Really dumb DCC question

[Blind Bruce]

If I have a circle of track, will I need a gap in the rails?
BB

[cacole]

No. The only time you need a gap in the rails is at a siding that you want to turn off so you can park something on it and not have the lights continually on, or if you have two loops of track you need a gap between them.

[MisterBeasley]

Even with DCC, you still need to electrically isolate a reverse loop at both ends. I would isolate the spur tracks at a turntable to allow for complete shutdown. I don't understand why you would have to isolate two loops from each other, though. Also, I don't see why (in a previous post) it was suggested that the feeder wires not be looped.

I don't have DCC, and I'm certainly no expert either, but I always like to hear from the pros when something strikes me as counter-intuitive.

[selector]

The feeders are one thing, Mr. B., but the thread was dealing with the 'bus', the heavier wires that emanate directly from the controller. The feeders come off the bus and are wired at the other end to each rail. I believe that if you close the loop on the bus wires, you create a closed circuit with no resistance, so the locos will see it as a short...and so will your controller. Not sure if I have that exactly technically correct, but closing the bus loop is a no-no. The trains are meant to do that, way up on the rails, where their motors will offer resistance to the current, and generate locomotion from it.

As for reversing loops, it isn't so much two loops that is the problem, but two rails looping back onto themselves via a turnout. If you draw a diagram of that situation, you'll see that the positive rail (left or right, your choice), and the negative rail loop back to their opposites, thus creating the same no-no that looping the bus wires would do. In any and every case on your road, you want the loco engine, or decoder, whether mobile or stationary, doing the loop closing because they are designed to do that. Anywhere else, and the road shuts down 'cuz the controller goes beep (or, if you are unlucky, the decoder goes 'puff').

In the case of a turntable, the loco leaves the approaching rails oriented with whatever polarity each rail has according to your wiring. It begins to transfer to the bridge rails, and they may, or may not, be oriented the same way...it depends on which wire goes to which rail on the bridge. Also, which end of the bridge are you presenting to the approaching loco? That will determine the polarity of the 'left' and 'right' rails from the loco's point of view.

Now, let us say you have positive aligned with positive, and the loco transitions onto the bridge nicely. You turn the bridge to align it with the roundhouse stall rails, and back or drive the loco into the stall. Can you be sure the stall rails are aligned positive with positive to the bridge rails? Well, the loco will short and stall very quickly if you haven't got them oriented for polarity. So, in DCC, an auto reversing mechanism senses the wrong orientation, and switches it so that the decoders and controller...and loco...don't know any better. They keep running as if you were on properly oriented track.

I hope this has shed even a bit of light on what I agree is a difficult subject to 'get'.

I agree, there is no need to gap a simple circle, because each rail is a loop, but a separated loop so that the loco engine is the only path through which the current can pass. That's where the electrical energy is transformed into rotational energy to the gears. Even a turnout will not compromise this integrity, except in the event that an all-metal wheel and axle set derails at the turnout and makes contact between the left and right rail at the same time. That creates the dreaded short of the 'do I close the loop on the bus wires?' type.

[MisterBeasley]

Electrically speaking, though, the bus should be the same as the rails. I'm assuming the bus is a 2-wire thing (let's call them black and red.) In a simple loop, the red wire would be connected by multiple feeders to the outer rail of the track loop, and the black wire would be connected to the inner rail. So in this case, what is the harm in looping the red wire back on itself, and the same with the black wire? Of course, you don't want to connect red and black. That would be a dead short, even from a DC standpoint.

Does DCC operate at a high enough frequency that looping the wires would create multipath phase errors at the far end of the layout if you did this?