I am getting ready to wire my first loop of track on a new layout. The loop will be about 130 ft in length. What gauge wire would you guys recommend?
Thanks in advance.
I used 12 ga. stranded for my bus and 16 ga. stranded for my feeders. I applied feeders to almost every piece of rail. Some people think my choices to be overkill, but I want ZERO problems with electrical feed no matter what the load or length, and I’d much rather err on the side of overkill.
Go with palallin’s suggestion, unless you might install DCS at some point. If you want to provide for future DCS you need to wire differently.
I used 12 gauge stranded for all track power off of TPCs (TMCC environment) with 14 gauge feeders. For accessory power bus I used 16 guage with 18 gauge / 20 gauge runs to the individual accessories.
Also consider fuses on all the buses. I used Scott’s Odds-N-Ends 10 amp breakers on the main track feed and fast blow fuses for all the accessory buses.
Finally, I soldered all connections.
These sizes are definitely not over kill. Putting heavier wire in is great insurance for trouble free operation and even greater for safety and worry free operation!
Thanks. 12 gauge sounds good. I will solder as well. Is it better to use stranded over solid for workability?
Whatever size you choose to reduce the voltage drop, be sure that it is no smaller than is safe for whatever your transformer can put out:
12 AWG 20 amperes
14 AWG 15 amperes
16 AWG 10 amperes
18 AWG 7 amperes
But, even if your present transformer is not very large, it is a good idea to use 14 or 12 AWG anyway so that you don’t have to redo it if you ever get a bigger one. These sizes are easy to find at “home-improvement” stores and cheap, especially if you buy non-metallic cable (“Romex”) on sale.
What does “AWG” stand for?
American Wire Gauge
With the bigger wire (i.e., smaller gauge) I suggest stranded as it is easy to work with. As the wire gets smaller, the solid wire is not as difficult to work with, but it can still be easier to use stranded.
Two benefits of solid:
- You can shape it better to “stay put”
- You do not have to tin it with solder if you are putting it in a screw terminal such as an Atlas switch machine or SC2. You may still want to tin even the solid to keep it from corroding - but after almost a year, none of my non-tinned copper wire has corroded.
Thanks all. This forum is always so helpful.
Luther, I will take issue with “700 circular mils per amp rule” quoted on that page. Ampacity generally goes as the 3/4 power of area, not linearly with area. The numbers given for the sizes we are likely to use in particular seem very low.
Yes they are low, or rather conservative, and the authors clearly stated that.
I also found two other sites that corroborated your suggestions so I edited my post, removing the link to avoid confusion. I still use their numbers as a rule of thumb as it is safer, but your recommended values hold value as well.
My philosophy with anything on the Internet today - do not accept it until you prove it either in practice or with alternate sources [:)]
For those interested in the theory behind the ampacity rule I mentioned:
In order for the wire’s temperature rise to be kept to a safe value, the maximum power P allowed to be dissipated in the wire should be proportional to the surface area available for it to get out, which in turn is proportional to the square-root of the cross-sectional area A:
P is proportional to A^(1/2)
where ^ denotes exponentiation. The power P in turn is proportional to the square of the allowable current I and inversely proportional to the cross-sectional area A:
I^2 / A is proportional to A^(1/2)
So
I^2 is proportional to A^(3/2) and
I is proportional to A^(3/4)
Since the cross-sectional area A of a wire doubles about every three wire sizes, this means that the ampacity I doubles every four wire sizes. So, for example, an AWG 14 wire is allowed to carry 15 amperes and an AWG 10 wire can carry 30 amperes.
It is often noted that the European use of 230 volts saves copper compared to our 120 volts, but with the assumption that doubling the voltage can simply halve the copper by halving the current. It actually saves a little more, since half the current can be carried by 40 percent of the copper, not 50 percent. Incidentally, they save some more by allowing wire that is about one AWG size smaller than we do for the same current. The bottom line is that, all else being equal, their system in principle needs only 32 percent of the copper that we use.
It never ever hurts to go larger … but bad to use undersized wire . I recommend stranded wire … a heck of a lot easier to run .
Thought I might re-visit this topic. I know that everyone seems to think that “bigger is always better” when it comes to wire, but what everyone seems to be overlooking is the way and location that the wire connects to the track vs the wire size. On MTH track it’s not that much of a problem because they use a track lockon that is versatile enough to allow large wire to be connected to it. And it connects OUTSIDE the track at the side of the roadbed. But Lionel Fastrak connects UNDER the track. Using wire that is nearly the size of jumper cable wire for cars is NOT going to be practical. Plus the method Lionel uses of a slip-on connector hardly makes it possible or practical to connect large gauge wire to the connector. I guess if everyone is using tubular track, this isn’t much of a worry. Heck, you can WELD car jumper cable directly to the track and you’re all set [:D]
I notice that Lionel uses 18 gauge wire and connectors for it’s track terminal sections. Obviously this is for rather small “under-tree” layouts with minimal power drain. But Lionel Fastrak is not exactly conducive to using super-size wire for it’s connections. Any advice?
Dep
There is another factor to consider, and that is the environment of the wire. The currents given in the post are for wire in an enclosure such as conduit where the cooling is not good and there may be other wires in the conduit that are carrying currents near their rated capacity. For wire in open air, you can go down one wire gauge safely. Also, for intermittent use, you can momentarily overload the wire without damaging it or causing a safety problem. If you really want to engineer your wiring close to the limit, look on the Internet for a wire gauge table that gives the ohms/foot for different wire gauges, and calculate the power dissipation in the wire. The power = current^2 * resistance (ohms).
The numbers that I quoted three years ago are not conservative. They are the maximum overcurrent protection that the NEC allows for those wire gauges (240.4(D)). Furthermore, the ampacities allowed for 18 and 16 AWG are only when the continuous load does not exceed 5.6 and 8 amperes respectively (240.4(D)(1)-(2)). When more than a few wires are run together, these ampacities are further reduced, to 80 percent for more than 3, 70 percent for more than 6, 50 percent for more than 9, 45 percent for more than 20, 40 percent for more than 30, and 35 percent for more than 40 (310.15(B)(2)(a)).
The ampacity of a wire is based not just on the heat generated in the wire, but must also take into consideration the surface area available to get the heat out of the wire. The ampacities that the NEC allows take this into account and are designed to limit the temperature rise in the wire to a safe level, no matter what size the wire is.
Short taps using wire smaller than the overcurrent protection would normally allow are okay if they are heavy enough to carry the normal load current. This provides a way to hook up Fastrack, using a piece of smaller wire spliced to the heavier feeder, no longer than necessary to mate with the Fast-On connectors under the track. I think that a couple of inches would be plenty long. Even though I don’t need this exemption with my tubular track, I use short taps anyway (although not smaller wire), soldered to the rails, with a wire-nut connection to the feeder under the table. This makes it a lot easier to remove the track when I need to.
Bob: Thanks for that clarification about the “short taps”. [tup]
I have 14g solid along the tracks. It will barely fit into a lockon. I used 14g stranded in other places. No more than 10 feet between places where power is to the track. Usually spaced every 6 sections. You can guess there are no places where the train slows down but I’m running MPC and newer and most of my cars are light 70s to 90s 027. Now I just need a bigger transformer. 14g feeder wires to the track is probably overkill.