Jump to content
 

Can DCC feed trains from hidden loops?


number6

Recommended Posts

I'd appreciate some advice on if it is possible to semi automate using DCC so a set of hidden loops on a circuit can feed trains out to [what I'm calling] my outer home. I'd quite like to replicate the receiving of trains from the next box down the line and that they are offered up to me by that virtual box. I have room between the exit from the loops for at least two blocks before the scenic section of the layout. Assume that the final block ends just on the scenic part of the layout.

 

Sequence would be as follows:

Lets say there are 3 trains in 3 loops.

At some trigger Train A sets off under DCC control from the loops and come to a halt at the outer home. From there I take over and drive it around the layout to eventually end up back in the hidden loops [not necessarily the same loop], or shunt a pick-up goods, loop it or generally do what we do when playing trains!

 

As soon as the section behind the outer home is free of TrainA, TrainB sets off from the loops crawls up to the outer home and the process repeats.

 

There is also the possibility that I could either let the up or down line run itself by leaving the outer home at clear so a procession of trains run a complete circuit without stopping apart from in the hidden loops. I've seen layouts working like this but its the semi-manual part where I want more control. This plan is for a long thin layout which I'd mostly operate on my own [sad eh?!] so it would also be good to have a signal box description system so I know what is where.

 

I like the idea that I could glance down the line to see something creeping up to the signal before I accept it if the next block is clear and then taking it over. It would also introduce an element of chance in terms of not really knowing what train is up next.

 

Where to start?

Is it possible?

What gubbins are involved?

Can it be signal controlled so the process is like operating a box? Accepting or not accepting trains depending on the state of the next section etc.

Can you take control while the train is in motion or does it have to come to a halt first?

Can I switch the auto-train-feed off [can I call it that?!] and run the whole thing manually?

 

Apologies if this the most basic and idiot of questions.

Cheers

Raphael

Link to post
Share on other sites

You could certainly do this with a computer. Otherwise there are some hardware systems such as the Lenz ABC, but they only work with certain makes of decoder and the logic needed to switch them on and off is a bit tricky. On the whole this is a bit harder to to in DCC than in DC.

Link to post
Share on other sites

I was wondering a similar kind of thing. The layout I'm about to build in my loft will basically be a big roundy roundy but the fiddle yard will be at a lower level than the rest of the layout. I then get scenery all the way around the loft with hidden fiddle yard beneath. Best of both worlds. However how would I know when my trains reach the fiddle yard and how can I leave the main lines running if I want to do some shunting in the yards etc with different trains entering and leaving the fiddle yard as I don't want say two trains constantly whizzing around. I could do this in DC but as I'm not too up on the DCC thing I too would be interested to know how!!

Link to post
Share on other sites

There are a few ways that you could do it, the most straightforward will be to use a computer with CTI or Railroad&Co software, but it will take quite a bit of setting up and be quite expensive for just a simple requirement.

 

If you don't want to use a computer you can use any DC method that just stops and starts the trains using simple isolating sections and point switchers and fit Lenz BM1 modules to the isolating sections. This will require you to have Lenz or Zimo decoders in your locos configured to use the Lenz Automatic Braking Control (ABC) feature and the constant braking distance will need to be set. You get to drive the trains when they are visible, then just leave them running in the hidden section where the electronics will stop and start them as required.

 

If you are not worried about the trains stopping suddenly in the hidden sections then you will not need the BM1s there, just on the outer home and any visible blocks where you want the train to slow and accelerate smoothly.

 

I would advise against using brake on DC or braking signal insertion modules as this is a lot more complex electrically to set up and limits the control you have of the trains while stopped. Only ABC or computer control will let you override a danger signal manually from the controller, and operate functions like tooting the whistle.

Link to post
Share on other sites

  • RMweb Gold

Yes I think it can be done without a PC but probably just as expensive.

Not too sure which would be the easiest route (pardon the pun) to take.

As a service engineer I like the electronic boards and wire up accordingly, others prefer to use software and learn how to use that, each to their own do what you feel comfortable with.

 

I have managed to automatically get trains to stop and start using Heathocte electronics IRDOTS as the detectors and the shuttle boards, video below.

I have used two of the dual timers and a relay board, the relay board switches the DC to the rails, it is all activated via the IRDOTS, there are two, one on the inner track and one on the outer track at eack end just under the tunnels.

The layout is controlled by DCC, the front section of both loops is isolated on both rails from the rest of the track, so when the loco passes the IRDOT dc is switched to the track it is running on which causes the loco to slow to a stop, going via the dual timer to keep the train stopped for an adjustable period of time. It also resets the relay board to put DCC back onto the other loop which causes the loco to speed up and drive off.

The acceleration and deceleration settings are controlled by CV's in the decoder.

 

th_Autostationstopstart.jpg

 

To use with DCC you need to isolate sections of track and via a relay switch DC to the isolated section.

Using the 'Brake-on-DC' function of the decoder.

 

In most decoders you can set DC on or off in CV29.

If you set it to off when the decoder detects DC on the track it will slow the train to a stop, when DCC is reconnected the train will speed up and carry on, hence the need to isolate the section of track where you will be switching DC and DCC to the rails.

 

The other alternative is to use DCC BIT Switches, I've got and I'm using the station stop board which works very well, these operate by again using the brake on DC function of the decoder.

Not cheap but do work well link to web site below:-

 

http://www.dccbitswitch.com/

 

Lots of very useful information of how it all works and all of the different boards that he makes and how to wire them up, also gives good information on setting up blocks which will keep trains apart on the same loop.

The can be activated by reed switches on the track and magnets in the Engine or coach.

You could even use the IRDOTS to operate them.

 

The other alternative is to use a PC, and any of the software programs, although I have no idea which software would be best suited to your exact application.

 

As I have the NCE Powercab I could now use the mini panel, which you can program to do all what you need, this wasn't available at the time I was playing the other stuff.

The mini panel can be programmed so that a certain loco will move off at the set speed pause operate a point and so on. Still need to get to grips with programning it though.

More info on the NCE website with a sample program www.ncedcc.com

 

Hope this proves a little helpful.

 

Cheers

 

Ian

Link to post
Share on other sites

Many thanks for the replies. Interesting that this would seem easier in DC! I can almost imagine how that would work.

 

The cost with the computer control is the machine or the software? I'm assuming the later. I notice these bits of electronics aren't cheap.

 

Thanks for the info on the dccbitswitch - looks like if I scale back the idea to just feeding trains out of the loops and then halting them it is entirely possible. Its the smooth take-over thing that will have to be something for the future maybe. Perhaps the way to go is just through train description in the hidden loops, with some form of block control in there leaving the scenic section 'all manual' - ie no blocks.

 

I would have thought something like what I was after would be fairly popular - reducing the need to be behind the layout and bringing operators around the front. Its one of my bugbears of many exhibition layouts that all the action and interesting bits are often around the back. I've wondered if there couldn't be a back to front exhibtion staged where all layouts have fiddle yards to the front? :lol:

 

I am keen on systems that have some relation to the real railway - always liked hand thrown points, working point rodding, cables to signals, wire in tube, old fashioned interlocking etc. The ability to 'drive' locos and trains is very liberating in DCC where before in DC it appeared to be under the control of the current - but which ironically is more like traditional signalling - where you can only move where directed by the signals.

 

Maybe DCC is better suited to [unsignalled areas] yards and depot models. It worked for me on my own small yard layout which needed nothing more complex than turnout control.

 

Some more investigation required.

Cheers

R

Link to post
Share on other sites

DCC works very well for signalled layouts but it does rather assume something with some intelligence is doing the work. The positive side of that is you can do things far more complex than anything in the DC world, the negative is it tends to want a computer sitting in it and then scripted. It also frequently means you can do cool new things without rewiring and stuff you couldn't do without it. Full interlocking for example becomes a small software problem not a bit electronics one.

 

One real weakness in this area with DCC has been transponding - or the lack of it. The newest setups allow the computer to read which loco is on a section of track. That makes it much easier to do things like automatic fiddle yards as the computer can switch to a track, read the loco address and then drive it to the signal for you to take over, while conventional computer DCC setups tend to rely on the right train being put in the right fiddle siding so it knows what it is driving.

Link to post
Share on other sites

The answer is certainly you can do it with DCC and a computer. I know, because I do it.

 

Choices for software are JMRI, RR&Co, KAM & probably a couple of others. JMRI has the advantage of being under GNU licence, so doesn't cost anything. That being said, RR&Co has a decent trial mode.

 

Expect to spend a fair amount to manage to do what you are wanting to do. For my "short" layout, I use something like 48 detection sections & 16 accessory outputs. To manage without the ability to operate under computer control across the front would take around 24 sections. Using the more current versions of RR&Co, setting stop indicators & so forth, it would appear as though as few as 6 detection blocks could be used...but, I have fears of bad things adding up quickly with it. (TTG can comment best of all on here...John Dew is another one with experience of RR&Co Gold)

 

So, the short answer is that yes, DCC will do it, but that to do anything involving automation reliably, you need to think about failure modes & what you don't want happening too...otherwise, when it goes in the pot, it will be spectatcular.

 

James

Link to post
Share on other sites

Yes I think it can be done without a PC but probably just as expensive.

Not too sure which would be the easiest route (pardon the pun) to take.

As a service engineer I like the electronic boards and wire up accordingly, others prefer to use software and learn how to use that, each to their own do what you feel comfortable with.

 

I have managed to automatically get trains to stop and start using Heathocte electronics IRDOTS as the detectors and the shuttle boards, video below.

I have used two of the dual timers and a relay board, the relay board switches the DC to the rails, it is all activated via the IRDOTS, there are two, one on the inner track and one on the outer track at eack end just under the tunnels.

The layout is controlled by DCC, the front section of both loops is isolated on both rails from the rest of the track, so when the loco passes the IRDOT dc is switched to the track it is running on which causes the loco to slow to a stop, going via the dual timer to keep the train stopped for an adjustable period of time. It also resets the relay board to put DCC back onto the other loop which causes the loco to speed up and drive off.

The acceleration and deceleration settings are controlled by CV's in the decoder.

 

th_Autostationstopstart.jpg

 

To use with DCC you need to isolate sections of track and via a relay switch DC to the isolated section.

Using the 'Brake-on-DC' function of the decoder.

 

In most decoders you can set DC on or off in CV29.

If you set it to off when the decoder detects DC on the track it will slow the train to a stop, when DCC is reconnected the train will speed up and carry on, hence the need to isolate the section of track where you will be switching DC and DCC to the rails.

 

The other alternative is to use DCC BIT Switches, I've got and I'm using the station stop board which works very well, these operate by again using the brake on DC function of the decoder.

Not cheap but do work well link to web site below:-

 

http://www.dccbitswitch.com/

 

Lots of very useful information of how it all works and all of the different boards that he makes and how to wire them up, also gives good information on setting up blocks which will keep trains apart on the same loop.

The can be activated by reed switches on the track and magnets in the Engine or coach.

You could even use the IRDOTS to operate them.

 

The other alternative is to use a PC, and any of the software programs, although I have no idea which software would be best suited to your exact application.

 

As I have the NCE Powercab I could now use the mini panel, which you can program to do all what you need, this wasn't available at the time I was playing the other stuff.

The mini panel can be programmed so that a certain loco will move off at the set speed pause operate a point and so on. Still need to get to grips with programning it though.

More info on the NCE website with a sample program www.ncedcc.com

 

Hope this proves a little helpful.

 

Cheers

 

Ian

 

 

If using Gaugemaster Prodigy you cannot run on DC

Link to post
Share on other sites

One real weakness in this area with DCC has been transponding - or the lack of it.

The newest setups allow the computer to read which loco is on a section of track. That makes it much easier to do things like automatic fiddle yards as the computer can switch to a track, read the loco address and then drive it to the signal for you to take over, while conventional computer DCC setups tend to rely on the right train being put in the right fiddle siding so it knows what it is driving.

Am I right in assuming you are referring to RailCom ?

RailCom combined with computer software would appear to provide a suitable solution for this sort of requirement. Hopefully the hardware will start to appear in the near future.

Link to post
Share on other sites

  • RMweb Gold

If using Gaugemaster Prodigy you cannot run on DC

 

What you are saying is correct, if using the Gaugemaster or many other systems they won't let you run a DC(no decoder fitted) loco, therer is no address zero to run a DC loco on these systems.

 

I think you may have misunderstood what is actually happening.

All of the locos have been fitted with a DCC decoder so they will all run on DCC (or DC via a standard DC controller)

 

Everything in the video is controlled by the NCE Powercab (which also can't run a DC non decoder fitted loco)

 

The whole layout is run on DCC, all I've done is isolated a section of track and switch DC into this section of track, the Decoder in the loco is set so that when it detects DC it will brake to a stop.

As this section is isolated when the IRDOT detects the loco, it activates a relay which switches the DCC off and applies DC.

This does mean you have no control of the loco in this section until the relay is reset and I do believe if it's a sound chipped loco the sounds will turn off.

All I was aiming for was to automatically slow the train to a stop for an adjustable time then allow it to continue running, without the need for a PC.

The DC is a fixed voltage about 1volt below the DCC voltage.

 

It's not perfect but it does work, and I just wanted to have a play about with the Heathcote electronic cards to see what I could do as they are much cheaper than the DCC Bit Switches.

 

It's much better now though because of the NCE mini panel, this you can program to do many things depending on what state the inputs plugged into it are at.

Select the loco you want to move, which direction, what speed, when it enters a block you can operate decoder functions such as sound the horn and you can throw points to send the train along a route. May take a bit of trial and error to get the program right but it does look fairly straight forward.

 

It does look limited in the amount of automation it can do, but for basic functions and movement looks like it could work very well.

The Macro Panel looks like it could go even further, but this is not available as yet, so I may wait until this is available and have a play with that.

The main issue is you would need to be using the NCE systems to use it!

 

Cheers

 

Ian

Link to post
Share on other sites

  • RMweb Gold

The whole ethos of using DCC for me is the fact that you drive a loco and have to obey signals etc. That fact is what makes what you want difficult. Having installed a system where you drive the loco you are then putting in something else to do the driving. If you want it automated DC can be simpler because you control the power before it reaches the track not within the loco. Jim Russsel devised a system using contacts and delay circuits to send out trains from his hidden sidings on his Little Great Western back featured in RM 1962. I also remember an article in Model Railroader for a microprocessor controlled hidden sidings.

Donw

Link to post
Share on other sites

The whole ethos of using DCC for me is the fact that you drive a loco and have to obey signals etc. That fact is what makes what you want difficult. Having installed a system where you drive the loco you are then putting in something else to do the driving.

 

Granted. I'll carry on investigating this but may end up with a mirror set above the sidings so I can see the hidden trains and some form of identifying the decoders sat waiting to go - perhaps written on a blackboard in chalk!

 

Is the reverse of what I described in the op any easier to devise - [forget the whole smooth take over thing]? Seems to me that some of what we've discussed is actually much more useful for the other end of this process where trains can be left to make their own way back into an empty hidden loop, to run in and come to a smooth halt. In my hair-brained scheme I was also going to have a long section of blocks before the loops so the trains queued before reaching them. The loops just give choice in what runs out next. These trains are going to have to make a smooth stop as my stock won't stand any emergency stops on a regular basis.

 

Have I got it right that 'automatic block working' can be both software or hardware based and that with train/decoder describing it would be possible to identify the available trains standing at the exit to the loops.

 

peach james, I love the description of a 'going in the pot spectacular'

 

cheers

R

Link to post
Share on other sites

Granted. I'll carry on investigating this but may end up with a mirror set above the sidings so I can see the hidden trains and some form of identifying the decoders sat waiting to go - perhaps written on a blackboard in chalk!

 

Is the reverse of what I described in the op any easier to devise - [forget the whole smooth take over thing]? Seems to me that some of what we've discussed is actually much more useful for the other end of this process where trains can be left to make their own way back into an empty hidden loop, to run in and come to a smooth halt....

 

Fairly cheap and simple with either Digitrax or NCE. I don't know for others systems but suspect its not hard.

 

Digitrax - use "dispatch" on the throttle, and then either computer script or CML Electronics "Shuttle" hardware board to pick up the dispatched train and drive it. Track detectors indicate to the automation when to slow/stop the train.

NCE - use the "MiniPanel" to drive the train. Depending on how the MiniPanel is scripted, may require one MiniPanel for each simultaneous train movement (ie. one panel can drive many trains, but if its in a "wait" loop, the script must finish that before running another script).

 

Fine details of both depend on local installations, how/where sensors are positioned, type of sensor used (IR and Reed switches give an instant notice of position, track current gives a longer area of detection), etc...

 

 

 

 

Back on train identification in the fiddle yard... If the layout is fully equipped with occupancy circuits, then once a train name/number is known to a computer system, that name/number can be tracked around the entire layout. Therefore, a computer system could display a list (in very large screen print!) of which trains are in which roads in the storage area. If prepared to tell the computer the initial state of all trains, either on boot, or as a train is driven, (or you always start from same position of trains), then its not necessary to have any form of train identification (RFID, RailCom, Transponding, Lissy, etc.).

 

 

- Nigel

Link to post
Share on other sites

Thank you Nigel.

Have been reading upon Digitrax and the CML Electronics "Shuttle", Need to have a read of the NCE stuff. Have also looked at the DCC-bitswitches - mentioned by Ian earlier. I think I'm getting the hang of how these things work.

 

These DCCBitswitches seem to do nearly what I'm after. If on entry to the hidden yard I had a sequence of Block Bitswitches followed by a couple of Yard BitSwitches trains would queue to enter free loops, then the exit from this yard would be controlled by a toggle switch which would feed a train at a time from selected loops to another block after the yard where they come to a halt on the mainline. Costly though...

 

But have I got this right?: if using any of these software or dccswitch methods any stationary trains are actually only being held waiting to 'run' opposed to being stopped and switched 'off' - they all rely on the last manual command to the loco decoder being: 'run'. I presume I need to readdress the decoder while the next free train waits at that final block signal in order to manually drive it away. At the other end of the layout I set it running into the hidden section to be automatically fed through.

 

cheers R

Link to post
Share on other sites

DCCBitSwitches have always struck me as an expensive way of doing things. They have three primary methods of stopping trains; two hold the train on the brakes using features supported in some decoders (Brake on DC (many decoders do this) and Assymetric Braking (Lenz, Zimo and some, but not all, CT decoders), the other sends a genuine DCC Stop packet (any decoder). But, all three revert back to whatever the DCC system was doing when released. So yes, the train will resume its original speed.

(You can build DIY Assymetric Brake modules for Lenz or Zimo decoders for 30p each, they need six rectifier diodes, or Lenz will sell you them for about £10).

 

 

The other ways I talked about (CML shuttle or NCE MiniPanel) are genuinely driving the train on your DCC system by sending commands into the command station. So, when stationary, the train will be stopped as far as the command station is concerned. If you want to drive a train again it has to be selected and driven, either by a manual throttle, or an automation device.

 

My own approach would be to throw a computer in the middle, and get it to drive the trains against occupancy detectors and signals (signals could be virtual ones in software only). RocRail and RR&Co can do this, its what they are designed to do. I think JMRI could do it as well.

 

 

 

I have a sketch design for the NCE MiniPanel to do roughly what you have asked without a computer. Its only a sketch until my friend who wanted to do this has built the baseboards and enough test track to prove it. We expect to be using a NCE Power-Pro system and three or four MiniPanels when finished, each panel dedicated to different tasks.

 

Rough scheme to park a train:

Panel with one button per loco. Picture of loco (and perhaps its DCC number) to explain to operators what button does.

When pressed, the panel macro drives the selected loco at a pre-set speed (or if the loco is running on a throttle, the macro will take over from the throttle), then wait for a "stop" sensor message.

The stop sensor is a IR (or reed switch) in the storage road.

Once the stop is detected, stop the train and end of macro.

The sensor used for "stop" depends on which storage road is selected. This can be done two ways, either one road per loco (ie. always use the same road), or user selects the road (another button on the panel to do this) and some relays/microswitches attached to the turnouts (same concept as frog polarity switching) determines which sensor is connected to a single input on the panel for all scripts.

 

Rough scheme to bring a train out of storage:

Panel with one button for each loco.

When pressed, the loco starts to drive forward at pre-set speed. Macro waits for "stop" sensor message.

Once stop sensor is triggered, stop the train and end macro.

In this case, there is one stop sensor which is common to all macros, at the end of the "feed" line going to the layout.

The turnouts in the yard are selected by the train driving up to the "rear" of them and tripping a sensor which then throws the turnouts before the trains short out. This could be done with Hare decoders for Tortoise motors (expensive, but the connections/capability are provided), or could be IR beams or reedswitches which trigger the turnout motors (either directly throw the motors, or indirect via DCC system, there are dozens of solutions for this).

 

In both cases, the panels would have an "emergency quit" button which would stop the macro (think its the 31st connection on the MiniPanel).

 

 

Does this help the comprehension ?

 

 

regards

 

- Nigel

Link to post
Share on other sites

  • RMweb Gold

Hi Nigel,

 

That sounds quite good.

Only problem is for those folks NOT using the NCE DCC systems, they can't use the mini panel!

 

So I agree the DCC bit switches are an expensive way of doing things but they can be used by any DCC system and suitable decoder.

If people want to use the NCE mini panel then they would need to buy a NCE DCC system.

If they want to use more than one mini panel then they would need to buy the Smart Booster if using a Powercab or they would need to buy the PowerPro system which can handle more than one extra controller.

 

I'm just making my mind up on which block detectors to use; the NCE BD-20 or Heathcote Electronics IRDOTS for the inputs.

Then I need the Snap-it's and Mini panel, Once I've got this little lot I'm going to have a play about seeing what I can get it to do.

 

Cheers

 

Ian

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...