MERG and the "CBUS"
Have been dreading the thought of wrestling with miles of cabling to signal the layout and how to interlock electrical and pneumatically-powered points and signals.
One big problem (ignoring the sheer quantity of wiring) was how to integrate three signal box positions which could be easily unplugged and removed when not in use .... without a mass of connectors and loose wiring which could potentially result in difficult-to trace breaks.
Ted Sadler has, on several occasions, suggested DCC as the solution and related how groups of MREG members attend Cliff's GTG, with the sole purpose of experimenting with electronic control.
I have never used DCC but assume it to be basically MUX/DEMUX as applied in computer networking.
This has the advantage of requiring just two wires to carry the signal to the demux unit fed from the two-wire BUS.
Well, today I have joined MERG and hope that their CBUS system will be the answer to my problem ..... although electronics and English weather are not an ideal combination!
One of the ideas being put forward by MERG is the use of "Raspberry Pi" micro-computers to run the CBUS....not sure I want to go that far.
Anyone have "thoughts" or suggestions?
John.
I did have a layout for a time using DCC control for trains and DCC control for points all from LGB, but having gone completely gauge 3 I've decided to go to DC only. Reason being I will only need 12 switches for the main station area split into two sections, 8 for the junction station and 4 for the branch terminus. I've decided to have signals to on the main loop will have the ability to be controlled by the train- reed switch control and a magnet fitted to the engines so it can run hands free. It might mean having more toylike LGB signals but better than no signals or even signals that stay at danger as attain passes through.
Even this will involve a lot of wiring so totally agree with your method for your layout.
My area of concern is trying to use LGB motors with cliff barker points- not sure it can be done.
Imagine I've answered no questions John, but thought I'd share my method of control.
John,
My experience goes back to the 1980's when I created a terminus to fiddle yard N gauge continental layout & had full point signal interlocking with mimic graphic diagram of signal & point status & block occupation on an old portable Sony B&W telly. The method I used was Hornby Zero 1 for the points & a microcomputer Z80 & 8085 based made by me with opto isolated input output ports driving small reed type relays & for inputs sensing switched on points to determine direction. But it still took a lot of wiring including butchering the side of the Zero 1 Master unit to patch the keyboard keys to the input ports on a ribbon cable. The whole interlock was burnt into a 4K EPROM with machine code. It worked like a dream. But even with the Zero 1 not using the track control facility with 2 twisted wires there was an abundance of other wires for other circuits & even for a small layout 12 feet long by 1 foot wide it took an age to accomplish & fathom out. Not exactly what you would desire.
Then I went to sleep electronically for a decade or longer after which LGB produced MTS/MZS which although costly produced a system which could be programmed from a manual which is 3 inches thick. Genuinely has limited 2 wire capability. So after amusing oneself with the basics & blowing out a loco chip when loco parked inadvertently on a point frog I decided that stumps would be drawn & live with simplicity. Then LGB turned belly up & Massoth took over the DCC side of things, this is where I went back to sleep again & have not awakened yet suffice to dole out this warning.
So after all that it is non-powered track for me, radio controlled locos & hand operated points.
Sorry I cannot be more positive John but you have a mountain to climb in terms of complexity & I wish you every success, will enjoy the outcome with much pleasure when you hold a GTG & look forward to admiring the result immensely.
Dave
454
Hi John,
I've been a member of MERG for a little while and the CBUS system is quite interesting. It is based on what is effectively a local area network (LAN) but one basically designed by the automotive industry for cars! This means that there is a lot of hardware and software around that supports CAN (the technology behind CBUS).
My understanding is that there two 'forms' of CBUS - one where the 'components' are 'pre-set' and another which enables everything to be configured on the fly. I'd think the simpler system would probably suit you and all the hardware and system software required is available from MERG (and the 'kits' are reasonably priced too). In theory CAN is good for half a kilometre, so it should be fine for your purposes.
You should receive a CD with past MERG journals on it - and all the key information for their various systems (they have a few - CBUS, DCC etc).
However, if I was implementing a signalling control system for G3 - CBUS is certainly the system I'd choose at this time.
Regard,
IanT
Thanks to you all for the replies........am pleased that nobody has said I am "stark staring bonkers"!!!
Glad to know that Ian T is thinking along same lines ..... I may be asking for suggestions with some aspects!
Last time I tried anything like this was on an indoor layout in the 1970's, using 30 relays (piggy-backed) to control a quadruple track junction, with four aspect colour light signals and correct automatic "sequencing" and "fail safe". The mass of wiring was horrendous .... but it did work, although the reliability of the relays was a maintenance problem.
I remember drawing copious "flow charts" and logic diagrams to work out the wiring for all the permutations of the track settings and signals (which included route indicators).
Until my MERG "welcome pack" arrives, there is not much I can do to further the proposals.
Regards,
John.
I spoke with MERG on a couple of occasions about a control system for colour light signals that was suited to larger scales and out of doors and CBUS was the main component. I was going to buy enough gear at Warley last year to do some tests but they didn't have it with them. I'll have to dig my notes out but the plan was to set up a block signalling system suitable for 3 aspect signalling - green then red then yellow then back to green in sequence for each signal as the train progressed through the blocks. The key thing appeared to be a train detector and the best bet seemed to be some sort of infra-red beam system or similar that the train would break as it passed through and so cause the signal to change aspect. Must try and find some time as the prototype colour light signal is gathering dust awaiting assembly.
Ian
One of the things that intrests me with gauge3 is the amount of electronics that can be used. Am experimenting with an arduino micro computer to control a motor via bluetooth with a mobile phone. Also used the same micro computer to control a servo to raise a signal to danger for a set period of time after a train has passed.
My feeling about state of art electronic systems & model railway applications is that it benefits the group layout or the club layout when an abundance of skilful talent is available to contribute. The sum total of man hours for everything involved with a model railway is immense after consideration of dreaming, planning, acquisition, building the models, making & laying of track etc etc etc.....
The number of man-hours to just get up to speed with a new "micro" is enough to put me off now. Then one's time is rationed to a proportion of the remaining time available to achieve the end result.
So for a one man band this is a tall order unless there are abundant resources to draw from.
The availability of off the shelf DCC with programming capability gives one a step up instantly but at what cost?
As modellers we have to get the balance right between making models, getting them to operate in a realistic manner & then when all is complete having all trains passing signals at danger & never stopping in stations & failing mid section. So any attempt to bring us into a state where realistic operation is the final goal, I welcome.
Sadly this is one reason why my garden railway consists of a simple 24 feet diameter circle.
I have been considering installing one signal just for appearances sake & make it a bit like Gavin_B described which is a good way of keeping things simple. It is essential it does not detract from the relaxation of watching trains go by.
Good luck with the project
Dave
454
As someone who has (far, far) too many "things on the go" myself - I pretty much generally agree with you Dave on how much an individual can (should?) take on and still hope to actually finish.
Very large projects are generally best left to groups, mainly because of the advantages of shared labour but also (I suspect) because individuals can then 'focus' on just a few things (rather than many). However, John seems to be demonstrating that given sufficient determination, an individual can achieve very remarkable results.
I have what I consider to be quite a wide range of interests but many of them fall more into the "armchair" category (e.g. mainly reading and 'imagining') than being activities I can actively implement. Some areas I do pursue a little further (let's call it 'playing' rather than 'experimenting'). For instance I have had a long time interest in micro's (dating back to my Z80 Nascom II system). Gavin has mentioned Arduino and I've had a Uno for some time. Arduino is a very interesting concept (and not expensive) and can be programmed directly from a PC. I've also dabbled with the Forth language for many years (not exactly a mainstream language) and can only explain this as being a better mental activity than doing Sudoku. Having said that, I was really pleased recently to get my Arduino 'chatting' to my PC's Forth system via USB - so I can now connect my laptop to the real world in a simple way.
But what does all this Arduino, Forth & CBUS stuff 'play' mean in the real (G3) world? Well nothing very much at all so far. But one day I may surprise (myself mainly) and come up with an on-board control system for live steam engines or a simpler (semi-electronic) gear hobbing unit. Maybe one day there will sufficient interest in 'modules' (here in UK that is) that there will even be a need for a flexible system to control train movement across larger spaces? Who knows - I don't smoke - so haven't got a pipe but I can still dream a little.
Back to John's 'real world', CBUS is well supported and John doesn't have to design and build everything himself. There are many others (often with expert qualifications) contributing to it's development. With this open source model, I think CBUS has far more going for it than a proprietary system which must be limited by commercial constraints. The open source nature of CBUS is also a factor in keeping costs down. People are already hooking up Android phones to CBUS systems - so Gavin, one day you might be able to control your engines and request a signal/turnout change - all from your phone - who knows what will be possible?
Anyway - the weather looks good so I have work to get done in the garden before the winter sets in. Back to Dave's real world then!
:)
Regards,
IanT
I don't like electronics. I know little about them and want to know even less. But, last Monday I was invited to a 7mm model railway which is all DCC and most of the 50+ locos had sound cards too. All were professionally built to the very best standards.
First thing I noticed was that when the mains power was turned on a loco started moving, straight through the catch points and derailed. The owner frantically picked up three controllers before finding the right one, then had to punch in which engine it was, and then control the power. When operating fully the effect was very good indeed, but for me its all far too complicated and uncontrollable.
I'll stick with batteries and maybe not even radio control.
Luddite Mike
I wasn't suggesting control of the locos via the CBUS Mike - just the signalling and turnouts.
Generally, I think (in G3) control of engines is very likely to remain on a "one-driver, one engine" at a time basis - but perhaps on some G3 railways in the future - you won't be allowed to climb out the cab and throw your own points (so to speak) - you'll have to await signals! :-)
My comments about 'Arduinos' & steam locos was not just about ways of 'driving' the engines but also other functions (such as sensing water levels & automatic bypass control for instance). We have very little 2-rail in G3 so apart from 'free running' , R/C control is pretty much the standard for us - but there are other alternatives around (infra-red & Bluetooth for instance) that may have advantages. And in the same way I don't know too much about the inners of the latest R/C kit these days, I don't think anyone will have to know too much about how any new alternatives work - it will all mostly be "black box" stuff. You'll plug it in and just use it.
Exactly what those 'black boxes' will do remains to be seen - but I'm sure it will be something tempting! :-)
Regards,
IanT
Quoteyou won't be allowed to climb out the cab and throw your own points (so to speak) - you'll have to await signals! :-)
That's exactly what I am aiming for, particularly in view of my complex track arrangement, which could result in serious and damaging collisions if left as a "free for all"!
The only involvement of power through tracks will be "track circuiting" to indicate line occupancy.....certainly no intention of using CBUS (or DCC) to control locos.
Regards,
John.
By the way MERG has quite a lot of technical information available to non-members on their website and a pretty detailed overview of CBUS is available here:
http://www.merg.org.uk/merg_resources/cbus.php
Regards,
IanT
This topic has been "dead" for a while now but it seems a good time to revive it!
There has been discussion within the Monkton Priors team about signal/turnout control and interlocking and Ian T has volunteered to do further research into possibility of a CBUS solution.
I have been discussing turnouts with Ken Cottle (he has volunteered to construct and pay for the six turnouts required for Monkton Priors station) and mentioned that the turnout blades should be pivoted (not biased or "sprung" in any way, as would be the case with Cliff's standard tie-bar arrangement) since we will probably use servo motors to power them. He said that he uses servo motors himself and has a PCB based digital control which sets the range of movement of the switch blades.
Has anyone here actually "dived in" and used CBUS since we last discussed this?
I have here a "Minx MicroDrive" package which consists of a pair of (postage stamp sized) actuators and a sophisticated digital controller and warning system. When the weather improves i will venture into the garden and install it...the big question is, how will it stand up to the damp?
John.
I am wrestling with the same problem for my layout. I too was thinking servo for point and may be signal control.
Whilst at the recent Scottish model rail exhibition I spoke to Fraser Black of 'Sig-na Track', he had a small unit that controlled the movement of the servo and could be DCC or CBUS. I use Crest radio control for my loco's and these have a facility for controlling 5 servo's via a RC unit.
Tim
Sorry to be negative here and whilst I have experience of full interlocking on a permanent layout, I have no experience of this system or on portable layouts, but ...
Monkton Priors is a very simple portable layout capable of being extended in different ways. I would keep it simple and have no interlocking and minimal electronics. Consider manual point control at this stage - wire and tube is tried and proven, cheap and quick.
How does the complex electronics work and physically connect up if you suddenly add another baseboard in the centre with a new point and siding, or a couple more boards at the end?
Mike
I agree with Mike; 'KISS'.
'The Pirate'
Well, I would normally go along with "KISS" (simple is best, I agree) but in this case, with lots of board joints (and longish distance to cover....the layout Monkton Priors layout will be 50 ft overall) there are difficulties with using mechanical means......not least "bridging" the board joints.
Whereas an electrical cable can simply be plugged and unplugged using multi-point (computer type) connectors, linking rodding is more complicated.
From the start, I had envisaged the turnouts and the signals being electrically powered by solenoid motors but (if it is affordable....and Ian T is working on that aspect) then using the CBUS (with servo motors of the type used with radio control) both simplifies the wiring and makes operation simpler (especially if the exhibition crew is limited in numbers).
There is also the fact that using digital control, the control panel diagram will indicate if a turnout has failed to "switch" fully thus avoiding unnecessary derailments. A bonus is that semaphore signals can simulate the "bounce" one sees when an arm drops.
Whether or not Monkton Priors goes digital remains to be seen...there are a lot of more pressing matters demanding attention.
In a garden environment, I am not convinced that CBUS (or any other PCB based system)would survive the weather but its use would very considerably reduce the amount of (long distance) wiring required.
Regards,
John.
John...
Speaking as a "computer person".... You are going to have to think very carefully about which type of connector that you use. I would suggest that you use sockets at the boards and connect via a group of leads NOT one!
If you are carrying power then the XLR connector or the Neutrik system is superb.
Data connectors that are robust -are rare. AVOID D25 and D15 connectors as they are too fragile -use instead a SCART system -these were originally TELCO plug together couplers.
regards
ralph
It's all very early days and my knowledge of CBUS is currently restricted to what I've read from the various MERG documents and their 'Forum' - so only 'theoretical' at this time.
A relatively 'simple' layout - such as MP (which is essentially pre-defined) would most likely be set up as a "SLIM' system - using pre-sets on the PCBs. If we ever need the added complexity of a fully self configuring system - then the SLIM version can be upgraded to the 'FLIM' version (full computer configure & control). It seems doubtful we will need a FLIM version in the foreseeable future though.
CBUS is a LAN system and signalling is carried over twisted pair. Obviously, the PCB 'units' that send/receive the control signals (and act on them) need more in the way of local wiring (and therefore connectors). However, signalling/switching a physically large layout will be problematic whatever system is used when that system has to be portable. There is no need for any layout to be fully automated or 'finished' on day-one - provided that we have some idea of the longer term design goals and can build in a bit of future proofing. I've always assumed that MP would evolve over time and hopefully be a good show-case for G3 best practice.
There is an active local MERG Group near to me and I do intend to contact them and seek their advice and support. I also intend to investigate CBUS (from a practical point of view) most likely by building a CBUS test rig. Once I've done this, I may be in a better position to understand the practical issues involved and to recommend the best way forward.
Regards,
IanT
seems to me that if you're going to have signals, then you may as well make a feature of them and operate them as they would in real life. Have someone in each signal box operating the points and signals that would be controlled by that signal box. Then the drivers, who are actually each in control of their own loco, should watch, and obey those signals, just like the real thing.
Then the control of the signals and points is a pragmatic one, rodding or electric motors depending on which is most practicable in that part of the layout.
I suppose the above assumes some kind of operational timetable which I think is a good idea, if your just running stuff round the oval like the G1 types do, then none of it is necessary.
With regard to connectors, we've been using the 9 pin equivalent of the D25 on our club layout for years with no problem, and we have to re-erect it every week.
That would certainly be my idea of how to run things Geoff....
Signalmen and Drivers :-)
Regards,
IanT
QuoteHave someone in each signal box operating the points and signals that would be controlled by that signal box
Absolutely .........my garden line will have at least three signalboxes and nothing will be permitted to move without a cleared signal and(where applicable) a single track token.
Probably will upset the "chase my tail" brigade without radio control but the layout complexity will require a disciplined approach or there will be some nasty pile-ups.
John.
HELP REQUIRED WITH MONKTON PRIORS!
I have recently posted a similar request in another thread but have revived this older topic since there were (at that time) a number of members thinking along similar lines : Have any of you progressed the research or even implemented a CBUS type system?
Ian T is looking into the matter (as am I, since we are both MERG members) in the hope we can incorporate the CBUS into Monkton Priors but we are both hard-pressed with the layout construction and have little time to pursue esoteric concepts.
To enable us (the Monkton Priors team) to provide the best solution to point and signal control (we are not into DCC...the layout is designed for battery power)can anyone, please, offer assistance to progress our understanding of the requirements in the context of Monkton Priors.
There are systems available "off the shelf" but they don't offer the same potential benefits as CBUS and are far more expensive.
A "push button" routing system which sets points and signals for the selected route with the push of a single button is the ideal (with warnings if conflicting settings are selected or a selected route fails to set correctly).
Regards,
John.