How does it work?

[Nearholmer]

Folks

 

Most of what I can find on here about DCC consists of very useful hints and tips about how to implement various proprietary DCC systems, but doesn't go very far, if at all, into how DCC actually works.

 

I've read some of the basic stuff on the NMRA website and elsewhere, but I'd be grateful for any pointers to simple introductory material.

 

Sorts of questions in my head:

 

- is the message format used common with any other (not model railway) applications?

 

- are the decoders of a generic-type that is used in other (not model railway) applications?

 

- to what degree can the system carry status information back from the decoder at the train/accessory to the user interface?

 

- can the message format be conveyed reasonably easily over links of forms other than that used in the normal two-wire system used between controller and train/accessory? (I know it can be conveyed over other links, but I'm wondering how much protocol conversion or wrapping is necessary along the way)

 

Is there, perhaps, a book about DCC that starts at "total innocent" level (where I am now) and builds up from there?

 

Many thanks in advance, Kevin

[Oldddudders]

While your questions are relevant to an interest in the further reaches of DCC, if you treat it as a sort of black box, obeying installation and user guidelines, DCC just works. It works just fine for those of us of a non-techie bent, and offers pretty much all things to all men and women if they have the desire to develop it further. The degree of implementation of accessories beyond the basic locomotive is down to individual ingenuity, while for many the simple ability to make each loco respond individually is quite enough.

[Talltim]

AFAIK DCC is only used in model railway applications. There's a fair amount of technical information on the NMRA website http://nmra.org/dcc-working-group

[Nearholmer]

Thanks Gents.

 

OD - I'm an engineer by background, which all starts with an innate tendency to want to take things to pieces to find out how they work. Whether I can then put them back together, and actually make them work again, is another question altogether!

 

K

[Reorte]

AFAIK DCC is only used in model railway applications. There's a fair amount of technical information on the NMRA website http://nmra.org/dcc-working-group

I assume that the digital version of Scalextric uses it too (it would make sense at any rate for Hornby to have used what they already have for model railways for that). I admit that that's not a very wide variety of applications!

[34theletterbetweenB&D]

It is an open standard system, so there should be a spec available. Starting from Wikipedia, (usual cautions!) there's plenty of references. https://en.wikipedia.org/wiki/Digital_Command_Control

 

I was told by a one time German colleague that Bernard Lenz developed his experience in this embedded signal control technique in applications for building management systems, aircon, lighting, etc. No way of quickly verifying that. Interestingly I have accidentally run across the lighting application, and found out that the low voltage control system is very vulnerable if exposed to high electric field changes, such as may occur in major thunderstorms. Normally such systems go into steel framed or reinforced structures which provide effective Faraday caging and thus protect the low voltage line from large induced currents. Put such a line along hundreds of feet of the ridge of a wooden framed roof though, and suddenly the cathedral's lights go out.

[Nearholmer]

Aha!

 

BMS was one of the parallel applications that I had in mind when I started wondering about all this.

 

Thank you 34C.

 

K

[teaky]

Isn't that true of any low voltage system?  In the case of our model railways we have a circuit(s) of track and cables linked to components (controllers, point motors, locomotives etc.) none of which could withstand the kind of voltages in lightning.

 

It's an interesting thought.  Should anyone with a layout in the loft be fitting a lightning conductor?

[34theletterbetweenB&D]

Since I learned this I have disconnected the DCC system from the track when not in use, and I wouldn't operate if a lightning storm was overhead - the flash and the bang following within a couple of seconds.

 

A typical lightning conductor doesn't help! It may locally depress the electric field, but doesn't eliminate it; and it is the change of the field that induces currents, so in some conditions the presence of an earth conductor has the potential to worsen the situation. What you need is a cage around the low voltage line: earth screened cable is most appropriate. But this is impractical for rails...

[Nearholmer]

And, what I was beginning to think about, how DCC seems a bit "yesterday" in the context of emerging "internet of things" technology, has already been thought about, by cleverer and quicker people than me, who are already on the case!

 

See here http://members.iinet.net.au/~backway/DWiC/DWiCSpecification.pdf

 

K

[Adrian Wintle]

And, what I was beginning to think about, how DCC seems a bit "yesterday" in the context of emerging "internet of things" technology, has already been thought about, by cleverer and quicker people than me, who are already on the case!

 

See here http://members.iinet.net.au/~backway/DWiC/DWiCSpecification.pdf

 

K

Remember that the basic standard dates from 1993. Since the volume and turnover of model railway equipment is very low (nobody replaces their DCC system and decoders every year), the technology is perforce limited to its late-'80s origins. There have been evolutionary extensions to the capabilities and the communications protocol but it does also maintain backwards compatibility. Provision for feedback from the decoder* is one of the newer extensions.

 

The specifications are here http://www.nmra.org/index-nmra-standards-and-recommended-practices

 

*during operations, basic feedback was always available during programming.

 

Adrian

[34theletterbetweenB&D]

And, what I was beginning to think about, how DCC seems a bit "yesterday" in the context of emerging "internet of things" technology, ...

 I perceive that as a benefit: it takes the very well established 12V DC two rail standard as a basis; and the DCC gear is all stock microprocessor and discrete component construction so indefinitely maintainable. Exactly as Adrian above: for the typical orientation of most model railway hobbyists it is a good fit.

 

I am going to sit on my hands a long time until a clear winner for an alternative control method wins wide acceptance. It could potentially be very good with a simple DIY way of building an on screen mimic diagram for the layout and much else. But that will require development by enthusiasts for model railway rather than softwariasts.

[Nigelcliffe]

Sorts of questions in my head:

 

- is the message format used common with any other (not model railway) applications?

 

- are the decoders of a generic-type that is used in other (not model railway) applications?

 

- to what degree can the system carry status information back from the decoder at the train/accessory to the user interface?

 

- can the message format be conveyed reasonably easily over links of forms other than that used in the normal two-wire system used between controller and train/accessory? (I know it can be conveyed over other links, but I'm wondering how much protocol conversion or wrapping is necessary along the way)

 

Is there, perhaps, a book about DCC that starts at "total innocent" level (where I am now) and builds up from there?

 

Many thanks in advance, Kevin

 

Attempting answers of the above:

 

Message format, it is what it is, fairly simple and at the DCC track level, slightly model rail specific.

 

Bi-directional messages.  

From the track, only with proprietary extensions to the standards.  There is RailCom, where the license is fairly open (though there are disputes as to whether it is really open).  There is RailCom+ which is proprietary to ESU/Lenz.  And Transponding, proprietary Digitrax.    Those offer various means to talk to devices (mostly locos) over the track interface.

But, practical implementations usually end up with another data bus to link trackside devices (accessory controllers for turnouts, detectors for track occupancy, etc.) back to the centre.  Example buses include LocoNet, S88, CBus, and there are others.

 

Communication over other medium.

No reason why one couldn't encode the DCC signal, but that is rarely done, and perhaps messy.   The various data buses (mentioned above) can usually be encapsulated in various other network protocols.

 

And, what I was beginning to think about, how DCC seems a bit "yesterday" in the context of emerging "internet of things" technology, has already been thought about, by cleverer and quicker people than me, who are already on the case!

 

See here http://members.iinet.net.au/~backway/DWiC/DWiCSpecification.pdf

 

K

 

Model railways as a hobby is fairly long term.   Layouts take 10-20 years to build in many cases.   So, stable, reliable, tested, and well supported by several companies (protection against one going out of business) are what people want and need.  

 

Yes, there may well be bidirectional things "around the corner".   As well as the above specification, you might want to read up on Protocab (Scottish company) and BlueRail (US company).    Protocab's plans included fairly advanced mesh network communication over radio frequencies, so a device need not be able to "see" the central system to receive a message (using stuff in the 802.something arena).   

 

I expect the next wave of communication for models will be battery, radio and bi-directional.  The hard part from here is spotting which of several possibles will be the "standard" of the future, and which will fall by the wayside as happened with products such as Zero-1 (launch to withdrawn by maker Hornby took 6 years, leaving the owners to fend for themselves).

 

 

- Nigel

[Nearholmer]

Many thanks Gents, lots more reading for me to find time for!

 

I particularly like your final paragraph, Nigel, because:

 

- I'm sure you're right; and,

 

- what I'm doing, out of no purpose other than curiosity, and an attempt to brush-up my knowledge in general, is to attempt to spot the probable "next generation winner".

 

My gut feeling is that "this time round" the technology will be much more general-purpose and freely available, so that most of the creativity will be more rapidly focused into functionality and into interface design/presentation.

 

All fine talk from a man whose main toy train-set is firmly stuck in the technology of c1950.

 

Kevin

[Edwin_m]

I think something like the direct wi-fi linked above will be the next generation, but I wouldn't like to say when. 

 

Standards are reasonably stable for wi-fi and if it can be fitted into a mobile phone then there should be no problem fitting it into something of similar size to today's decoders.  Because the signal isn't going through the track it can co-exist on the same layout with other control systems including DCC or alternative wireless systems.  It could take its power from the track, whether a simple constant voltage or rectifying a DCC waveform, or from onboard batteries or any other source anybody could devise, maybe in the larger scales even a diesel-electric using a model aero engine! 

[davetheroad]

And, what I was beginning to think about, how DCC seems a bit "yesterday" in the context of emerging "internet of things" technology, has already been thought about, by cleverer and quicker people than me, who are already on the case!

 

See here http://members.iinet.net.au/~backway/DWiC/DWiCSpecification.pdf

 

K

 

See bluerail technology

[Phil S]

"Dutch_Master" wrote: Roco offers camera-equipped trains that allow the operator to control the loco via the z/Z21. The live video feed comes via Wireless. Currently, prices are quite steep, but I expect these will fall in the coming years, in a self-strengthening circle: more units made means more choice of loco, meaning more people will buy then, meaning more need making which reduces cost, which enables more models to be equipped so more folk will buy them., etc

 

Phil: The Roco ICE video camera set, which I bought last year, comes with an 18vdc power supply, and 'no controller'.....you can run it directly via WiFi from your phone or tablet...... But note that to get the pictures, the phone needs a display of at least 640 pixels.... I had to hijack an iPad to see the pictures. A suitable computer would probably do.... But is less easy to Walk around with. 8-)

The loco can be remotely selected to be controlled from the wifi Or from dcc.... In which case the phone talks via the Z21 or z21 and its router to control the camera loco as part of a normal dcc system.... But with perfect live pictures available.... No analogue link breakup.

 

I have not downloaded the manuals of the other camera-loks to see if their options are the same.... I expect so.

 

So, wifi links are already in use... Probably using dcc protocols .... Giving the advantage of direct control with low cost Wifi, as also demonstrated by the new low cost multimaus wifi handset for about 80gbp 99eurp rrp. A side effect being that battery, dc , or dcc or even mfx track power could be used

[kevinlms]

Many thanks Gents, lots more reading for me to find time for!

 

I particularly like your final paragraph, Nigel, because:

 

- I'm sure you're right; and,

 

- what I'm doing, out of no purpose other than curiosity, and an attempt to brush-up my knowledge in general, is to attempt to spot the probable "next generation winner".

 

My gut feeling is that "this time round" the technology will be much more general-purpose and freely available, so that most of the creativity will be more rapidly focused into functionality and into interface design/presentation.

 

All fine talk from a man whose main toy train-set is firmly stuck in the technology of c1950.

 

Kevin

I think that your conclusions are somewhat suspect.

 

Just because Zero-1 died after 6 years, doesn't mean something else is about to replace DCC, every few years. Some continuity is always required. The main problem with Zero-1 were is limitations, especially limited to 16 locomotives (or other remote device), its a by product of being a pioneer. Other similar vintage systems (there were many others besides Zero-1) and all have died, but Zero-1 is the one that cops criticism, presumably on the basis that it was exclusive to one major toy manufacturer.

 

DCC is supported by many manufacturers & so will not suffer the same problem of obsolescence. For some years, it has been stated on sites like RMweb that decent & flexible radio control is 'just around the corner' & is true of some usages, such as outdoor large scale models. Their key advantage, is the space for bigger batteries.

 

It ought to be remembered that the level of communication between device & controller (bandwidth) isn't that high and won't need to be.  Why, because we are just controlling power to devices & not controlling graphics, which as any computer games person knows, this is what drives the need for ever faster & more powerful computing devices.

 

I suppose we could forget about building model railways & just have perfect graphics of a layout, stock & scenery, but somehow, I don't see it catching on. But I could be wrong.

[peter220950]

There's a good intro to DCC in the MERG Electronics for Model Railways book that can be downloaded here

 

http://www.merg.org.uk/ebook.php

 

The DCC section is towards the end of Part A (page 212). Just don't to try and print it all out, there's hundreds of pages!

 

Peter

[Nearholmer]

KevinLMS

 

I'm not expecting something to replace DCC "every few years".

 

As has been pointed out by one other others, DCC is late-80s tech, standardised in 1993, so if a successor emerged over the coming five years, as remote control and monitoring of all sorts if things becomes much more common, DCC would have had 25-30 year "reign", which is pretty good going.

 

I think a lot depends upon whether remote control and monitoring really does become commonplace, or whether all that talk turns out to have been hype ........ If it doesn't, then the market for small embedded controllers (decoders) with the right functionality may be too small to make them cheap and readily available enough to cause "next generation" technology to replace DCC.

 

Kevin

[Nearholmer]

KevinLMS

 

I'm not expecting a successor to DCC "every few years".

 

If one emerges, as I'm guessing it will, from tech that came to market about three years, in say three years time, DCC will have had a 25 year reign, and won't fade out instantly.

 

Kevin

[kevinlms]

KevinLMS

 

I'm not expecting a successor to DCC "every few years".

 

If one emerges, as I'm guessing it will, from tech that came to market about three years, in say three years time, DCC will have had a 25 year reign, and won't fade out instantly.

 

Kevin

Companies like North British said something similar. "Steam locomotive construction would gradually fade out over 20-25 years, as 'The Colonies' would still require something cheaper to buy and simple to maintain and run". Was the general thought, but probably not those exact words. How wrong they were.

 

If something new & better comes along, DCC will disappear quite quickly, IMO. That is the challenge. Systems such as Red Arrow have been around for 10 years or so & a statement from their Home Page, says:-

 

<RED ARROW is suitable for scales of 4mm and larger, the limiting factor being only the insertion of modules and power batteries into the motor unit.>

 

http://www.redarrowcontrol.co.uk/

 

So there you have it, the batteries & equipment are the problem & it excludes scales smaller than 4mm, although I suspect HO would be fine, especially larger prototypes like the US. This of course excludes N Gauge, which is a very popular scale (not by me I hasten to add - too small for my liking, so I'm not arguing on this basis). Also excluded are small 4mm prototypes, models of which frequently appear within RMweb.

 

If a new standard comes along & starts to erode DCC, I'll happily eat my words. Until then though!

 

The Other Kevin.

[davetheroad]

Companies like North British said something similar. "Steam locomotive construction would gradually fade out over 20-25 years, as 'The Colonies' would still require something cheaper to buy and simple to maintain and run". Was the general thought, but probably not those exact words. How wrong they were.

 

If something new & better comes along, DCC will disappear quite quickly, IMO. That is the challenge. Systems such as Red Arrow have been around for 10 years or so & a statement from their Home Page, says:-

 

<RED ARROW is suitable for scales of 4mm and larger, the limiting factor being only the insertion of modules and power batteries into the motor unit.>

 

http://www.redarrowcontrol.co.uk/

 

So there you have it, the batteries & equipment are the problem & it excludes scales smaller than 4mm, although I suspect HO would be fine, especially larger prototypes like the US. This of course excludes N Gauge, which is a very popular scale (not by me I hasten to add - too small for my liking, so I'm not arguing on this basis). Also excluded are small 4mm prototypes, models of which frequently appear within RMweb.

 

If a new standard comes along & starts to erode DCC, I'll happily eat my words. Until then though!

 

The Other Kevin.

 

Parts of the redarrow site do not work including the pricing and detailed technical information but I was able to gain some insights into what it can do.

 

We are talking about basic locomotive control here, there are no paths for advanced or automatic control but there are options to change stuff such as starting voltage. however I believe basic control is all most DCC users want anyway.

 

The receiver board is a reasonable size at 36mm x 14mm but DCC and radio control receivers are already available in smaller sizes.

 

IMO what eliminates it as a serious contender is the use of infra red rather than radio. Even the radio option uses infra red repeaters as the final link to the locomotive. This technology is a dead end especially as several reliable industry standard radio communication systems are available 'of the shelf'. OK, maybe not several but I can think of Bluetooth, WiFi and 2.4Ghz band model plane,boat,car, drone systems.

[Edwin_m]

I think the continued existence of DC 20+ years after DCC came along illustrates points made earlier in the discussion.  Many people are quite happy with what the have, don't want to invest the time, effort and cash into something new and allegedly better and may not wish to get to grips with the complexities of a new technology.  For the same reasons I believe DCC will continue for many years alongside any new control system.   

[peter220950]

So there you have it, the batteries & equipment are the problem

 

The Other Kevin.

 

And also the lack of a credible mainstream producer, after all battery remote control has been around in O for some considerable time, but still remains relatively obscure.. Small cottage type industries are unlikely to have much impact on the market as they tend to come and go depending on the health or enthusiasm of the person running them. Unless a Hornby or a Lenz come along and invested heavily it's always going to be a minority thing, and even then it might not catch on, witness the original Zero 1.

 

One of the greatest barriers to any of these 'revolutions' (bluetooth sound, battery remote control, etc.) is the lack of substance of any  suppliers. - Something like 12 years ago I went DCC with ZTC, we are now on about the fifth owner, with several intervening spells when there was no product available or product development. In comparison the likes of Lenz and ECoS have developed the products beyond recognition, to make inroads and keep market share requires considerable investment.

 

As time passes many of the advantages of remote control start to be eroded by advances in the competition, one of it's main strengths is the reduction in stalling on dirty track, the stay alive has dented this significantly, why would I therefore want to throw a functioning system away, convert all my loco's, and substitute it with something broadly similar. There were significant reasons to want to change to DCC, but even then it took a lot of heart searching to make the investment in time and new kit.

 

I have a feeling r.c. control is a bit like 3D and curved screen LED television, an interesting technology but not a viable alternative for most in the forseeable future.

 

Peter