Maybe I am overthinking this, but I just built a small layout about 2 ft x 8 ft double loop (made using T-Trak modules) using Kato Unitrack and I am using the Unitrack feeders which go to the Unitrack 3-to-1 connections and then ganged multiple 3-to-1 connectors to a DC powerpack. Everything seems to be running smoothly. Since all of the Unitrack wiring is 24 gauge (from I can read on the wires) should I instead run the feeders to a thicker gauge bus to the DC powerpack? Hans
Hi Hans, I suggest using 16 gauge wire to the power pack. 24 gauge is fine for short runs of a few inches. But the resistance build-up for longer runs would limit the power available to your engines.
This is a relatively small DC N scale layout, not a sprawling club layout with hundreds of feet of track, or dozens of HO or larger locomotives to be powered. Thus, heavy bus and lighter feeder wires are not really necessary, and the Unitrack 3-to-1 wye cables and Unijoiner feeders will work great. However, connecting all the wyes in a long line, like a "vine", with each wye cable feeding two Unijoiner feeders and the next wye cable, is not as effective or "balanced" as connecting the wyes in a broad "tree", with feeders primarily only at the "leaves." Wyes should primarily feed wyes, until you have enough open sockets to plug in every feeder. This ensures that every feeder "sees" close to the same total length of wire (and therefore voltage drop) back to the power pack. Hint: No matter how you plug them together (vine or tree), N 3-way wye cables will supply 2N+1 feeder cables. In practice, some middle ground between a vine and a tree may be more practical, but the closer you can emulate a tree than a vine, the more uniform the total resistance from the feeder back to the power pack.
Thanks for your advice, Andy. My suggestion was based on being a DC Dinosaur unfamiliar with Unitrack and its wiring. Your recommendation sounds spot-on.
I certainly would not refer to heavy bus wiring with periodic lighter-gauge feeders as prehistoric. For larger, more permanent layouts, that is still a preferred method. But small layouts of N scale Unitrack are a different application, with a more convenient wiring solution, necessitated by their Japanese home market, where space for a persistent layout was typically not available. The need to dismantle and reassemble (and re-wire) their layouts necessitated a convenient, flexible and reconfigurable wiring solution, and Kato provides it. Modular layouts (even those using Unitrack) typically have at least one feeder per module, and some use heavier duty Anderson Powerpole connectors for connecting the individual modules to a power bus for the layout.
Let's do the math... First, 24 AWG wire has a resistance of 88 milliohms per meter at 86F (30C). 1 yard (36") is 0.9144 meters. The length of a Kato wye cable is 1 yard. The Kato feeder cable (spade lugs on one end, to single Kato receptacle on the other), should he use one, is another yard. Thus, the room temperature total (round trip) resistance for the 24 AWG feeder cable and the first wye cable is 322 milliohms. That's 322 millivolts (not quite 1/3 of a Volt) of voltage drop per Amp of current (~4 N scale locos). If you want to waste more copper on it, knock yourself out... Or you could just double (in parallel) the feeder cable and first wye, and divide up the remaining wye cables amongst the two, if you want to avoid soldering. I wouldn't bother with either. Standard disclosures: YMMV, CCPD, OIMACTTA, yada yada yada.
For such a small layout, you really only need 1 feeder per track for the whole layout. I personally would do two feeders per track, one on each side of the layout. For 4 feeders, I wouldn't worry about the wire size at all. If I wanted to add a buss wire, I would use a larger gauge to help with short circuit protection. I had a small layout that used 24awg wire and the power supply breaker didn't trip during a quarter test. I upgraded to 16awg wire and the protection passed the quarter test. Note - I only used 16awg because I had some on hand.
I had minor issues using DC on a 36x80 HCD Unitrack thrice-around with only one powered Unijoiner (pair) per loop, more or less equally spaced. Locos visibly/audibly slowed down when far away from the nearest feeder. This was in a spare bedroom that had continuous climate control. Similarly, a ~6' long oval test loop on my dining room table needed two power connections (roughly equally spaced) to avoid obvious speeding up and slowing down in different parts of the oval, using DCC. This was continuously climate controlled. IINM, the THERR layout uses roughly one feeder (powered Unijoner pair) per ~30" of track (three long straight pieces), and performs very well, even in a non-continuously-climate-controlled environment (parked RV trailer). These examples might serve to define a "works/doesn't-work" spectrum for Unitrack power feeder spacing. Somewhere in there can likely be found a good "sweet spot." Powered Unijoners are easy to apply at module boundaries as needed. If the modules have their own power connections, by all means use those.