20040418 - In order to help promote my company http://www.electric-clothing.com I've decided to adopt a number of gorilla tactics. The first being to make clothing for myself and others that can be worn all over so as to invite people to ask what the items being shown are all about. This would present a opportunity to hand out business cards. Jexe will be my first friend to receive a piece of clothing after this bag is completed. I will be modifying a existing back pack. Its a european over the shoulder xbox bag I picked up in Seattle. I liked its rugged material and silver look. However I hate that it says "xbox" which is a Microsoft made gaming console. As much as I'd like to use newer technologies like AVRs and C for this project its somewhat overkill for the simple blinky EL pattern I wish to create for this project. I'm going to take what I believe to be "the right approach" and use the simple following materials. Parts: ------ o PIC 16F84 o 9V battery o 9V regulator o 5V regulator for PIC o 10pin ribbon cable for cleaner EL to PCB hookup o capacitors to cleanup power o 20Mhz crystal o 8 triacs (TO92 / 400V / 600mA) o 8 resistors (1000ohm) o 18 pin DIP mount for PIC to PCB holding o Buttoneer to hold EL o Scotch tape around EL to prevent sliding o Wire wrap for GND ribbon connection from EL (fine hairs) Order of Operations: -------------------- o Hookup angel hair to ribbon cable (aqua/yellow EL) (5hrs) o Schematic Design (2hrs) o Pattern Coding (3hrs) o PCB Design (1hr) o Board Etch, Cut, Solder (3hrs) EL Layout: ---------- \ | / -- -- / | \ o Pipes and double dashes are aqua. o Slashes are yellow. o Piece zero is the center top pipe. o Count up to seven going clockwise. o 4" of usable EL per piece (cut 6") Patterns (repeat 2x then go to next): ------------------------------------- 1. Rotate Clockwise slowly starting from top piece, then come back quickly. 2. Rotate Clockwise slowly only aqua on. Leaving aqua on slowly rotate clockwise turning yellow on (every other). 3. Rotate Clockwise two pieces simultanious across from each other. 0,4 1,5 etc. 4. Flash allow Aqua, Flash all yellow. 20040425 - Started and Completed Schematic today. It took abut two hours. This board is really just a conjunction of the led_jumpsuit and el_pants so I used those schematics for reference which great speed things up. I had to change gschem2pcb. The paths for /usr/X11R6 were hardcoded into the script. Originally it looked like: /sw/bin/gschem2pcb [...SNIP...] #include(/usr/X11R6/lib/X11/pcb/m4/connector.inc) #include(/usr/X11R6/lib/X11/pcb/m4/dil.inc) #include(/usr/X11R6/lib/X11/pcb/m4/misc.inc) #include(/usr/X11R6/lib/X11/pcb/m4/plcc.inc) #include(/usr/X11R6/lib/X11/pcb/m4/to.inc) #include(/usr/X11R6/lib/X11/pcb/m4/qfp.inc) [...SNIP...] Now it looks like: [...SNIP...] #include(/sw/share/pcb/m4/connector.inc) #include(/sw/share/pcb/m4/dil.inc) #include(/sw/share/pcb/m4/misc.inc) #include(/sw/share/pcb/m4/plcc.inc) #include(/sw/share/pcb/m4/to.inc) #include(/sw/share/pcb/m4/qfp.inc) [...SNIP...] /sw/share/gEDA/scheme/gnet-gsch2pcb.scm WAS: [...SNIP...] (define m4-pcbdir "/usr/X11R6/lib/X11/pcb/m4") [...SNIP...] /sw/share/gEDA/scheme/gnet-gsch2pcb.scm NOW IS: [...SNIP...] (define m4-pcbdir "/sw/share/pcb/m4") [...SNIP...] Also tried the big cheat. Adding a single symlink to /usr/X11R6/lib/X11 called "pcb" pointing to /sw/share/pcb. No love. gschem2pcb still fails. Error: the footprint CONNECTOR for the device CONN1 does not exist Error: the footprint TO92 for the device D8 does not exist Error: the footprint TO92 for the device D7 does not exist Error: the footprint TO92 for the device D6 does not exist Error: the footprint TO92 for the device D5 does not exist Error: the footprint TO92 for the device D4 does not exist Error: the footprint TO92 for the device D3 does not exist Error: the footprint TO92 for the device D2 does not exist Error: the footprint TO92 for the device D1 does not exist Error: the footprint R025 for the device R9 does not exist Error: the footprint R025 for the device R8 does not exist Error: the footprint R025 for the device R7 does not exist Error: the footprint R025 for the device R6 does not exist Error: the footprint R025 for the device R5 does not exist Error: the footprint R025 for the device R4 does not exist Error: the footprint R025 for the device R3 does not exist Error: the footprint R025 for the device R2 does not exist Error: the footprint TO92 for the device CR1 does not exist Error: the footprint R025 for the device R1 does not exist Error: the footprint RADIAL_CAN for the device C2 does not exist Error: the footprint RADIAL_CAN for the device C1 does not exist Error: the footprint RO25 for the device S1 does not exist Error: the footprint TO220STAND for the device U2 does not exist Error: the footprint DIL for the device U1 does not exist 'gsch2pcb' seems to be working a lot more reliably. $ gsch2pcb el_ring-bag_display.sch 0 file elements and 24 m4 elements added to el_ring-bag_display.pcb pcb was able to bring up the elements. I had the ground and trigger connections backwards on the triacs - fixed. xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx I wanted to etch this board using using the floodfill method to speed up the board making process and cut back on the amount of etchant required. I just checked in a fix for this. To apply to your existing boards, do "select all" then hit Ctrl-K, then "unselect all" and you should be able to plonk down a rectangle. It doesn't clear text, but it will clear lines now. You can toggle the "clear" feature of individual lines by putting the cursor over them and hitting "j". This worked before, just pointing it out. For new boards, you should select the "new lines, arcs clear polygons" option in the settings menu before you add anything. xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx 20040427 - Drilled holes in PCB and worked on assembly code. The patterns are a little crazy right now, but what is amazing is that hte code works at all. I've been home for work for the last two days. Whatever I got made me really sick. I woke up at 5am this morning and wrote over a 100 lines of assembly code with only two small corrections to fix at build time. 20040428 - Hooked up the board. Looks like the microcontroller is sequencing (barely), but the EL is not lighting up. The SW1 (switch) was being used by my incorrectly. I really needed to aditional VIAs in the PCB for power and GND. Two more for the inverter would have been clearer as well. Probably would be best next time to use four terminal block holes or even a ribbon cable. At this point I can see the sequencing clearly, but it looks like I might have blown out the inverter. 20040429 - Final hookup of EL-Wire to bag. Used buttoneer tool (as seen on TV) to adhere EL Wire. Worked very well. If I would have done the EL Pants project with this tool I probably would have saved myself 30hrs of labor.