Newsflash
I've set up some message forums.  Use the link in the main menu to get to them.   Eventually, I'll integrate all this into the CMS.
 
This site looks best when viewed with a secure, standards-compliant browser like Firefox or Opera
LED Driver Board Redesign A Success PDF Print Email
 Written by Administrator   
Friday, 14 April 2006
As expected, UPS had a package for me today...my redesigned LED driver boards!

Image
The newly redesigned LED driver board prior to assembly.
I opened up the box, disposed of the biodegradable packing peanuts, and found two plastic bags containing the boards I had ordered. On my last order from PCBPro, they had sent me 55 boards for the price of 50, and I was eager to find out if they had done the same thing this time. I was not disappointed as they had sent me 56 boards this time. Thanks, PCBPro! I inspected one of the boards more closely and compared it to my drawings...looked good. If it didn't work, I would have only my own ineptitude to blame.

Eager to get started, I set about gathering my soldering iron and the other parts. My assembly area is one end of our dining room table, much to the chagrin of my wife who already has to put up with my laptop there 100% of the time. I lost track of time, but I think it took me about an hour to completely assemble the stack.

Image
The assembled LED driver board attached to the LED board.

There were two main changes that I made to the layout of the LED driver board. The first was to connect any unused pins on the PIC to ground through a resistor. Without pulling the pins to ground, the driver could behave erratically. It was very sensitive to the proximity of my hand. Whenever I touched the chip case (not the pins) in the vicinity of the unconnected pins, the PIC would reset.

Image
The assembled stack inside of the enclosure.
The second change was to move the spot where the power/data cable connected to the driver board to the center of the board. The original layout had the cable connecting near the edge of the board with the microcontroller in the center. The opening on the enclosure where the cable would enter was in the center. This arrangement made it impossible to run the cable into the enclosure given the thickness of the cable. Along with that change, I enlarged the holes where the wire would be soldered to allow the larger gauge wire to fit.

The first test I normally run (besides the smoke test Laughing) is the color ramp. I connected the device up to the lighting controller and took it for a spin. It executed the color ramp and the other effects perfectly! It also no longer exhibited the instability of the previous revision of the boards. I'm just not sure what I'm going to do with those old boards. I may find a use for them someday. For now, I have the assembly of 25 (maybe 30) of the fixtures in front of me. It's time consuming, but I quite enjoy it.

Higher resolution versions of the images shown here are available in the photo gallery .

Last Updated ( Saturday, 15 April 2006 )