14gage wire for light switch?

[mbrooke]

The BIG question is, How do we test for it?

:roll:

 

[Electric-Light]

Well, the fixture has been approved by Underwriter's Laboratories ;-) Your branch wiring has not! :lol:

Exactly my point. A luminaire that connects to 10 AWG branch approved as a unit with three LED ballasts inside can utilize 18 AWG wires to interconnect within the fixture if that's how it was listed but you're not allowed to thread 18 AWG to interconnect three luminaires that use the same ballast in tandem.

This explains it.
http://ecmweb.com/code-basics/cords-and-fixture-wires

As a practical point, fault let through current can heat up the 18AWG wiring beyond the insulation rating before the breaker trips, but the spread of damage would be entirely contained within the fixture.

 

[DrSparks]

As a practical point, fault let through current can heat up the 18AWG wiring beyond the insulation rating before the breaker trips, but the spread of damage would be entirely contained within the fixture.

Precisely, which is why I would never do this with romex against wood framing!!! :happysad:

 

[mbrooke]

As a practical point, fault let through current can heat up the 18AWG wiring beyond the insulation rating before the breaker trips, but the spread of damage would be entirely contained within the fixture.

It doesn't in most cases.

Precisely, which is why I would never do this with romex against wood framing!!! :happysad:

You do know that in motor applications the same wire is allowed to have an OCPD 250% larger + the next size up then would normally be allowed? #14 can legally be put on a 40amp breaker. I don't think fault current would be a problem here.

 

[DrSparks]

It doesn't in most cases.

Where is the evidence of this? Do you work for UL?

You do know that in motor applications the same wire is allowed to have an OCPD 250% larger + the next size up then would normally be allowed? #14 can legally be put on a 40amp breaker. I don't think fault current would be a problem here.

I do know this. But that only applies when you are dealing with a dedicated circuit where you know the total load being connected. This would certainly not apply to a GL branch circuit! Fault current is not the issue. A direct short will trip even a 200amp breaker. The issue is overloading, which is why we have the 15 and 20amp requirement for 14 and 12 AWG wire ;-)

 

[mbrooke]

Where is the evidence of this? Do you work for UL?

The temperature rating of fixture wire is much higher then NM, and often of such short length the impedance introduced into the circuit is negligible in limiting current to the effect of increasing breaker operating time. The same science that allows smaller EGCs and up-sizing OCPDs for motors also allows 18 guage fixture wires to hold together during a fault on 15 and 20 amp residential circuits.

If I did work for an NRTL I might not be allowed to disclose it on a public forum.

I do know this. But that only applies when you are dealing with a dedicated circuit where you know the total load being connected. This would certainly not apply to a GL branch circuit! Fault current is not the issue. A direct short will trip even a 200amp breaker. The issue is overloading, which is why we have the 15 and 20amp requirement for 14 and 12 AWG wire ;-)

True, however 2 points:

1. The load is guaranteed not to be exceed if feeding say a keyless socket. Yes a person could screw in a 300 bulb, but the load would not exceed 15 amps.

2. If the load did exceed 15 amps there is still no danger as NM is already good for 20amps under the 75*C column. #14 NM is allowed to be loaded to 20amps for certain applications in the CEC.

Of course Im not saying the NEC allows this, but no real danger exists either.

 

[DrSparks]

The temperature rating of fixture wire is much higher then NM, and often of such short length the impedance introduced into the circuit is negligible in limiting current to the effect of increasing breaker operating time. The same science that allows smaller EGCs and up-sizing OCPDs for motors also allows 18 guage fixture wires to hold together during a fault on 15 and 20 amp residential circuits.

If I did work for an NRTL I might not be allowed to disclose it on a public forum.





True, however 2 points:

1. The load is guaranteed not to be exceed if feeding say a keyless socket. Yes a person could screw in a 300 bulb, but the load would not exceed 15 amps.

2. If the load did exceed 15 amps there is still no danger as NM is already good for 20amps under the 75*C column. #14 NM is allowed to be loaded to 20amps for certain applications in the CEC.

Of course Im not saying the NEC allows this, but no real danger exists either.

I agree mostly, but I've seen first hand the results of continuous overloading of branch wiring (i.e. window air conditioners) and I understand why the NFPA enacted these limitations. For the motor load example you provided, you are allowed to exceed the normal OCPD requirement to facilitate motor start current, which is very brief. The circuit depends on the overload circuitry integral to the motor to prevent continuous overloading.

 

[mbrooke]

I agree mostly, but I've seen first hand the results of continuous overloading of branch wiring (i.e. window air conditioners) and I understand why the NFPA enacted these limitations. For the motor load example you provided, you are allowed to exceed the normal OCPD requirement to facilitate motor start current, which is very brief. The circuit depends on the overload circuitry integral to the motor to prevent continuous overloading.

You are correct, however, if the load is fixed in place like a light fixture it will not pull more current since more of them can not be added to the circuit. In such a case all that would be of concern is short circuit and ground fault protection. The NEC doesnt allow the practice, but the CEC and other codes do if the load is hard wired and fixed in place like electric heat.

 

[iwire]

You are correct, however, if the load is fixed in place like a light fixture it will not pull more current since more of them can not be added to the circuit. In such a case all that would be of concern is short circuit and ground fault protection.

Or larger lamps ....

The NEC doesnt allow the practice, but the CEC and other codes do if the load is hard wired and fixed in place like electric heat.

The fact that the NEC prohibits something does not mean the practice is unsafe.

The fact other codes allow something does not mean the practice is safe.

When it comes to conductor ampacity there is no black and white, all the codes make judgments and estimates on safety.

The conditions of use have much more to do with it than the ability of copper to carry current.

There are ways you could safely use 14 AWG to carry 50 amps, there are ways you could unsafely use 14 AWG to carry 10 amps.

Your continuos pointing out that the CEC allows this or some other code allows that is nothing but a waste of out time.

If anything is likely to change it is other codes tightening up.

 

[mwm1752]

Table 310.106(A) shows the minimum of #14 copper however section 240.4(D)(1) allows smaller conductors. But I don't know if we can use #18 or #16 as branch circuit conductors.

Doesn't meet 240.4(D)(1)(2)

If you would run #18 fixture wire from the switch to the fixture, it would be code compliant.

Are not fixture wires part of the luminaire or a listed extention for that use?

You can't use them as branch circuit conductors, but you can tap fixture wire in those sizes to the branch circuit conductors to feed a fixture. 240.5(B)(2)

IMO using a switch makes it part of the branch circuit, now where you junction fixture wires to branch circuit might be a tap. but don't taps per 240.21 need to be protected at either end? I see no branch circuit lighting in Table 210.2

 

[mbrooke]

Or larger lamps ....

Not if its fixed modul LEDs, or a single keyless socket.

The fact other codes allow something does not mean the practice is safe.

Time has proven otherwise.

When it comes to conductor ampacity there is no black and white, all the codes make judgments and estimates on safety.

Of course, its their job.

The conditions of use have much more to do with it than the ability of copper to carry current.

Of course, thermodynamics is the foundation on which current limits are built.

2.5mm2 T&E (NM) is legally allowed to carry 27 amps when clipped direct, but only 13 amps when imbedded within thermal insulation inside a wall, hence why regs restrict current carrying capacity based on conditions of installation.


http://www.batt.co.uk/upload/files/currentratingstable4d5_1312876143.pdf

There are ways you could safely use 14 AWG to carry 50 amps, there are ways you could unsafely use 14 AWG to carry 10 amps.

That true, but 14NM is not likely to encounter conditions that restrict its ampacity to that level other then serve bundling which would be a code violation in any case.

Your continuos pointing out that the CEC allows this or some other code allows that is nothing but a waste of out time.

You said I was expressing opinions rather then facts, I provided fact where NM is used at 20amps.

If anything is likely to change it is other codes tightening up.

I doubt that will happen. The CEC just allowed this in the last 5 years, and other code making bodies see no point in adding extra safety factor to the point of no return on investment.

 

[peter d]

If I did work for an NRTL I might not be allowed to disclose it on a public forum.

So in other words, you're not an electrician.

 

[DrSparks]

Some points I'd like to add:

1. There's no reason for a residential lighting circuit to be fused at 20amps, ESPECIALLY if you're installing fixed LED fixtures. So just put in 15amp breakers... problem solved.

2. I've actually seen a dehumidifier in a basement plugged into a socket adapter! So if there is an Edison socket, we have to assume any load could be plugged into it. You can argue that someone could pull apart an LED fixture and wire in an outlet, but that's not easily accomplished by the average home owner. Everything can't be idiot proof but it's pretty easy to screw in a light socket adapter.

This is the very fact why the NEC limits the ampacity of branch wiring #14 at 15a and #12 at 20a

 

[user 100]

Some points I'd like to add:

1. There's no reason for a residential lighting circuit to be fused at 20amps, ESPECIALLY if you're installing fixed LED fixtures. So just put in 15amp breakers... problem solved.

2. I've actually seen a dehumidifier in a basement plugged into a socket adapter! So if there is an Edison socket, we have to assume any load could be plugged into it. You can argue that someone could pull apart an LED fixture and wire in an outlet, but that's not easily accomplished by the average home owner. Everything can't be idiot proof but it's pretty easy to screw in a light socket adapter.

This is the very fact why the NEC limits the ampacity of branch wiring #14 at 15a and #12 at 20a

This is spot on. Many tend to forget about the lack of true convienence outlets from the early 1900s. The only "receptacle" many people had was an edison socket paired with a hubbell adapter. People would utilize that single socket for all kinds of stuff-photos from the era usually show all kinds of black spaghetti draped near the ceiling and inevitably fires would result. I *think* I had read once that those kind of conditions was the catalyst for the 10 amp fusing of lighting circuits at the time.

 

[DrSparks]

This is spot on. Many tend to forget about the lack of true convienence outlets from the early 1900s. The only "receptacle" many people had was an edison socket paired with a hubbell adapter. People would utilize that single socket for all kinds of stuff-photos from the era usually show all kinds of black spaghetti draped near the ceiling and inevitably fires would result. I *think* I had read once that those kind of conditions was the catalyst for the 10 amp fusing of lighting circuits at the time.

Reminds me of "A Christmas Story". One of my favorite movies ever!

 

[mbrooke]

So in other words, you're not an electrician.

Why does it matter? Iwire took something out of context and everyone else seems to be fixated on it now. :roll:

 

[mbrooke]

Some points I'd like to add:

1. There's no reason for a residential lighting circuit to be fused at 20amps, ESPECIALLY if you're installing fixed LED fixtures. So just put in 15amp breakers... problem solved.

2. I've actually seen a dehumidifier in a basement plugged into a socket adapter! So if there is an Edison socket, we have to assume any load could be plugged into it. You can argue that someone could pull apart an LED fixture and wire in an outlet, but that's not easily accomplished by the average home owner. Everything can't be idiot proof but it's pretty easy to screw in a light socket adapter.

This is the very fact why the NEC limits the ampacity of branch wiring #14 at 15a and #12 at 20a

Its just a guess, but I think 240.4(D) might also be influenced by a typical breaker's trip threshold. Per UL standards a typical breaker can legally hold 125%+ indefinitely with most trip curves starting beyond 100%. So a 15 amp breaker could easily be loaded to 19 amps as well as the wire, so restricting the current rating would make sense.

 

[DrSparks]

Its just a guess, but I think 240.4(D) might also be influenced by a typical breaker's trip threshold. Per UL standards a typical breaker can legally hold 125%+ indefinitely with most trip curves starting beyond 100%. So a 15 amp breaker could easily be loaded to 19 amps as well as the wire, so restricting the current rating would make sense.

True. Most breakers will allow overcurrent for a period of time but will eventually trip, which is generally ok for brief periods. The real problem is continuous overloading over time, which, over time can cause fires.

 

[mbrooke]

True. Most breakers will allow overcurrent for a period of time but will eventually trip, which is generally ok for brief periods. The real problem is continuous overloading over time, which, over time can cause fires.

Thats true, people will sometimes load outlets as much as they can without getting the breaker to trip.

 

[don_resqcapt19]

Thats true, people will sometimes load outlets as much as they can without getting the breaker to trip.

Assuming that the OCPD is correct for the wire size, I don't see that as a problem.