Full Ring Bus or break it?

[Augustus Caesar]

I am not sure if this has been asked before but I did do a search and found virtually nothing on it so apologies if it has appeared... maybe I could be pointed there if it has to save going over it all again.

 

Anyways... I have just read an article in the latest Hornby mag (May) about layout wiring. As my own track has been lying as a test track for a few months now while trying to figure a design where I wouldn't be mucking about with it over time to a large extent, I was about to take it all up and begin wiring the ring bus underneath the boards.

 

The bus I was going to use was a solid ring with no breaks but having read this article I wondered how many of you have had the same question I am about to ask...

 

Is it better to complete the ring or break it evenly either side of the controller (eLink) so the lengths of cable would be, say, 20 feet either way left to right?

 

In the mag the author says don't use a full ring quoting thus:

 

"When wiring a circular layout, it is tempting to run the power bus around the layout as a 'ring main'. This should be avoided for a variety of reasons, including potential problems with data transmission and localised power problems.

The rails should also be double-gapped where the power bus is terminated so a 'ring main' is not created through the layout rails. In effect, the power bus wiring should radiate from the base station in a linear pattern rather than form a complete ring."

 

Now, I do understand all that so no questions about the meaning etc. But, I do wonder WHY there has to be a break and not a complete ring bus... as far as I am concerned I know of plenty layouts where a complete ring is used with no problems.

 

So, is there anything anyone here can add to convince me not to use a complete ring and go for the break with terminators? If the latter how would you terminate the ends and what do I use?

I would PREFER to just use a complete ring for simplicity but IF errors arise and commonly because of it the it may need to be looked at again.

 

Therefore... can anyone advise on this? As I said, I need to know only this as wiring and the rest I do understand... I just cannot remember seeing arguments for the breaks in the ring.

[96RAF]

@AC

I also wonder about the WHY - as normally your track would be laid as a complete ring without a break and no one would query that, so why not ditto for your bus.

 

I appreciate if there is a bus break, then you can end up with an 'aerial' broadcast effect so a bus terminator/damper is recommended.

 

Maybe the smart people like FB, Chris and Fishy who understand all this harmonic resonance stuff could enlighten the rest of us.

[mjb1961]

Interesting as I have recently re-wired my track adding droppers and this time going for a full circuit instead of a break ,I assume you mean that the bus just comes to an end ,at the last dropper ,if this is the case I can say I've not had any problems ,yet! 

[Chrissaf]

The maths can show that "to ring or not to ring" is a function of the total ring length. As the DCC signal propagates along the Bus there will be a propagation delay. If the total length of the ring is long enough then the signal that arrives back at the starting point of the ring can be slightly out of phase. Purely a mathematical example but if the phase delay was 180 degrees, then the DCC signal would be cancelled out. In reality of course, to get a 180 degree phase delay with a 8 Khz DCC signal would take a ring of many 1,000s of metres. But even relatively small phase delays can cause distortion of the original signal.

.

If you 'T' a Bus (or use a Star topology) , and leave the ends of the 'T' (Stars) open circuit, then the risk is that signal reflections will return back along the Bus and also potentially (on very long Buses) cause phase delay distortion. Thus, it is recommended that the end of the 'T' (Stars) include a terminator to sink the DCC signal, thus reducing the reflection phenomena. This is all fairly basic AC frequency 'Transmission' theory stuff that I was taught as part of my electrical engineering apprenticeship. These terminations could be the RC filter discussed in another recent thread or more commonly just a basic resistor of about 1,000 ohms value. The 1,000 ohms obviously needs (wastes) far less current than the RC filter termination. But then the RC Filter is giving far better 'short circuit' protection. Swings and Roundabouts, take your choice

.

Small and medium layouts will probably not notice any difference whether the Bus is 'Ring' or 'T' or 'Star'. It appears to me that most published pundits favour the Bus 'T' & 'Star' approach to adopt a 'best practice' philosophy.

.

What is important though, is if the Bus 'T' approach is adopted it is important to include the 'gaped' rails referenced in your bold highlighted extract statement . Else the benefit of the Bus 'T' approach is negated, because the physical track rails will short circuit the gap introduced in the Bus. The rail gapping is obviously far more onerous to design with a 'Star' Bus topology.

.

Mark Gurries site will give you plenty of bed time reading to send you off to sleep in no time at all. Save you having all those nights working through to the early hours with your computer support business.... 😉

.

PS - In case your wondering. My own layout is a Bus 'T' of about 30 feet in each direction, properly gapped rails, plus RC Filters as 'T' end terminators. Maybe complete overkill, but I have no DCC operational issues or problems.......so far.....mind you, I probably wouldn't have had any issues or problems with a basic 'Ring' either.

.

EDIT: Typical....when I started writing my reply, mine would have been posted directly below AC's original query. Perhaps I should have posted a quick holding statement post, then spent my time editing it instead.

[poliss]

But as the previous posters have pointed out, most layouts and certainly begginers layouts, will have a continous track circuit with no breaks. I couldn't actually find anything on Mark Gurries site about not having a ring bus for the reasons mention in earlier posts. Twisted pair wiring would seem to be much more important, but may cause problems with block detection so I read.

Breaks in a bus are very useful for fault finding.

[Chrissaf]

I totally agree, that twisting the Bus provides far more benefit than the Bus wiring topology deployed.

[RDS]

My Circular layout (4 concentric ovals, with 16 sidings in the middle) is wired using a full Circular Bus using 2.5mm Single Strand Copper with droppers made from 1.5mm Single Strand Copper and I have had no problems at all.  There are no insulated gaps in the ovals, so I guess I have 5 circular Bus's altogether!

[poliss]

5 buses at once. Isn't that typical of buses. 😆

[Augustus Caesar]

@RAF, mjb, Chris, poliss and RDS...

 

Great replies guys. Plenty food for thought there and apparently there is no real reason why I should not use a full ring... apart from the technical data from Chris's post, which I actually fully understand Chris, which spells out why terminators are a good thing.

 

Now, my layout will be 8 feet x 8 feet and 2 feet deep all round meaning an operational well, to begin with anyways, of 4 square feet. I would have used a single bus with a full ring but may be persuaded to look at the terminators. I will check the Mark Gurries site shortly (thanks for the link Chris) and see if there is any expansion on what I have read here. At the moment I plan 3 or 4 maximum circles of tracks with a good few sidings. No loops or whatever though. I'll use plenty droppers around the bus no matter whether 'T' or ring so this at least gives you all a picture of what I intend to do.

 

I suppose one more thing may have come from this... is there ANY worth in having more than one ring or 'T' shaped bus, say one for each circuit? My own thoughts say no but this is relatively new to myself and have not even thought of that before.

 

@poliss... having worked as a conductopr for four years in the late 70's early 80's I can honestly see where you are coming from when they all turn up at once... I used to instruct my driver to hang back so the next bus could pass, take the load and we could stroll to the terminus in Newcastle City Centre almost empty and ready for a cuppa in the canteen at Worsick Street!!! Very relaxed too... but I must also add we only did that if the driver behind us was a moaner and deserved to be 'taught a lesson'! :-)

 

Thanks again guys... any further comment on my question above will be of great help. Off now to check that site out...

[Augustus Caesar]

@Chris

 

Have been looking at, or more precisely, glancing around, that site you gave the link to...

 

You weren't kidding about plenty reading to send one to sleep at night... trouble is... I can't read it all tonight (well, the bits I need to read) because I do actually have some laptops to sort... I'll let you know how I get on with the texts though later... zzzZZZ!!!!! 😀

[Chrissaf]

Mark Gurries is a member of both the NMRA, PCR and the OpSig group and attends many NMRA conventions around the country where he presents his NMRA DCC Clinics.  He is also involved in his local NMRA Coast Division and PCR events.

.

So presumably he knows what he is talking about with regard to DCC technicals.......

.

However, as Poliss pointed out there does not appear to be any of his pages specifically expressing a preference with regard to Bus topology. That said, any references to Buses seem to be focused on 'T' and 'Star' ones. Maybe he doesn't mention 'Rings' because he assumes nobody will use them.......just a thought........I openly admit that I have made this statement without committing every page of his site to memory. So there may be a mention of 'Ring' Bus that I just haven't spotted yet

[Augustus Caesar]

I've had a good look around the site and while extremely interesting in parts there is no specific mention of bus system wiring.

I also see his qualifications and admire his enthusiasm for his work which seems outstanding.

I'm not, as yet, totally committed to any system and may just use the simple full bus ring for now. I say simple because if faults or irregularities occur then I can just snip the wires and terminate both ends in a nice and simple operation. From what I gather I only require a capacitor and resistor for each end of the 'T' bus?

Still, one learns all the time. Each one of us has expertise in their own fields per say and we can learn from each other on here all the time... now, on with my work tonight and maybe a further check tomorrow for new info etc...

[Chrissaf]

I've had a good look around the site and while extremely interesting in parts there is no specific mention of bus system wiring.

.

This is the page on Marks site that discusses Bus topology. He discusses 'T' Bus that he terms 'Linear' and he discusses 'Star' Bus that he terms 'Point to Point'. He makes no mention of 'Ring'. Interestingly, the web page in question is a sub page of a main page called 'DCC Best Practices'. This would seem to infer, that in his view a 'Ring' is not best practice as there is no mention of it.

.

I know from past experience, discussion on Bus topologies generates heated debate. For other's reading, please do not have a go at me if you disagree, aim your comments at Mark Gurries site. I'm just providing the link to the relevant pages in response to AC's quoted reply.

.

Also, bear in mind that the site has a USA rather than European bias. They use controllers (they call boosters) with much higher current capabilities than we do.

 

 

[RDS]

With my full ring, it would be a very simple modification to break it at some point and put terminators on, if it ever seemed likely that it would be better.

I guess then, that terminators would then be required on each of my ovals as well, making 10 terminators in total? 

[Augustus Caesar]

@Chris

You're quite right about the bus systems mentioned except that of the ring bus. My mistake was to generalise where I should have noted the ring type in my first paragraph.

I am always aware of American differences in their power supplies, controllers and the rest... simply because I do get some American clients with laptops and different power supplies where they don't use our 200 odd voltage.

Reading the texts does sort of imply that he may not necessarily like the ring system or, more unlikely, that he deliberately bypassed it. Who knows....

 

@RDS

I assume your layout is run from 5 cable pairs for 5 circuits but are they running from one pair from the controller or are they seperate in terms of power? Hope that reads ok because being this early in the morning after working my hrain is frazzling...  😎

[RDS]

@AC.  It is just one loop running from and back to, my Elite Controller.  I estimate I have between 40 and 50 droppers and I just ran a single ring using 2.5mm single strand insulated copper (stripped from 2.5mm T&E) around them all.

[2e0dtoeric]

Mine is an assymetrical Tee bus, because the control area isn't central to the layout, which is a double roundy tail-chaser.

I used excessively heavy 8mm silicon wire for it, because that is what I had to hand, and no other reason. The droppers go from each and every track piece to the bus, using Maplin's dead cheap stranded stuff. Pointwork is/will be controlled separately by the toggle-switch and press-button system I've mentioned previously. (The toggle lever shows the direction, and the press-switch fires the motor).

Power from the Elite to the bus uses an old curly telephone wire, with the cable from the psu run up the middle of the curly - saves them getting into a tangle!

So far (taps wooden head) I've had no problems with this arrangement.

[Yelrow]

Tis interesting thread this, as EEC, have banned ring mains now, as being unsafe, due to feed back. john

[Augustus Caesar]

Tis interesting thread this, as EEC, have banned ring mains now, as being unsafe, due to feed back. john

That's interesting John... can you elaborate further? Not come across that one as yet.

[Yelrow]

Well, in EEC, ring mains are  forbidden and have been for many years. Everything goes back to the box.As i understand the new regulations, existing installations are fine, but new ones need to comply with  EEC regs. I would not have known, but my cousin was here on a visit  when i was wiring up my consumer unit. Its the same for us now, in uk, he said, all back to the box, ring mains, apparently had had some problems, hence the ruling. No doubt, i am going to be told to the contrary, but certainly my cousins firm had stopped installing them. john

[Flashbang]

Not the 1st of April is it?  😮

 

There is nothing in the current IET wiring regulations 17th Edition BS7671 and the amendnent 3 for 2015. Of which some came into force in 2016.

 

The UK ring main (240 volt power outlet sockets) are used only because the plug has a cartridge fuse fitted. Most, if not all, Eurpeon electrcial installions do not have fused plugs and therfore only a single radial circuit is permitted. 

 

I would like to know the BS or IET reference number for this please, as its unknown curently! 

[poliss]

I can't find anything in BS 7671, Requirements for electrical installations. IEE Wiring Regulations. Seventeenth edition, (as ammended), that says ring mains are banned. There's a lot about armoured cables, RCDs and changing the standard wire colours for the mains.

[Chrissaf]

John, could your cousins reference to UK wiring be aimed at commercial premises rather than domestic. I'm with FB and Poliss on this one. I like to keep up with the IEE wiring regs, and this ban is news to me.

.

I know the EU have completely different rules, but so far their bureaucracy hasn't impacted yet AFAIK on UK electricals else we would probably have to standardise on the Euro plug. Heaven help us if we do.

.

Roll on June 23rd.

[poliss]

I read that radial circuits are preferred for commercial premises, but couldn't find anything about a ban.

Not that it makes any difference to a DCC ring bus. Electrickery travels at almost the speed of light, so the possibility of the DCC 'signal' being corrupted because of a ring bus is remote in the extreme.

[Fishmanoz]

Poliss, electromagnetic radiation travels at the speed of light in the medium in which it is travelling (fastest in a vacuum, slightly slower in anything else) but not electrical circuits.  Chris made the point earlier in this thread about phase changes of the DCC signal around a large bus and interference when this occurs.  In fact, it is this phase change becoming a problem which is the significant factor in defining a bus to be large.