One bus or two?

[JohnR]

I am currently designing my next layout - the track plan is finalised, and now am turning to control and wiring. 

 

My plan at the moment is to have both points and signals motorised, and with Electrofrog pointwork, to include frog polarity switching. Ultimately i want to go for DCC, and while a few of my locomotives are chipped, by no means are all of them. My thoughts are to have DC control at first, but designed to be easily upgraded to DCC as and when my fleet gets fully chipped. 

 

Reading up on the DCC Concepts website, I like the look of their products, especially the Cobalt point motors. They suggest it is possible to have a separate accessory bus for points etc, with a DC controller for the locos. How easy would it be to wire this up, and how easy would it be to upgrade? With DC control, I would need to have section switches - is installing these more hassle than theyre worth if I'm going to upgrade to DCC in a couple of years time?

 

 

 

 

 

[WIMorrison]

Using 2 busses at this stage makes it very easy to run DC on the track and DCC for accessories and you can use the SPDT switch on the Cobalt Digital IP motors to change the polarity of the frogs.

 

When you want to go full DCC later it will be extremely easy with the connection of the DC bus to the DCC controller through an automated current protection device such as a NCE EB1, PSX or MERG DCO. This will give you a properly constructed DCC layout using 2 busses which is best practice

[RFS]

The reasoning behind having a separate bus for accessories is that in the event of a track short circuit, most often caused by running a train into points set the wrong way, is that you can still change the points in this situation. This is usually done by having a circuit breaker on the track feed which the accessory bus bypasses. The CB cuts power to the track, leaving the command station and accessory bus still running. 

[RAF96]

Even if you need section switches now for the DC track, upon going DCC you just throw all switches on or remove and bridge the switches.

[Phil S]

Not forgetting, that if, in analogue, you used commercially supplied power connectors (eg Hornby. Roco...):  to remove the interference suppression capacitors contained within them - as these would appear as a short circuit for the ac dcc signal.    Not to be confused with any discussion about the suppression capacitors / inductors located adjacent to the motor in a loco.     if you made your own power connections before - you are extremely unlikely to have added interference suppression capacitors ... so no problem.

[Phil S]

On the subject of making it 'analogue' and then converting it to dc...   No problem making the pointwork/signals all dcc to begin with  - it simplifies their wiring, and reminds you that a separate bus is best for accessories - so that, as stated above by others, you can still control points even if the track has come to a halt due to a short circuit.

The complexity and excessive amount of work to 'fully sectionalise' the layout in analogue form, is, to my mind, counter productive !! and costly in time and effort and money - requiring a central control panel - even though you can avoid that for the accessories by using dcc for them from the start ... the benefits really come when EVERYTHING CAN (optionally) be controlled from a single tablet/controller/handset ...  so why double the cost of your layout by spending further on analogue redundant systems.  [I have been digital since Zero-1 was launched ... by which time I had 32 locos, or 2x its numbering system - so I conceed I am biased !]

 

HOWEVER - as a 'intermediate' way of operating  - as this type of layout lends itself almost to '1 engine in steam' operation - you could create perhaps just 2 (possibly 3) sections in the scenic part - the passenger platform area(s), and the goods area -

Confining/isolating the goods engine in that area ( or using the lower entry/exit track as a headshunt off screen )

Passengers coming from the upper entry/exit track into the platform and escape road run around, with a 3rd section for the other platform .....  

....but then you could say ... well that's only 3 locos at a time ... so I could convert those to digital now, and save all that wiring and operational restriction .... and the the other locos over the coming months  ...

( 3 switchable areas which can be isolated during fault-finding is a reasonable approach for digital - so not necessarily a duplication of work in creating the 3 basic areas,  plus  traverser)

The worst of all worlds, for diagnosing problems, is a layout designed for both analogue and digital, giving the best advantages of neither.

[Phil Bullock]

Echo whats been said - definitely two buses. Why not go the whole hog and go straight to DCC for traction as well - avoid all those section switches!

 

Phil

[John K]

Absolutely agree - Two buses every time

John K

[Ian Morgan]

DCC as an accessory bus is far from ideal, as the DCC specification is, essentially, a one-way communication channel from the command station to the accessories. If you want to get feedback from the layout (train on track detection, RFID identifcation, point position confirmation, etc), then you will need 2-way communication. Some manufacturers have their own, propriety, feedback systems, such as Railcom, but this ties you into that supplier.

 

The NMRA, who published the DCC specification now adopted by all the major manufacturers, have finally published a specification for a separate Layout Command Control  (LCC) based on the OpenLCB groups Layout Control Bus (LCB) which has been adopted by just a couple of manufacturers so far.

 

Merg have a Layout Control Bus system called CBUS that has been used on a large number of layouts for many years now. This specification was offered to the NRMA as a basis for their LCB specification, but seems to have been rejected as NIH (not invented here).

 

Both these layout control buses as based on the CANBUS system that is present in all newish cars, so benefit from cheap, widely available components made in vast numbers for the automotive industry.

 

[WIMorrison]

A DCC Accessory bus is very sensible and works perfectly well. Accessories do not need to provide feedback over the instruction channel (DCC), but can do where needed though the preference is to use LocoNet, XpressNet, S88 or RS Bus.

 

RailCom is within the NMRA DCC Specification, has been for a long time, and is supported by many manufacturers with a almost all European systems supporting it.  It does provide the 2 way communications as does Railcom+ now supported on mainland EU as is BiDIB. It is only the American market |(and to an large extent the UK market) that is lagging behind with little or no innovation nor adoption of standards and comms methods that are operated in EU. 

 

I would strongly advise that anyone entering the hobby uses freely available, mainstream commercial products that use open standards thus ensuring that he/she is not tied into any specific make, manufacturer on non standard comms method in the future. 

[peach james]

2 hours ago, WIMorrison said:

A DCC Accessory bus is very sensible and works perfectly well. Accessories do not need to provide feedback over the instruction channel (DCC), but can do where needed though the preference is to use LocoNet, XpressNet, S88 or RS Bus.

 

RailCom is within the NMRA DCC Specification, has been for a long time, and is supported by many manufacturers with a almost all European systems supporting it.  It does provide the 2 way communications as does Railcom+ now supported on mainland EU as is BiDIB. It is only the American market |(and to an large extent the UK market) that is lagging behind with little or no innovation nor adoption of standards and comms methods that are operated in EU. 

 

I would strongly advise that anyone entering the hobby uses freely available, mainstream commercial products that use open standards thus ensuring that he/she is not tied into any specific make, manufacturer on non standard comms method in the future. 

 

I don't see any non compatable with NMRA DCC Spec systems...all the systems which use RS-232 as a basis,  CAN BUS, and those which use RJ-12 based connectors, have open source elements to them, and the information exchange is certainly wide enough.  RailCom is a red herring at best, would be my thought, in that it provides decoder:command station 2 way transmission.  It is a bit of a kludge on the original DCC signal, as is the competing option of using Digitrax's Transponding system.  Neither one is needed for accessory decoders, as they can generally be wired back (and should be wired back !) via a separate wire, and usually fed a separate signal too on larger layouts.  Certainly, that is what I have found I need on Long Marton, with the booster/command station being on a separate signal loop than the accessory decoders & throttles.  Mind, that's with a 24x36 layout, with >180 detection blocks, >180 accessory outputs & a fairly complex layout.     
 

I would argue that one should buy what one is going to need fairly close to in one lump go, and pick a maker who has all the elements you want available off the shelf, not vaporware, but other than that, it should be OK.  Colour me a happy Digitrax user with Loconet providing the background communications system for the fixed elements of my DCC layout.

 

James

[Ron Ron Ron]

8 hours ago, JohnR said:

 

.......Reading up on the DCC Concepts website, I like the look of their products, especially the Cobalt point motors.

They suggest it is possible to have a separate accessory bus for points etc, with a DC controller for the locos.

How easy would it be to wire this up, and how easy would it be to upgrade?
 

 

If not planning to purchase your main DCC system at the outset, then as you have probably read, DCC Concepts offer a very low cost way of installing a separate DCC accessory bus by using their "Cobalt Alpha DCC Power Bus Driver and SNIFFER Adapter".

 

This adapter is essentially a stand-alone DCC system to operate the accessories. 

r.r.p. = £24.95

It can be found for as little as £22.

 

You will still need some means of inputting DCC commands to change points and signals.

DCC Concepts offer a number of options, along with the kit needed for each option.

 

1. Cobalt Alpha - They provide all the kit needed to create your own working mimic control panel.

Alternatively....

2. Cobalt Alpha Central - A self contained point switching panel (plug and play)....or,

3. Cobalt-S point levers - Create your own miniature lever frame.

 

As an option, the Alpha Central and Cobalt-S levers can also be used with the Alpha Mimic display system, to provide a visual indication of route settings.

 

There are YouTube videos demonstrating how easy this stuff is.

 

 

Later on, when it comes to converting the running of trains to DCC and purchasing your main DCC system, you can then do one of two things.

 

1. Integrate the existing DCC controlled accessory bus into the new DCC system. Using  the existing accessory bus as your separate accessory bus on the new system (as described by others above).

 

2. Keep the control of the points and signals on the original, separate, dedicated accessory bus and only run the trains from your new DCC system.

 

The downside from option 2. is that control of points and signals may not be possible through the main DCC system's throttle (handset or console) and it makes it more difficult if you intend to progress to computer control, or automation.

Many DCC users prefer to keep control of points and signals separate from the control of the trains anyway.....and your diagram suggests that computer control probably isn't on the cards?

 

 

However, as pointed out already by a couple of guys......

Consider the cost and more importantly the time and work involved in wiring the layout for DC operation, with the attendant section switches etc.

All this will have been wasted once you convert the layout to DCC and throw open all those section switches.

Might it be more cost and time effective to just go straight to running the trains on DCC from the outset?

 

 

.

 

 

 

 

[Junctionmad]

If you have some interest in electronics, them the MERG cbus system is worth a look and also has its own very low cost DCC controller system. Full source code , circuit schematic etc are available so no closed proprietary system here.  You need to be comfortable building simple electronics as the systems come as kits 

 

disclaimer : I am a current MERG member 

[Andymsa]

8 hours ago, peach james said:

 

I don't see any non compatable with NMRA DCC Spec systems...all the systems which use RS-232 as a basis,  CAN BUS, and those which use RJ-12 based connectors, have open source elements to them, and the information exchange is certainly wide enough.  RailCom is a red herring at best, would be my thought, in that it provides decoder:command station 2 way transmission.  It is a bit of a kludge on the original DCC signal, as is the competing option of using Digitrax's Transponding system.  Neither one is needed for accessory decoders, as they can generally be wired back (and should be wired back !) via a separate wire, and usually fed a separate signal too on larger layouts.  Certainly, that is what I have found I need on Long Marton, with the booster/command station being on a separate signal loop than the accessory decoders & throttles.  Mind, that's with a 24x36 layout, with >180 detection blocks, >180 accessory outputs & a fairly complex layout.     
 

I would argue that one should buy what one is going to need fairly close to in one lump go, and pick a maker who has all the elements you want available off the shelf, not vaporware, but other than that, it should be OK.  Colour me a happy Digitrax user with Loconet providing the background communications system for the fixed elements of my DCC layout.

 

James

 

I have to agree with James here on railcom and transponding. Having tried transponding I gave up on it as I found it very temperamental. But if a user is thinking of these reporting systems means there looking at some sort of computer control and traincontroller does train tracking very well, and I believe I-train also does this. So feedback is not really needed. As to feedback for other devices yes a feedback bus is needed but not really needed if your not having feedback, the specific system dictates if you require a feedback bus or not. As to the original question yes go for two dcc busses as suggested one for accessory and one for your track, unless your using multiple boosters where you need one for each booster. And go dcc now and save the hassle of section switching.

 

Andy

[JohnR]

Thanks Ron, thats very good advice. 

 

With regard to the other points raised, while the layout is fairly simple, at times, on Sidmouth (on which it is based) two engines were in the station at once - and at one point 3 on a Summer Saturday. I think this means the loco stud does need to be upgraded to DCC, and my old Bachmann split-chassis Ivatt 2MTs will need to be retired. 

[34theletterbetweenB&D]

23 hours ago, RAF96 said:

Even if you need section switches now for the DC track, upon going DCC you just throw all switches on or remove and bridge the switches.

 

25 minutes ago, JohnR said:

... two engines were in the station at once - and at one point 3 on a Summer Saturday. I think this means the loco stud does need to be upgraded to DCC, and my old Bachmann split-chassis Ivatt 2MTs will need to be retired. 

You won't need many section switches. For all that I am wholly DCC for operation, there is a little DC style sectionalisation on my layout. It means I can switch in DC supply to defined sections to test run mechanisms ahead of decoder fitting; and aids location of a short circuit on track tripping the DCC system. (And even with that helping narrow down the search area, it recently took me an hour to find the Morton brake gear crossbar that had fallen neatly between the rails to create a dead short!)

[melmerby]

54 minutes ago, JohnR said:

Thanks Ron, thats very good advice. 

 

With regard to the other points raised, while the layout is fairly simple, at times, on Sidmouth (on which it is based) two engines were in the station at once - and at one point 3 on a Summer Saturday. I think this means the loco stud does need to be upgraded to DCC, and my old Bachmann split-chassis Ivatt 2MTs will need to be retired. 

If your split chassis locos run Ok why retire them?

Converting to DCC is not difficult as I have converted two Mainline locos (Manor & 43XX) and it was straightforward

[JST]

I converted my old Bachmann split chassis 2MT to DCC and it runs beautifully.