Mastodawn

Okay so let's write this down to think it though.

The latest smoke detector to howl pointlessly into the night sends out 11.3V DC onto the signal pin when triggered but running off batteries. I suspect that's 12V nominal, and it probably delivers 12V when operating on AC power.

Most importantly, it's not like... 500mv. Or AC. Or complicated. It's simple DC.

When acting as a non-reporting _satellite_ node, it triggers when _receiving_ 4V DC on the signal pin (4.0 exactly), and that voltage is polarity sensitive. -4V doesn't trigger the alarm.

(1/n)

#electronics #SmokeDetectors #why

Checking for DC on the signal line, I get functionally nothing. 20ma DC at most, and even that's something I'm picking up out of noise floor shift rather than direct measurement.

My _thought_ was that if the signal line was somehow floating in whole number volts (for whatever reason) than maybe somehow the right RF noise could kick it over.

The problem with _that_ is that I can now _also_ confirm that non-detecting units go off _exactly as long_ as a detecting device keeps saying it's detecting by putting voltage on the signal line. If that voltage goes away, so do the satellite alarms - and immediately.

And that's not what happens. We have to manually intervene and shut the alarms off ourselves.

(2/n)

#electronics #SmokeDetectors #why

The reason I paid meaningfully more than baseline for this particular set is that they report exactly which detector went off and why. That way, if it were the signal line somehow triggering the alarms, none of them would claim to be the originating unit; they'd all report it came from the signal backbone.

But they don't. There's always a unit claiming to be the active detector and it's always smoke (and there is *never* actually smoke), and none of them shut up until we shut off that unit, which sometimes seems to require removing it from power.

So today's afternoon check was basically just another way of confirming what we already knew, and I guess I've done that now, but...

(3/n)

#electronics #SmokeDetectors #why

All _that_ does is get us right back to where we started, which is, "we have alarm after alarm after alarm of different makes, methods (ionisation, photodetector), and models which just in this house are determined to go off randomly, usually but not always at night, for absolutely no detectable fucking reason, and then pass self-test just fine afterwards."

And no, regular cleaning - even weekly cleaning - does not help. I do _all the things_. None of it stops the problem.

(4/?)

#electronics #SmokeDetectors #why

If you're new to this adventure, I have heard this exact same story from _many other people_ at this point - though nobody I've talked to has said they've literally taken metres to the signal wires to verify that way.

Regardless, I know it is not just us.

What I've been told from others who deal with this is to RMA individual units that trigger randomly one at a time until you end up with a set that doesn't. And I guess that's what I'm gonna do, but

holy shit, team

holy shit

(5/6)

#electronics #SmokeDetectors #why

this is the opposite of fire safety

this is the opposite of how anything like this should ever work, I mean

what if all the RMAs are getting you are a set that won't go off even when they should?

but whelp

guess i'm gonna find out

'cause this sure ain't workin'.

(6/6 fin)

#electronics #SmokeDetectors #why

Well, the good news is that _unlike_ the makers of the previous smoke detectors we've had who refused to honour their warranties, BRK/FirstAlert gave absolutely no pushback on this. They don't even want the original unit mailed back, they told me I could just toss it out.

(Since it's optical, I can recycle it cleanly.)

#SmokeDetectors

@moira It sounds like your alarms are actually alarming, as if they sensed fire, at night.

There's this info about marginal battery power and colder temps leading to low battery chirping at night.

https://www.kidde.com/support/smoke-alarms/nighttime-alarm-chirps

Could something like that be involved?

I also have had problems with false alarms from a variety of units, models, technologies, and brands. I'd love to have a better solution than the one you outlined.

1/

Why Do Smoke Alarms Always Chirp in the Middle of the Night?

@moira I have stopped buying networked devices that need every device individually silenced and devices that take wired rather than battery power. These are both aimed at making it simpler to find the source of the false alarm and silence it.

I know that having alarms that don't trigger every alarm when one senses fire is slightly more dangerous, but the usability trade-off is terrible.

2/

@david42 Yes, as per thread, one specific unit always reports having detected actual smoke. Not chirping, full alarm, due to smoke detection. Also, I check batteries, and it is _not_ a battery issue regardless.

(Curiously, they never report having detected carbon monoxide.)

Also the reporting unit is not difficult to find as it announces itself, and when between announcements, flashes a reasonably visible red LED. You do have to get close enough to see it, but the offender is reasonably clear.

As for here, networked is a code requirement and we also get an insurance discount for networked alarms.

Rue Mohr 5d ago

@moira ok, Here is my other idea (but it doesn't explain everything either) maybe there is a significant DC voltage difference on the neutral (or reference) line in this building. If these detectors are in stand-alone (network line not connected) does this still happen?)

@RueNahcMohr ...like a -3 or -3.5 DC ground to neutral or something? that would be fucken weird but... it's something I can check I guess... but why would it matter? I mean, reference is reference and ground isn't connected to these alarms, the only reference they have is neutral.

Rue Mohr 5d ago

@moira yea, like bad splice and a load that somehow creates a DC bias (??)

nightmare on an intermittent problem.
what brand you have in right now?

Rue Mohr 5d ago

@moira or, ya know... unregistered random device on the network? between hot and signal? :]

@RueNahcMohr No. I am routinely in the attic, because I have antennas up there, and I know what animal infestations look and sound like in part because we _have_ occasionally had rats in the crawlspace. But this circuit doesn't go there.

Rue Mohr 5d ago

@moira
hmmmm.....

Rue Mohr 5d ago

@moira .......

does your area have shallow bedrock and your house has a deep basement?

@RueNahcMohr Our house is on glacial till. Insofar as we _have_ bedrock, which we do not, it's far, far below. xD We are on relatively _stable_ fill, however - by our standards it's quite firm.

The house is _very_ underground, _sort of_. It's built into the side of a hill. Three and a half storeys (five levels if you count the halves), and in front the bottom level is even with ground, and in back without the giant retaining wall only the _top_ floor would be above ground. _With_ the giant retaining wall, the second-from-top level is 2/3rds above ground in back.

@RueNahcMohr So deep basement: ...sorta?

What else can I talk about? The reason I wanted to confirm how the inter-unit signalling worked is because we have _freakish_ RF around here. Not just us, but us in particular, because that giant retaining wall is kind of a reflecting dish complete with angled in sections at both ends.

Older neighbours tell me that back when garage door signals were simple, garages would just spontaneously open and close. And those are people who don't even have retaining walls.

When I had my little recording studio here too, I ended up having to put RF chokes and filters on _everything plugged in the entire building_ to stop getting KUOW and BBC - both! - audibly on ribbon microphone amplifiers.

@RueNahcMohr BBC World Service - that, I understand. That's easy. But KUOW is _FM_. How did _that_ work without a demodulator? I have _no idea_.

Rue Mohr 5d ago

@moira OOOH, well, ok!

{Rue puts the tinfoil hat on the smoke detector}

@RueNahcMohr Well, I managed to get a measurement for DC offset neutral/ground, and ... nope. Not there. Not on a nearby general-purpose power circuit, and not on the removed detector mount point either, so that's two different circuits. Same 20mV noise and that's all I see.

Which is good to check anyway, I suppose. But, well.

Rue Mohr 5d ago

@moira yep, I'm going to go with the heavy RF as the cause then, feel like foil wrapping your detectors?

@RueNahcMohr I felt silly enough putting RF chokes on the hot supply. YEAH I TRIED THAT xD

Rue Mohr 5d ago

@moira well, I will only briefly put on the tinfoil hat and say "radon", but without immediate bedrock, I dont think so.

@RueNahcMohr oooooh that didn't even occur to me! But yeah, we mostly don't have radon issues around here, because mostly, we can't.

The tradeoff is earthquakes and volcanoes! Cascadia is a very dangerous place to live, over time. xD

@moira I'd check your wiring and see if you can find some weird behaviour or loose wiring or something. I couldn't imagine the why or how but given symptoms perhaps it's the environment and not the alarms.

Also, I'd absolutely give in for a while and switch to battery operated just to have some peace at night.

@cereal_cable I don't think so, because not included in today's writeup is that I've inspected and verified literally every available connection and also replaced literally every harness. While I'm not an electrician, I was an electrician's assistant for a summer in grad school and did read code, so I have a reasonably good idea what it should be like.

(Also, I've done my own electrics - permitted and inspected, of course - recently. The inspectors say my work is still quite good. Don't ask me to touch high voltage tho', I've forgot _all_ that shit xD)

@moira I declare that this must certainly mean you have ghosts.

@cereal_cable where's egon when you really need him

everything_bagel

@moira the opposite of fire safety, aka: fire fun!

Man aging with madness 5d ago

@moira all I can say is yike!

@moira I scrapped our cheap smoke detectors that had a lot of false alarms and got expensive name brand smoke detectors instead (like 40 bucks a piece). No more false alarms - only real detections (like bread stuck in the toaster).

Its really not good to get a call from a neighbor stating "your house beeps, but it sounds like the temperature warning of a fridge or something" just for me to confirm "ah thats just one of the smoke detectors. If this is going on for a few hours now and the house is still fine, just ignore it".

I don't know what would be a worse outcome: There is a fire but it is ignored as the smoke detectors sound like a forgotten alarm clock or its a false alarm and the front door gets kicked in to check what's going on. Both cost more as a few of the expensive smoke detectors.

@stereo4x4 Yes, that's what we did too, as per the thread. It didn't change anything _other than_ confirming that it is in fact individual detectors going off.

(These have voice messages, they have origination identification, they have CO maximum count reports, etc. Quite nice. Or would be, if they didn't go off at random, like every _other_ make and model we've tried.)

NilaJones 5d ago

Your thread makes me realize that this stopped being a problem for me when I started running air purifiers all day every day

I didn't get new smoke alarms, and I didn't start cleaning the house better. So I think this is the only thing that changed

I have a cat, and I suspect it's a cat hair in the air that gets them, but I don't know

@NilaJones I wish that could solve it, but we have both whole-house air filtration and a number of additional HEPA filters. We have _extremely_ clean air. So I do not think it's filtration.

@NilaJones see also:

#SmokeDetectors

NilaJones 5d ago

I mean seriously, wtf?!

@NilaJones It really just doesn't seem to have anything to do with what's in the air. Not as far as I can tell, anyway.

Jigme Datse 5d ago

@moira Can confirm this is a known issue in this house. Oddly several continued to signal until both removed from line, and battery pulled. Always, always wear hearing protection when working with these devices. Fortunately, this for us was on install, not random afterwords. All since signals have been due to proper detection, with the one device complaining of low battery. Sounds like we may have been relatively lucky...

@moira Why not ditch the fancy smart system and get a bunch of battery powered smoke detectors?

Then whatever is haunted about your wiring won't be an issue?

They make 'em now with ten-year batteries. Which is the replacement cycle for smoke detectors anyway.

@apLundell I don't know that anything is haunted about our wiring. I do know that the whole _building_ is haunted by RF, and that's not just wiring. It's ambient.

Isn't that all the more reason NOT to be relying on long runs of low-voltage wiring?

@apLundell Only if the RF is doing something actually directly to the circuit board through the low-voltage line which is making an individual unit think it is optically detecting smoke.

Which, I mean, I can't rule out, but... these are going to be built to expect that, right? And mains power is going to bring in more... though I guess mains do go through a rectifier and that's literally designed to convert AC to pure DC, so.

(Also potentially relevant - and something I've looked at a lot - is that it started with the large-scale switch from leaded to unleaded solder. At first, I absolutely thought problems with that process were the cause. But those problems got solved and units from 2024 are still doing it.)

@moira If it was the solder, it'd be a large-scale systemic issue and not just your building.

Most people are certainly not replacing smoke detectors on the regular.

(Except in the sense that they expire every 10 years and we replace them then. In theory.)

@apLundell Again, as per the thread, I have talked with _many other people seeing the same issues_. There are replies _on this thread_ from other people who have seen the same thing.

It is

NOT

just this building.

The advice I've received from building managers (plural) who have run into this has been, "RMA units that go off until you have a stable set." And that's what I'm starting to do.

(1/2)

@apLundell I'm also _not_ saying that there's nothing about this building making it _more likely_ to trigger this issue. I think there is.

My hypothesis is that there is something borderline about smoke detector design (possibly made worse post-no-lead-solder) that can but usually does not get triggered by local environments. And is here.

To wit, I'm also ordering a new set of most-suitable RF chokes and I'm going to re-choke everything. I tried this before, but with chokes I had on hand. This new set is the closest I can get to exact without spending hundreds of dollars.

And yes, I do in fact have specific and good reasons for not going wireless. It's the _same_ reason, and also why our wifi topography is bizarre.

Anyway. Back to work.

Rue Mohr 5d ago

@moira did you open one of the tripped ones up? *most* detectors can be set off with particular bugs. Here, we have "spider breeding season" those little guys get in detectors all over the place and set them off. it sort of sounds like you have a different type of bug :] (spiders are easy cause their just everywhere in there when it goes off)

Rue Mohr 5d ago

@moira "from the neutral line to the signal line"

right?

@RueNahcMohr Yeah? I mean, you're not gonna use AC mains hot as a reference xD

Rue Mohr 5d ago

@moira actaully, there were old detectors that did!!!

@RueNahcMohr for a third-wire fire signal? holy shit okay

i guess thinking about it... if you're talking retrofit on two-wire electrical systems i guess you better be ready to work either way... but wait no you still need the signal wire so you're rewiring anyway so you should fix that shit. unless you want to piggyback the fire signal directly onto hot. is that what they were doing?

Rue Mohr 5d ago

@moira 3 wire, old as hell model detector, but when you connected a neutral-ref detector (first alert) to the network it would trigger them all. "sorry, I know you only had one detector technically fail (of your ion detectors made by Davinci himself, and should have been changed millennia ago) but we have to change ALL your detectors."
I'd like to hear the 'stand alone' results! I bet between the detectors the ref line has a bad splice, 4V diff is a lot!

@RueNahcMohr You'd be waiting a while for a "stand alone" result because this is like five months between problems.

Also, these are detectors which identify who set the network off, and for what reason. And in all cases, the guilty party claims responsibility, and the other units agree.

Current units are FirstAlert, after trying two kinds of Kidde first.

Also something I didn't say is that I noticed the when-we-bought-the-house Kiddes were 20 years old the first time it happened, after 10 years in the house. I replaced them all for age. Since then, it's been 2-4 times a year.

I have also checked _all_ the accessible connections and updated _all_ the harnesses. No change.

Rue Mohr 5d ago

@moira oh, sorry, I got brands mixed up, it was kidde I was referring to in the neutral ref units.

hmm

@RueNahcMohr Well, foo.

I'll still check it just to see, but after dinner.

Michael Dwyer 4d ago

@moira
I doubt this is you, but the networked ones at my moms house set themselves off because they were setting themselves on fire. !

They were AC powered and had a big multi-watt resistor on the board. This heated up enough to blacken itself and set themselves off one after another over the period of a few years.

They were of an age to be replaced anyway, but it certainly rattled my opinion of smoke detectors.

@mdwyer I've opened several, once Kidde declined to honour their warranties. I've found some bad solder joints, some improperly-applied wax as conformal coating (which also didn't cover the entire intended area), and so on - but ever anything like that. So.