How DCC Works

	Home | Products | Distributors | Documentation & Support | DCC | CV Calculators | Links
What is DCC? Getting Started with DCC A Starter System Mobile Decoders Command Buses How DCC Systems Work Making Your Own DCC Components Decoder Installation	This page provides technical information about how DCC systems work. It is not necessary to understand this to use DCC - it is provided for information to those that are interested only! DCC Packets on the Rails DCC is a digital control system. It operates by encoding messages as a sequence of 1 and 0 bits onto the two rails. remote decoders can then decode those messages and decide what action to take. Bit Encoding DCC packets are encoded by modulating the two rails A and B to one of two voltages: VL & VH. VL is nominally 0v and VH is usually between 12v & 14v. The rails are modulated such that if rail A is at VH, rail B is at VL and so on. 1 and 0 bits are encoded by varying the timing of the transitons. This ensures that: The difference between the two rail voltages is always constant (ignoring the changeover periods); The 1 and 0 encoding does not require particular voltages to be measured, making it very immune to noise; The average DC voltage between the two rails is zero, because each rail spends equal times in the VL and VH states.   Individual bits are encoded by a pair of signal levels as shown. A logic 1 is encoded if two consecutive levels are present with a period of nominally 58us. A logic 0 is encoded if two consecutive levels are present with a period of nominally 100us. There is a minor complication caused by support for DC locomotives without a decoder. To allow a single DC locomotive to be run, it needs to be possible to give the signal an "average" DC value. This is done by stretching one of the two halves of a "0" bit. This period can become up to 9.9ms. Packet Encoding   Message Decoding The whole decoding process is normally carried out by a microcontroller. Its software is typically written in assembler or in "C" and looks at the DCC rail inputs to decoder individual bits, then forming them into packets. For a typical "PIC" series microcontroller this will take much of its time. Some microcontrollers are significantly faster - for example the Atmel AVR series - and can perform this as a background, interrupt driven task. Bit Decoding Typically individual bits are decoded by looking at one rail only and looking at the periods when the signal is above a threshold or below a threshold. Note that there is no guarantee that the bits will arrive with the 1st half of the bit being above threshold (as drawn for rail A): if the decoder was in a locomotive, and it is removed and put back on the rails the other way round, it will now see rails A and B swapped over. Instead, the decoding software will need to synchronise so that it is seeing Packet Decoding Mobile Decoders Accessory Decoders DCC System Architecture Command Bus Interfaces The two common command buses are XpressNET and LocoNet. These are both based on asynchronous serial communication and can be interfaced using a UART.
	©CML Electronics Limited 2008 Page Last Modified on: