DCC Wiring the DCC power bus - short circuit protectors?

Dog Star

Western Thunderer
Steph, thank you for providing info on wiring for DCC - you just knew that there would be more questions, did not you!

To answer your questions:

1. Pass. I'll look it up - they're not tricky or even critical in their component selection.
Please try to make the answer understandable to a silly old bear(e).

To answer your questions:

3. Again, it's not. If you think/plan every layout as an end-to end then the snubbers (if required) go on the ends of each/the bus.
Got that... arrange for the snub stubs to be at the outer ends of each power bus. What happens if we then need to test just (say) the middle of three boards and for which there are no snub stubs?

In general there's no great difference between 'analogue' and DCC wiring. The issue is merely how well you do it. The main thing to remember is that under DCC there's the potential for some very high currents to be winging their way around your layout; using 5A rated cable is an absolute minimum. Even then Richard's using a 10A booster, so should be using something rated at least 20% over that (i.e. 12A rated). Same for any/all sockets and switches.
My intention was to use 1mm diameter copper stripped from twin and earth. What current can that carry? What diameter copper cable is needed for 10A plus contingency?

If you've got the points controlled through the DCC it's worth having them switched from the track supply (also switched). That way, should you end up with a short-circuit across a point (we've all driven into a 'closed' point haven't we?) you can switch out the track supply, change the point and restore the track supply with most people remaining completely unaware of what's happened.
I know what you are talking about here... just do not understand what you mean about "having them switched from the track supply (also switched)", a diagram might help here.

One other quick thing to mention is that running all the buses as twisted pair is highly recommended. If you do this, there's a chance you won't need the snubbers as they'll do a great deal to reject any interference.
My immediate reaction was balan/unbalan... and then thought, silly old bear(e), this is an AC circuit. Twisted pair is a sure recipe for a short if I use bare copper wires!!!! If I use 15A rated twin, multistrand, cable then I can strip the outer insulation and wind the two insulated wires to create a twisted pair (others might think that this discussion could only be between a twisted pair of modellers). A power bus created as a twisted pair is going to be a pain in the proverbial with droppers every 21cms (30' prototype rail length) - removing the insulation from a twisted pair, keeping the un-insulated wires apart and then making the feed joints looks like a mind-numbing activity.

regards, Graham
 

Steph Dale

Western Thunderer
Steph, thank you for providing info on wiring for DCC - you just knew that there would be more questions, did not you!
Yep!

Please try to make the answer understandable to a silly old bear(e).
No problem - it's only a resistor and capacitor after all. Actually, you'll probably find chapter and verse in the Heyside thread...

Got that... arrange for the snub stubs to be at the outer ends of each power bus. What happens if we then need to test just (say) the middle of three boards and for which there are no snub stubs?
Don't worry about it. The need for the snubbers increases with size and complexity of the layout. It's related to the ability of the layout wiring to either reflect the DCC signal around so it gets jumbled or act as an aerial.

My intention was to use 1mm diameter copper stripped from twin and earth. What current can that carry? What diameter copper cable is needed for 10A plus contingency?
Pass. I buy my cable from either motor factors or RS. In either case it's a breeze to rock up and say "can I please have a reel each of black and red insulated flexible cable that's rated around 5Amps?"

I know what you are talking about here... just do not understand what you mean about "having them switched from the track supply (also switched)", a diagram might help here.
Sorry, lack of clarity: Get two f**king big switches (technical term, but see comments about current ratings). Wire one into the track feed bus, the other into a separate bus that's connected to the accessory/point decoders. Normal position for both is 'ON', under fault conditions flick the track feed one to 'OFF', set the point(s) through your handset and then reset the track switch to 'ON'

My immediate reaction was balan/unbalan... and then thought, silly old bear(e), this is an AC circuit. Twisted pair is a sure recipe for a short if I use bare copper wires!!!! If I use 15A rated twin, multistrand, cable then I can strip the insulation and wind the two insulated wires to create a twisted pair (others might think that this discussion could only be between a twisted pair of modellers). A power bus created as a twisted pair is going to be a pain in the proverbial with droppers every 21cms (30' prototype rail length) - removing the insulation from a twisted pair, keeping the un-insulated wires apart and then making the feed joints looks like a mind-numbing activity.

Actually, not a bad analogy to compare with balanced/unbalanced cable architectures. Twisted pair speaker cable works because it stops the cable acting as an aerial and picking up external interference.

So, a thought in terms of architecture; run a pair of twisted-pair buses down the boards. Each is switched at the booster, one for 'track' the other for 'points'. Both may have a snubber at the end furthest from the booster, if it's needed. Each bus has one connection on each board. From that connection a series of spurs can be taken to each track section or point/accessory decoder.

Your track is wired normally - with a bare copper wire roughly following each rail, under the boards. each of these wires is connected by a spur to the track bus. I'd suggest using tinned copper at a high enough rating to take the 5A or so suggested, where they cross I'd suggest putting heat shrink insulation on both wires.

Take care with the selection of whatever means you use to switch the polarity of the crossing vees on your points (sorry, I only just thought of this). These too need to be able to carry sensible current - a couple of Amps might be enough here, depending upon how many locos you could have in each track section.

Steph
 

Dog Star

Western Thunderer
In general there's no great difference between 'analogue' and DCC wiring...
Steph,

When you explain the way in which snub stubs work for DCC - you did say that you would explain did you not? - please let us all know what impact/ effect a DCC snubber has when the power bus is used for 12V DC.

regards, Graham
 

Dikitriki

Flying Squad
Good question Graham - and not one that had occurred to me. Heyside is switchable, but we haven't used DC since we went DCC.

Richard
 

Steph Dale

Western Thunderer
Steph,

When you explain the way in which snub stubs work for DCC - you did say that you would explain did you not? - please let us all know what impact/ effect a DCC snubber has when the power bus is used for 12V DC.

regards, Graham

Graham,

To quote myself back from the Heyside thread:
"Maplin http://www.maplin.co.uk/ components will be fine: Resistor 100 Ohm, 0.6w would be item M100R at 24p each. Capacitor 0.1uF, 50v would be item BX03D at 31p each. These are pretty expensive prices, but you'll only need a few of each, not thousands."

So that's pretty simple and cheap. They're connected in series across the ends of the bus. Slightly simplistically what happens is that the capacitor allows passage of the high-frequency AC component in the DCC signal to the resistor which converts it to heat. So the signal can travel down the bus and then gets turned into heat, so can't reflect back to mush up your next instruction. And, as I said before you may not need them anyway.

In terms of 'safe' use with analogue they'll be fine as the capacitor will act as an open circuit. There's no AC component to worry about. However; if you're using feedback controllers expect some trouble as the feedback will act as AC, the signal will pass the capacitor and the resistor will work to clamp the feedback quite nicely. This may lead to runaways under analogue control. Potentially the resistors might get a little warm too.

Steph
 

3 LINK

Western Thunderer
Good question Graham - and not one that had occurred to me. Heyside is switchable, but we haven't used DC since we went DCC.

Richard

Evening Richard,

Hmmm, so I could of tried my 2-8-0 tank through your 31.5 turnouts after all :rolleyes::thumbs:.

Or do you not like the sound of flange squeal........

Martyn.
 

Dog Star

Western Thunderer
Hmmm, so I could of tried my 2-8-0 tank through your 31.5 turnouts after all :rolleyes::thumbs:.

Or do you not like the sound of flange squeal........
More like the scream from tortured timber as the consolidation tank teaches a lesson to the lanky track by straightening every curve :oops: ;) :thumbs: .

regards, Graham
 
Top