Wire Sizes and Spacing
Proper wiring is one of the most important aspects of Digital Command Control.
Summary: Proper wiring will result in less problems later, such as poor operation, intermittent operation, or even runaway locomotives. Of even more importance, a robust wiring scheme for you layout can prevent damage caused by a short circuit, as poor wiring can have a direct impact on the operation of circuit protection devices in your booster or power management devices.
Wiring is one part of layout construction which is easier to do correctly the first time than to change or upgrade it later.
Some of the information presented may seem overly complex or unnecessary. The wiring practices described are based on Best Practices which have been demonstrated to work, both in a home layout and large modular layouts.
- 1 Wiring Issues
- 1.1 Selecting an Appropriate Wire Gauge
- 1.2 Can You Use Wire Which Is Too Heavy?
- 1.3 Proper Operation of Overcurrent Protection Devices
- 1.4 Harmonics
- 1.5 Recommended Wire Gauges by Scale
- 1.6 Metric Equivalents to American Wire Gauge
- 1.7 Power (Track) Bus Wiring
- 1.8 Terminating Bus Wires
- 1.9 Feeder Wires
- 2 See Also
Selecting an Appropriate Wire Gauge
Good or Cheap. Pick One.
Proper Wiring is Central to the Proper Operation of your Digital Command Control System.
But no topic seems to elicit more debate or misunderstandings than wiring. This article will attempt to explain some of the rationale behind DCC Wiring Practices. Most of the recommendations are based around technical reasons and Best Practices which have been found to work reliably.
- Power Bus wires must be the correct size (gauge).
- The amount of power available to locomotives will be reduced if inadequately gauged wires are used.
- Trains will run slower in areas which are poorly wired
- Consists or multiple trains in the same Power District will run at reduced speeds if the wiring is inadequate
- Wires which are too small can be a fire hazard when used with high current boosters (8 or more amps).
Can You Use Wire Which Is Too Heavy?
Experience has shown that since you can run more trains with more locomotives using Digital Command Control, you will.
This means your electrical loads will be higher for a given layout. In addition, a voltage drop of just 2 volts is a big issue with Digital Command Control, and you can't compensate by cranking the throttle open a little more. Keeping these factors in mind, it is clear you will need heavier gauge of wire. A small table top layout can reduce these sizes slightly without problems, but larger home or club layouts should adhere to these suggestions - you'll appreciate it in the long run. It's cheaper to do it right the first time than it is to tear it out and do it over!
The important issue with wire gauge is resistance. Heavier wire (a smaller gauge number) has less resistance than light gauge wire (with a larger gauge number).
The resistance causes energy loss when current flows through the wire. This is lost through heat. The result is a drop in voltage. This is expressed as I2R loss (I Squared R). When current flows through a resistor, a voltage drop is created, equal to the current multiplied by the resistance. The energy loss equals the current squared, times the resistance.
Basically, if a loop of wire has one ohm of resistance, and you pass 1 amp through it, you will see a loss of one volt. If you pass 5 amps, your loss is 5 volts. You will also convert 25 watts of energy to heat. If the resistance is doubled, that voltage loss increases to 10 volts, and the power loss increases to 50W.
For the most part, the resistance of copper wire is so small that it is expressed as ohms per 100, or 1000 feet. For a short run of a few feet, the resistance is negligible. It becomes an issue with a long run of wire. Remember, an equal length of wire is needed to complete the circuit, doubling the resistance of the circuit. High impedance caused by inadequate wiring can prevent proper operation of the booster's over current protection circuit, resulting in damage to track, rolling stock, and even your booster.
- Keep in mind that one of the reasons for using heavy gauge wire is to minimize the effects caused by the nature of the Digital Command Control signal. While the DC resistance (R) is quite low, the AC Impedance (Z) is not the same, and may be higher. In fact, a heavy AWG 12 wire may be electrically equivalent to a 20 AWG wire when used with a Digital Command Control signal. (Remember, electrical codes are written with 60 Hz (or 50 Hz) alternating current in mind. Also, thou shalt not exceed 80% of the rated current capacity of the circuit in actual use.) This will introduce new problems in terms of voltage drop.
Proper Operation of Overcurrent Protection Devices
- An important concept which is often ignored is the way circuit protectors work with many DCC systems. They are not sensitive to the amount of current. They react to the rate of change in current. A sudden spike will trigger them. This allows them to cut off current flow much sooner than a device that relies on a set value. Which means power will be cut off long before significant damage occurs from excessive current flow which would not trip a device acting on a set value.
- Poor wiring will interfere with their operation. Which could result in damage to your locomotive, or its decoder, because too much current was flowing, resulting in excessive heat. Or a short occurs, and something melts because the power was not interrupted. This happens because the wiring isn't heavy enough, interfering with the rate of change the booster sees. Or the voltage drop demands more current to maintain the same amount of power. A change from 1A to 1.5A in draw will also increase the power dissipation by more than twice.
- Good heavy wiring goes a long way to preventing problems like that. Using AWG 14 or heavier is not overkill for a power bus, and track feeders of AWG 18 are not that heavy. Choosing wire because it is cheaper or easier to hide is just asking for problems.
- Automotive tail lamps are not recommended as current limiting devices.
- Always test your wiring using the quarter test. The booster should cut out immediately. If not, there is a problem.
This is an important issue. Harmonics are multiples of the fundamental frequency. They cause distortion of the waveform. Harmonics Do Not contribute to the available power. This is mentioned in the NMRA S-9 Electrical Standard years before the NMRA Digital Command Control Standard was written, warning of possible damage to your motor by harmonic currents.
Why is this important with Digital Command Control?
The nature of the DCC waveform. A DCC signal consists of a square wave. Square waves are, by their nature, loaded with harmonics. Harmonics waste energy by not contributing to the power available to you. In extreme cases, half the power available will be wasted in the form of harmonic currents.
In DCC you probably will never see that happen, yet energy will still be lost in this manner. That is why the wiring is much heavier than that used for analog operations. Analog power packs will also waste energy when using pulse power modes, the difference being they supply one low current load, whereas your DCC booster and wiring supply multiple loads demanding a lot more current.
Recommended Wire Gauges by Scale
- Recommended Wire Gauges Shown in an Ampacity Table Are For Copper Wire Only, at 60 Hertz, AC, Sine wave. Keep that in mind when making comparisons.
- This chart is based on a 5A maximum current on the bus. For an 8A booster, move to the next larger gauge (smaller number).
According the Big Book of DCC by Digitrax, runs up to 100 feet are possible with 12AWG wire, and 50 feet with 14AWG. This is for one leg, in theory it is possible to have a 100 foot run (14AWG) with the booster in the middle (50 feet on each side). Don't skimp on wiring, plan for future power needs.
These guidelines are based on technical requirements and Best Practices.
Wire Size Table
|Wire Size Guidelines for DCC Power Bus|
|Wire sizes are American Wire Gauge (AWG)|
|Scale||Bus Length||Feeder Length|
|<21 ft||21 - 40||>40 ft||<5 ft||<10 ft|
|<6m||6 - 12m||>12m||<1.5M||<3m|
|H0||1:87.1||14||12 - 14||18 - 22||18 - 20|
|N||1:160||16||14||12||18 - 22||18 - 20|
|N-Trak||16||14||12||18 - 22||18 - 20|
|N-Trak is based on the N-Trak Wiring RP for DCC http://ntrak.org/ntrak_powerpole_rp.htm|
Metric Equivalents to American Wire Gauge
American Wire Gauge (AWG) with Metric Equivalents
|AWG||Diameter (Inches)||millimetres||millimetres squared||Nearest Size (mm^2)|
Power (Track) Bus Wiring
Bus wires carry the power from the booster to the track, but don't directly connect to the track. Feeder wires handle that task.
Below is some information to get the most of your system but using the correct wire types, gauge, and installation methods.
You can use either solid or stranded wire. Stranded wire is more flexible and will better handle repeated bending. (Repeated flexing will anneal the copper and make it brittle, which can break).
To Twist, or Not to Twist
This topic is fiercely debated on internet forums and DCC related mailing lists, endlessly and without any real solid conclusions. Seems everyone has an opinion, just like they do with wire sizes…
To Twist or Not to Twist, That is the Question…
There is always a large debate on the twisting of track bus wire. Keep in mind that DCC track buses are an Unbalanced Pair: One wire (A) is held to ground while the other (B) is energized, then they flip when A carries a signal and B is held to ground. This is happening constantly.
Twisting: Is It Necessary?
Insert Heated Debate Here
You should consider twisting your bus wires if they are going to be 30 feet or more. Twisting is more important on outdoor railroads where runs can sometimes be a 100 feet or more. If you compare two long and widely spaced wires (inches apart) with two long but closely spaced wires (1mm apart), the latter have a lot less inductance which is better. The amount of inductance you have in your wire directly relates to the degree of DCC waveform distortion and other problems such as large voltage spikes. Due to the harmonic content of the DCC signal, it has a lot more in common with alternating current than direct current.
Large voltage spikes are created during intermittent short circuits caused by derailments or other electrical track issues. Inductors store current in a magnetic field, and when the current flow decreases, it resists that change by inducing a voltage in the wire. That voltage, according to Lenz’s Law, will be positive when the current increases, or negative when it decreases.
Should a short occur, the current will suddenly increase and ‘’charge’’ the inductor. When the current is cut off by the overcurrent protection circuit, the magnetic field collapses and the inductor rapidly discharges into the circuit. This causes a voltage spike to appear. This is exactly what happens in your car’s ignition system: The ‘’coil’’ is in fact a transformer, and current flowing through the low voltage coil creates a magnetic field, which is storing energy. The ‘’points’’ or a switch then opens, the current stops, and the magnetic field collapses, which creates a huge voltage in the secondary coil. This voltage is thousands of times greater in magnitude than the 12V used in the primary circuit, which has enough potential to break across the gap in the spark plug. Televisions with a CRT used much the same trick to create the high voltages needed to energize the CRT.
Relays and solenoids do the same thing, which is why they have circuits added to prevent/reduce the damage the kick back can cause.
If your DCC system is causing electrical noise, this is one way to reduce it. Since many people don’t listen to AM Radio anymore, you might be radiating a large amount of RF energy without realizing it. Since TV is now in the digital domain, you won’t be yelled at for introducing noise into the picture either.
Interference could be interpreted by loco decoders and could cause havoc on the system. This interference can come from outside the layout, or be caused by the bus wires themselves or nearby signal buses. It also can reduce any interference the track bus may cause by inducing a signal in a low power signalling bus, such as your throttle network or occupancy detection system. If you do twist the wires and have a detection system, the section of bus being monitored should not be twisted, and the wires spaced apart, to reduce capacitance between them. Otherwise a small leakage current can cause issues by raising the noise floor..
When a pair of wires make a long run parallel to each other, the one wire induces currents in the other. This was discovered in the early days of telephony. In fact, Alexander Graham Bell discovered that by twisting the two wires together, interference and inductance were reduced, meaning the signal was stronger and clearer at the other end of the line.
Inductance and Impedance Reduction
One way to decrease the inductance is to keep the wires close. The closer the better. The most effective solution is to twist the wires, with about three to five twists per metre. Doing so alters the phase relationships between the two wires, reducing any induced currents in the wire. Twisting is very effective in reducing the inductance and the resulting impedance.
Keep this in mind when routing wires: You don't want to induce signals in another bus, such as the throttle or LCC wiring.
The negative side of the coin is that attaching feeder wires to the bus wire could get complicated if you are not consistent with your colour coding, however, the wires are not twisted a great deal so it shouldn't be too difficult.
Twisting the wires will also create a capacitor, with a value of 1 to 2 pF per inch. Doesn't sound like much, but that can be as much as 480pF over 20 feet. This creates a leakage path for the DCC signal, which will cause issues with current sensing block detection systems.
Power/Track Bus Impedance
Impedance Measurements on a loop made of two 12AWG wires, ~36 feet each for a total loop of ~72'. Average impedance of each wire is 0.8Ω, L = 12.6μH, resistance 0.09Ω. Measurements made at 10kHz.
|Bus Wires||Z (Impedance, Ohms)||Inductance (μH)|
|Parallel, >1 foot spacing||1.38||22|
Rail Resistance, Nickel Silver
The following table gives the impedance of various codes of rail. The impedance was found with 1A (at 60Hz) current flowing through the sample, using a comparator feeding a detector set at 50μV. By injecting a negative impedance, the impedance of the rail is found when the measuring system is brought into balance.
|Code of Rail||Impedance per metre, mΩ||Equivalent Wire Gauge||Strands/Gauge|
- The wire used for an equivalent is stranded. Since the measurements were made at 60Hz, impedance better reflects the results.
- Rail resistance measurement uncertainty is >50ppm. Actual resistance will vary by manufacturer due to alloy and profile.
Loss of control is often caused by excessive power bus impedance, resulting in motor current pulses being superimposed on the DCC waveform.
Some decoders have increased sensitivity to waveform distortion because:
- The manufacturer has designed them to reject out-of-specification waveforms.
- The default setting is to continue as it was in the absence of a valid waveform (many decoders simply shut down after a packet timeout period). This can make some brands of decoders less sensitive to dirty track resets, at the increased risk of loss of control when waveforms are badly distorted.
- Failure to remove the EU-mandated interference suppression inductors from the loco motherboard is another cause of loss-of-control at speed.
To isolate this issue:
- remove the vehicle from the track,
- place in a cradle or on a short section of isolated flex-track (restrain the vehicle if needed),
- power with jumpers directly from the command station's track output.
- disconnect the command station from the layout's power bus first!
This process will isolate whether the problem is layout wiring or vehicle wiring.
- "The track power bus wires should generally be parallel to each other. Slightly twisting the track power bus wires together will virtually eliminate radio interference, but this is not absolutely necessary. Avoid non-parallel wiring which might be tempting when running wires through and around various obstacles. This prevents unnecessary electronic emanations. The trains will not care, but reception of distant AM radio stations might experience some interference if track power bus wires are neither twisted nor parallel."
- DIGITAL COMMAND CONTROL; Stan Ames, Rutger Friberg, and Ed Loizeaux, Alt om Hobby AB, 1998, page 38, paragraph 4.1.1
Digitrax doesn’t require it, and suggests that proper selection of wire gauge and feeder lengths kept to a minimum are essential to reducing resistance and power loss.
Terminating Bus Wires
This is another DCC topic that gets a lot of ink (or maybe electrons moving) on a regular basis.
In general, bus wires should not need termination. However, you may find it beneficial on pre-installed long wire runs and/or in situations in which your experiencing control problems, such as decoders losing their programming or worse, a decoder blowing up. Refer to the your system manual to see what is recommended. The RC Network can absorb some of a voltage spike by giving it an alternate path, instead of your decoder’s front end.
Some manufacturers recommend the installation of a bus terminator, others do not. Digitrax doesn’t recommend them, while some command stations may have the equivalent built in. If the output of the booster is a high impedance, this usually minimizes the issue. A low impedance output driving the track bus with moving higher impedance loads on it will create a transmission line, and ringing is quite possible.
As a rule, they should only be installed if needed, and after proper investigation such as measuring the waveform with an oscilloscope.
Feeder wires are wires that connect the track to the bus. That is, every few feet, a set of wires run from the bus to the track. The goal is to make sure that there are no voltage drops and that the train has full power available to it. The benefits are that the train will not slow down. Also, this helps to ensure that the booster's short circuit protection will work.
For a trouble-free railroad, it's recommended that you follow these guidelines for feeder wire spacing.
|G (1:20.3-1:29)||Every 12-20 feet (4m-6m)|
|HO (1:87.1)||Every 3 to 6 feet|
|TT (1:120)||Depending on size of the layout:|
up to 250ft of mainline: every 4ft
250ft-450ft of mainline:every 3 ft
more than 450ft of mainline: every piece of track, for short pieces (up to 5") keep on connected to a bigger piece
|N (1:160)||Every separate piece of track should have its own feeder.|
Track pieces over 18" should have a feeder near each end.
Never rely on rail joiners for electrical connections!
|Z (1:220)||Every separate piece of track should have its own feeder.|
Never rely on rail joiners for electrical connections!
Don't Place Feeders at the End of a Short Section
If you have a very short block or track section, and will only have one set of feeders, place it in the middle instead of at either end. Don't worry if you can't get it exactly in the middle. There is the ideal and then there is the practical: aim for the ideal, but keep the practical in sight.
Feeders can be installed in a variety of ways. Marrettes, splices or IDCs (Insulation Displacement Connectors). When choosing a mechanical method of joining the feeder to the track bus, make sure it can handle the difference in gauge. IDCs are made for joining two wires, and they are available for different gauges.
IDCs are also known as ScotchLoks (manufactured/invented by 3M) or sometimes called 'suitcase connectors'. The 3M devices are usually superior in design and construction than the generic knock-offs.
3M makes a T-Tap, an IDC you crimp onto the wire, which has a tab which will accept an appropriately sized crimp-on female connector. These can be useful to make connections for feeders as well, with the advantage of being able to be disconnected.
- Wiring - Primary wiring article.