I did some measuring of currents in the various fuses, as well as some open circuit voltages...

At the time it was sunny, and the panel was producing an open circuit voltage of 19.5v.

By removing the load fuse and putting the meter in its place, I could measure the current of the load in various configurations... Bearing in mind we are referring to the gazebo and 5 post caps, here are the numbers...

Full on (255,255,255) = 5.8A, so I replaced the 10A load fuse with a 7.5A one. On the bench, one postcap at full bore was measuring between 400mA and 450mA.

Full off (0,0,0) = 0.133A, the individual cap on the bench measured 22mA at 32Mhz, and when I turned down the clock, it measured 20mA. I decided that it was not worth the effort to change the clock speed depending on output. The regulators are causing havoc. I am toying with building a couple of the new design controllers, with the new FETs and seeing what happens with a single regulator.

Charging the battery, I did a monitor run on charging the battery, I have not yet plotted the results, but I will. First glance showed about 800mA-1.1A under full sun, with it dropping to 200mA-400mA with cloud obscuration. At some point I want to run the battery down and measure the whole charging cycle properly, but I can see that it is working as expected.