FireAngel smoke alarm stupid engineering

[bd139]

So I woke up this morning with a reminder to check my smoke alarm. It was 100% dead! This is a FireAngel SI-610 apparently good for 10 years. It has lasted just under 2 so far.

Took it to bits to work out why as this is one of the new sealed ones without replaceable batteries. Checked batteries expecting them to be dead. Nope all good.

Well I found the issue and it's a fucking stupid bit of engineering. The entire thing has a reed switch which enables or disables it based on whether or not it is attached to the base. The base however the magnet had fallen out of and fell out onto the floor when I took it down. I found it in the shoe cupboard. I looked at the plastic and it's actually flexible as hell and doesn't hold the thing in properly so just a tap on the base will knock the magnet flying. So the entire smoke alarm actually operating properly is based on a flimsy bit of plastic hook that doesn't work properly 



I've replaced these units with two Ei Electronics units which are relatively old style ionisation units. Also dirt cheap as there's no fancy shit.

[MrMobodies]

So it would be tested and working but if it fell down or got knocked down and there was a carbon monoxide leak and that didn't go off it maybe too late.

I don't think the FireAngel ST-622T one I have has that mechanism as they are just plastic mounting holes in the back and the screen is always on.

[bd139]

In this case I think it was because the button gets whacked regularly on it due to burning toast etc. It probably worked its way out. Definitely not suitable for a domestic situation. Also it appears that it rotates on the frame very easily so it's possible for the magnet to actually be misaligned with respect to the reed switch. Also I notice the reed switch was just glued on and was quite loose.

So to fix the issue of having batteries that go flat, they made it more complicated by putting lithium batteries and a reed switch in it that fuck up earlier than the batteries would have done. Stinks of design by committee. 

Also it came from Amazon and they are refusing to refund / return. I've just cancelled Prime. That usually kicks them into dealing with it 

[MrMobodies]

Amazon at their prime when it comes to returns like that but I like the name Fire Angel.

When there is a fire and it doesn't work you might even meet one.

[amyk]

A smoke alarm that doesn't work unless it's perfectly mounted, even if it has batteries? That's certainly stupid. A simple on-off switch would be far better, since they clearly wanted the unit to be switchable on/off. (If they only wanted it to be turned on once, a pull-tab over a battery terminal would be appropriate.)

[bd139]

Yeah exactly that.

Pull tab would make more sense.

[floobydust]

Contact the manufacturer and let them know of the problem, and whoever the approvals agency is.
They will open a file and investigation. In North America, they issue a product recall overanything that causes it to not detect smoke.

FireAngel SI-610
• "Independently certified to BS EN 14604:2005 by the Loss Prevention Certification Board (LPCB) Certificate No. 651d/02"

I've done it before, as smoke detectors get imported from china now and have fake approvals and parts that do not last.
I had one the mains capacitive dropper started smoldering and false triggering the smoke alarm.
Most of the board was half baked due to heat. But UL and the manufacturer swept it under the rug, as technically it didn't fail to detect smoke.

[mnementh]

seems to me the least assache solution is to assemble the stupid thing to the base & put a dot on the detector with a Sharpie through the magnet hole; use that to locate and epoxy the magnet to the detector.

Then rehang it and fucking get on with your life. 

mnem
Alternately, just burn the house down and use that as an excuse to move to Canada.

[bd139]

I think the best solution turns out to be either to jump the reed switch or buy a cheaper unit without this turd in it . I did the latter

[tggzzz]

Am I the only person that tests fire alarms by holding smouldering paper near them?

Having said that, without my deafaid in I wouldn't hear one anyway

[bd139]

No you're not the only one although I suspect very few people bother. I test mine properly as well. I take some plastic wrap off some bread, stick it on a fork and set fire to it.

[forrestc]

I actually have an aerosol can of smoke tester spray.  Yes, it's a real product, made by several vendors.   One can lasts almost forever for residential use. 

It's particularly nice since it squirts the "smoke" far enough away that I can test ceiling mounted units without climbing on anything.

A bonus is that it doesn't require creating a fire risk.  It also dissipates quickly and doesn't smell up the house.

[mnementh]

Yes, but do you KNOW the detector is actually any good at detecting SMOKE...?

mnem
confidence is low...

[station240]

I was expecting FireAngel to be some slang term used by firefighters for human remains found after the fire.

But no, it's a crappy smoke alarm where poor design and crappy choice of materials made it worthless.
I would suggest a rename is in order, change to AngelOfFire, same train of thought as anything called AngelOfDeath.

I wonder if the magnet used the same crappy brown glue that eats electronics ?

[bd139]

FireBastard is about right.

I sent them an email. I expect a canned response.

[tom66]

Having worked on the design of smoke detectors FireAngel is not too bad, but they have a lot of reliability issues. They are making £10 detectors - you cannot expect too much for the money. The optical detectors are generally a lot better than the Ei ionising ones which are built to even lower price points. 

[tom66]

I actually have an aerosol can of smoke tester spray.  Yes, it's a real product, made by several vendors.   One can lasts almost forever for residential use. 

It's particularly nice since it squirts the "smoke" far enough away that I can test ceiling mounted units without climbing on anything.

A bonus is that it doesn't require creating a fire risk.  It also dissipates quickly and doesn't smell up the house.

You do need to be careful with these cans. Always spray from several foot away. The moisture can condense inside the sensor which will cause an offset in the zero reading. This may either reflect itself as a false alarm, a fault condition or just a desensitised or oversensitive alarm.  Usually it tends to slightly desensitise our alarms.  On the test devices we have built, I have to recalibrate after more than 10 sprays - it seems the moisture doesn't dissipate quickly (it takes at least a day or so before they begin reading correctly again).

The above applies only to optical detectors - I can't comment on ionising ones not having experience with them.

[bd139]

On optical vs ionisation, I have both for reference. The optical one has never gone off even when I've tried testing it and when my kitchen was actually on fire. I do not trust it.

I suspect the optical ones mostly exist so they can be farmed out to non specialist bulk manufacturing as there are no radioactive components.

[tom66]

Optical alarms are better than ionising alarms for -most- applications.  Optical alarms do not do well with smoldering fires, as the wafting smoke does not produce enough obscuration/scatter for long enough such that the alarm will actually reliably trigger.  However, on almost every other test, they are superior. Therefore the general recommendation is to use optical or hybrid optical/ionisation sensors.

It takes around 0.1dB/m of smoke to trigger an alarm. Obscuration of 3dB/m is half light attenuated transmitted per metre.  0.1dB/m doesn't sound like much but it's a lot once it reaches the sensor;  the room has to be seriously smoky. Most detector designs don't trip below 0.05dB/m because the stability of the sensor (especially when you get people smashing into them and knocking them off the ceiling) over 10 years is hard to guarantee, and then dust and other things can set them off.    Note most optical sensors work on a scatter principle instead of obscuration, as it tends to be easier to design and stabilise, but the behaviour is similar.

The relevant standards are fairly strict about false alarms and alarming too soon, as they tend to be the detectors that get covered up or have their batteries removed. It is judged to be better to be slightly less sensitive than over-sensitive, as with a real fire the devices will still trip well even if they are 20% less sensitive, but they will not trip at all with the batteries removed.

You aren't wrong about the ionising detectors being a pain, our consultant supplier for test tunnels doesn't like to deal with them because they invoke another tonne of paperwork regarding radiation and storage of components.  But you can still buy cheap ones.  In mass production they are cheaper to make.  Optical sensors require discrete components in the smoke chamber, as well as a photodiode amplifier, usually some software to compensate for drift and the battery performance is a bit worse as the sensor is active and not passive.

[Red Squirrel]

Seems to me that whole reed switch thing is over designed, if it's just to activate it after you buy it, some kind of pull tab that closes the battery contacts would be much simpler.  The more complicated something is, the more likely it is to fail, and you don't want that with a fire alarm.

[Mr. Scram]

Am I the only person that tests fire alarms by holding smouldering paper near them?

Having said that, without my deafaid in I wouldn't hear one anyway

Most manuals I've seen explicitly warn you not to do this.

[jogri]

A local public broadcaster (SWR) here in germany ran a piece on those smoke alarms (model ST-630s), it seems like there is a major problem with their lifetime: Out of 70.000 smoke alarms (total, they didn't mention how many of those were from FireAngel) that were installed in one city a few hundred from that manufacturer failed each year (battery was low). They also interviewed a chimney sweeper who had to swap over half the units he installed because they failed after two years and FireAngel stopped providing free replacements for the faulty units.

Here is a link to the video (it's in german and it doesn't have subtitles):

[tggzzz]

Am I the only person that tests fire alarms by holding smouldering paper near them?

Having said that, without my deafaid in I wouldn't hear one anyway

Most manuals I've seen explicitly warn you not to do this.

Do they say why, and how to test them?

[Mr. Scram]

Do they say why, and how to test them?

They don't say why but I suspect it has to do with contaminating the sensor or offsetting the calibration. It's generally recommended to use the test button.

[tom66]

And the test button doesn't test the sensor. Well, not on every device at least.  It depends - the Nest devices are very good and thoroughly test their hardware but many Chinese detectors just sound the siren, and that goes for devices with a low battery too.
 
Your best bet is the aerosol cans, just limit tests to once every few months, and spray a good distance away from the detector (at least 3 foot).  They should alarm within 30 seconds with a small cloud, if you spray about 5-10 seconds worth.  Excessive spraying will contaminate the sensors but as long as it is fully aerosolised this should be fine. The test tunnels that are used to calibrate the detectors essentially pass a fine aerosol through the detectors at a constant 0.8 m/s, but the concentration is very low.
   
The problem with burning paper etc is that you could contaminate the sensor with soot, generally the recommendation is after a house fire, even a minor one, you should replace your detectors because they might be sooty and less effective. So I can imagine this is where manufacturers are coming from.