Author Topic: Bad Printed Circuit Board, BAD, BAD!  (Read 1195 times)

Offline egenoup

  • Sr. Member
  • ****
  • Posts: 359
Bad Printed Circuit Board, BAD, BAD!
« on: April 26, 2011, »
[disclaimer] this is not a rant, complaint or in any way putting down the great effort that RJ and family do in bringing us Xmas lighting equipment for an affordable price [/disclaimer]

With that out out the way, I just wanted to post this issue I had with one of the LE's from the first COOP.  It was actually the first one I built from that COOP.  By no means am I new to soldering, so once I was done assembling the LE, I was confident that it would work.   Boy what a let down!  

I attempted to load the firmware into the PIC and right from the start, the Pickit 2 software could not recognize the device.  If I enabled the vdd output, it would sometimes correctly identify the pic.  Other times, I had to have both the vdd output and the /mclr to get it to identify the pic.  Though it could identify the pic, it would not read or program the chip.  I would get an error "... fail" (i did not write it down).  I got one of my 2010 LE's and hooked up the Pickit to it and I was able to read and write the pic, so my pic programmer was not the problem.

I got my meter out and began taking readings and was able to get continuity for all the ICSP pins to the pic , with the exception with the ground pin (#3).  I did not have continuity from the pin to the ground trace.   Check check check, I could trace ground from the large cap all the way past the via that connected the ICSP pin to the ground trace located just above the PIC pin 28.  I decided that maybe I had a cold solder joint, which would be my first in many years, but one never knows.  I retouched all pins on the pic and all pins on the ICSP.  Retested for continuity from the pin to the cap ground and nothing.  I read continuity from the ICSP pin  on one side of the board to the same pin on the other side, so it is not the solder joint... HMMMMMMMMM, lets test the via...... and wahlah! there was no continuity from one side of the via to the other.  I just removed some of the solder mask from the ground trace and also from the ICSP trace on the other side.  I put a piece of wire (from on of the resistors) through the via and soldered it to both sides.  Problem fixed!  Sooooooo I must have gotten the one in 10,000 bad boards and just happened to be the first one I built!   I also went and bought the lottery just in case I is a sign of good luck!

So in summary, if you are having trouble with programming the pic, something else to also test is that the first five pins of the ICSP header should have direct connections to one or more pins on the pic!

Once again RJ, Thanks for all the effort you put into bringing these devices to us.

Egenoup

Offline mcangeli

  • Full Member
  • ***
  • Posts: 63
  • Original Gangsta
    • Mcangeli
Re: Bad Printed Circuit Board, BAD, BAD!
« Reply #1 on: April 26, 2011, »
I thought that if you bought the LE in a coop, you didn't have to program the pic? (I know I didn't program any of the ones with my three LE's)

Not that it negates the problem of the board....

Glad you were able to get it working.

Mark

Offline egenoup

  • Sr. Member
  • ****
  • Posts: 359
Re: Bad Printed Circuit Board, BAD, BAD!
« Reply #2 on: April 26, 2011, »
I had one of those DUH moments.... I had already began trying to program the chip when I realized that the chip should be programmed.  It did not matter, since I tried to run it and it would not come up until I fixed the via (tiny holes that, in most cases, connect traces on one side of the board to traces on the other side of a double sided board).  They are little tubes that are placed in the board after it is drilled.

Egenoup
« Last Edit: April 26, 2011, by egenoup »