New 2017.12.23 PCB Bringup

[Maurice]

This latest batch of PCBs got delayed by almost 2 months, but they are finally here.

• Couldn't upload to sam3x.  Found the issue was the varistors used by my pcba house had too much capacitance.  I removed them and I could upload code to the sam3x and it seems to work.
• Loading the esp8266 through the sam3x upload method doesn't seem to work.  Need to investigate this.  Uploaded with ftdi cable and that works on one, but not another one.  Need to investigate why ftdi uploads failing sometimes.
• Uploaded firmware to working esp8266 module via OTA update and it worked well.
• Verified webpages work and used test trigger button which seemed to work except one flash port didn't light up.  Tried another CA6 board and a different cam port led didn't light up.  Need to determine why all the cam leds aren't lighting up on all the boards.
• Ran the aux, module, and cam port tests with test jigs.  Everything passed except aux 12,13.  This means the cam ports that were lighting up do work.  Need to investigate aux failures.

Overall I'm happy it's basically working, but I was expecting a cleaner bringup since the previous version was working.  I was not expecting so many issues, but over the next few days I should be able to diagnose these issues.

[Ernie Hatt]

Bad luck Maurice. but I am sure you will work it out.

[Maurice]

I figured out the problem with uploading to the esp8266 through the sam3x.  That was a software bug.  I can upload 3 of 5 esp8266s with this method now.  Oddly there are some esp8266 modules where I can't upload them with the ftdi cable even though I can upload them through the sam3x.  That's add because the ftdi path is so simple...  Anyways I'm going to need to start probing the board at points of failures to figure out why I'm seeing all these different failures.  It's going to be a long process, but I'm determined to figure out exactly what is causing all these different failures.  I'm afraid it's going to be lots of issues with assembly which is something I've never had problems with, but this is also the most complex board I've ever had assembled.  Once I have a list of the diagnosed problems I'll have to talk with my assembler to see if they can help me figure out what is going wrong.

[Maurice]

The assembler things the issue was a failure they had in the solder paste mixer.  They are sending me 2 more PCBs to make sure that fixes it.

I found a fundamental flaw in the way I was doing my IR emitter module.  I wouldn't ever get it to have a stable enough 38 Khz output so I'm changing the design.  This is the only module that I've found doesn't currently work.  I tested the ir detector and it does work great.

I also found the layout for the light module just doesn't work so that physical design needs updating.

There are also lots of other small adjustments to module layouts I want to make in the next version, but I'm going to keep testing with the existing modules.

I'm working on a new case design for most of the modules.  It will leave the top of the circuit boards exposed, but I can't come up with a better option without really expensive cases.  I'll share an update on my plan for the module cases soon.

Overall things are going pretty good, but I'll need some updated modules before I can make the kickstarter videos.