Friday, April 29, 2011
My issues with smoke detectors is the signal-to-noise ratio; i.e. the number of times it goes off or starts making noise when there's no actual fire versus the number of times it goes off or starts making noise when there is a fire.
I bring this up, of course, because I recently had a smoke detector start beeping in my bedroom at 5:00 AM. Due to its design, it started emitting an 85 dB beep every minute due to a low battery alert. There's no way to acknowledge the alert, either. I had to find something to stand on, remove the alarm from the wall, unplug it from the A/C power and communications interconnect, pry open the battery case, remove the dying battery, then discharge the alarm capacitors to get it to stop making noise.
And herein lies my largest objection to modern smoke detectors. First off, it was on A/C power, so the battery should barely be used unless there's a power outage. Secondly, the status light was still green on detector, meaning there was no visible indication that anything was wrong. Finally, and - in terms of good design - it takes me a while to get past that first and second points, it should have a way of acknowledging that the battery's low and that it doesn't need to be changed immediately; it can wait 24 hours. Really, that's all I'm asking. I mean, let's look at some numbers:
Number of times, in the past year, my smoke detectors have gone off because of cooking: 15Now, of course, that last number is a little misleading. Ideally we want it to be zero, because I've had 0 fires in my house. Still, we want the other numbers to be as low as possible and they're not. The noise level is pretty high; there's got to be a better way.
Number of times, in the past year, I've had to get up in the middle of the night and change a smoke detector battery: 2
Number of times the smoke detectors have actually detected unwanted smoke: 0
I don't have an answer to this problem yet, but I assure you, I'm working on it.