Seems that I underestimated the problems that would result from attempting to assemble the PCB without the benefits of a reflow oven and a pick and place machine. After several attempts using a hot plate I have had to concede defeat as the QFN device (MCP73871) proved impossible to position and solder with the crude equipment at my disposal. Lessons learnt are to ensure that you have a good quality stainless steel stencil, preferably with a jig, and a reliable controlling the heat setting whilst monitoring the temperature of the hot plate.

So I admitted defeat and had half a dozen boards assembled professionally.

Testing is now underway but it seems that there is a problem that prevents uploading of the firmware. I have had much experience with ESP8266 devices although this is the first time I've used the -F version. Believe me, I am aware of most of the pitfalls to be avoided when setting up the correct conditions for 'boot mode' but currently, in spite of my best efforts, the devices fail to yield their MAC addresses which leads me to believe that the system has not defaulted to the correct mode on power-up. Thankfully I have a, cheap and cheerful, digital oscilloscope that should help get to the source of the problem.

Hopefully a solution will be found soon as the gardening season is rapidly approaching and there is much to do before I can deploy the system.