Why not

Why not, Michael. Your area is almost exactly the same size as mine and I've had no trouble with my bus wires or feeders running parallel.

Roy

Why / why not

Why?

Others can explain better than I, but the short version is that you're running AC on those wires, and two parallel conductors will form a rudimentary capacitor.  Capacitors are pretty good at blocking AC (imaging replacing the main rods on a steam engine with giant shock absorbers).  For some reason that I do not fully understand, twisting the wires reduces this capacitance significantly.  Thus, for a long run, it makes a big difference, especially when we're only talking about a few volts to begin with.  You could get the same result by physically separating the bus wires as much as possible, but most people find it more convenient to twist the wires together.  The greater the separation, the less the capacitance.

For what it's worth, it's the same reason that telephone and network cables have twisted pairs inside.

How?

I cheated.  I bought a long extension cord at the local hardware store.  The cord cost a good bit less than the individual wires would have.  Then I stripped off the insulation and found three wires twisted together - black, white, and green.  I used the black & white wires for my bus, leaving the green for a spare and if I ever need a third wire for some reason.  Then I connected my feeders with suitcase connectors.  Instant low cost low labor twisted pair!

Given the small size of your layout, I would not be overly worried about it.  I seem to recall reading somewhere that it only matters at greater than 20'-30' for a given run, and maybe more than that.  I'm sorry to be vague here, but it's been quite a while since I did any research.  Like I said, I twisted mine, but I bought them pre-twisted.  I really don't think you'll have a problem, especially if you keep them a few inches apart.

Capaacitance and interferance

Capacitance increases as the wires get longer with the twist.  Capacitance also increases as the conductors get closer together as when they are twisted together.  Parallel conductors normally have more space between them reducing the capacitance-see the formula for capacitance, specifically plate area and plate separation.

Twisting reduces the possibility of interference as the conductors act as as an antenna.  One way is the inductance created reduces the amount of energy in the antenna.  The phase also shifts and cancels with the twist.  The amount of interference created by the DCC signals is also reduced by the twisting.

I found that I did not need to twist the leads for my 20 x 20 layout.  I had wired it for block detection so I had lots of paired feeders just about like DC block wiring which gave me current flowing only to areas that had locomotives running in them. 

I did not take any precautions about keeping them separated nor preventing them from twisting.

Twist where it is easy, but don't worry to much about it in the area where track drops are connected and if you miss a spot, don't worry about it.

Terry

Capacitance and AC

  To Ken,

  Actually capacitors PASS A.C. and BLOCK D.C. - just an observation...

Opps

Thanks for the clarification, guys.  That's what I get for not doing my homework.

Tony, I always thought a capacitor would block AC.  I've seen circuits which superimposed sound on the rails (before the days of DCC), and they always included a capacitor across the motor terminals to prevent the AC sound signal from getting to the DC motor.

Do the Twist..

The idea of twisting as already stated by Terry is to block interference in the wires. Long runs of wire can act as an antenna for radio signals in the air, and also just plain old lines of magnetic force radiating from transformers, picture tubes, etc. So a wire 20 feet long can be quite a collector of noise on the lines, and these lines are carrying the DCC signal, so noise and hum on the lines is a bad thing. By twisting, it creates a natural filter for hum and noise on the lines. Lets say you twist your lines so a total twist is about 5 inches long, so only 5 inches of wire being cut across by lines of magnetic force, or picking up radio noise is as bad as it will get, not the whole 20 foot, or longer, wire. Because, as you look at the first 5 inches, you have positive on the left and negative on the right, creating current flow of noise on the line. Then the next 5 inches you have negative on the left and positive on the right, reversing the current, and causing the noise to cancel out! A very neat and simple way to combat noise on the lines.

With that said, in practice you can get away without twisting your wires as long as your runs aren't that long. Remember that DCC  is carried on a square wave form from about 12 to 16 volts. So noise and hum that amounts to a few millivolts will not be enough to actually cause locomotive decoders to not be able to read the signal. In fact, it would take quite a bit of noise to interfere with a 12 volt signal, much higher than computer signals on a standard Ethernet network. So unless you've got runs of wire that exceed 50 feet or so, probably even a lot longer, you will not suffer signal interference bad enough to make any difference to warrant twisting your wires.

Personally, I just twisted my wire to keep it neat and easy to manage. If you don't twist it too tight, then it's easy enough to pull them apart far enough to make your connections. A half twist every 5 to 6 inches is probably more than necessary, and very easy to part, you could have one twist per foot even and still gain the benefits of filtering interference. But I've seen some very nice and neat shelf layouts that have one conductor on the rear of the shelf and the other on the front, and I've seen some that runt them about 2 inches apart, hanging on hooks, and both operate very well.

So IMHO it's just about theory, what is best theoretically, and not what actually makes any difference in practice.  I doubt anyone could even measure the difference without super sensitive equipment.

Everything you need to know

Everything you need to know about why you should twist. Even the smaller layouts can benefit  from it.

http://www.wiringfordcc.com/track.htm

Scroll down the right side menu to twisted pair.

      Pete

Question

Since my rails carry basically the same voltage and info as my buss, should I twist the track, too? DJ.

Yep....

That's always been my thought about it too. I saw the suggestion a few years ago and the consensus then seemed to be it didn't really matter but now it seems to be looked upon with great favor. I can still find an article on the Digitrax site stating it doesn't matter.

The funny thing is, I have my amateur radio gear right here in the same room as the layout and I don't recall hearing any noise from the DCC system in my receiver. Probably was down below the noise floor. Of course, I'm between DCC systems right now so I can't check if my recollection is correct.

Michael

AH! Now I see the confusion...

  A capacitor across the leads of a motor WILL keep the AC out of the motor, but it does not BLOCK the AC.  It appears as an OPEN to the DC used to drive the motor and looks like a short to the AC on the line.  I thought you were referring to a capacitor IN-LINE with a load...

This was a joke...

Twisted Track

Google Images showed me that Poland already is implementing this idea:

Large layout and no twist.

30x32 triple deck HO layout and did not twist. No issues found, no interference found either.

Ive read about this for years and unless you have interference from the DCC signal with home electronics, I cant see it effecting anything. It would be possible for a resonant length to come into play but at the frequencies used I would say its very slim.

Im sure someone somewhere has had a problem and twisted and fixed something, but as a tech guy, I dont think its a issue worth the worry.

Just my 2 cents with a tech background. Electronics guy, radio guy and a recovering broadcast engineer.

Twisted track refererence

Yep, It was joke  but the  question it presents is not and it's one I've never quite found a suitable answer to. Why go to the trouble to twist your track buss wires when the rails on the track itself are running parallel to each other about 3/4 of an inch apart. Your track is a huge antenna and any interference cancelling you might do by twisting the buss is defeated by the track "radiating" a signal. At least in my mind it is. All that being said, I just twisted em  up with a drill and I'm going to see how easy it is to connect feeders to them. I still think it might be "much ado about nothing" though.

Michael

Partially a joke

Michael, that was my point. I have a couple hundred feet of 12 ga. wire with not twists, and no problems. Digi-Trax Super Chief Radio. DJ.

Why All the Resistance?

I've read, it seems, dozens of these discussions since DCC became popular.  What amazes me are the number of folks who are just adamant about not twisting their buss wires.  I mean, some folks get so up-tight about it.

No one said it was a law that it must be done.  It is simply a recommendation that you do.  It is true that you might have a layout the size of WalMart with plain old wire runs side-by-side with nary a trouble.  And that's cool.

Others might have a relatively small layout and have problems that a twisted buss might help.

The folks that promote twisting the buss wires are merely listing the potential advantages and benefits of doing so.  If you don't want to, THAT IS OKAY.  It's your layout.

Can we give it a rest now?

-Jack

Interesting posts about DCC

IF the idea is valid and it solves problems then thats when it needs to be broadcast as a information.

DCC has had its share of jokes and misguided ideas too. One post years ago was from a guy claiming to be giving up the entire hobby because he wired his layout only to find out the DCC signal would not go thru a green wire. Funny how smart DCC is with being able to control a layout but I didnt know DCC was smart enough to discover the color of wire insulation?

Others have claimed it wont work with solid wire, stranded wire, wire smaller than 24 gauge, wire larger than 12 gauge, and dont even get me started on the ENDLESS comments about what wire gauge to use for the buss lines, crossed or not.

The KISS idea works very good with model railroading, even better with DCC.

I'm just trying to understand why it's supposed to work

That's all.  I get all the arguments about twisted pair and noise cancelling but it still seems to me that all that is immediately mitigated when you start adding feeders that are connected to two parallel conductors that very much electrically and physically emulate an untwisted buss. I'm not trying to resist anything, I've twisted mine now, just for the heck of it. I'm just looking for some logic to explain the why....

Michael

Allow me to put my electronics teacher hat on....

You hear a lot of strange things about wire and cable. (If you want to experience the deep end, just look at material put out by the audiophile cable companies.) Most of them are true for some kinds of signal in some situations but inconsequential for other situations. For instance, if you are working with old-style video or cable TV, you have to make sure the length of cable between connectors does not set up a harmonic resonance (ghost) given the cable type and signal frequency. This is seldom a problem with audio signals because the wavelengths involved are huge compared with the amount of wire in the typical living room. However, old-timers will remember echoes on long distance phone calls.

Twisting wires reduces interference, both broadcast and received. Twisting is absolutely required for small signal applications such as microphones, long distance telephony and multi-pair Ethernet cable. Otherwise you will pick up power line hum, radio stations and even astronomical sources. Twisting is a good idea if you are running high voltage, high frequency signals such as in a TV flyback circuit. This will reduce the amount of broadcast interference.

Twisting wires together does increase the capacitance between them. When you buy commercial cable, the capacitance per foot is specified and you can decide if it will work with the associated circuit. The key here is the amount of current we are moving around. If you are dealing with tiny amounts of current (as for instance with an electric guitar) the amount of current that can leak through the capacitance is significant, especially as there is more leakage as the frequency goes up. This removes the sparkle from a guitar and will make Ethernet lose data. If we are talking about loudspeaker wiring where the currents are much higher, capacitance is not significant until it reaches the point where the amplifier goes unstable. Since DCC moves enough current to power a locomotive, loss of high frequency (which messes up the edges of those nice square waves shown in the books) is not noticeable. In fact, since the decoders react to the width of the dancing waveform, the actual shape has no effect at all.

The DCC spec says "The exact shape of the NMRA digital signal shall be designed to minimize electromagnetic radiation such that a large layout operated using this standard can meet applicable United States Federal Communications Commission interference requirements." Running afoul of the FCC is such a PITA that the manufacturers (who will get the blame) may be specifying twisted buss wires to cover every possibility of  interference. Twisting the buss wires pretty much eliminates them as a source of interference.

It is true that the rails themselves are an antenna, but they are not very efficient at the DCC frequency. No radios are tuned to that frequency, so interference picked up will be on the internal wiring. My C Crane radio will pick up the DCC signal if it is within a meter of the track. (The buss wires on my test setup are twisted, shielded and about eighteen inches long.) A well shielded condenser microphone picks up nothing, but the simple mics on some video cameras may. The radio picks up interference at a similar distance from my computer, printer, and TV set. These all have FCC approval, so a DCC layout is probably in compliance. The instructions for the mentioned devices all say in case of interference, move your radio.

Interestingly, holding two wires parallel is also a fairly effective way of reducing interference (remember TV twin-lead?), and track certainly does that. However, my guess (I haven't done the math) is that it would work better for Z gauge than for HO.

pqe

The biggest offender ....

...within my amateur radio station for interference was the CRT monitor of my PC. Now having gone to an LCD, not so much anymore, it's pretty quiet. As I said earlier I don't recall hearing any interference and I have a pretty sensitive receiver in my Flex SDR . Doesn't mean it's not there, I just don't recall having heard it. I need to do some research and see if a DCC system is an FCC Part 15 device. Would be interesting to find out. I'm saving my pennies for a Digitrax Zephyr Extra so I'll experiment once I get one again for my own curiosity's sake.

FWIW, twisting the buss does make it much more difficult to strip it and solder feeders to it. Maybe it's not a problem with IDC's but I don't trust anything but soldered connections.

Michael

If you have ever operated on

If you have ever operated on a layout where something unusual happens like a loco not responding or horn blowing when not pressing F2 or other oddity? Ever have a key fob in the proximity of your layout and see lights blink on a loco? Ever thought that the track was dirty because you just rear ended the train in front of you but found that it was not that dirty and everything works now that your spouse is not on the cordless phone?

 The reason that twisting the buss lines works relates to how the DCC signal is transmitted through the wiring. Think of it as square AC sine wave. The ups and downs and the width of the ups and downs of the waves is what transmits the commands to the decoders. The wave goes up and down very fast (milliseconds) through the round copper wire. The nickel silver rail is not round and does not act so much like an antenna as does a round copper wire. When a broadcasting device like a key fob or cordless phone (Cell phones use a different broadcast system), or even some remote controls are in the proximity of the DCC buss lines they have a tendency of creating spikes in those waves. These spikes in the square sine wave is what confuses the decoders. The twisting of the copper buss tends to cancel out or diminish the spikes by having the ups and downs in so close of a proximity (inductance). The downs will spike the same time as the ups and the decoders will not see such a tall spike because the twisting buss will cancel or diminish the spike. Allan Gartners wiring for DCC and Marcus Amon's DCC for dummies explains and shows oscilloscope readings of the spikes. The recommendation is three twists per foot but even one per foot is very helpful. Keep your feeder wires as short as possible. I use 6 inch long 22 gage feeders soldered to 14 gage runners to the 12 gage buss. I also twist the 14g runners. Even with 3 twists per foot they are still easy enough to separate to solder and insulate. 

            Pete

Some Nice Explanations

The last couple posts have some nice explanations of things that could potentially happen without twisting buss wires.  Notice, I said potentially.  You may have problems, you may not.  You won't know until you build the thing.  Then it becomes much more difficult and inconvenient to implement corrections if you do have problems.

Michael, my earlier rant wasn't really directed at you . . . I realize that you're just asking for the "why's".  Rather I was referring to those who always feel compelled to defend their position on the matter, with the attitude that those that do it differently are either wasting their time and effort, or they're going to interfere with worldwide communications (depending on their viewpoint).

I personally feel that adding the twist probably isn't that big of a deal, so I plan to add them in my new layout.  In fact, I've purchased my buss wire and will be twisting and hanging it over the next couple weeks.  Do I need to twist?  Dunno, but I figure it won't hurt things.  But frankly, I couldn't care less if somebody over in Flatbush doesn't twist his wires.

-Jack

Hmnnn....

Maybe, maybe not. Most anything metal will radiate if it's impedance is matched well to the transmitting source. I've seen a guy hook up a light bulb to his transceiver and have a conversation with another guy  a couple of hundred miles away. He did it just to show it could be done. I can't speak about DCC signals  but generally  RF could care less what shape the radiator is. It doesn't run through it, it runs on the outside of it. I won't argue it anymore though, I twisted mine just for the heck of it but I'm still not convinced it has any real effect.

Michael