almost immediately. I found a file system object and MP3 player object on the Propeller Object Exchange. They weren’t exactly what I needed but I was sure I could modify them to be. It took a lot more work than I expected but I finally had my file system and MP3 object.

The Real Time Clock


Another component I needed was a Real Time Clock (RTC). This would be necessary to have the controller start and stop autonomously. The supplier I used for the two breakout boards also had RTC chips available. I ordered a Phillips PCF8583P RTC and the required 32.768 kHz crystal. Once it arrived I realized I wouldn't be able to breadboard this circuit up because the leads on the crystal were too small to make contact with the sockets. This would come back to bite me later, but I'm getting a head of myself.

The Line Drivers


One of the good things that came out of my DIYC post about the Prop 128 v1 was a question about the maximum distance between the daughter boards and the main board. They pointed out that since I was sending the data signals to the daughter boards at logic level voltages, without any sort of drivers, that the maximum distance must be fairly short. In my 2007 Christmas display I had a daughter board located about 50′ away and connected to the main board with a 75′ Cat 5e patch cord. It worked perfectly but the question got me thinking. In my rush to design the Prop 128 v1 I didn't even think about line drivers. I decided that it would be a requirement for the Prop 128 v2.

Hand Soldering SMT


After some more research I settled on a couple of chips from Maxim-IC. The first was a quad RS-422 driver, MAX3032EESE+. This was the only RS-422 driver with at least three transmitters I could find. However, it had one big drawback: it is only available in a SOIC/16 surface mount package. I found a video showing several different techniques for hand soldering SMT. I decided I would have to try hand soldering of SMT chips.

The second chip I chose was the quad RS-422 receiver, MAX3093EEPE+. It was available in a 16 pin DIP chip.

A Better AC Zero Crossing Circuit


Another aspect of the Prop 128 v1 I didn’t like was the AC zero crossing circuit. It worked but not like I expected it to. Then in May 2008 we held the local Mini-PLUS at my house. I set up my Mega-Tree in the backyard and was prepared to show off my custom controller to 20+ Christmas lighting enthusiasts. Wouldn’t you know it, that is when the zero crossing circuit decided to fail. Luckily I was able to quickly change the firmware and bypass the need for the zero crossing, of course the only control over the light channels I had was full on or full off. That incident sealed it; I had to find a new zero crossing circuit. After even more research I found an AC zero crossing circuit that was simpler and better than the one I had used. I tested it with the Prop 128 v1 board and it worked great.

Wrapping Up The Design


The last two things I wanted to change were to include the power supply on the main board and design a custom daughter board. I didn’t really like having a separate power supply but since I used the same board design for both the main board and the daughter board on the Prop 128 v1 I needed to minimize the board size. There was already a lot of wasted space on the daughter boards. So I thought if I redesign the daughter boards to be as compact as possible, I could increase the size of the main board to be able to accommodate the power supply. After redesigning both boards I was pleased with the results.

Parallax World Wide Design Competition


By this time it was getting to be late summer. That is when I discovered that Parallax was having a worldwide Propeller design competition with a $2000 first prize. This provided a lot of motivation for me to finish the Prop 128 v2. The deadline for entry was Oct 21, 2008 and I didn’t even have my PCBs ordered yet. I quickly finalized the design of the PCBs and ordered them. After about two weeks they arrived.

The Assembly


Once again I set about assembling the boards. It took a little longer than I expected to hand solder the three surface mount RS-422 transmitters, but overall I was able to do it. Once I finished the assembly of the main board I started slowly testing it. First I tested the power supply without any of the chips installed. Next I installed the Propeller chip and its EEPROM chip. These first two tests worked out great. Then I added the XBee radio to the board. Again everything worked. Next I added the RTC. I plugged the board in and nothing happened. I had lost communication with the Propeller chip. After removing the RTC, everything started working again. It took some digging but I finally found the problem. The RTC and the Propeller’s EEPROM both use the I2C communication protocol. As such I put them on the same bus. The problem was that both chips used the exact same device address. When both chips were installed neither one could work properly since they were both trying to respond to the messages. Luckily it wasn't a fatal error. I was able to scratch out one of the traces on the PCB and solder in a jumper wire to change the address of the RTC. After that everything else worked like a charm.

Contest Deadline Approaches


I spent the next couple of weeks tweaking the firmware and software to get everything running smoothly. It took me a couple of more weeks to put the project report together for the Parallax contest. I finally finished everything and submitted my design with a full day to spare. For the next several weeks I waited for news from the judges.

Debut on DIYC


Toward the end of Oct 2008 I created a post on the DIYC forum to debut the Prop 128 v2 to the DIY crowd. It immediately received a lot of attention. This time I did feel it was almost finished enough to be able to share it with other users. Since it was close to the elections and Prop 128 v2 sounded too political, Bob Lilly, one of the first people to show interest in it, suggested I change the name. I asked for suggestions and John Wilson from Melborne, Australia suggested Helix. He said Helix means Propeller in Catalan (it is a Romance language, the national and official language of Andorra). There were several other great suggestions also but that one struck a cord with me so it stuck.

Contest Results


On Nov 25, 2009 I finally received an email from Parallax. I had won Honorable Mention in the Propeller design competition. It wasn’t first place but I was very proud that my design had received recognition. It also came with a $100 prize.

The 2008 Christmas Display


Due to a variety of factors, the 2008 Christmas display was not as grand as the 2007 display. There were problems with the new LED lights that I ordered. I tried to increase the size of the Mega-Tree but one of the couplers on the support pole started to fail. So I ended up taking the Mega-Tree down before I could even get it fully up. I did get some lights around the perimeter of the roof and the lights in the windows. I was able to control those with my new controller. Everything went great.

Back To The Drawing Board


While the Prop 128 v2 board was a much better design than the Prop 128 v1, it was still designed to meet my needs only. Since it was receiving interest from a wide variety of people from as far away as Australia, it became apparent that the design needed some refining before it would be ready for wide spread use. Thus I went back to the drawing board to work on the new and improved Helix.