This is my initial blog post for the Solar Power Contest. I thought I would provide some background on my Solar PV installation and the motivation for my design entry.
In December of last year I had solar PV panels installed on the roof of my house in Oregon. I had been debating whether to do it or not, but a number of factors motivated me to go ahead with it last year. I've been trying to be more earth friendly by reducing my overall energy consumption and thereby my carbon footprint. I've bought newer more efficient Energy Star appliances, added solar powered attic and crawlspace fans, added more attic insulation, use a programmable thermostat, and converted all my lighting to LED. As a result I got my annual electric power consumption to 5 MWhrs (14 KWhrs/day average) for a 3000 sqft two story house. Portland General Electric in Oregon has a net metering program that allows you to only pay for net energy consumption on an annualized basis (they do not pay for excess energy produced). Along with Federal and State tax credits and a State renewable energy rebate it became economically feasible to install a grid-tied solar PV system that would allow me to zero out my net energy consumption from the power company on an annualized basis (I'll provide detail in a future blog). The incentive for doing it last year was the availability of some older technology equipment at reduced cost and the fact that the rebates and tax credits were expiring or being reduced. As a result of doing it last year, my installed cost break-even point is about 4 to 4 1/2 years.
Being new to solar power, I've been closely monitoring the amount of power my system is generating. I've attributed most of the variation that I've seen to the total irradiance on the panels (sunny/cloudy/rainy days, change in the incident angle of the sun, shadowing, etc). I did notice in the spring that I was losing performance over a string of days with very similar sunlight conditions. It turned out that I had gotten a lot of tree pollen on the surface of the panels. Luckily a few rainy days fixed that. As we moved into the dog days of summer I noticed another performance drop off that correlated to the heating of the panels due to self-heating in sustained sunlight and high ambient temperatures. I realized that it would be nice to have a remote sensor that would report irradiance and temperature so that I could verify that the panels were operating properly (not dirty or damaged). Hence the idea for my project - Solar Panel Efficiency Checker (SPEC).
Here are some of my initial thoughts on requirements:
1) Use standard WiFi for communication
2) Use solar cell(s) for power (with battery backup)
3) Use solar cell for irradiance measurement
Use a monocrystalline cell which is similar to the panels
Mount in same plane and close proximity to panels (must not be prone to shadowing)
Must be "calibrated" against known standard
4) Use simple microcontroller for processing
5) Use temperature sensor on panel remote from processor
Analog or digital?
How many sensors?
6) Use cloud for monitoring
Here is my initial thought on implementation: (I'll come up with a block diagram and schematic soon)
Sparkfun 8266 Thing board for the remote processor
1) 802.11 WiFi
2) LiPo battery interface/charger (adapt to run off solar)
3) A/D converter input for irradiance sensor
4) Pads for a TMP102 I2C Temp sensor (would this be reliable on a long cable?, more than one in series?)
5) I have used this board before with Cayenne IoT Project Builder