|
|
|
poring over inverter data
Monday, June 24 2024
location: rural Hurley Township, Ulster County, NY
It was cooler today, so when I took the dogs for a walk atop the escarpment west of the Farm Road, Neville definitely wanted to come. I stopped at the permanent part of the bog on the homeward walk northward on the Farm Road so Neville and Charlotte could get in the water and cool down, though all Neville did was dunk his face.
A little before 5:00pm, our neighbor A and her boyfriend J were to pick me up and take me to Woodstock so we could have dinner at the Garden Café. As I waited for that to happen, I weeded the garden (most of the weeds being sedges and grasses). I don't really know A and J very well, so I felt a little awkward on the ride to Woodstock. I was also somewhat stressed out about having left my phone and wallet in the Bolt, which Gretchen had likely left unlocked in Woodstock all day. But I made smalltalk about things like J's photography career, my trailmaking in the Adirondacks, and the beautiful Ashokan Reservoir as we drove past it. We drove to directly to the bookstore, where we met Gretchen and A bought some books. Then we walked over to the Garden, where we dined indoors (a strong wind had blown down a big tree branch into the outdoor dining area). There wasn't much on the menu that I liked, so I ordered the black bean burrito (even though it's pretty boring; with hot sauce and jalapeño peppers, one can make it fairly good). I didn't have much to contribute to dinner conversation, which dwelled for a long time on an ongoing recession in the New York City photography industry, which affects J. Also, since the run of a broadway show she was in is over A herself could be considered "unemployed." Gretchen asked if perhaps she was "taking a beat," and I chimed in that "taking a beat" was definitely what I was doing. (I've been unemployed for nearly a year.)
Later this evening, I went through the serial data I'd logged from the SolArk inverter trying to find the hexadecimal versions of values in screengrabs of the data that had been displayed on the SolArk inverter data page at the same time. The easiest datum to do this with was the percentage charged of the battery, as this changes slowly and never required more than a single byte. I eventually found where in the data this seemed to be transmitted: in a block of 372 bytes after the ASCII string "MB_real data,seg_cnt:3{carriage_return}," it was byte number 303. I then found strong indications that a two-byte representation of the inverter load value began at byte number 304 and the solar power value was at the two bytes beginning at byte number 307 (which were consistently zeros for data collected after dark). Those are three of the four values I currently care about, and it seems likely I will be able to find the fourth (the battery charge/disspation wattage) when looking closely at live data. Once I can collect that data entirely within the cabin, I can tweak my system so it doesn't even require an internet connection to do its automation. Also, the data I'll be gathering will be a lot more granular. (I was already seeing finer variation in the values just in the recorded data I was poring over.)
For linking purposes this article's URL is:
http://asecular.com/blog.php?240624 feedback
previous | next |
