What are the most reliable hardwired smoke alarms?

[Del91574]

You need to be clear on if you are discussing smoke alarms or smoke detectors.   The two are very distinct and different items and vastly different QC and requirements. To be clear, I install, service, maintain and test extremely large systems and sites for fire alarm.
 
The 10 year milestone is ONLY for detectors that can't be tested for sensitivity in the field. The reason why the 10 year mark is used is because if you can't read the sensitivity of the chamber or the drift compensation your device may be out of the manufacturer's spec and may or may not alarm. If the detector is within spec and can have the values pulled then the detector may remain in service.
 
Depending on the detector and technology involved, the test switch can simulate an actual chamber alarm, not just activate the circuit.....it depends on the panel and intelligence in the field. Many older units had obscuration cards that were installed for testing purposes before test switches came in vogue. The same holds true for CO detectors.
 
I would have to see the documentation and what is specified in the latest adopted NFPA 72, but technically Elk's 2 way smokes do not supercede the building code requirements for hardwired fire alarm and design elements. I'd have to see what the actual verbiage is, however if the detectors do not annunciate a trouble condition on a fail to communicate to the host panel (which is what dictates the logic here) that is going to be the nail in the coffin. There is no fail-safe to allow the detectors to tandem ring in any of the documentation that I have seen come to market so far.
 
FWIW, one of our sites has upwards of 20K installed detectors and failure is less than a tenth of a percent in a commercial building occupied 24/7 with all different enviromental criteria. Most of the times it's a detector falling out of spec that necessitates it's replacement (easier than cleaning in mission critical applications vs. price of detector).

 

[NeverDie]

Thanks, DELInstallations.  I'm really glad you weighed in on this, because your insight sounds well informed.  So, if I were to use a supervised smoke detector based system, where sensitivity and drift compensation can be read from the panel and compared against spec, then it sounds like I can bank on all the detectors actually working, as compared to the crap shoot of smoke alarms, where evidently around 40% might not be working properly but there's no way to know because the sensitivity and/or drift information can't be read.  Am I reading you right?
 
If it's of any help, here's free online access to NPFA codes and guidelines:
http://www.nfpa.org/freeaccess

 

[NeverDie]

Since the Elk-6050 isn't hardwired (it runs on batteries), then maybe it will be denied on that basis alone.  Here's a quote from a 2014 NPFA paper (http://www.nfpa.org/~/media/Files/Research/NFPA%20reports/Fire%20Protection%20Systems/ossmokealarms.pdf) which says:
 
"Most homes do not yet have the protection required in recent editions of NFPA 72®.
NFPA 72®, National Fire Alarm and Signaling Code® and NFPA 101®, Life Safety Code®,
require new and existing one- and two-family dwellings to have smoke alarms in every bedroom,
outside each sleeping area, and on every level. They should also be interconnected so that when
one sounds, they all sound. New homes should have hardwired smoke alarms."  [emphasis mine]
 
Unless I'm confusing things (which I may well be), this quoted position is what makes it all tricky.  It would seem to imply that even supervised detectors would need to act like interconnected smoke alarms (= detector plus sounder).  Or, would having supervised detectors from a suitable alarm panel supercede the need for all that?  That's one of the things I'm rather confused about.

 

[NeverDie]

The other issue that's unclear is what counts as a wireless interconnect?  One of the two Nest Protect versions is hardwired into power with battery backup, but it uses 802.15.4 to provide the wireless interconnect.  The Nest Protects can be queried using WiFi, but their wireless interconnect is not based on WiFi.  Nest claims to be NFPA 72 compliant for new construction, but I don't know what proof they have of that.  The nice thing about the Nest Protect is that it does seem to act like a smoke alarm (i.e. detector and sounder, plus interconnecting, plus hardwired with battery backup) to meet the NFPA 72 requirements, yet provide access to supervisory information like a regular detector would but that seemingly no typical smoke alarms offer. As far as meeting the code, yet providing the critical supervisory info that's missing in typical smoke alarms, maybe it is a very good hybrid solution?  The Nest Protect is priced comparable to the Elk-6050.
 
In contrast, I'm not sure what the Elk-6050 uses for interconnecting.  I get the impression it is a hub and spoke arrangement, but maybe there's more to it than that.  If it is purely hub and spoke, would that qualify, or would the lack of meshing disqualify it?
 
In any case, from the standpoint of verifying functionality by access to supervisory information, while seemingly meeting NFPA 72 requirements for new construction, the Nest Protect looks like a strong contender.

 

[RAL]

"... They should also be interconnected so that when one sounds, they all sound. New homes should have hardwired smoke alarms."

 
There are a couple of ways to interpret these statements.  It's not clear from the document you linked to whether these statements are quotes from the NFPA or just the author's own words. 
 
For one thing, "should be" does not always mean the same thing as "must be."  So you could read the first statement as meaning it is desirable to have one alarm trigger all the other alarms, but not required.  And the means of interconnection isn't stated.  It could be wireless or wired.
 
The second statement could be interpreted as meaning that the alarms be hardwired to a power source, rather than simply battery powered.  In the context of other parts of the document, I suspect that is what it means.
 
I don't claim to be an expert on the codes that govern smoke alarms, and of course they are what really counts, as DEL points out.

 

[newalarm]

we have double the smokes in our place
 
Did not realize that Fire Marshall/permitting would require one smoke in each room. So those are 120v. Then we wanted monitored smokes in case we were not home. so installed 1-2 per floor additional. So we are completely covered and have redundancy. Elks are in corridors. Added a bunch of heats as well (kitchen, furnace rm, garage, and will add to attic.)
 
If I were to do it again, I would have tried to do it all through the Elk. Live and learn.

 

[lleo]

Interesting discussion. I have come to have similar questions on the functioning of the ELK detector and raised these with ELK.
Here are the answers they provided.
 
- will the smoke detector work (detect smoke/heat and sound the built-in sounder) without the panel present? i.e. if there is no power for extended period, and panel shuts down due to exhausted battery.
[SIZE=11pt]The ELK-6050 will sound if there is smoke/heat present, but with no power to the panel, the alarm will not be recognized or reported.  The  Sound All™ feature also cannot be activated.[/SIZE]
 
- in such 'unsupervised' operation will the sound-all feature work or that is dependent on the panel?
[SIZE=11pt]The Sound All™  feature requires the M1 and wireless transceiver to be functional[/SIZE]

 

[NeverDie]

Newalarm,
 
I'm glad you brought this up, because my situation is similar.  Ideally, I'd like to leverage the code required smoke alarms with the monitored panel, rather than have redundant detectors for each smoke alarm.  It seems the easiest path to doing that might be to have one monitored detector (plus integrated sounder, making it double as a smoke alarm) that is wirelessly interconnected with the code required smoke alarms. However, to get the full benefit of that approach, you'd want your code required smoke alarms that can be supervised.  I don't think there is a way to do all that as yet (?), but it seems newer detectors/alarms are inching toward that possibility.  My code required alarms are the cheap type, so I would just toss them out and replace with new units when the time comes, thereby eliminating redundant footprints on the ceiling.  I'll be replacing them regardless, so I'd rather do it with something that can fit this paradigm if at all possible and thereby avoid redundant footprints on the ceiling.

 

[newalarm]

Well, actually in practice, my set up is not bad. The other day, we burned something in kitchen and 120v smoke went off (non monitored). It avoided call to monitor company that if not intercepted in time can actually dispatch fire dept unnecessarily.
 
The downside, if there is a fire in a bed room for example, it would take additional time to hit monitored smokes. That delay could mean much much more damage or complete loss of home. Additionally, two systems is confusing. People naturally freak out when alarms are blaring. it is confusing and disorienting. So you need to make sure everyone living with you is comfortable with system and understand the duel aspect. This is where placement of equipment is key to avoid false alarms.

 

[NeverDie]

Interesting discussion. I have come to have similar questions on the functioning of the ELK detector and raised these with ELK.
Here are the answers they provided.
 
- will the smoke detector work (detect smoke/heat and sound the built-in sounder) without the panel present? i.e. if there is no power for extended period, and panel shuts down due to exhausted battery.
[SIZE=11pt]The ELK-6050 will sound if there is smoke/heat present, but with no power to the panel, the alarm will not be recognized or reported.  The  Sound All™ feature also cannot be activated.[/SIZE]
 
- in such 'unsupervised' operation will the sound-all feature work or that is dependent on the panel?
[SIZE=11pt]The Sound All™  feature requires the M1 and wireless transceiver to be functional[/SIZE]

 
Thanks IIeo.  That seems to answer the question about the Elk topology: it's hub and spoke.  DEL can probably confirm, but my cursory reading leads me to strongly suspect that, regrettably, that way of doing it won't meet the NFPA 72 requirements for a wireless interconnect.  That said, I don't think it would preclude Elk from developing/selling a wireless smoke detector (with integrated sounder) that would mesh and meet code.  If the Elk could interface with the Nest Protect using 802.15.4, maybe that would do it?

 

[lleo]

Given that the wireless interconnect is not working if the panel is removed, based on this bulletin it looks that at least in NJ the ELK-6050 will not be acceptable.
 
http://www.state.nj.us/dca/divisions/codes/publications/pdf_bulletins/b_08-1.pdf

 

[NeverDie]

The downside, if there is a fire in a bed room for example, it would take additional time to hit monitored smokes. That delay could mean much much more damage or complete loss of home. Additionally, two systems is confusing.

 
Exactly.  That's why leveraging the code required smoke alarms, if it were possible, would be such a big win.  I'm certain that at a technology level it would be possible, but at the code level would it be allowed?

 

[NeverDie]

Given that the wireless interconnect is not working if the panel is removed, based on this bulletin it looks that at least in NJ the ELK-6050 will not be acceptable.
 
http://www.state.nj.us/dca/divisions/codes/publications/pdf_bulletins/b_08-1.pdf

 
The part of that letter which caught my eye was the reference to the IRC code, where they say "Section R313.1 prohibits the use of low-voltage smoke-detection systems in lieu of hard-wired smoke alarms.  Hard-wired, interconnected smoke alarms are required to be installed in all cases."
 
The result of that, of course, is the crazy situation that you are in, where to have a monitored system with good coverage you may end up with double redundancy in your smoke alarms/detectors.  Howeever, you don't even get the potential benefits of redundancy.  Instead, you get two different systems that are completely oblivious to one another.
 
Given that the IRC has wide influence, I would imagine this catch-22 scenario has ensnared a very large number of homeowners, not just the ones in NJ.
 
I wonder what NJ's position would be on the use of hardwired Nest Protect smoke alarms and whether they would allow a panel to monitor them?  Technically, it would not be a low voltage system, so maybe it would be allowed?

 

[tmbrown97]

This also makes me wonder how one would weigh in with a home that has fire sprinklers in the sense of having smoke alarms and rate-of-rise heat detectors and all that - or if one should just tap the flow sensor in the sprinklers?  Once that goes off, you pretty well know there's a problem... but would earlier detection serve much purpose?  Sure, the smoke alarms would hopefully save lives by alerting you the fastest and getting you out of the house - but for purposes of alerting the authorities - how much faster would a monitored detector or sensor go off before a sprinkler is triggered?  Any reason NOT to monitor the sprinkler's flow sensor?
 
When my community was built, any house over a certain size was required to have sprinklers - only affects maybe 1/4 or less of us - but by default they put in a siren/flasher on the outside of the house and I guess rely on neighbors to call it in.  Now all new homes built in CA are required to have sprinklers so I imagine this will come up more.

 

[NeverDie]

  Sure, the smoke alarms would hopefully save lives by alerting you the fastest and getting you out of the house - but for purposes of alerting the authorities - how much faster would a monitored detector or sensor go off before a sprinkler is triggered?  

 
 
 I've read that smoke kills more people than heat or flames, and lethal smoke can develop from smoldering long before a flash point is reached.  I would guess this is the main reason for emphasizing smoke detectors over heat detectors.  That said, if the purpose of a monitored system is to save property, not people, I can see your point, although isn't smoke damage a kind of property damage, and thus worth detecting and monitoring?
 
Likewise, I would guess NJ is assuming that line voltage, interconnected smoke alarms will more reliably provide earlier warning to building occupants than a similarly configured monitored system based on smoke detectors.  Is that assumption true?  If so, why?  Supervised detectors would seem to have a significant advantage in that, as DEL points out, their sensitivity and drift can be monitored, and needed replacements flagged, whereas it seems 40+% of your smoke alarms might be barely functioning (equals reduced early warning, equals reduced egress time, equals more death) and you'd never know it.  If that's so, then what is it about a comparable system of monitored detectors and sounders that makes it inferior to the line voltage (with battery backups) interconnected smoke alarms?  Is it the single point of failure, like the panel or the wiring?  What if the monitored detectors with sounders also had their own individual battery backups and they were interconnected?  Wouldn't that be at least as good, and probably better, than the line voltage interconnected smoke alarms that NJ appears to be mandating?