Hacker News new | ask | show | jobs
by dcchuck 498 days ago
Thank you for sharing. I particularly enjoyed the call out where you highlight what you learned from the incident. Took the post from fear-brain to information-brain for me (don't mind my curiosity haha).

I had a much less intrusive incident but equally "alarming" ;). The apartment I live in has hard wired smoke/carbon monoxide alarms (with backup battery). One night, ~2am 1 of them starts beeping that the battery is dead. I confirm the breaker didn't flip or anything...finally decided to complete disassemble it and get some sleep. 4 of these in the apartment. Head hits the pillow and a second one goes off! Take all 4 down and live dangerously for an evening.

In the morning I do my research - turns out these things just start doing that after 10 years. "Get a new one!" it peeps incessantly.

Shame me please - as I bought replacements. I'll save this post for myself in 10 years so at least I can be the one to say "I told you so".

5 comments

Fixed lifetime suggests ionization rather than optical sensor? If so yeah, they have a specific lifetime, it's just physics. There's an Americium-241 source, a tiny, tiny amount of radioctive material in a sensor which is decaying constantly, and once it has decayed a certain amount too bad, buy a new smoke detector.

There are arguments for or against both types, but ultimately "Make sure you replace it every ten years" doesn't feel like a huge problem. If you really can't do that, then the optical sensors don't have this property - they will still need new batteries, but of course you can just swap those out on a schedule when it suits you, they're typically a cheap household 9volt battery (yes even for a mains smoke alarm, fires don't magically stop when the power goes out)

> Fixed lifetime suggests ionization rather than optical sensor?

Not at all. The 10 year lifetime is something that was set by people concerned by the lifetime of the electronics. The half life of the isotope in smoke detectors allow them to last well beyond 10 years. (The half life is 432.2 years)

I was aware that the half-life was much longer but I'd assumed the 10 years was based on the fact that nevertheless we're talking a degraded sensor performance over time, however it does seem like it's actually 10 years because they're worried the electronics will crap out, so you're right.
I installed some gas sensors like this a while back. It says in the manual that some date after installation they trigger and stay that way permanently.
No, but you'd think they could charge a battery/capacitor during the good times and use that charge during the outage.
10 years, you're lucky, one of mine just started doing it after 4 years, it comes with a 5 year guarantee but of course I don't have the receipt and even if I did I don't have the will to fight for a replacement. I think these new alarms are designed to fail early I've never had one fail before but after checking up on different makes and models they all seem to fail randomly.
This is by design, and it's for a good reason. The carbon monoxide sensor chemically degrades over time. Good CO detectors will perform self-tests periodically, and will go into an end-of-life mode when those tests fail, or when the sensor's expected lifetime has expired.
Yep, and that's a big reason why combined Smoke + CO detectors are a bad idea. The smoke detector side doesn't wear out, the CO detector side does.

The beep patterns for "dead battery" and "detector worn out" are usually different. Generally they're printed somewhere on the unit, in a tiny font that's hard to see bleary-eyed at 3AM when it inevitably starts.

Smoke detectors usually have something like a 10-year usable life, and should have a sticker to that effect along with its manufacture date.

I've had a couple that when they aged out, went straight to full-on alarm beeping.

True, this does depend on the detector type. Photoelectric detectors last pretty much indefinitely, but ionization smoke detectors do age. And you ideally want both, since they detect different sorts of combustion. Also there are smoke detectors you can buy with a 10-year life lithium primary cell, so no need to change batteries. These are perhaps a bit wasteful (the cell isn't replaceable and the rest of the detector could be fine) but also don't require throwing away lots of 9V batteries.
Oh, true enough. Or at least, by the time the battery gets low, the whole thing should be replaced anyway. As you say, depending on detector.
I have a plug-in CO detector that's probably 30 years old. Has never given any "end of life" alerts but maybe they didn't have them back then. Guess I should replace it.
You definitely should. 30 years is almost certainly beyond the useful life of the sensor.
FWIW some smoke alarms have a manufacture date printed on the side, so a receipt might not be necessary.
But... you have to go to a shop to get the replacement. I suppose you could mail it by yourself to the manufacturer, though I must admit I never tried that. It requires too much effort, contacting manufacturer, describing the problem, etc. Do people do this?
How would you prove it came from a certain store?
I would love the wired ones to just have a built in super capacitor; enough to give it 5-10 min (minimum) runtime during a power outage. No Battery. ONLY alarm loudly for sensor detects possible smoke (known positive detection), very softly for sensor broken (failed self test).
The battery isn’t for incidental power outages. It’s for when a fire starts somewhere in the building while you are asleep and melts some wiring, knocking out the power to the smoke alarm. An hour later you wake up in a house full of smoke and die.

That’s why smoke alarms have batteries.

That is also why smoke alarms have interconnects, and there should be smoke detectors all over. The fire should be detected and sound an alarm long before it melts wiring.

Though smoke detectors only detect 40-50% of all fires in time to get out (dual sensor detectors can do much better!). A lot better than nothing, but still not very good.

Two in the same night? I wouldn't have been so quick to dismiss it as "it just happens". Scary.
Hmmmm, we had two cheap quartz wristwatches with a drift of +/- 15 seconds per month, and they were allowed to run for 5 years, drifting in opposite directions, we might expect a difference between them of 0.5 hours.

So having two of these detectors going wonky with "just" a 4-hour difference doesn't seem too far outside the capabilities of some kind of "replace me" timer circuit.

Obviously it could also be Invisible Evil Gas Traces, and the impact of being wrong is severe, so I wouldn't blithely dismiss the alarms as planned obsolescence either.

> Obviously it could also be Invisible Evil Gas Traces, and the impact of being wrong is severe, so I wouldn't blithely dismiss the alarms as planned obsolescence either.

And that is why I only buy CO alarms which display the ppm they measure.

"I'm out of sensor lifetime / battery, please replace me" is a very different level of priority than 300ppm CO. And how I react to 300ppm is very different from how I would react to 3200ppm.

same thing happened to me at about 3 am one morning. I'm thinking the house is burning down. Eventually find out there is no fire. Get the ladder, take down the smoke alarm. Go to remove the battery. On the back it says "non replaceable battery, unit can not be serviced". So sledge hammer + trash can it is!

The next day I tell my landlord I destroyed the thing and need a new one. His reply was that I had vandalized his property.