Thread: Smoke detector --AFCI?

Should the smoke detector be wired to the AFCI curcuit? I have seen it both ways and it gets a little confusing sometimes.

Similar Threads:

• AFCI and GFCI Question
• AFCI in conversion to condo
• Smoke alarm placement
• smoke detectors
• smoke alarms wired together

All bedroom outlets (includes smoke detectors) should be AFCI protected.
Jim

That is what I thought I even pulled the battery out to see if the light went off but no luck so it was not AFCI protected. I put a call into the county just to make sure but no call back yet.

Thanks for the fast response.

Mike

I can't quote the code Tim but that one won't work except for Mobile or Modular homes... got another one?
Jim

Thanks Jim,

Reading one while copy/pasting another. duh.

Let's try this one.

From the 02 NEC:

(B) Dwelling Unit Bedrooms. All branch circuits that supply 125-volt, single-phase, 15- and 20-ampere outlets installed in dwelling unit bedrooms shall be protected by an arc-fault circuit interrupter listed to provide protection of the entire branch circuit.

The definition of arc-fault circuit interrupter given in 210.12(A) explains its function. The basic objective is to de-energize the branch circuit when an arc-fault is detected.
Arc-fault circuit interrupters are evaluated to UL 1699, Safety Standard for Arc-Fault Circuit Interrupter, using testing methods that create or simulate arcing conditions to determine the product's ability to detect and interrupt arcing faults. These devices are also tested to verify that arc detection is not unduly inhibited by the presence of loads and circuit characteristics that may mask the hazardous arcing condition. In addition, these devices are evaluated to determine resistance to unwanted tripping due to the presence of arcing that occurs in control and utilization equipment under normal operating conditions or to a loading condition that closely mimics an arcing fault, such as a solid-state electronic ballast or a dimmed load.
UL 1699 is the standard covering arc-fault devices that have a maximum rating of 20 amperes intended for use in 120-volt ac, 60 Hz circuits. These devices may also include the capability to perform other functions such as overcurrent protection, ground-fault circuit-interruption, and surge suppression. UL 1699 currently recognizes five types of arc-fault circuit interrupters: the branch/feeder AFCI, combination AFCI, cord AFCI, outlet AFCI, and portable AFCI. Placement of the device in the circuit and a review of the UL guide information must be considered when complying with 210.12. The NEC is clear that the objective is to provide protection of the entire branch circuit. (See Article 100 for the definition of branch circuit.) For instance, a cord AFCI could not be used to comply with the requirement of 210.12 to protect the entire branch circuit.
Section 210.12 requires that AFCI protection be provided on branch circuits that supply outlets (receptacle, lighting, etc.) in dwelling unit bedrooms. The requirement is limited to 15- and 20-ampere 125-volt circuits. There is no prohibition against providing AFCI protection on other circuits or in locations other than bedrooms. Because circuits are often shared between a bedroom and other areas such as closets and hallways, providing AFCI protection on the complete circuit would comply with 210.12.

Mike,

You have to be a little careful with the requirements, as directed by the NEC.

If the home was built before Jan 1, 2002 (and after 1999)... more than likely, you are going to have the AFCI's installed at the receptacles only.

The 2002 NEC changed to include "the entire branch circuit", which includes the receptacles, lights and smoke detectors.
___________________________________________
The 1999 N.E.C., 21-12 Arc-Fault Circuit-Interrupter Protection
a. Definition. An arc fault circuit interrupter is a device intended to provide protection from the effects of arc faults by recognizing characteristics unique to arcing and by functioning to de-energize the circuit when an arc fault is detected.
b. Dwelling Unit Bedrooms. All branch circuits that supply 125 volt, single phase, 15 and 20 amp receptacle outlets installed in dwelling unit bedrooms shall be protected by an A.F.C.I. This requirement shall become effective January 1, 2002.
___________________________________________

The 2002 N.E.C., Per 210.12 Arc-Fault Circuit-Interrupter Protection,“All branch circuits that supply 125-volt, single-phase, 15- and 20- ampere outlets installed in dwelling unit bedrooms shall be protected by an AFCI listed to provide protection of the entire branch circuit.” The 2002 NEC removed the word receptacle from the 1999 version, thus, making the requirement so that all “outlets” (see Definition) shall be AFCI protected in bedrooms.

Definition of “Outlet”: the NEC defines "outlet" 'a point on the wiring system at which current is taken'.Rich

As others have said Yes, smoke alarms should be on the rooms AFCI circuit. IMO, this is a stupid rule or code. Just my opinion for what it is worth, I just don't see the logic behind it. You are putting a life saving device on a circuit that could be turned off by.

Smoke alarms on an AFCI circuit should have an integrated Battery Backup.

If not, it is noted as Deficient.

Ok thanks this is a new home and does have battery backup So I have recommended that they be AFCI protected but will also talk to the county to get it sealed for this area.

Thanks guys it is a pleasure being on this board and talking to you guys.

Mike

Originally Posted by Mike Huppi

Ok thanks this is a new home and does have battery backup So I have recommended that they be AFCI protected but will also talk to the county to get it sealed for this area.

Thanks guys it is a pleasure being on this board and talking to you guys.

Mike

No problem...

Originally Posted by Joseph P. Hagarty

Smoke alarms on an AFCI circuit should have an integrated Battery Backup.

If not, it is noted as Deficient.

Joe, that is a very good point.

I just got off the phone with washington county in Oregon and they said that the smoke alarms do not have to be on AFCI protected curcuits it is optional. So Thats the story from my part of the world in Oregon.

Not required but is optional as long as it has a battery backup.

Originally Posted by Mike Huppi

I just got off the phone with washington county in Oregon and they said that the smoke alarms do not have to be on AFCI protected curcuits it is optional. So Thats the story from my part of the world in Oregon.

Not required but is optional as long as it has a battery backup.

Mike,

Tell them to go back and "read" the NEC.

It is not "optional", it is "required".

Of course, if they cannot read, then the entire NEC is, I guess, "optional".

Originally Posted by Scott Patterson

Joe, that is a very good point.

Scott,

I've been explaining (trying to explain) that for the past few years, guess Joe said it better.

Thanks Joe.

Originally Posted by Jerry Peck

Scott,

I've been explaining (trying to explain) that for the past few years, guess Joe said it better.

Thanks Joe.

No, you did well! I still don't like it. But, folks need to listen to those that agree with it, as that is the proper way.

Originally Posted by Scott Patterson

No, you did well! I still don't like it. But, folks need to listen to those that agree with it, as that is the proper way.

Scott,

I don't understand what you could not like about it.

First, if there is an arc in the smoke detector circuit, and that is the only way the AFCI will trip off, then that arc could be the cause of the fire which you would rely on that smoke detector to detect.

Trip the AFCI and you cut power to the circuit which is having the arc, thereby removing the energy source which would cause the fire. Thereby not needed the smoke detector to detect anything.

Second, if the arc had 'already' caused the fire, the smoke detector is still functional and will still detect that fire.

Third, if the AFCI trips off and there is no fire, the smoke detector goes into battery backup mode anyway. When the battery runs down, it chirps like it would when the battery runs down.

I cannot figure out or understand what you, and others who agree with you, "don't like" about having smoke detectors on AFCI breakers.

It really is the best of both worlds. Offers the best protection from an arc causing a fire and offers the best protection for detecting that fire.

Consider this: Should the fire originate somewhere else and burn through the insulation, shorting or ground faulting that same circuit out, thereby removing power to the smoke detector circuit, you are relying on the battery backup for detection, right? That's always been the intent anyway. So, now we have that same scenario, except that we are not trying to remove power to that circuit which is arcing and potentially causing the fire which would short or ground fault that circuit out anyway.

What's the difference?

What's there not to like about it?

I know it makes perfect sense and is logical. It's a quark I have. I wonder how many smoke alarms have been linked to fires due to arcing?

I think it is 2008 or 2010 that the NEC wants to make all outlets in the home AFCI protected. I believe I heard this from Douglas Hansen in 2005 at the ITA show in Vegas.

[QUOTE=Scott Patterson;1494]I know it makes perfect sense and is logical. It's a quark I have. I wonder how many smoke alarms have been linked to fires due to arcing?

Scott,
Not many, I would guess, but wouldn't it be the height of irony if the house was to burn due to a short in the smoke detector?

Wouldn't there have had to be a history of arc related causes for fires for there to even be a device to protect us from them?

Originally Posted by Richard Stanley

Wouldn't there have had to be a history of arc related causes for fires for there to even be a device to protect us from them?

Bingo!

Give that man a seegar.

That's what drove the research which lead to the requirement to develop AFCIs.

And I know it's old hat now and than many think knob and tube is still the best thing since sliced bread, but the origins of that problem started on old homes with ... you guessed it ... knob and tube wiring ... and the older rubber insulated wiring (which some also say is not in need of replacement - because it has not caused a fire --- yet).

Installing AFCIs on new homes with new wiring? What to heck for? Certainly not because new wiring gets old ... er ... that could become a problem later on, couldn't it?

Which gets us back to ...

The arc does not need to be "in" the smoke detector, just "in" the circuit the smoke detector is on for that scenario to exist.

And, yes, the 2008 NEC will require all 120 volt 15 and 20 amp circuits to be AFCI protected.

Since the AFCI protects the circuit, not just the device (smoke detector) I am guessing the rate of failure would be very similar to the overall rate which caused the need for AFCI in the first place. Why would it be less? Of course it will become a moot point in the near future when everything has to be protected (unless some one declares AFCI as "optional")
Jim

I just pulled some notes out from a class that I took a couple of years ago. The instructor said that around 200 to 300 home fires a year could be traced to arcing in the walls. He went on further to say that around 50% were in the kitchen and laundry rooms! Class was with Mike Casey and Douglas Hansen. They went on to say that it was a common feeling in the electrical engineering field that the electrical industry pushed for the AFCI's, not for need but for income.

Now how many of those I wonder go back to a faulty smoke alarm? I would have to guess not very many if any.

Originally Posted by Scott Patterson

Now how many of those I wonder go back to a faulty smoke alarm? I would have to guess not very many if any.

Scott,

You are the one who brought up "the" faulty *smoke detector*.

We have been talking about "the" smoke detector *circuit*, not "the" *smoke detector*.

How many fires would have, could have, been stopped with AFCIs?

How many fires on smoke detector circuits (that pesky word again "circuits") could have been stopped by AFCIs?

No knows either answer.

The question has not been "defective and arcing smoke detectors", but rather "arcing on the smoke detector circuits".

....

Here is what we have here in oregon from the state. It says AFCI's are not required for smoke alarms, I guess this is my answer at least here

http://www.bcd.oregon.gov/programs/e...ult_interp.pdf

Mike, you have to go with your AHJ, so that is the answer for you. I might make a note of the disagreement to my clients though. Also, the smokes are not required to be on AFCI, but they would be allowed to be AFCI. As previously noted, Code is a minimum.

Feb. 10th, 2003
Over 4 years old, I think it is time for your state to revisit the issue as they mentioned in the document.
They are in direct contradiction to the latest IRC (if they use that) and the NEC.
I can see a special exemption for "lifesaving" equipment, but the rest is just silly. IMO
Jim

Here is the 2005 interpretation that I got from the state of Oregon it just has a different date on it but it is recent.

http://www.bcd.oregon.gov/programs/e...s/arcfault.pdf

That's not what that says.

That says, and refers only to, 'dedicated smoke detector circuits' for new construction.

For remodels, it is 'do not need to if wiring is not exposed', but, 'if wiring is exposed, you do not need to *on dedicated smoke detector circuits*'.

Now the question becomes (and the answer is in there): Is the smoke detector on a dedicated circuit?

Because the answer to that is almost always "No.", then an AFCI would need to be installed 'if wiring was exposed' during a remodel.

How on earth are you, or anyone, supposed to follow that and know whether it was a remodel which did, or maybe did not, require AFCIs?

Talk about confusion.

What it says is 'Yes, AFCIs are required for smoke detectors, except ... ' and then it gets fuzzy.

My understaanding of that is that they are required for new construction and in remodels or additions if the area (ceilings/walls) is readily accessible. Destruction of these areas is not required ie; exposed wiring.
Mike - you might also want to explain to your clients that the local AHJ doesn't give a sh** for their safety, but, to be sure and pay their taxes. I've done that a few times when I hear "the city doesn't enforce that".

Originally Posted by Jerry Peck

Of course, if they cannot read, then the entire NEC is, I guess, "optional".

Dude, for the AHJ everything IS optional. Every model code says so.

Originally Posted by russell strahan

Dude, for the AHJ everything IS optional. Every model code says so.

Russell,

You need explain that one to me - show me where "everything IS optional" in every model code, in any model code.

I am wondering why someone added to a thread that was dead over 4 years.

Originally Posted by Jim Port

I am wondering why someone added to a thread that was dead over 4 years.

Stupidity on my part. Some others may have contributed.

Article 90 of the NEC, Section 104 of IRC. ANY model code says the AHJ has responsibility for interpretation and application. They also explicitly indemnify the AHJ. All model codes use similar language. Always have.

This is the way of the world, no point in arguing. What the AHJ enforces is code. What we read in books is a model. Authority and jurisdiction are pretty clearly defined.

Originally Posted by Jerry Peck

Bingo!

Give that man a seegar.

That's what drove the research which lead to the requirement to develop AFCIs.

And I know it's old hat now and than many think knob and tube is still the best thing since sliced bread, but the origins of that problem started on old homes with ... you guessed it ... knob and tube wiring ... and the older rubber insulated wiring (which some also say is not in need of replacement - because it has not caused a fire --- yet).

Installing AFCIs on new homes with new wiring? What to heck for? Certainly not because new wiring gets old ... er ... that could become a problem later on, couldn't it?

Which gets us back to ...

The arc does not need to be "in" the smoke detector, just "in" the circuit the smoke detector is on for that scenario to exist.

And, yes, the 2008 NEC will require all 120 volt 15 and 20 amp circuits to be AFCI protected.

Speaking of Knob & Tube, here are more pictures from yesterday's inspection worthy of the "What Were They Thinking File."

Originally Posted by russell strahan

Article 90 of the NEC, Section 104 of IRC. ANY model code says the AHJ has responsibility for interpretation and application. They also explicitly indemnify the AHJ. All model codes use similar language. Always have.

This is the way of the world, no point in arguing. What the AHJ enforces is code. What we read in books is a model. Authority and jurisdiction are pretty clearly defined.

Actually, if you were to read Article 90 of the NEC, it does not say that. (bold and underlining are mine)
- 90.4 Enforcement.
- - This Code is intended to be suitable for mandatory application by governmental bodies that exercise legal jurisdiction over electrical installations, including signaling and communications systems, and for use by insurance inspectors. The authority having jurisdiction for enforcement of the Code has the responsibility for making interpretations of the rules, for deciding on the approval of equipment and materials, and for granting the special permission contemplated in a number of the rules.
- - By special permission, the authority having jurisdiction may waive specific requirements in this Code or permit alternative methods where it is assured that equivalent objectives can be achieved by establishing and maintaining effective safety.
- - This Code may require new products, constructions, or materials that may not yet be available at the time the Code is adopted. In such event, the authority having jurisdiction may permit the use of the products, constructions, or materials that comply with the most recent previous edition of this Code adopted by the jurisdiction.

That does not give them the authority to enforce, or not enforce, the code willy-nilly as they so chose.

The IRC is even more explicit that the AHU (the Building Official) DOES NOT have the ability to enforce, or not enforce, the code willy-nilly as they so chose.
- R104.1 General. The building official is hereby authorized and directed to enforce the provisions of this code. The building official shall have the authority to render interpretations of this code and to adopt policies and procedures in order to clarify the application of its provisions. Such interpretations, policies and procedures shall be in conformance with the intent and purpose of this code. Such policies and procedures shall non have the effect of waiving requirements specifically provided for in this code.

The AHJ as the RESPONSIBILITY of ENFORCEMENT and INTERPRETATION.

NOT ...

Originally Posted by russell strahan

Dude, for the AHJ everything IS optional.

Words on paper. Please cite one example where an AHJ was materially sanctioned for failure to enforce a requirement in model building code.

"Willy nilly" is an excellent description of code enforcement. Inspectors working in multiple jurisdictions see it every day.

Originally Posted by russell strahan

Words on paper. Please cite one example where an AHJ was materially sanctioned for failure to enforce a requirement in model building code.

The codes themselves are only "words on paper" ... so what is your point ... we ARE talking about THOSE WORDS, and THOSE WORDS state what I pointed out, not what you said.

"Willy nilly" is an excellent description of code enforcement. Inspectors working in multiple jurisdictions see it every day.

Incorrect again.

Each inspector inspector to the code (unless the inspector does not know what they are doing - but there are those types in all professions), and there is nothing "willy nilly" about it.

Does each inspector inspect for the SAME EXACT thing? Yes - the code, no - regarding each different section of the code, but it is all code.

Do home inspectors inspect homes "willy nilly"? Of course not, but in your world (based on the above posts by you) that answer would be "yes".

You don't have a basis for any opinion I hold regarding HI, nor is it what we were discussing.

I repeat the request:

Please cite one example where an AHJ was materially sanctioned for failure to enforce a requirement in model building code.

Heck, I asked twice. Cite two.

Originally Posted by russell strahan

You don't have a basis for any opinion I hold regarding HI, nor is it what we were discussing.

I do have a basis for that opinion regarding you and home inspections/home inspectors ... and I stated it ... and it is totally relevant to what we were discussing.

And no citation is needed as your posts are your opinions, while I posted actual code requirements regarding what was being discussed.

In Canada The 2009 CEC does not want them on AFCI or GFCI circuits.

32-110 Installation of smoke alarms and carbon monoxide alarms in dwelling unit
The following requirements apply to the installation of permanently connected smoke alarms and carbon

monoxide alarms in dwelling units:

(a) smoke alarms and carbon monoxide alarms shall be supplied from a lighting circuit, or from a circuit that

supplies a mix of lighting and receptacles, and in any case shall not be installed

(i) where prohibited by Rules 26-720 to 26-724; and

(ii) where the circuit is protected by a GFCI or AFCI;

(b) there shall be no disconnecting means between the smoke alarm or the carbon monoxide alarm and the

overcurrent device for the branch circuit;


In Canada you only have to protect the receptacles in bedrooms.

26-722
(f) branch circuits that supply receptacles installed in sleeping facilities of a dwelling unit shall be protected by

an arc-fault circuit interrupter; and

(g) for the purpose of Item (f), “arc-fault circuit interrupter” means a device intended to provide protection

from the effects of arc-faults by recognizing characteristics unique to arcing and functioning to de-energize

the circuit when an arc-fault is detected

Statistics quoted from a recent IAEI online article.

1,348,500 fires in the United States for the reporting year.
3000 deaths associated with those fires.
17,050 injuries.
$12.5 Billion in damages.
A fire occurs every 87 seconds in the U.S.
Past statistics reveal approximately 1/3 of fires are electrical in origin.
So that basically equates to 1000 deaths from electrical fires.
A device that can ultimately prevent even one of those deaths.....priceless.
Canada needs to wake up to the facts if they specifically do not allow smoke detectors on AFCI/GFCI circuits.
Mr. Peck explained the situation very concisely.
Battery backup is required.
If AFCI trips on smoke detector circuit then the need for the smoke detector is un-neccesary (even though the battery will leave such still functional.)
When the lights/receptacles on the circuit are renedered non-functional due to an AFCI trip, investigation will ensue.
If there is no AFCI on the circuit, then the smoke detector will most likely be called upon to perform its design function. It will more than likely sound the alarm but the subsequent damage will have already occured.
Maybe the smoke detector will alert occupants and save lives.
Unless of course it's too late.
I recall an article pertaining to the state of Vermont with data reflecting a 36% (approx.) reduction in electrical fires since inception of AFCI requirements.
Apply that percentage to the afforementioned extrapolation of 1000 deaths related to electrical fires and its a no-brainer.
I'm sure somebody will find differing statistics elsewhere.