On the Busses

[Ian J.]

I have been thinking of all the busses required for a layout with a decent set of power requirements (specifically my S&P concept) and this is what I think I may need:

 

S&P DCC Buss Arrangement, using a SPROG-based system

 

---

 

These are the board feed busses, that get power and signal to stock on the track and accessories on the boards:

 

Buss 1: DCC Track Power & Comms, 2 wires, +/-

Buss 2: DCC Powered Accessories & All Accessories Comms, 2 wires, +/-

Buss 3: 12V DC Power Accessories (no Comms), 2 wires, +/-

Buss 4: 16V AC Power Accessories (no Comms), 2 wires, +/-

Buss 5: Continuous Power 1 (for model lighting, etc), 2 wires, +/-

Buss 6: Continuous Power 2 (for model lighting or maybe layout lighting, etc), 2 wires, +/-

 

---

 

These don't connect to anything on the boards, instead using the boards as a conduit to get from one dcc district box to the next instead of having wires loosely connecting directly between district boxes haphazardly on the floor:

 

Buss 7: SPROG to SBOOST Comms, 2 wires, +/-

Buss 8: SBOOST Common, 1 wire, -

 

---

 

15 wires total

 

===

 

I'm intending to use an existing stock I have of Bulgin 8 way plugs/sockets to implement, two pairs per board, 6A per pin to connect all the boards together, and power districts will be employed, most likely using SBOOSTs in combination with the SPROG, one (or two) districts for the fiddle yard, and one for each scenic 'vignette', though it's likely only one 'vignette' would be in place at any one time.

 

---

 

Any thoughts, constructive criticism, etc? I'm particularly concerned I may be overdoing the number of busses, but I'm also concerned about power type separation and not having one need drain power from another.

 

TIA

[Edward]

Speaking as an absolute ignoramus of electricity etc  but as someone with 9 or so layout boards in a NCE system,  I think you would have an awful lot of wires and connections across!  I just have 4 supply wires going from the DCC output station to the 4 power districts following the "Star" principle.   My rule is "no Connections" across boards.  (But have had to put in separate 12v power for panels etc)

Yes the wire is mostly on the floor cos I've yet to lift it up!

The control part is of course is the  RJ (?) cable which does go from socket to socket so there's a convenient plug in point where needed.

 

[jamespetts]

I do not know about SPROG systems, so cannot advise on whether you need busses 7 and 8.

 

I do not think that you need buses 5 and 6 - bus 3 should be able to deal with this if it has sufficient amperage. If you need to distinguish between different lighting, local switches powered by your communications bus would be better than lots of different power buses.

 

Bus 4 is only necessary if you actually have 16v AC accessories. I am not aware of any, but you might have some. You may wish to consider whether 12v DC versions of such accessories can be procured to eliminate bus 4 entirely.

 

Make sure that all of your wiring is rated for a current no less than the maximum current that the power supply supplying that wiring, as limited by protection devices such as fuses, can supply.

 

For my own layout, I use the LocoNet bus instead of a DCC accessory bus, as this is a better general communication bus and is capable of two way communication, which is important for feedbacks. If you do not need any feedback (e.g. block occupancy) and will never need this in the future (e.g. for computer automation), a DCC accessory 'bus will suffice, but, if you have not already considered this carefully, you might want to give thought to an alternative data bus.

[Ian J.]

6 hours ago, Edward said:

Speaking as an absolute ignoramus of electricity etc  but as someone with 9 or so layout boards in a NCE system,  I think you would have an awful lot of wires and connections across!  I just have 4 supply wires going from the DCC output station to the 4 power districts following the "Star" principle.   My rule is "no Connections" across boards.  (But have had to put in separate 12v power for panels etc)

Yes the wire is mostly on the floor cos I've yet to lift it up!

The control part is of course is the  RJ (?) cable which does go from socket to socket so there's a convenient plug in point where needed.

 

 

Good points, I'll need to have a think about them, but I don't have a problem routing wiring across boards as my knowledge currently stands.

 

1 hour ago, jamespetts said:

I do not know about SPROG systems, so cannot advise on whether you need busses 7 and 8.

 

I do not think that you need buses 5 and 6 - bus 3 should be able to deal with this if it has sufficient amperage. If you need to distinguish between different lighting, local switches powered by your communications bus would be better than lots of different power buses.

 

Bus 4 is only necessary if you actually have 16v AC accessories. I am not aware of any, but you might have some. You may wish to consider whether 12v DC versions of such accessories can be procured to eliminate bus 4 entirely.

 

Make sure that all of your wiring is rated for a current no less than the maximum current that the power supply supplying that wiring, as limited by protection devices such as fuses, can supply.

 

For my own layout, I use the LocoNet bus instead of a DCC accessory bus, as this is a better general communication bus and is capable of two way communication, which is important for feedbacks. If you do not need any feedback (e.g. block occupancy) and will never need this in the future (e.g. for computer automation), a DCC accessory 'bus will suffice, but, if you have not already considered this carefully, you might want to give thought to an alternative data bus.

 

I don't believe the SPROGs have a comms buss such as LocoNet or the like. My impression is they are limited in that respect. Different return system but I don't think they don't support RailCom either. Layout automation by computer is something I'd like to do in future, but I'm not entirely sure how that is done with SPROGs.

[Ian J.]

Thought: how many LEDs can be powered of 1 amp of power (either 12V DC or 16V AC)?

[jamespetts]

If you are going to do computer automation, you definitely need a data bus that can do feedback. I don't know much about SPROGs, I am afraid, so you will have to investigate this, but if you think that you ever want to do computer automation, you need to be planning right from the start to have block detection and feedback, both in how you set up your track feeds to insulate at the right points and how you set up your data 'bus to be able to get the feedback back to your command station and then onto the computer.

 

As to LEDs - LEDs vary, so you will need to check the amperage for each and multiply it by the number that you need, allowing a margin of error.

[Ian J.]

I won't bother with automation until there's a better way to do it than with predefined block sections. It needs something more flexible like some kind of triangulation, and since that may never happen then I'm happy never to bother with it.

 

Re LEDs, I need simple examples, so how many LEDs suitable for say the light in a room in a house could be powered off 1 amp? Another might be how many street lights off 1 amp?

[jamespetts]

15 minutes ago, Ian J. said:

I won't bother with automation until there's a better way to do it than with predefined block sections. It needs something more flexible like some kind of triangulation, and since that may never happen then I'm happy never to bother with it.

 

Re LEDs, I need simple examples, so how many LEDs suitable for say the light in a room in a house could be powered off 1 amp? Another might be how many street lights off 1 amp?

 

An alternative to block occupancy detection is to use infra-red or optical point detectors. These can be used for full automation, but they have drawbacks - large and visible holes need to be cut into the track bed at multiple points and there are issues with many items of rolling stock not having sufficiently reflective undersides to be detectable without paint based modification. Generally, it is easier to use occupancy detection if one is starting afresh. I should note that dividing track into individual block sections is exactly what happens with track circuits in real life; if you are modelling a real location you can even go so far as to obtain the signalling diagrams and base your occupancy sections on real life track circuits, but it is not necessary to go that far to get automation working.

 

If you are content never to automate, then you can leave out block sections and not worry about a feedback bus, but it will be much more difficult to add automation later if you change your mind. Only you can decide whether you might change your mind.

 

As to LED current draw, I am afraid that I do not know without looking it up - it is likely to vary from LED to LED depending on brightness, efficiency and the values of resistors used (LEDs must always be connected to a power source via a resistor no matter the voltage). There may well be people more knowledgeable on the topic who are able to assist - otherwise, you might compute the necessary current draw by finding the specifications for relevant LEDs online.

[Crosland]

3 hours ago, Ian J. said:

I don't believe the SPROGs have a comms buss such as LocoNet or the like. My impression is they are limited in that respect. Different return system but I don't think they don't support RailCom either. Layout automation by computer is something I'd like to do in future, but I'm not entirely sure how that is done with SPROGs.

 

Watch this space...or, rather, our website

 

Andrew Crosland

www.sprog-dcc.co.uk

[Ian J.]

9 minutes ago, jamespetts said:

 

An alternative to block occupancy detection is to use infra-red or optical point detectors. These can be used for full automation, but they have drawbacks - large and visible holes need to be cut into the track bed at multiple points and there are issues with many items of rolling stock not having sufficiently reflective undersides to be detectable without paint based modification. Generally, it is easier to use occupancy detection if one is starting afresh. I should note that dividing track into individual block sections is exactly what happens with track circuits in real life; if you are modelling a real location you can even go so far as to obtain the signalling diagrams and base your occupancy sections on real life track circuits, but it is not necessary to go that far to get automation working.

 

If you are content never to automate, then you can leave out block sections and not worry about a feedback bus, but it will be much more difficult to add automation later if you change your mind. Only you can decide whether you might change your mind.

 

As to LED current draw, I am afraid that I do not know without looking it up - it is likely to vary from LED to LED depending on brightness, efficiency and the values of resistors used (LEDs must always be connected to a power source via a resistor no matter the voltage). There may well be people more knowledgeable on the topic who are able to assist - otherwise, you might compute the necessary current draw by finding the specifications for relevant LEDs online.

 

DCC needs to get far more advanced to allow such things as triangulation for accurate position detection to allow flexible coupling up in a station platform without relying on dead reckoning, so I'm content to leave the automation matter for another time (and maybe even another life, bearing in mind that technology in model railways seems move at a pace so slow it makes glaciers look embarrassed at how fast they go).

 

I'm sure I've asked the LED question (or something like it) on RMweb before, but it may be so long ago as to no longer be retrieved. It may even have been in one of the previous incarnations (pre 2009).

 

 

Edited July 14, 2021 by Ian J.

[jamespetts]

1 minute ago, Ian J. said:

 

DCC needs to get far more advanced to allow such things as triangulation for accurate position detection to allow flexible coupling up in a station platform without relying on dead reckoning, so I'm content to leave the automation matter for another time (and maybe even another life, bearing in mind that technology in model railways seems move at a pace so slow it makes glaciers look astoundingly fast).

 

I'm sure I've asked the LED question (or something like it) on RMweb before, but it may be so long ago as to no longer be retrieved. It may even have been in one of the previous incarnations (pre 2009).

 

 

 

Flexible coupling in a station platform can work reliably with dead reckoning - have a look at the "Coupling by Computer" video here for a demonstration (note that the server seems to be down at present so you may have to check back on this to-morrow).

[ISW]

2 hours ago, Ian J. said:

Thought: how many LEDs can be powered of 1 amp of power (either 12V DC or 16V AC)?

Ian J,

 

I bought a 5m length of 12v LED tape (the sticky-backed type) that you can cut at intervals along its length. I used it to light the underside of my Upper Level of Baseboards, so that I can see the Lower Level. It's on the underside of 10 baseboards with a common 2-wire bus. The whole lot is powered by a cheap 12v 2amp power supply I bought at a car boot for £2. It'd easily power much more, and I plan to power any scenic lighting from the same bus.

 

As an aside, I have 6 power districts, 1 lighting, and 1 MegaPoints buses on my Upper Level. On most boards that means 6x 2-wire cables to attach to the underside of the baseboards. At baseboard joints I have plug-n-socket 'chocolate block' connectors (12-way) and this has proven top be easy and reliable.

 

Here's the underside of a 'typical' baseboard:

 

And this is a close up of a typical baseboard connector and distribution veroboard, with the LED tape in view:

 

Hope this helps.

Ian

 

[Ian J.]

Reading up and asking a few questions of friends, I get the impression that a rough average LED power might be 10mA. Taking into account some advice to halve the amount of such into any available amperage value, it's suggested that about 50 such LEDs could work off 1 amp. If I had a town + shed 'vignette' for S&P (which will likely be the first vignette I do), then the amount of LEDs to cover street lamps, buildings, yard lights, etc, could easily be 200+ across the four boards that might eventually make up the vignette. That would work out at 4+ amps. If I have only the one 12v accessory buss, buss 3 (so no busses 4, 5 or 6), then that's going to leave somewhat less available amps for other accessories like turnout motors, turntable motor, etc.

 

I think in previous years when I've worked on this I came to the conclusion it would be better to separate the lighting from the other accessories so that each could be comfortably powered without detriment to the other, hence the reasoning for buss 5. Buss 6 could very easily be used for overall layout lighting using the kind of LED strips as previously mentioned by @ISW. That's the reasoning for having those busses.

 

The 16v AC buss is really a 'just in case' provision, I'm not sure it would ever actually be used, and might end up being repurposed.

[Ian J.]

2 hours ago, Crosland said:

 

Watch this space...or, rather, our website

 

Andrew Crosland

www.sprog-dcc.co.uk

 

Silly question I suppose as you would probably have said if you felt comfortable saying so, but are you hinting at a feedback buss, RailCom, or something else to do with computer control...?

[jamespetts]

14 hours ago, jamespetts said:

 

Flexible coupling in a station platform can work reliably with dead reckoning - have a look at the "Coupling by Computer" video here for a demonstration (note that the server seems to be down at present so you may have to check back on this to-morrow).

 

I should note that the video hosting platform is now back up, so you can see how dead reckoning works in practice.

[Michael Hodgson]

On 11/07/2021 at 23:21, Ian J. said:

 

Buss 1: DCC Track Power & Comms, 2 wires, +/-

Buss 2: DCC Powered Accessories & All Accessories Comms, 2 wires, +/-

Buss 3: 12V DC Power Accessories (no Comms), 2 wires, +/-

Buss 4: 16V AC Power Accessories (no Comms), 2 wires, +/-

Buss 5: Continuous Power 1 (for model lighting, etc), 2 wires, +/-

Buss 6: Continuous Power 2 (for model lighting or maybe layout lighting, etc), 2 wires, +/-

Buss 7: SPROG to SBOOST Comms, 2 wires, +/-

Buss 8: SBOOST Common, 1 wire, -

 

 

Yeah, its always the same.  You wait ages for a bus they eight come along at once.

 

Nos 5 & 6 seem to duplicate each other and No 3 or No 4 .  And AC isn't +/- of course.

If you put a bridge rectifier wherever you need DC you could dispense with No.3 and run it off no No.4

 

You have forgotten one though 

Bus 9 - Rail Replacement Service !

[Crosland]

13 hours ago, Ian J. said:

That would work out at 4+ amps.

 

If you are wiring your own LEDs (rather than using the ready made strip) then you can connect at least two LEDs in series across 12V, depending on the LED colour. Four or five red LEDs would work. The same current flows through all the LEDs so you still only need 10mA for each such series connection. You use a single, lower value, resistor so you also save on resistors and waste less power.

 

 

13 hours ago, Ian J. said:

Silly question I suppose as you would probably have said if you felt comfortable saying so, but are you hinting at a feedback buss, RailCom, or something else to do with computer control...?

 

We have two new products in the pipeline that have a connector for CBUS. This is a CAN bus based layout control bus that was developed by members of MERG. They have been delayed by post-pandemic component shortages, so I don't want to make too many promises (and don't want to do too much advertising on RMWeb). The R-Pi based 'Pi-SPROG 3 Plus' are being manufactured now and should be with us in a few weeks. For layout interfaces you will need to join MERG and build some of their kits, until we get our own in production, hopefully in the not too distant future. Some details are already on our website.

 

To head off a couple of questions, CBUS is not compatible with Zimo's CAN bus, nor the NMRA's LCC/OpenLCB

 

[Ian J.]

6 hours ago, jamespetts said:

 

I should note that the video hosting platform is now back up, so you can see how dead reckoning works in practice.

 

You've done some considerable work there and it does look impressive. However, it's still using fixed track positions for coupling and uncoupling, and I'm looking to be able to couple or uncouple anywhere, not dependent on pre-arranged magnet positions. The tech isn't in place in the model railway world to achieve that yet.

 

5 hours ago, Michael Hodgson said:

Yeah, its always the same.  You wait ages for a bus they eight come along at once.

 

Nos 5 & 6 seem to duplicate each other and No 3 or No 4 .  And AC isn't +/- of course.

If you put a bridge rectifier wherever you need DC you could dispense with No.3 and run it off no No.4

 

You have forgotten one though 

Bus 9 - Rail Replacement Service !

 

5 & 6 aren't direct duplications, it's not impossible that I'd use 6 for overall layout lighting, while 5 would be for the model lights (street lights, etc).

 

The +/- is to indicate number of wires, rather than polarity. That's still true of AC.

 

I'm not confident that I'd want to share sources for different power types by getting a 12V DC off the 16V AC, hence why it's treated as a separate bus, however as I've mentioned in another post, it could be that the buss for the 16V AC could be repurposed at a later date.

 

4 hours ago, Crosland said:

 

If you are wiring your own LEDs (rather than using the ready made strip) then you can connect at least two LEDs in series across 12V, depending on the LED colour. Four or five red LEDs would work. The same current flows through all the LEDs so you still only need 10mA for each such series connection. You use a single, lower value, resistor so you also save on resistors and waste less power.

 

 

 

We have two new products in the pipeline that have a connector for CBUS. This is a CAN bus based layout control bus that was developed by members of MERG. They have been delayed by post-pandemic component shortages, so I don't want to make too many promises (and don't want to do too much advertising on RMWeb). The R-Pi based 'Pi-SPROG 3 Plus' are being manufactured now and should be with us in a few weeks. For layout interfaces you will need to join MERG and build some of their kits, until we get our own in production, hopefully in the not too distant future. Some details are already on our website.

 

To head off a couple of questions, CBUS is not compatible with Zimo's CAN bus, nor the NMRA's LCC/OpenLCB

 

 

Thanks for the LED info, I need to remember such things as series wired LEDs in order to better use the power available.

 

I'll look into the MERG CBUS, to see how it works and see if I can take it into account in my buss arrangement.

[Ian J.]

OK, so I've looked up a number of pages on t'internet regarding the MERG CBUS, but none seem to explain what the + and 0 connectors do on a CBUS module, all of which seem to have a four port connector, called mostly either J1 or J2. L and H I can see are the actual CBUS pair, but I don't know what the others are?

 

@Crosland Will existing SPROGs be usable with your implementation of CBUS, bearing in mind they don't appear to have any way of connecting to it?

[melmerby]

On 14/07/2021 at 19:26, Ian J. said:

I won't bother with automation until there's a better way to do it than with predefined block sections. It needs something more flexible like some kind of triangulation, and since that may never happen then I'm happy never to bother with it.

 

Re LEDs, I need simple examples, so how many LEDs suitable for say the light in a room in a house could be powered off 1 amp? Another might be how many street lights off 1 amp?

Isn't that what the real railway has? With signals controlling them.

In that way computer automation is very prototypical.

 

Triangulation seems a right faff with it unlikely to ever be practical enough to allow the flexibilty of fixed block working.

[jamespetts]

7 hours ago, Ian J. said:

 

You've done some considerable work there and it does look impressive. However, it's still using fixed track positions for coupling and uncoupling, and I'm looking to be able to couple or uncouple anywhere, not dependent on pre-arranged magnet positions. The tech isn't in place in the model railway world to achieve that yet.

 

There are such things as DCC controlled automatic uncouplers, but they are a little marginal with current technology. I looked into them but decided that it would be far easier (and possibly more reliable) for my own purposes to use fixed magnet uncouplers.

 

So long as you are content that you will not want to automate this layout, you should be fine not to install a feedback bus and to use a DCC bus. Whether you want to automate is a matter for you.

[AMJ]

There is way too much duplication.  From what I have seen in Larry Pucketts videos on YT you could get away with just your DCC and AC feeds as sub busses for lower voltages AC/DC can be taken from the main higher voltage one.  You can make the decision do you need a certain voltage on this section?

See

https://www.youtube.com/results?search_query=dcc+guy+power+bus

 

[Ian J.]

9 hours ago, melmerby said:

Isn't that what the real railway has? With signals controlling them.

In that way computer automation is very prototypical.

 

Triangulation seems a right faff with it unlikely to ever be practical enough to allow the flexibilty of fixed block working.

 

You're missing my point. I am looking for mm-accurate, anywhere-on-layout coupling / uncoupling. That simply can't be served by any kind of sections. The usage of 'block sections' on the real railway is for the movement of trains along their journey, not for knowing precisely where a train is on the track.

 

Triangulation is for being able to know precisely where a rail vehicle is. Think of it as like using a highly accurate version of GPS on the real railway so that you know not just what section a train is in, but precisely where it is in that section. This has to be known so that coupling up by automation can be achieved anywhere (not just at fixed magnet positions as in the 'current' dead reckoning systems).

 

As I've said, the tech within the model railway world isn't there to do this yet.

[5BarVT]

10 hours ago, melmerby said:

Triangulation seems a right faff with it unlikely to ever be practical enough to allow the flexibilty of fixed block working.

Something similar was allegedly so simple that it would be in place on WCML in time for Virgin’s Pendolino revolution!

Paul.

[melmerby]

30 minutes ago, Ian J. said:

 

You're missing my point. I am looking for mm-accurate, anywhere-on-layout coupling / uncoupling. That simply can't be served by any kind of sections. The usage of 'block sections' on the real railway is for the movement of trains along their journey, not for knowing precisely where a train is on the track.

 

Triangulation is for being able to know precisely where a rail vehicle is. Think of it as like using a highly accurate version of GPS on the real railway so that you know not just what section a train is in, but precisely where it is in that section. This has to be known so that coupling up by automation can be achieved anywhere (not just at fixed magnet positions as in the 'current' dead reckoning systems).

 

As I've said, the tech within the model railway world isn't there to do this yet.

Obviously never heard of onboard dcc controlled uncoupling then.

You can couple and uncouple anywhere on the layout.