|
|
|
the end of a multi-day ceiling fan installation
Saturday, March 23 2024
Gretchen had to do some actual work today at the Woodstock Bookfest, so she left this morning at around 10:30am and was gone for most of the day, though we'd already queen beed in the New York Times Spelling Bee before she departed. Still, I made myself my customary Saturday french press of coffee and sat briefly on the couch in front of the fire with a Chromebook on my lap. But then I resumed work on the ceiling fan installation project. Yesterday I'd managed to touch up all the places needing that sage-green paint I'd had color-matched yesterday. So now I could go about the work of actually installing the fan.
But before I got going on that, I had to figure out how to automate the ceiling fan's behavior based on the temperature at the top of the living room's cathedral ceiling. The way I did this in the past was that I had the power to the ceiling fan connected via a thermostat using a mercury tilt switch. When the temperature got to something like 85 degrees, the power to the fan turned on. And if I'd had the fan running when it last had such thermostat-provided power, it would automatically start spinning. This was great because it ran the ceiling fan whenever the woodstove was burning, helping to keep all the wood-heated air from uselessly clinging to the top of the living room. With the old ceiling fan (the one that crashed to the floor in November while we were in Portugal), the thermostat operated upstream of the control mechanism, meaning that none of it worked (including the light feature) when it was too cold. I'd found a fix for that, though, by routing a separate lighting circuit to the fan's light for all but one of the sockets, allowing me to control them from a conventional light switch. The one still controlled by the controller had a color-changing bulb that would give its little lightshow whenever conditions got sufficiently warm. For the new fan, I just wanted the thermostat to control the fan function. This meant it would have to intercept one or more of the three wires running from the controller to the fan motor. But what signaling system was this? I tested all three wires, and there was a 120VAC between any two of the three wires when the fan was running at full speed and 80VAC when it was running at its lowest speed. With this information, I asked ChatGPT what the signals might be. Initially it proposed that two of the wires were probably hot and neutral and that the third wire was probably a control wire. But that didn't make sense based on the voltages I was seeing. So then it proposed that maybe the wires were hot, neutral, and that the third wire would be connected somehow to a capacitor. I researched that option and found that if that were the case, the resistance between one pair of the three wires would be twice that of resistances I would measure across the two other pairs. I also noted that the three wires were cryptically labeled "U," "V," and "W." When I included this information, suddenly ChatGPT decided that perhaps the fan controller was producing a three-phase signal, which is often labeled with those wires. Obstinately, though, ChatGPT continued to insiste on labeling the wires "hot," "neutral," and "control," even though those words have no relevance to the three wires used to power three-phase devices. So the fan motor was probably a three-phase motor. I wondered if perhaps I could find a different controller for it that would give me more options for hacking (if only because then I could have two remotes for it), but based on my internet searches, three-phase ceiling fan motors are not common devices.
By this point I'd already found an issue that was going to make the new fan impossible to control with a simple thermostat: it couldn't remember its last state through a power outage. This meant that once the thermostat turned the power off, I still had to push a button on the remote to get the fan working again, because the next time it had power it would be unaware that it had lost power while on. If I wanted to have the temperature control this fan, it was going to have to involve a hack to the remote, one that detected the temperature at the top of the cathedral ceiling and then decided, based on that, which buttons to bridge on the remote (sort of like the hack I use to have a microcontroller reset the Moxee cellular hotspot at the cabin).
Of course, had I known this particular fan wasn't able to remember what state it had been in when its power fails, I never would've bought it. But how is it possible to know such things when buying a device? Such a seemingly-trivial detail is never in the manual and isn't known to the sellers. But it makes a huge difference in terms of how I can use it. It's just another thing to have to keep in mind.
Instead of letting this stuff bother me, I soldiered on with the ceiling fan installation. Using four small carriage bolts, I attached the hanging socket to the underside of the little collar tie and began doing all the wiring I could from the dining room table, as I didn't like working up at the top of the ladder, where I mostly had to work one-handed to ensure a safe connection to the ladder. In addition to the ceiling fan and its built-in light, I was wiring up a separate light perched on the topside of the collar tie that I could
control with the dimmer switch that had once controlled the extra light sockets I'd hacked into the old ceiling fan.
Gretchen returned from her workday in the midst of this, when the dining room table was a mess of different components.
Gretchen had been to the Garden Café and they'd had that special I like, eggplant rollatini, and gotten me that. So I had a delicious meal waiting for me when I needed a break.
By this point I was ready to start attacking the metal electrical box that the old ceiling fan had once hung from. Somehow the ball of the fan's root had slipped free of the socket it had been resting in, and that socket had to be removed, since it was jutting out of the box well proud of the ceiling drywall. I could now see how the installers of the original ceiling fan (whose socket I'd used for the one that had fallen) had managed to get the socket to hang a lamp 135 degrees (on the uphill side) from the angle of the ceiling (even though such large angles are usually impossible to achieve). What they'd done is hack the socket attachment so that the top part was even with the dywall layer while the bottom part was attached to a screw at the bottom (or, actually, since it was facing downward, top) of the electrical box. This allowed the 135 degree angle to be something closer to 115 degrees. But getting that socket out was not easy. The screw holding it against the inside of the electrical box was stripped and couldn't be turned, and I couldn't get any tools in there to bend it back and forth. So I was forced to break the socket apart bit by bit as if I was performing a dilation and curettage abortion. Fortunately, the socket was made of soft aluminum, and once I attacked it with an angle grinder and a drill, it softened up enough to be broken apart with a pair of channellock pliers.
I'd decided to work in this box while it was still live, which was probably a bad idea. But it was now dark outside and I needed the other lights that were on its circuit to see what I was doing. This wouldn't've been a problem had I not made a couple mistakes. The first was accidentally letting the bare ground wire poke up into the skirt of a wirenut atop the red wire that had live power from the extra-light dimmer switch. This short was downstream of the dimmer and fried that instead of tripping the breaker, which was a boneheaded thing to allow to happen. The other rookie move came when I stupidly grabbed the bare end of two white neutral wires downstream of the already-connected-to-hot ceiling fan. This gave me a shock in my fingertips, though it was tempered by the power-limiting load of the fan. The shock wasn't a bad one, but it made me instinctively fling away the orange wirenut I'd been holding. (I should mention that experiencing such shocks is a very rare thing for me and has happened not more than four or five times in my entire life.)
Once I had the fan blades on, the fan finally looked as attractive as it had been depicted on eBay. It had just three plastic blades that, form a distance, looked like they'd been carved out of a single block of wood. I spent the next half hour or so cleaning up all the mess I'd made and finally getting the ladder out of the living room. Somehow the installation of one fucking ceiling fan had taken days of my life. But this had been a tricky case, what with the need for a collar tie and all the research I'd had to do to see if I could control it with a thermostat.
The new living room ceiling fan.
Click to enlarge.
For linking purposes this article's URL is:
http://asecular.com/blog.php?240323 feedback
previous | next |
