Define 12G building wire. How is this different than type NM cable?

I wouldn't worry about what you have, it should be fine. A speaker signal is low voltage and the wire shouldn't even begin to get warm. I'll defer to Ron, an electricity expert, for further comment. A downside of a long run with perhaps a too small gauge is that the resistance can cause some volume loss. I have one run that's at least 70 feet, with (I believe) 12 gauge wire. When I use those speakers, I need to adjust the volume upward a bit.

To answer a question you didn't ask, it's important to have impedance protection provided by using a speaker switch that does that, if you play more than one station at a time. I learned the hard way and blew out a power amp I really liked (that I can't replace because it's no longer made). The one I have is a Niles brand, it works well and provides peace of mind.

I'm still trying to figure out what he has that he's concerned about. There's nothing (code wise) wrong with using any wiring that would be legal for 120V power for audio systems. In addition to these you can use the class 1/2/3 as appropriate at a smaller gauge.

Of course, the code is only concerned with making sure the voltages and amperes used aren't going to cause problems. When it comes to the efficacy of your audio system, you need to go into details not expressed here. Voltage drop usually isn't a concern because the currents involved are low. unless you're driving big assed speakers from a remote amplifier. Many home audio systems are just wired up with cat 5.

Sorry, Ron, I meant single-conductor (THNN?) wire designed to be pulled through EMT, but just running loose in the wall - for speaker use only. The issue (brought up on another forum devoted to audio) is that if you had a fire and the insurance company saw that wire, they might blame it, even though it wasn't the cause.

Ron, homeowners will often run two-two strand wires loose, not in conduit and not shielded, through attics or crawlspaces for speaker pairs. The connections at the amp and speaker sides are typically nut on terminal post things to tighten or push in clip holds. I'm not sure how Cat-5 would be used unless the end connectors were taken off and the wires stripped and twisted together.

In my experience, signal drop due to resistance in a too small wire is experienced at anything more than about 15 feet. As a kid, I had a neighbor who was a Hollywood sound engineer (which also meant he was an equipment tinkerer) and he told me that other than the obviously overly large gauges meant for high voltage and distance, there was really no such thing as using too large a wire for speakers. He said pick the wire you think you want to use, and then go one or two sizes larger (smaller numbers, of course).

Kudzu, thanks, that's what I thought.

The issue (brought up on another forum devoted to audio) is that if you had a fire and the insurance company saw that wire, they might blame it, even though it wasn't the cause.

This sounds like one of those scenarios that someone dreamed up and pretended that it happened in order to sound like they know what they're talking about.

"There was this friend of a friend that had a dog that they adopted who's previous owner once had a fire and their insurance company combed through the ashes of the house and concluded conclusively that the speaker wire was the cause. They spent thousands of dollars trying to fight it but that inspector's case was rock solid and now they live in a van down by the river."

BTW - Insurance can't deny coverage based on accidental code violations by the home owner. You are protected from your own ignorance. Unless they can prove that you purposely made those violations in an effort to purposely burn down the house (ie. arson), they will pay.

Code violations by a professional is a different matter and the insurance company will attempt to get the pro's insurance to pay for that. But either way, the homeowner is in the clear.

Anything legal for running power wire is legal for speakers by the code.
Otherwise, you can run the Class 2 (or 3) wiring. THHN outside cable or conduit isn't code.

Cat 5 is a cable spec. The termination is a different standard (TIA-EIA-568). One doesn't mandate the other.

Isn't Cat 5 a multi stranded, individually insulated wire bundle? I'm not sure why one would bother with that for a speaker run. Resistance-wise, what gauge two wire wire would it be equivalent to?

mtvhike, I think the gauge is what matters for such low voltage for adequate service. I wouldn't worry about using what's a thicker version of lamp cord so long as it's neatly run and away from heat, sharp edges, and other physical factors.

Somehow this discussion morphed into using Ethernet cable as speaker wire; I never intended that. Ron, you say "Anything legal for running power wire is legal for speakers by the code." I was asking if it's OK to run wires which are NOT legal for power, as speakers. I ran THHN cable as speaker wire NOT inside conduit, 30 years ago.

Those are not legal (and probably weren't legal 30 years ago). You can run something that is legal power wiring (Type NM would work) or one of the classified cables. Frankly, after 30 years, I personally wouldn't be too concerned. The biggest issue would be if the wiring would chance become electrified. Since this is already good insulation (THHN) it would be safer than other things like bell or thermostat wiring that doesn't have insulation rated for 120V. As long as all your terminations don't occur inside boxes also used for power, I'd not worry.

Thanks for the focused comment, Ron. mtvhike, you're good to go.

My experience:

30 years ago, I put in 12 or 10 gauge wire to 2 different speaker locations. 20 years ago, we did a remodel to add space and two more speaker locations were wired by a licensed electrician (using the same 10 or 12 gauge wire at my request). One run (as I said above) 70+ feet.

Three pairs are in a stand up crawl space, carefully stapled to wood structure and away from hot water pipes. One pair is in the attic. All continue to function perfectly well.

Be careful with the assumption that audio is a low-voltage application. A modern AV receiver could output close to 100V if speakers were turned all the way up. In practice, this is not very likely to happen. Most people never drive their speakers anywhere close to maximum volume. But accidents do happen, and amplifiers do malfunction. So, plan for the worst case and pick adequately insulated wires.

A modern AV receiver could output close to 100V if speakers were turned all the way up.

That would really surprise me. By Ohm's law, power = voltage squared divided by resistance. Doing some hand waving (since a speaker isn't a resistive load nor is its impedance the same at all frequencies), to get an instantaneous peak of 100 volts across an 8 ohm speaker (load), you'd need a peak amplifier power output of 1250 watts!

I'm not up to date on what's happening in the home theatre business, so maybe such hairy beasts now exist. However, in my experience, claims of that kind of output power for audio amplifiers ("peak music power") are pretty much advertising fantasy from the marketing department, not from the engineers.

Even if such peaks should happen at ear-splitting volume levels, any UL approved wire typically used for either home wiring or appliance wiring will be rated to withstand at least 300 volts, and more likely 600 volts. So I honestly don't think this is a real concern, unless you're buying super cheap unrated imported-garbage wire. Probably not even then. (For the record, I don't recommend using super cheap unrated imported-garbage wire for anything.)

With anything that isn't DC, things get more complicated. I based my math off of what I found here: https://electronics.stackexchange.com/questions/123390/calculating-the-peak-output-voltage-from-rms-power-of-an-audio-amplifier

And my comment is with regards to people saying that audio is always low voltage so it would be ok to use pretty much any type of wire.

Yes, if you use Romex or anything similar, there is no concern. But if you use some random crazy thermostat wire that you jury-rigged for more amps, then all bets are off. Don't do that.

Or simply buy speaker wire that is rated for in-wall installation. Then all of this is moot.

The voltage isn't the problem. The voltage rating is largely determined by the insulation and all these wiring have good enough insulation. The issue is current. Even 100W full bore into 8 Ohms is less than 4 amps.

Thermostat wiring is BAD because not only because of the lack of ampacity (typically right about one amp) but because it's insulation isn't rated that high. This is why you can't run it through boxes, etc... where 120V terminations are done.

The category 2 and 3 wires that the original poster have ampacities around 10 and good enough insulation. Obviously, even 14g conductors rated in art 300 would be plenty as far as safety goes.

We're not even going to start on the "magic ear" investigations of properties of the wire.

I'd want to make sure that those speaker cables are clearly labeled at both ends, not so much for you, but for the next owner of the house.

I'm not sure what would be so BAD about thermostat wire. But I wouldn't use CAT5 Ethernet wire as someone suggested - you might end up losing half the power going to the speaker with that small a wire.

I explained the issues (see my last post) with thermostat wire. I didn't "recommend" CAT5, I just pointed out that it is often used and is not illegal. But what YOU fail tor mention is POWER capacity isn't determined by wire size. AMPACITY is. As I pointed out that if you have a low power system operating into an 8ohm load, the current flow won't be that high.

Modern amplifiers regularly exceed 100W per channel. No, they don't do this at all times. But their power supply, their capacitors, and the current-capacity of their circuits all can do this, and most importantly, they can easily do this in case of faults. And that's what we try to design around. We don't want things to catch fire, just because of an otherwise minor fault somewhere else in the system (e.g. a toddler that fiddled with the AV controls or a pet that chewed on the wiring).

We also don't worry just about the 8Ω load, but with thin wires we worry about damaged insulation -- especially so since 100W (or 12A) is enough to overheat wires and let insulation melt. That can also lead to arcing, and that's why voltage is a concern as well. With longer runs, this is potentially less of a problem (heat is dissipated over the larger volume, and wire resistance limits the overall current), but you can't really count on exactly where your insulation is going to be damaged. It could happen a few inches away from the amplifier just inside the wall.

Resistance, power, current, voltage, heat dissipation, ... are all related to each other. You need to compute all of them for a specific system to get a sense of how it behaves.

I've seen people use scraps of Cat 5 wire as speaker wire, but they twist all the conductors together at both ends to make one cable, and use two of these per speaker. I forget what the equivalent wire gauge is, but it's pretty decent.

If you're running 100 Watts through your home speaker cables for long enough to heat them up, you're probably deaf by now. That's a stinking LOT of continuous power, even with moderately inefficient speakers.

When I was a pup, older engineers still considered a 20 Watt amplifier entirely adequate for a medium sized auditorium. Those were tube watts, though, so even if it was clipping now and again, the resulting distortion wasn't all that obnoxious.

The reason for using large gauge wire for speakers is that you need low resistance so as not to excessively limit peak power. That buys you good transient response, especially at low frequencies. If you were only concerned with average power, AWG 24 to 18 would be plenty for as much volume as most sane people would ever want at home. And, in fact, speaker wire of that size is used in most mass market mid-fi systems.

Actually, how loud 100W is depends on the efficiency of the speakers. I've had some folded horn PA speakers that were deafening at 20W. I've had others that I could run 100W into a small living room comfortably. However, most in wall/ceiling home speakers are pretty efficient. My amp system puts out 100w a channel (supposedly), and I've never turned things up over half way. Note each channel is only serving a small area (there are 24 channels).

Ron, you're absolutely right that it depends on the speakers' efficiency. But if you measure the actual power output of your amplifier at normal listening levels, I think you'll find the average output even with sealed acoustic suspension speakers isn't more than a few watts. That extra power is there to handle big peak demands, like a bass drum hit.

Bar perhaps Fisher, the typical console stereo of my youth (remember The Voice of Music? Capehart?) could seldom produce more than 2 to 5 watts per channel of continuous power, but it managed to do a more or less adequate job for undemanding music in the average living room. You could turn those boxes up enough to make mom cover her ears, and not even be at full volume yet. :)

A little anecdote might serve here. The old timers who said 20 watts was plenty for a midsize auditorium were right. Many years ago, I volunteered to provide audio for a youth dance concert. I reserved a PA from a local rental agency, but I misunderstood their closing time on Friday, so I missed picking it up! Any port in a storm, so I took a late 60s vintage integrated amplifier (noisy bipolar transistors, undersized linear power supply) and 2 Dynaco A25 speakers (a moderately efficient ported bookshelf design). I knew the amp was good for just under 10 watts of continuous power per channel into 8 ohms at clipping, because I'd measured it.

The dance was in a high school auditorium. We're not talking about the gigantic megachurch sized auditoriums of today's "learning centers," this was a more modest sized place built, I would guess, in the first third of the 20th century or a bit after. The auditorium was typical of that era, a rectangular box, orchestra and balcony seating, bare concrete floor, unpadded wood seats, big wood windows, solid plaster walls and ceiling. So it was fairly live, thank goodness.

It worked mostly fine, at least for the audience. There was plenty of volume out in the house, though without stage monitors, the dancers complained that they had some trouble hearing.

So, 20 watts indeed can be auditorium power. And average power in home listening is seldom more than a few watts.