I chose to make a RGB LED board for my outputs assignment.
I milled this board along with the networking boards to save time. I had an issue with the board not being level, but it simply meant that I had to run some milling paths again at a deeper depth, and clear some channels with a blade. In the future it would be sensible to mill a flat surface in MDF before taping the FR1 board to that, that way ensuring that the surface the workpiece is being placed on is registered with the horizontal axes of the milling machine.
Here is the board:
You'll notice that I have a big old jumper wire soldered on here. I had a time trying to find out what was wrong with this board: the quality of the milling wasn't as good as previous boards, due to that unlevel surface problem. I found one short between two pins of the chip, which was easily fixed, one bad trace which was easily repaired, and still there was a problem (the symptoms: communication errors, and the chip getting VERY hot). I eventually realised that one of the programming pins was not connected to its pin on the chip, but could not for the life of me find where the trace was broken - presumably as it passed under the headers. In the end I resorted to jumping the whole trace using the yellow wire shown. At first this still didn't seem to make a difference, with the same symptoms occurring as before. On the second programming attempt everything went fine however. It seems the board is intermittent in programming, but reliable in operation, suggesting that there is still a problem with that programming channel that I jumped I reckon..
Here is a video of the RGB LED working however:
I milled this board along with the networking boards to save time. I had an issue with the board not being level, but it simply meant that I had to run some milling paths again at a deeper depth, and clear some channels with a blade. In the future it would be sensible to mill a flat surface in MDF before taping the FR1 board to that, that way ensuring that the surface the workpiece is being placed on is registered with the horizontal axes of the milling machine.
Here is the board:
You'll notice that I have a big old jumper wire soldered on here. I had a time trying to find out what was wrong with this board: the quality of the milling wasn't as good as previous boards, due to that unlevel surface problem. I found one short between two pins of the chip, which was easily fixed, one bad trace which was easily repaired, and still there was a problem (the symptoms: communication errors, and the chip getting VERY hot). I eventually realised that one of the programming pins was not connected to its pin on the chip, but could not for the life of me find where the trace was broken - presumably as it passed under the headers. In the end I resorted to jumping the whole trace using the yellow wire shown. At first this still didn't seem to make a difference, with the same symptoms occurring as before. On the second programming attempt everything went fine however. It seems the board is intermittent in programming, but reliable in operation, suggesting that there is still a problem with that programming channel that I jumped I reckon..
Here is a video of the RGB LED working however:
No comments:
Post a Comment