JMB,
I run HO, usually have 6 -8 locomotives running at anyone time using Lenz DCC with a single power district. I have a RRamp meter and hardly ever pull more than .5 - 1.2 amps. I run NS flex track with most of the joints soldered and I do have a DCC bus but it is only 16 gauge stranded wire. No regrets, no issues with current and I have feeders every 6 - 8 feet.
My layout is using CTI block detection as well to provide computer automated operations and signalling.
I use 2 terminal barrier strips where ever I connect into my buss and I use crimp on spade connections.
IMHO, 12 or 14 gauge is overkill. Issue is usually track rail joiners
Joe
See my model railroad blog at http://wwwjoe-daddy.blogspot.com/
My 13 lessons learned by a Newbie are posted there.
Good luck and happy modeling.
1. You could split the bus into a T, with the stem connected to the power source/controller, and the two arms going in opposite directions to each extreme of your layout. No probs.
2. Leave them tidied up wth a retainer of some sort, and there is no reason for them to be covered if they are held up tight to the nether side of something and are highly unlikely to come into contact with another bare wire. Wrapping a small piece of tape around the ends wouldn't hurt, though.
Hi John,
I think you said cold, not old, right? :)
Q1 - Yes, or a 'T' as was already answered.
Q2 - As I said, I use barrier strips with 2 postions, (terminal strips) for connections to the buss. This means that at the end of the buss is a terminal strip where the wires are terminated with a connection to the required block feeder(#20). When I expand the layout, I add a new piece of #16 wire.
Good luck and we look forward to pictures of your layout soon!
I am glad that 16 gauge wire works fine. 12 and 14 seem so big. I am beginning to wire my layout soon and plan to use 16 for the bus and 22 for feeders.
I also wondered about what to do with the end of the bus wire so this has been an informative thread for me.
grayfox1119 wrote:I would also advise you to coil the bus wires, about 3 to 4 turns per running foot. DCC is an alternating signal in the 900 Mhz frequency range, as such, it is possible to pick up radiated interference. Is it common? No, Is it possible ? Yes.....it is very easy prevent, but your choice. It is NOT necessary to coil the track drops, those runs are too short to cause any problems.
This topic has been covered before and proven wrong.
Personally I think its a waste of time. Just wire it straight and leave it at that.
Driline wrote: grayfox1119 wrote:I would also advise you to coil the bus wires, about 3 to 4 turns per running foot. DCC is an alternating signal in the 900 Mhz frequency range, as such, it is possible to pick up radiated interference. Is it common? No, Is it possible ? Yes.....it is very easy prevent, but your choice. It is NOT necessary to coil the track drops, those runs are too short to cause any problems.This topic has been covered before and proven wrong. Personally I think its a waste of time. Just wire it straight and leave it at that.
Jeff But it's a dry heat!
Driline & Jeff......buss wires are like antennas, and the longer they are, the more susceptable your layout is to problems. If you have an O scope, you can observe the waveforms of fast rising and falling pulses, and your booster can be considered as a transmitter. Decoders have been blown out by this sort of inductive reactance, and people attribute it to other causes, not knowing what caused the problem.
If you want to see some of these waveforms, check out "wiring for DCC", by Bill Gartner I believe, he has good info on this subject on his web site.
Small layouts will probably never see a problem, however, as you buss length approaches 30 feet and greater, you now increase your chances of having this problem.
As for frequency, I was referring to RF freq of wireless transmitters from NCE or Digitrax, sorry for the confusion, my fault.
Yes, a bus line is an antenna, but I am exceedingly skeptical that anything picked u pin the air is going to blow a decoder, and it is going to have minimal impact on the signal. The specs and requirements on the waveform are such that a little noise on the edges doesn't matter, and the signal is large enough that a little interference isn't going to change it. The rails are not twisted, and are just as much an antenna as the bus lines. The biggest source of interference is the little motor spinning around right next to the decoder. I understand the thinking behind the twisting, but I remain absolutely convinced (as an EE) that for this type of signal it is not an issue. The point of twisting isn't to prevent the wires from picking up the interference, you can't stop that. Twisting is done to try to get both wires to pick up the same interference, and to essentially cancel it out when it is received. That's way more important at high speeds and lower signal levels (say a network cable). But, if you do twist, it isn't going to hurt anything, so feel free.
Yes, this is urban legend. This has been discussed previously. There are a number of reasons why this doesn't wash. First, the signals on the track bus are a mirror image of each other provided by opposite polarity of the DCC pulse stream. The height of the pulses is generally 10 volts and higher depending the booster, load, distance, wire gauge etc.. So to induce a signal onto another wire at close range with enough amplitude to impact the DCC pulse stream already on the wire would take very large waveform that could not possibly be generated by a 10-12 v pulse on a wire. I am not going to post the math and show how much power would really be required but I assure you it is not trivial. Second, even if such an event were possible it would induce a signal of the same polarity as the source, which would be opposite of the wire it is being induced onto (this assumes we have wires running in parallel to each other). Thus, if this event happened, it would tend to cancel out the waveform on the receiving wire and not add to the voltage alreaedy there. This would be akin to a lower voltage condition. Lastly, inductance and antenna theory state that longer wires tend to favor lower frequencies, not higher frequencies (we'll avoid SWR discussions for now). So if this condition were possible then it would tend to occur on shorter lengths of wire and as the wire length gets longer the inductance of the wire would oppose the current induction. Now you could argue that the capacitance between the wires would increase and thus favor the higher frequencies as the wire gets longer. This is true but now the capacitance effect would couple the signal over (at a much reduced level) to the other wire but remember that the same thing would be happening in the opposite direction (since we have equal and opposite polarity signals) so against they would tend to cancel each other out. The more likely result of the increase wire length and wire to wire capacitance is a slight rounding of the leading edge of the pulses. Again, doubtful it would have any impact at the signal levels we are discussing and the relative short distances of wire.
As another EE I agree with you. I just took a peak at Alan Gartner's website and he agrees that twisting the wires on a long bus will actually reduce any self-induced voltage into the same wire. All of his talk about blown decoders and such is what he calls an unterminated bus where the booster output onto a single wire acts as a transmitter and induces an in phase voltage back onto the source wire and that since there is no load, you have an induced voltage across a very high impedance. But the reason there is a high impedance is because there is no load. As soon as there is a load, the induced voltage goes away because its source impedance is high and cannot source any amount of current. He recommends terminating a long bus with an RC load circuit. Sure, that will work but I'd recommend just breaking the layout into smaller blocks and not try to drive lengthy runs of wire on monolithic blocks. The one thing I didn't see Alan mention, that you do, is that the rails aren't twisted and they too carry the same signal as the track bus. I guess if someone is especially concerned, add the terminating circuit Alan recommends. It won't hurt anything and won't take away from the power or voltage delivered to the locomotives.
Engineer Jeff NS Nut Visit my layout at: http://www.thebinks.com/trains/
Wow...2..countem 2 double EE's. And I was only a lowly electronic tech for 17 years in broadcasting.
Now that I'm working in Education, I will say that I use a lot of CAT5E cable and of course its twisted because of the long runs. Twisting there makes sense. Not on a train layout with 12 to 14 gauge wire in much shorter runs.
Three.
Best practices would keep the bus lines as short as possible - ie if you have a 20x75 room, don't cluster all your boosters at one corner and tun hundreds fo feet of heavy bus wire. Distrubute the boosters and keep the bus runs as shoort as possible and there will be no issues with twisting or termination.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
rrinker wrote: Three. Best practices would keep the bus lines as short as possible - ie if you have a 20x75 room, don't cluster all your boosters at one corner and tun hundreds fo feet of heavy bus wire. Distrubute the boosters and keep the bus runs as shoort as possible and there will be no issues with twisting or termination. --Randy
Absolutely agree. And don't twist your rails <g>...
jbinkley60 wrote: rrinker wrote: Three. Best practices would keep the bus lines as short as possible - ie if you have a 20x75 room, don't cluster all your boosters at one corner and tun hundreds fo feet of heavy bus wire. Distrubute the boosters and keep the bus runs as shoort as possible and there will be no issues with twisting or termination. --Randy Absolutely agree. And don't twist your rails <g>...
Unless you are building a Wild Mouse ride and not a railroad.