A: The goal of this committee is to set a standard for the communication protocol that is used to transmit information from the command station to the receiver (also known as a digital decoder) that is contained within a locomotive such that multiple manufactures can build competing systems that interoperate with each other.
Q: What is covered by the proposed Standards and Recommended Practices?
A: There are many components to a digital command control system (see figure at rear of FAQ). The proposed standards only cover the signal sent over the rails, tolerances allowed in its generation, tolerances which must be accepted by the decoders, and a basic communications protocol used to convey information to digital decoders. Recommended practices have been drafted that cover more advanced decoder features, basic operating characteristics, and an internal locomotive plug. Future work may include recommendations for other communication busses in the system.
Q: What is not covered by the proposed Standards and Recommended Practices?
A: Neither the design and implementa-tion of system components or the opera-tor interfaces are addressed by the proposed standards and RPs. The standard does not require that components be built in any particular fashion, nor does it require that any particular circuit or technology be used. Each manufacturer is free to develop a system that conforms to the style of operation desired. Only the basic communications protocols are covered to ensure interoperability.
Q: Can one standard cover both simple and complex layouts?
A: Yes. One standard can. It is unlikely that one command station will meet the needs of both simple and complex layouts, but both low end and high end command stations can conform to the same standard.
Q: Why use digital communication instead of analog?
A: Digital communication is much easier to enhance and extend than analog. A digital system encodes information to be sent from the command station to the digital decoder (receiver) as numbers, grouped into packets. New packet types can be easily added to send more information. In short, a digital system is almost infinitely extendible, without causing backward compatibility problems. An analog system is inherently more restrictive.
Q: Aren't all command control systems digital? They all seem to have lots of little chips inside.
A: Strictly speaking, most existing command control system are not digital. The addressing techniques for the "channels" varies from the use of audio frequency tones to the use of "multi-plexing". These techniques result in the comparatively small number of channels that have been available so far. In a digital system, both the address of the locomotive and the speed value are sent as digits (that is as numbers) over the track.
Q: If the system is digital will it be able to pass stringent FCC requirements?
A: Yes. Part 15 of the FCC rules regulates the RF interference potential of digital devices such as those used in digital model railroad equipment. Manufacturers who produce turnkey systems will be required to verify that their equipment adheres to these rules. Experience to date indicates that this is not a significant problem.
Q: Why is the standard based on the Lenz system instead of one of the other existing systems?
A: In drafting the standard, we looked at various alternatives including existing approaches and completely new designs. After careful study, the committee found that the signaling techniques used by a system designed by Lenz Elektronik, a German electronics firm, far exceeded our list of requirements and employed a communication protocol that offered the greatest future growth potential. While we based the signaling techniques on the Lenz design, we made numerous improvements and created a packet format that is much richer in features provided. See the questions in the operation section for more details.
Q: Why is the NMRA considering a standard based upon proprietary equipment?
A: We are not. The NMRA requires that adherence to standards does not require you to gain a license from a vendor. Further no copyrighted, patented or proprietary information can be in a standard unless it has been released to the entire industry. We have a letter to the NMRA from Lenz confirming that there is nothing in the draft Standards and Recommended Practices that infringe upon any Lenz copyrighted, patented or proprietary information. One should also note that we have made significant extensions to the original Lenz system. In short, the standard is not based on proprietary equipment. It has its roots in the Lenz system, which has been made public.
Q: What components in the standard are based upon custom or proprietary equipment?
A: It is not necessary to use any custom or proprietary components to build a system that complies with the standard. Some manufacturers may develop custom parts to allow them to build smaller digital decoders.
Q: Won't a command control standard stifle technological advances? As an example, what would have happened if the NMRA made standard GE's ASTRAC system 15 years ago.
A: It is important to note that the standard only standardizes the protocol on the track. This protocol is a very flexible one - it is easy to add new packet formats to meet new situations. The standard does not require any particular hardware, so the command stations and digital decoders take advantage of new technology as it becomes available. In contrast, GE's ASTRAC system was not flexible enough to base a standard on. Good communications protocols (and we believe that the standard is a good protocol) tend to remain constant for a long time. For example, the RS- 232 standard for communicating between pieces of computer equipment has been around for a very long time, and is still useful even though computers have advanced far beyond the point where they were when the standard was developed.
There is a definite cost to not having a standard. You have a limited choice of equipment, and the cost will stay higher since the market for any given system is not as large as if there was a standard.
Q: In reading the list of manufacturers and committee members I fail to see several prominent command control manufacturers. Why?
A: We have contacted all the manufacturers that we are aware of. Some of the manufacturers believe that they have too much invested in their existing systems and choose not to participate. The standards process has been open to anyone who desires to participate. Several existing command control manufacturers are working to produce products that conform to the standard as well as to continue to support their existing system and product base.
Q: Why not just standardize on an internal locomotive plug that allows any command control system to be installed?
A: The committee is working on an RP for just such a concept that should be of benefit to any command control manufacturer. A standard plug, while very needed, does not answer the issues of basic interchange of equipment, S9 conformance, nor is applicable for all scales (it will likely be hard to fit the plug into Z scale). Standardizing the signal on the track will allow you to choose a command station that best suits your needs without regard for whether the company that makes it also makes good digital decoders. You will also be able to choose from a much wider range of digital decoders than a single manufacturer could make.
Q: The standard seems to benefit Marklin.
A: Up until recently Marklin marketed two digital systems: Digital= and Digital~. The Marklin Digital~ system was initally developed exclusively for 3-rail equipment. It is not compatible with the proposed NMRA standard. The Marklin Digital= system was developed jointly by Lenz and Marklin for control of HO gauge 2-rail equipment. The draft standard uses many of Lenz's contributions to Marklin's Digital= system. The vast majority of Marklin sales to date have been in support of their 3-rail market . Marklin has recently upgraded the Digital~ system to work with either 3-rail or 2-rail equipment, and has decided to concentrate its future development on furthering the Digital~ system (with its large installed user base). Marklin has announced that it will no longer manufacturer the Digital= system.
Q: Does this mean that the standard only applies to 2-rail DC?
A: No. The standard does not care how many rails you use. It is also possible to develop digital decoders that work for AC motors as well as DC motors.
Q: I am a current Command Control user. Won't the existence of a standard devalue my investment and put the existing manufacturers out of business?
A: The proposed standard describes a communication protocol for digital communications. It does not apply to analog command control systems nor will all digital systems comply. Existing systems will continue to function as they do today. There is no requirement that you use or transition to digital systems that conform to the NMRA standards. All the existing manufacturers we have talked to are committed to continue to support their existing installed base and we support this commitment. In time, as existing manufacturers upgrade their products, they will hopefully develop components that can work with either their existing systems or with the proposed standard digital approach. The NMRA does not require this transition nor will it do anything to inhibit manufacturers from introducing products that do not conform to its standards.
Q: What is the rationale behind putting some things in the standard and some things in a recommended practice?
A: By placing the electrical information and the low end packet format in the standard and other packet formats, including high end formats, in a set of RPs and data sheets we address the entire market in a fashion compatible with the philosophy of the organization that chartered us. As an existing example consider wheels, clearly an important interchange issue. The standard only describes the minimum dimensions but references RP25, which gives the very important curvature specs. NMRA definitions:
NMRA STANDARD: "As charged by the NMRA CHARTER and CONSTITUTION, NMRA STANDARDS provide the primary basis upon which interchange between equipment and various North American Scale model railroads is founded. Under this requirement NMRA Standards include only those factors that are considered vital to such interchange. For less critical matters see NMRA RECOMMENDED PRACTICES."
NMRA RECOMMENDED PRACTICES: "The RECOMMENDED PRACTICE category was established by the NMRA Board of Trustees in January 1957, to: a. Promote maximum interchange between and within units. b. Specify the details of major components to improve design and function. RECOMMENDED PRACTICES are only less mandatory than STANDARDS by virtue of their slightly less critical subject matter and/or the fact that deviation for specific reasons is permissible."
DATA SHEETS: More specific Information on such topics as Scenery, Trackwork, Operation, Electrical, Motive Power, Rolling Stock, Structures, and other general items of interest.
Q: Can I build it myself?
A: Yes. In additional to commercial kits, the working group is developing simple kits both for the command station and for various forms of digital decoders. Plans for constructing these kits are being distributed for the cost of mailing and duplication.
Q: What does the locomotive over bar graph design Logo mean?
A: In order to help users identify manufacturers who are attempting to develop compatible equipment that conforms to the draft documents we have constructed a logo of a steam locomotive on top of the command control signal. The use of the logo is voluntary and does not guarantee compatibility as the testing program for conformance has only just begun.
Q: Is there a European standard being developed?
A: The Europeans are actively working on a locomotive plug standard. We will attempt to keep our plug Recommended Practice compatible with the emerging European standard. We are not aware of any other group working on a standard for the digital communication protocol.
Q: Where can I get a copy of the standard?
A: The standard and related items, such as this FAQ, are available
electronically on Compuserve, and by anonymous FTP from ftp.mcs.net
in mcsnet.users/dsdawdy/Cyberspace/NMRA/ directory. They are also available by US mail from any working group member.
Q: What is required in order to Join the NMRA Command Control Working Group?
A: Anyone who is willing to spend time in assisting us develop standards/ recommended practices for compatible digital command control or participate in the testing of products as they become available is welcome to become a member of the working group. Requests should be sent to Michael Greene, NMRA Command Control Working Group, 16 Falmouth Drive, North Grafton MA, 01536. E-mail: greene@edisto.lkg.dec.com
Return to the Table of Contents
A: No. We have tested this on some really large layouts that have such a nightmare of wiring that no command control system in the past has been able to operate. The NMRA tentative standard system did fine.
Q: How are reverse loops and Y's wired?
A: Reversing loops can be done in the classical way for command control. The polarity is correct when the train enters the loop. While in the loop the polarity of the rails is reversed. Note that unlike conventional systems, it is possible to reverse the polarity of the reverse loop itself while the train is running in it - the locomotive will keep moving in the same direction. There is another technique that utilizes the ability of the digital decoder to run on traditional power. Use a separate analog power supply to power the loop. When you enter the loop it is a normal digital to analog transition and when you exit the loop it is a normal analog to digital conversion. When transitioning between digital and analog power a digital circuit breaker (see following question) or a choke should be used on the analog power supply to prevent the power supplies from combining.
Q: What is a digital circuit breaker?
A: Should you want to combine analog and digital operations on the same layout, a device needs to be placed on the analog side to prevent the voltages from combining when you go over a gap. This can be accomplished by using a large choke or by a digital circuit breaker. A digital circuit breaker must be used if the "0" bits are elongated (the DC component of the signal is non-zero). Arnold has developed a digital circuit breaker (they call it a Trenmodule) for the N market. As such its power capacity is rather small. Circuit diagrams are available and it should be relatively easy to build a 4-6 amp version.
Q: How do you do signaling with digital command control?
A: Signaling still depends on block-occupancy detection. You can do this with optical or magnetic sensors, or with the more traditional current sensors. In the latter case, you still need separate electrical blocks for determining if a particular stretch of track is occupied, even though you do not them for the purpose of running multiple trains.
Return to the Table of Contents
A: A packet based system is very flexible and easy to extend in the future. Basically, each packet corresponds to a command for a particular digital decoder (receiver). Each packet contains the address of the digital decoder it is intended for, and some data that tells the digital decoder what to do - for example change to speed 10. It is flexible because you can easily add new packet formats if needed, even ones which can address more digital decoders that the baseline packet format. These new formats will be published as recommended practices. There is currently a recommended practice being drafted for more advanced operation that can be supported with the baseline packet format. The extended packet format RP also defines packets that allow you to control auxiliary devices like switch machines, etc.
Q: What is meant by a bi-polar signal?
A: Each bit is divided into two halves, which are "mirrored" around zero volts:
Note that the bottom half of the signal (below zero volts) is a mirror image of the top half, but shifted over half a bit width.
Q: Why use an bi-polar signal instead of DC with a superimposed signal?
A: There are a number of advantages to using a bi-polar signal: - The signal is the power, instead of just being superimposed on the power. This means that if you can get power to a locomotive, you can control it. Wiring loops and resistive loss, which can cause problems in carrier systems, do not cause problems with this technique.
Q: The standard implies that the track voltage can range from +/- 7 to +/- 20 volts. Is this how speed and direction are controlled?
A: No. During operation, the voltage provided by the command station does not vary. The digital decoder within the locomotive controls speed and direction by varying the amount of power provided to the motor. The voltage range of the signal described in the standard is to allow for the different power needs of the various scales. For example, a typical Z scale command station will place +/- 10 volts on the rails, and a typical HO scale command station will place +/- 14 volts on the rails and a typical G gauge command station will place 20 volts on the rails. If you were controlling live steam locomotives, you would probably use the minimum +/- 7 volt signal. The +/- part is because the signal is mirrored around zero volts.
Q: If a packet is lost due to dirty track, what happens?
A: Command stations retransmit packets to mobile digital decoders as often as possible, so if one packet is lost, another one with the same information will arrive soon after (how soon depends on system load). If power to the locomotive is interrupted for long enough, the digital decoder goes into reset mode - and stops the locomotive. When power is restored it will not move until it gets a new packet. See the section on fail safe operating characteristics in the standard.
Q: Doesn't retransmitting packets all the time use up too much bandwidth causing slow response to throttle changes, etc.?
A: Although simple low end digital command stations that only provide for control of a limited number of engines simply cycle through all the addresses in use repeatedly, higher end systems may use a priority based scheme so that packets that contain values that have changed are sent before packets that are repeats. Thus repeating packets does not necessarily impact the effective bandwidth of the system.
Q: How many locomotives can one control using a compliant system?
A: The basic standard calls for independent control of 99 locomotives. RPs increase this up to 32,000 locomotives.
Q: The standard baseline packet only allows for 14 speeds. Is this enough?
A: Most people's first reaction is that you need much more than 14 speeds. However, most people who have actually used the system say that 14 speeds is OK. This is partly due to the fact that you can adjust the start voltage so that the engine starts to move on the first speed step by setting the appropriate register value. The extended packet format RP provides for 28 speeds, and there is a proposal for a command that allows 128 throttle settings.
Q: Can you control things other than locomotive speed and direction?
A: Yes. The baseline packet in the standard only provides for basic locomotive control since that is all that is essential for interoperation. The extended packet format RP defines packets that can be used to control 32,000 different accessories such as lights, sound, turnouts, etc.
Q: How does the standard work with computer control of the layout?
A: Computer control can be accomplished by using a command station with a computer interface, or by generating the signal directly with the computer and using a power station to drive the track. The interface between a computer and a command station is outside the scope of the standard.
Q: Can the digital decoder send information back to the command station?
A: In service mode the digital decoder can acknowledge a packet with a brief pulse of current. This can be used to find out the digital decoder's address, for example - send a packet asking the digital decoder if it is address 1 and wait for a response. If no response, try address 2, etc. There is currently no defined way to send information back to the command station in normal running mode, although some work is being done in this area to determine what is feasible. See the extended packet format RP for a description of feedback in service mode.
Return to the Table of Contents
A: This is an optional feature of digital decoders provided for by the standard. If a digital decoder supports analog operation, and does not see a digital signal on the track, it reverts to analog operation and can be controlled as if no digital decoder were present. While this works with most power packs, decoders can be confused by some forms of very narrow pulsed power and extreme care must be taken to ensure that the packs maximum output is less than the decoders 24 volt maximum.
Q: Can you run an un-modified locomotive together with a locomotive equipped with a digital decoder from a digital command station at the same time?
A: The standard allows for this, although some care needs to be exercised in its use. We refer to this type of operation as analog compatibility mode.
The signal is symmetric around the 0 volt level which provides a 0 volt DC component. By expanding the length of the zero bits on the positive side of the signal a positive DC component can be added. Likewise by lengthening the bits on the negative side of the signal a negative DC component can be added. Only the 0 bits can be lengthened in this manner.
The result is a non-zero average DC voltage which will run an unmodified locomotive. However, since the complete signal gets to the motor, the stretched zero side causes the motor to turn, and the following instant the unstretched side (the opposite polarity) tries to reverse the motor. The longer (stretched) side wins, but motors do run more noisily and generate more heat. Some high precision can motors (the ironless core type) may be permanently damaged. (see following question)
Q: How is a locomotive that does not have a command control digital decoder installed affected by exposure to the command control signal?
A: It is not recommended that you leave locomotives not equipped with a digital decoder on track powered by a digital command control signal for extended periods of time. Since the signal has an average DC level of zero (unless you are using analog compatibility mode - see the previous question), the locomotive will not move. However, the motors will make some noise. If the amplitude of the digital signal is greater than the maximum stall rating of the motor, the motor may experience permanent damage. We have found that common open frame motors, and many common can motors are not damaged by the digital signal even after extended exposure. Ironless core type motors can be damaged by extended exposure to a digital command control signals that are greater than 12 volts due to excessive heat buildup.
Q: What affect does analog compatibility mode (zero stretching) have on normal digital operation?
A: None, unless you use it. When you are running an analog engine, the zero bits are stretched, which reduces the bandwidth of the system.
Q: Why allow zero stretching at all?
A: Analog compatibility mode is only one use for zero stretching. Other possible uses include the following:
There are probably other uses that will become apparent as time goes on. Note that the above mentioned uses are still very much in the experimental stages. The point is that preserving the ability to stretch zeros allows for the possibility of some interesting things.
Return to the Table of Contents