The abnormally-high temperature reading on my pool’s SWG that I noticed 10 days ago is back. This morning, it was reading 94°F when the actual water temperature was 80°F. The reading was consistently about 14-15° too high every time I checked today, all the way up until 7:00 or so, when I went out to swim. I’m going to try disconnecting and re-seating the cable that connects the tri-sensor to the controller, and I’ll take a look at the contacts to see if they’re dirty or oxidized. Maybe it’s just a bad connection. I went ahead and ordered a tri-sensor simulator, which is a little widget that plugs into the controller and generates a fixed temperature, salt, and flow reading. If cleaning and re-seating the cable doesn’t do the trick, then I can use the simulator to determine whether the problem is with the tri-sensor or the controller.

In other news, we are preparing to get rid of our landline phone service, but before we can do that, I have to convert our security system to use cellular and/or internet monitoring. The problem is, the panel’s firmware was too old to work with the new communicator, and to make a long story short, I had to swap out the entire panel. It’s kind of amazing that the panel is still in production, and is practically identical to the version that was produced 21 years ago when I originally installed it. All I had to do was move all the wires over and then reprogram the new panel (a tedious, but not difficult, job). That was today’s project. Now, the weird thing: the new panel works perfectly, except for one thing: for some reason, it thinks it doesn’t have AC power. The alpha keypad flashes “AC LOSS”, and the fixed keypad says “NO AC”. This is normally what it does during a power outage, when it’s running off the battery. Get this, though: it does this even with the battery disconnected. So, if it doesn’t have AC, I don’t know how it can possibly be working. Induction? Hamsters? Who knows. The voltage from the transformer is fine: I read 16 volts AC at the transformer itself and the connection terminals on the panel. I’m wondering if maybe I can just ignore this, as everything seems to be working, but I’m concerned that it may not be charging the battery. If it’s not, I’ll find out in the next few days once the battery runs down. As a last resort, I may try unplugging the panel to simulate an actual “NO AC” condition, and seeing if maybe it resets itself when I plug it back in. But I’m not holding my breath, as the problem has persisted through several power cycles. I’m also going to reread the manual and see if there’s anything I missed with the initial setup. If all else fails, I’ll contact my alarm company, and see if they have any ideas.