basically black magic
Tuesday, July 14 2015
Early this afternoon I went back to that skeletal oak about a quarter mile from the house on the Gullies Trail, though this time I went a little down the slope to another, similarly-sized fallen skeletal oak and bucked a couple pieces out of it. Unfortunately, they'd been in contact with the ground and were damper than I would have preferred, so I ended up only taking one of them as well as smaller pieces already cut up from the other tree. As I was strapping these onto my backpack, I noticed a tiny Poison Ivy seedling growing right there in the trail. I didn't have any way to extract it, but at some point I will. That would be a bad place for another foothold for this still-rare plant on Hurley Mountain. The load felt light on the walk back home, and sure enough, it only weighed 94.5 pounds. Most would consider that a heavy load for a 170 pound man, but my body is accustomed to heavy loads. Indeed, it seems my body didn't forget or unlearn or lose its familiarity with such loads over the near-year of not gathering firewood that happened between July of 2014 and June of 2015, because getting back into the routine seemed easier than starting the routine back in March of 2014.
Today I installed the little AmbientWeather-data-collecting assembly into the chassis of my barometric windvane. It was so tiny that all I needed to secure it was a stiff piece of wire pointing upward that I could impale it upon using one of the pre-drilled holes in the breadboard I'd built it on. I attached it to one of the many four-pin I2C headers on the barometric windvane's motherboard and then tested it to see how well it received data from AmbientWeather sensors. Oddly, sometimes it seemed to pick up the data from disparate sensors (even the one down in the greenhouse basement, 90 feet and several thick walls away) without difficulty, and other times it wouldn't pick up data even from sensors right there in the laboratory. I kept tinkering with it, trying different orientations of the little coiled antenna on the superhet RF board. Coping with marginal radio reception is basically black magic and you can drive yourself crazy tinkering with it; I remember spending hours trying to improvise antennas out of extension cords in the almost-fruitful effort to pick up WHSV Channel 3 (the Shenandoah Valley's ABC affiliate) on my teevee when I was a teenager. In the recent past when experimenting with this AmbientWeather sensor reception, it wasn't as obvious how well individual sensors were being received. But now that I'm seeing an indication of the staleness of the data next to every reading, I can quickly see which sensors aren't being heard, which then causes me to tweak the antenna obsessively.
Eventually I gave up on the antenna and tried upping the voltage to the AmbientWeather reception board from 3.3 volts to 5 volts. The board could run at either voltage, but I'd had it connected to an I2C header designed for a barometric sensor, and those need to run at 3.3 volts. To get five volts, I had to build a new five volt I2C header on the barometric windvane's increasingly-crowded motherboard. Getting all the wires to it was tricky given all the existing wires already soldered in groups to individual Atmega328 pins, but I managed to do it. Interestingly, though the increase in voltage didn't seem to help with sensor reception, a reorientation of the board on the support wire did. It seems that moving it away from the Atmega328's 16 MHz crystal helped a lot, making the AmbientWeather receiver receive data from all existing sensors with their every broadcast. I hadn't really thought much about RF interference within the barometric windvane itself, but obviously that could be a problem. There's absolutely no RF shielding of any sort in there.
The tiny AmbientWeather daughter assemblage (with its coiled antenna pointing upward) can be seen atop the little steel wire that supports it.
This evening there a culinary first in our household: a meal comprised entirely of stuff harvested from the garden (with the exception of things like spices and oil). Gretchen fried up a bunch of Ragged Jack Kale with green Northeaster Pole Beans cut into segments. There was also a salad made with a bunch of Spotted Trout Lettuce, which, in our garden, has begun to bolt. As simple and carb-poor as it was (I love carbs), it was delicious.
I had to stay up late tonight for a meeting with web development colleagues in Los Angeles. The meeting began at midnight and lasted until 1:00am, which is kind of late even for me. Such is the fate of someone working remotely in an eastern timezone with others in a western timezone who are themselves working on the project after-hours (because they have day jobs).
For linking purposes this article's URL is:feedback
previous | next