Video: wires under the baseboards

[Andymsa]

24 minutes ago, jamespetts said:

 

The amount of track will affect the number of wires as each occupancy section must have its own pair of droppers going to the occupancy detector. The more track that one has in any given area, the more sections that there are in that area, and thus the more wires that there will be. The smaller the scale and gauge, the more track that will fit into a given area for any given track layout normalised as to scale, and thus, all other things being equal, the smaller the scale, the greater the number of dropper wire pairs in any given area of baseboard that one will have. Certainly, there are not enough wire colours for all the sections on this layout.

 

The District Cut-Outs are not completely necessary in the sense that the model railway can run without them, but it does make it an awful lot easier to know approximately whereabouts that a short circuit is and for that short circuit to affect only part of the layout rather than all of it. Also, these add relatively little to the amount of wires.

 

Of course, the whole wiring could be simplified drastically without occupancy detection - but then there would be no automation. As previously written, it is a matter of balancing function with the additional work involved - just like the extra work in things like weathering and detailing models or any other sort of modelling.

Having around 860 detection sections and climbing that would need a lot colours. I assign 16 different colours to each board. Also the wires run loosely run in cable trunking which makes tracing easy. As there can be multiple droppers for one detection section the amount of wires in HO can be the same as in N most likely.
 

as I have 7 boosters that only shut down that specific zone when a short occurs in that zone I see no need in my case for additional cutouts.

 

i have to disagree with the wiring could be simpler without occupancy, I have shown it can be simple and have occupancy as each board has defined wire that goes to it. In your example of the railcom detector you just show red and white wires going to the detector and there appears to be no identification of these wires, so if the detector had to be changed how would it be obvious where each wire goes.

[jamespetts]

11 minutes ago, Andymsa said:

Having around 860 detection sections and climbing that would need a lot colours. I assign 16 different colours to each board. Also the wires run loosely run in cable trunking which makes tracing easy. As there can be multiple droppers for one detection section the amount of wires in HO can be the same as in N most likely.
 

as I have 7 boosters that only shut down that specific zone when a short occurs in that zone I see no need in my case for additional cutouts.

 

i have to disagree with the wiring could be simpler without occupancy, I have shown it can be simple and have occupancy as each board has defined wire that goes to it. In your example of the railcom detector you just show red and white wires going to the detector and there appears to be no identification of these wires, so if the detector had to be changed how would it be obvious where each wire goes.

 

I am having great difficulty in following your reasoning. Why does having multiple droppers in a section mean that the amount of wires in HO will be the same as in an equivalent N gauge layout? What you write is somewhat ambiguous - you write that it "can" be the same. Of course, it is possible for a given N gauge layout to have the same amount of wiring per square meter than a given H0 gauge layout; but that does not negate the fact that, all things being equal, any given N gauge layout will tend to have more dropper wires per square meter than any given layout of a larger scale. Can you clarify exactly what point that you are trying to make to what end here? I am unclear what one is supposed to take from what you have written on this subject.

 

As to boosters and cutouts - I have only one booster on this layout, as N gauge locomotives draw less current, so use the DCOs instead to separate the layout into districts rather than boosters. Again, can you clarify what point that you are ultimately trying to make here and to what end?

 

As to the level of complexity relating to occupancy, I am afraid that I do not understand what you have written on the subject. Are you trying to claim that the wiring on my specific layout could be simpler with no loss of function? If so, then I suggest that you have insufficient information to make that judgment (as demonstrated by the lack of any clear explanation as to precisely what specific wires can be omitted from my specific layout with no loss of function). If you simply mean that a different layout with a different scale and a different density of track may well have fewer wires, then that is no doubt true, but I do not understand what reason that you have for posting this. Can you elaborate?

 

I am afraid that it is extremely unclear what you are trying to achieve by these posts - it would help if you could explain your purpose in posting much more clearly. At present, you appear to be coming across as insisting that my specific layout ought to have been wired differently without having any real understanding of how it is currently wired and why it is wired thus, and not being able to give any clear explanation of what specific things ought to be different and for what specific reasons that those specific things ought to be different, which comes across as irrationally hostile. Is this what you are trying to suggest, or do you mean something else? If something else, what is that something else, exactly?

 

If I have misunderstood your intention, my apologies: but I am really struggling to understand what the purpose of your posts this evening is.

[Andymsa]

10 minutes ago, jamespetts said:

 

I am having great difficulty in following your reasoning. Why does having multiple droppers in a section mean that the amount of wires in HO will be the same as in an equivalent N gauge layout? What you write is somewhat ambiguous - you write that it "can" be the same. Of course, it is possible for a given N gauge layout to have the same amount of wiring per square meter than a given H0 gauge layout; but that does not negate the fact that, all things being equal, any given N gauge layout will tend to have more dropper wires per square meter than any given layout of a larger scale. Can you clarify exactly what point that you are trying to make to what end here? I am unclear what one is supposed to take from what you have written on this subject.

 

As to boosters and cutouts - I have only one booster on this layout, as N gauge locomotives draw less current, so use the DCOs instead to separate the layout into districts rather than boosters. Again, can you clarify what point that you are ultimately trying to make here and to what end?

 

As to the level of complexity relating to occupancy, I am afraid that I do not understand what you have written on the subject. Are you trying to claim that the wiring on my specific layout could be simpler with no loss of function? If so, then I suggest that you have insufficient information to make that judgment (as demonstrated by the lack of any clear explanation as to precisely what specific wires can be omitted from my specific layout with no loss of function). If you simply mean that a different layout with a different scale and a different density of track may well have fewer wires, then that is no doubt true, but I do not understand what reason that you have for posting this. Can you elaborate?

 

I am afraid that it is extremely unclear what you are trying to achieve by these posts - it would help if you could explain your purpose in posting much more clearly. At present, you appear to be coming across as insisting that my specific layout ought to have been wired differently without having any real understanding of how it is currently wired and why it is wired thus, and not being able to give any clear explanation of what specific things ought to be different and for what specific reasons that those specific things ought to be different, which comes across as irrationally hostile. Is this what you are trying to suggest, or do you mean something else? If something else, what is that something else, exactly?

 

If I have misunderstood your intention, my apologies: but I am really struggling to understand what the purpose of your posts this evening is.

 

the point about DCOs is why use them on a smallish layout as by your own omission you have one booster. My reasoning is that trains will still be running on other parts of the layout and could result in you taking your eye off the ball so to speak, whilst you deal with the short. I speak from experience. Also the additional joins in wires from device to device could be an issue if a wire let’s say became dislodged. I have not said about removing wires but removing hardware to simplify things.

 

ok to make it clear the wiring looks like a rats nest with no logical structure, loconet cables are cable tied to other cables, loconet cables should be kept away from other cables. With all the complexity of your wires these wires should be more identifiable which is the main point I’m trying to make. And I’m willing to bet you will have problems at some point, we all do

 

im not being hostile at all, It’s your layout at the end of the day, just trying to give the experience I have gained and avoid the pitfalls I have made. I may not know everything but I am always open to alternate ideas, hence why I have started to post my layout updates for positive or negative feedback. But if you don’t want to hear the experiences and feedback I’ll go away, and if I have offended in way I apologise.

 

 

[jamespetts]

I am sorry if I have misunderstood: I have found it quite hard to follow in places, especially as in your earlier post this evening you wrote specifically about the amount of wiring,

 

...this amount of wiring is just making things hard

 

whereas you now state that you are referring not to the amount of wiring, but rather its layout and the amount of other hardware (seemingly the DCOs).

 

I am not entirely sure that I understand the suggestion that it is always sensible to stop trains on other parts of the layout while a short circuit occurs; it is not clear that this should necessarily be the case (especially as the running is automated, and thus intended to be unattended in any event), and could be quite disruptive. Trains can quite easily be stopped manually either on the computer using the "ESC" key or with the emergency stop buttons in various locations around the layout. Also, as previously written, being able to track the fault giving rise to a short down to an area one seventh the size of the whole layout makes fault finding vastly easier than it might otherwise be. It is difficult to see why this latter advantage alone does not outweigh the relatively modest amount of additional wiring to which the DCOs give rise - can you elaborate?

 

The DCOs also have another important advantage. The booster's output is 3A. The DCOs are each set to 1.5A. The dropper wires used are all 7/0.2, and thus rated at 2A. Upstream from the DCOs, I use much thicker wiring (32/0.2), rated for 4.5A (and I also use 16/0.2, rated for 4A between the DCOs and the occupancy sensors). The 16/0.2 wire is too thick to solder neatly to the rails, but the 7/0.2 is a good thickness. Using the DCOs, I can use the 7/0.2 wire to the track and ensure that it is  protected from ever conducting a current in excess of its rating; otherwise, should a fault condition occur on the layout causing a section of track to draw just under 3A, the 2A rated wires might overheat without the booster registering a short circuit, which, if left unchecked for long enough, could cause a fire in the flammable wooden baseboards and thus be very dangerous. (This latter point, incidentally, is based on advice given to me by a club member who worked in process control/automation in the oil and gas industry for decades; this issue seems to be poorly understood among railway modellers). The use of DCOs thus allows the safe use of 7/0.2 wire whilst still allowing locomotives in all seven districts between them to draw up to 3A, so long as not more than 1.5A be drawn in any given district. a

 

As to the organisation of the wiring, the difficulty is the limited amount of space under the board to run the wires; it is difficult to see how there could possibly be space for totally separate wire runs for different types of wires. If there is a better way of organising the wires that will not enormously increase the workload and will fit into the space available, do let me know so that I can consider it for future projects. As I wrote above, your wiring is very neat, but it is difficult to see how I could replicate that given the density of trackwork that I have  and also the different hardware that I use.

 

In relation to LocoNet wires being run next to other wires, may I ask whether there is any literature on this that I can study to understand this issue more thoroughly from a robust, empirical perspective? This is not a suggestion that I have seen before, and it is not written in the manuals for the LocoNet products that I use, so far as I recall.

[RFS]

7 hours ago, jamespetts said:

I am not entirely sure that I understand the suggestion that it is always sensible to stop trains on other parts of the layout while a short circuit occurs;

 

Any short circuit that occurs when using automation is always going to be disruptive. If this is a home layout, and you are most often the only operator, do you really want trains running on the rest of the layout when one part has shut down because of a short, most likely caused by a derailment? I don't because I need to focus on the fault. Other things to consider are

 

(1) Your automation software will not be aware and will continue scheduling trains even for the section that has been stopped. When you clear the short and power the section back up, trains may not be in the positions the software thinks they are.

 

(2) If your layout is a roundy-roundy then trains running on the other parts of the layout will eventually arrive at the dead section and stop. All very well for a train with a single loco at the front, but there will be problems with consists, and especially consists of DMUs and EMUs.

 

(3) If a short occurs whilst a fast-running DMU/EMU consist is straddling the boundary, then this can cause real issues when one half loses power and the other doesn't. I had real problems with this.

 

I ran my layout for a while divided into two power districts each with their own circuit breaker, but quickly found it was too much trouble. Whenever a short occurred, I needed to deal with that without worrying about other trains still running. All too often I had to hit emergency stop to halt all trains to prevent worse situations arising. 

 

I now run my layout with a single booster (Lenz LZV200 5-amp) but the layout is divided by switches for fault-finding. I do not use CBs any more so any short trips the LZV200, and Traincontroller is aware immediately. Restarting after the short still has problems, and I have a flagman in TC that locks all blocks and schedules when power is returned. This allows me to check all trains are in the correct positions before allowing running to resume.

[Andymsa]

8 hours ago, jamespetts said:

I am sorry if I have misunderstood: I have found it quite hard to follow in places, especially as in your earlier post this evening you wrote specifically about the amount of wiring,

 

 

 

 

whereas you now state that you are referring not to the amount of wiring, but rather its layout and the amount of other hardware (seemingly the DCOs).

 

I am not entirely sure that I understand the suggestion that it is always sensible to stop trains on other parts of the layout while a short circuit occurs; it is not clear that this should necessarily be the case (especially as the running is automated, and thus intended to be unattended in any event), and could be quite disruptive. Trains can quite easily be stopped manually either on the computer using the "ESC" key or with the emergency stop buttons in various locations around the layout. Also, as previously written, being able to track the fault giving rise to a short down to an area one seventh the size of the whole layout makes fault finding vastly easier than it might otherwise be. It is difficult to see why this latter advantage alone does not outweigh the relatively modest amount of additional wiring to which the DCOs give rise - can you elaborate?

 

The DCOs also have another important advantage. The booster's output is 3A. The DCOs are each set to 1.5A. The dropper wires used are all 7/0.2, and thus rated at 2A. Upstream from the DCOs, I use much thicker wiring (32/0.2), rated for 4.5A (and I also use 16/0.2, rated for 4A between the DCOs and the occupancy sensors). The 16/0.2 wire is too thick to solder neatly to the rails, but the 7/0.2 is a good thickness. Using the DCOs, I can use the 7/0.2 wire to the track and ensure that it is  protected from ever conducting a current in excess of its rating; otherwise, should a fault condition occur on the layout causing a section of track to draw just under 3A, the 2A rated wires might overheat without the booster registering a short circuit, which, if left unchecked for long enough, could cause a fire in the flammable wooden baseboards and thus be very dangerous. (This latter point, incidentally, is based on advice given to me by a club member who worked in process control/automation in the oil and gas industry for decades; this issue seems to be poorly understood among railway modellers). The use of DCOs thus allows the safe use of 7/0.2 wire whilst still allowing locomotives in all seven districts between them to draw up to 3A, so long as not more than 1.5A be drawn in any given district. a

 

As to the organisation of the wiring, the difficulty is the limited amount of space under the board to run the wires; it is difficult to see how there could possibly be space for totally separate wire runs for different types of wires. If there is a better way of organising the wires that will not enormously increase the workload and will fit into the space available, do let me know so that I can consider it for future projects. As I wrote above, your wiring is very neat, but it is difficult to see how I could replicate that given the density of trackwork that I have  and also the different hardware that I use.

 

In relation to LocoNet wires being run next to other wires, may I ask whether there is any literature on this that I can study to understand this issue more thoroughly from a robust, empirical perspective? This is not a suggestion that I have seen before, and it is not written in the manuals for the LocoNet products that I use, so far as I recall.

RFS has explained much better than I have regarding stopping train after a short, so further explanation is not needed.

 

as to wire size for the droppers 7/0.2 this is far to inadequate. I use 16/0.2 to get around neatness I solder to a small stiff copper wire which is moulded to shape and the 16/2 is wired to that. If there is a risk of the booster not shutting down during a short then you need to revisit your wiring putting in heavier gauge wire as resistance will be to high using smaller gauge wire, or another method have a 2/3  7/02 wire directly on the track and then go to 16/2 after that. 

 

as to wire routing I would of drilled a series of horizontal holes in layout framework, so for instance looking at the quad there be holes on all sides. If memory serves on your layout build thread you have draws on castors underneath. I would mount vertically from the layout behind these draws plywood panels to mount hardware on. One obvious benefit you can look at them directly instead of looking up at them.
 

here is an extract from the digitrax users group on proximity of loconet cables to other cables. This was written by a very experienced user and there is a fare amount about this topic on the group but it would take hours to find it all.

 

“Should note as long runs of Telco/Data cable come in to play, it is a good
idea to avoid long, [close parallel runs] of LocoNet cable to current
carrying buss wires. That is do not bundle the Loconet cable in with your
power buss wires. Keep some distance between the LocoNet cable, and your
power buss wires. Here a few inches goes a long way. Again note: this only
applies to long parallel runs, not angled or perpendicular runs. And in
cases where there is no other choice but to have long parallel runs in close
proximity give the LocoNet cable a twist every now and then to avoid any
possible coupling between LocoNet Data and High Current noise.”

 

 

Edited December 7, 2020 by Andymsa

[jamespetts]

Thank you both for your thoughts. As to stopping trains in any given short, given that the DCOs provide feedback to the computer, it would not be hard for me to configure them to pause the automation whenever a short occurs if I find that this is necessary or desirable. The other reasons for maintaining DCOs remain.

 

As to the suggestion that 7/0.2 is inadequate for droppers, can you elaborate? In what way is it inadequate given that it exceeds by a 0.5A margin the current rating of the DCOs? What is the current rating of the "small stiff copper wire" that you use? Do I understand correctly that this will have to carry a maximum of 5A on your layout before the booster will cut out? Do not forget that not all short circuits are a dead short; it is possible to have a resistive short that draws a lot of current but not enough to trip the breaker in the booster (e.g. a resistive short drawing 4.5A). If this should exceed for an extended time the current rating in any of the dropper wires in the section(s) in which the resistive short occurs, then this is a serious fire hazard.

 

As to drilling multiple horizontal holes, I am not sure how this would deal with wire routing issues - can you elaborate? I am not sure that I understand where the holes would be drilled, how these would fit around the existing circuit cards or how the wires would relate to them. Do you mean holes for passing wires through from quadrangle to quadrangle or holes for mounting P clips to take wire runs? If the former, these holes were put in by the people who built the layout. I do not have equipment to drill holes this large. If the latter, there is simply insufficient vertical space to get in two stacked wire runs plus circuit cards with the circuit cards having enough space on both sides for connexions, and that is without considering the fact that servo mounts are often right next to the structural members dividing the quadrangles in a number of places, making wiring runs along the tops of the quadrangles impossible.

 

As to the suggestion apropos wire runs, that is interesting, although I note that no data are cited and this does not appear to be official documentation. I will nonetheless note the need to twist the cables, as, as noted above, I do not have space for multiple parallel wiring runs separated by a distance.

[Andymsa]

1 hour ago, jamespetts said:

Thank you both for your thoughts. As to stopping trains in any given short, given that the DCOs provide feedback to the computer, it would not be hard for me to configure them to pause the automation whenever a short occurs if I find that this is necessary or desirable. The other reasons for maintaining DCOs remain.

 

As to the suggestion that 7/0.2 is inadequate for droppers, can you elaborate? In what way is it inadequate given that it exceeds by a 0.5A margin the current rating of the DCOs? What is the current rating of the "small stiff copper wire" that you use? Do I understand correctly that this will have to carry a maximum of 5A on your layout before the booster will cut out? Do not forget that not all short circuits are a dead short; it is possible to have a resistive short that draws a lot of current but not enough to trip the breaker in the booster (e.g. a resistive short drawing 4.5A). If this should exceed for an extended time the current rating in any of the dropper wires in the section(s) in which the resistive short occurs, then this is a serious fire hazard.

 

As to drilling multiple horizontal holes, I am not sure how this would deal with wire routing issues - can you elaborate? I am not sure that I understand where the holes would be drilled, how these would fit around the existing circuit cards or how the wires would relate to them. Do you mean holes for passing wires through from quadrangle to quadrangle or holes for mounting P clips to take wire runs? If the former, these holes were put in by the people who built the layout. I do not have equipment to drill holes this large. If the latter, there is simply insufficient vertical space to get in two stacked wire runs plus circuit cards with the circuit cards having enough space on both sides for connexions, and that is without considering the fact that servo mounts are often right next to the structural members dividing the quadrangles in a number of places, making wiring runs along the tops of the quadrangles impossible.

 

As to the suggestion apropos wire runs, that is interesting, although I note that no data are cited and this does not appear to be official documentation. I will nonetheless note the need to twist the cables, as, as noted above, I do not have space for multiple parallel wiring runs separated by a distance.

regardless of the type of short, you want the booster to shut down. The Americans call it the quarter test, I will call it the 2p test put that across the track in multiple locations if booster shuts down then it’s a pass. Having the inadequate wire can introduce resistance and that’s why the booster might not see the short, it’s not about what a wire can carry but it’s resistance.

 

as to loconet cable the extract I gave was from a larger posting of this topic and went into more technical aspects. I posted the conclusion. There are a lot of things manufacturers leave out there manuals and these gaps are covered by users with years of experience. 
 

yes I mean quad to quad, they don’t need to be big holes, a 25mm spade bit would do the job.

 

personally I would get all the hardware out of the quads and mount on separate boards. Looking at the video it looks like there is plenty of space for wires alone, I’m not sure what you mean by stacked wires

Edited December 7, 2020 by Andymsa

[jamespetts]

I am not sure that I follow - where do you envisage that these separate boards would be in relation to the layout? How would they be secured?

 

As to short circuits, I am afraid that I do not follow the reasoning here - that a high resistance connexion to the track can in theory mask a short does not negate the importance of having wires with the capacity to take as much current as the booster can give (so as to prevent a fire hazard), nor does it explain why it is important that the booster, rather than the DCO, should cut out. I have tested the DCOs with the 7/0.2 wire, and they do in fact all trip when shorted with the equivalent of what you call the "quarter test" - you can even see this test being done in the video. Do not forget that the 7/0.2 wire goes only from the dropper to the occupancy detector. 16/0.2 wire is used between the occupancy detectors and the DCOs, and 32/0.2 wire from the DCOs to the booster.

 

In relation to the LocoNet cables, do you have a link to the posting?

[Andymsa]

44 minutes ago, jamespetts said:

I am not sure that I follow - where do you envisage that these separate boards would be in relation to the layout? How would they be secured?

 

As to short circuits, I am afraid that I do not follow the reasoning here - that a high resistance connexion to the track can in theory mask a short does not negate the importance of having wires with the capacity to take as much current as the booster can give (so as to prevent a fire hazard), nor does it explain why it is important that the booster, rather than the DCO, should cut out. I have tested the DCOs with the 7/0.2 wire, and they do in fact all trip when shorted with the equivalent of what you call the "quarter test" - you can even see this test being done in the video. Do not forget that the 7/0.2 wire goes only from the dropper to the occupancy detector. 16/0.2 wire is used between the occupancy detectors and the DCOs, and 32/0.2 wire from the DCOs to the booster.

 

In relation to the LocoNet cables, do you have a link to the posting?

im am sorry but I have given you my thoughts and I have tried as best as I can to explain and help, but with every reply you seem to question even more. The example here is you now require a link to the loconet posting as if my word is not good enough.  The wire sizing needed has been recommended by experts better than me and are proven, take a look at the wiring for dcc website.  I shall withdraw further from this thread and wish you luck.

Edited December 7, 2020 by Andymsa

[jamespetts]

2 hours ago, Andymsa said:

im am sorry but I have given you my thoughts and I have tried as best as I can to explain and help, but with every reply you seem to question even more. The example here is you now require a link to the loconet posting as if my word is not good enough.  The wire sizing needed has been recommended by experts better than me and are proven, take a look at the wiring for dcc website.  I shall withdraw further from this thread and wish you luck.

 

I am afraid that I cannot make anything useful of suggestions if the information is very incomplete, which is why I ask questions.

 

The suggestions about the wire layout in particular cannot be parsed so as to be able to be accommodated in the space that I have available, and I simply do not understand what you meant about additional boards or how those would relate to the layout baseboards.

 

I do not think it in the least unreasonable to ask for a link to the source for highly technical information of this nature.

[Alan Kettlewell]

I find it most interesting to see how fellow modellers build their DCC layouts for automatic train control and what equipment they use.  To the uninitiated it must seem like a minefield.  There are of course several options to choose from and a variety of equipment to achieve it.  I got into this about 15 years ago and have built several automatic layouts since then, the latest, but now discontinued, effort is shown in the link at the foot of my post. 

 

Regarding train tracking, a consideration which bears a cost factor and varies the degree of complexity, is whether to opt for block  'Occupancy Detection' or 'Passing Contact' feedback.  A while ago I chose to use the much simpler 'Passing Contact' feedback method using reed switches in the blocks (in the track) and magnets under the locos.

 

I have not regretted this decision as I find it quite adequate for fully automatic control.  My layout is a fair size housed in a purpose built shed 29 x 14ft, about 300 yards of H0 track, 3 hidden storage yards with about 40 trains.  One complete circuit round the layout is about one scale mile. 

 

I haven't found it necessary to use the perhaps more robust method of 'occupancy detection' nor do I use feedback on the turnouts or indeed the Railcom feature as I find Train Controller takes care of all train tracking effortlessly.

 

I have much respect for fellow modellers who take a more belt and braces approach than I have so the purpose of this post is just to show that there are different ways of skinning this particular cat - and I appreciate there are many pros and cons of the various methods - each to their own and all that.  

 

For interest and comparison, I've listed in the table below the range of equipment used on the layout:

    

 

Cheers .. Alan

[WIMorrison]

Alan - very sorry to hear that you have discontinued Hufeisental as that was truly one of the starts in the automated layout world. 

 

Iain

[Alan Kettlewell]

Just now, WIMorrison said:

Alan - very sorry to hear that you have discontinued Hufeisental as that was truly one of the starts in the automated layout world. 

 

Iain

 

Thanks Iain, a compliment indeed and appreciated.  I'm into my next project which is 0 gauge where I plan to take automated operation even further with automatic shunting, uncoupling etc.  It'll be a much simpler layout but equally challenging to get the auto shunting and automatic forming up of trains to work!

 

Cheers ... Alan

 

[RobinofLoxley]

On 11/12/2020 at 10:32, Alan Kettlewell said:

I find it most interesting to see how fellow modellers build their DCC layouts for automatic train control and what equipment they use.  To the uninitiated it must seem like a minefield.  There are of course several options to choose from and a variety of equipment to achieve it.  I got into this about 15 years ago and have built several automatic layouts since then, the latest, but now discontinued, effort is shown in the link at the foot of my post. 

 

Regarding train tracking, a consideration which bears a cost factor and varies the degree of complexity, is whether to opt for block  'Occupancy Detection' or 'Passing Contact' feedback.  A while ago I chose to use the much simpler 'Passing Contact' feedback method using reed switches in the blocks (in the track) and magnets under the locos.

 

I have not regretted this decision as I find it quite adequate for fully automatic control.  My layout is a fair size housed in a purpose built shed 29 x 14ft, about 300 yards of H0 track, 3 hidden storage yards with about 40 trains.  One complete circuit round the layout is about one scale mile. 

 

I haven't found it necessary to use the perhaps more robust method of 'occupancy detection' nor do I use feedback on the turnouts or indeed the Railcom feature as I find Train Controller takes care of all train tracking effortlessly.

 

I have much respect for fellow modellers who take a more belt and braces approach than I have so the purpose of this post is just to show that there are different ways of skinning this particular cat - and I appreciate there are many pros and cons of the various methods - each to their own and all that.  

 

For interest and comparison, I've listed in the table below the range of equipment used on the layout:

    

 

Cheers .. Alan

Just a question Given the size of the layout im a bit surprised that you have no more than 80 blocks. In fact you have more detectors than blocks. Generally I'm all for simplicity and I hope my layout when finished will have the same kind of equipment list but with current detection. And work half as well

[Andymsa]

3 hours ago, RobinofLoxley said:

Just a question Given the size of the layout im a bit surprised that you have no more than 80 blocks. In fact you have more detectors than blocks. Generally I'm all for simplicity and I hope my layout when finished will have the same kind of equipment list but with current detection. And work half as well

it is possible to have more detectors than blocks, the term block may mean something different to you. In Traincontroller a block can have more than one feedback sensor, the same applies to I train, the more detectors you can have the better automation will be.

[Alan Kettlewell]

11 hours ago, RobinofLoxley said:

Just a question Given the size of the layout im a bit surprised that you have no more than 80 blocks. In fact you have more detectors than blocks. Generally I'm all for simplicity and I hope my layout when finished will have the same kind of equipment list but with current detection. And work half as well

 

Some blocks have two detectors - ie momentary contacts (reed switches) - particularly useful if a block or section is bi-directional.  On long runs of track there may be just one block to control that whole section, which could be up to 20ft.  Most block sections are a minimum of the longest train length which is about 8 ft. except for the loco shed area where there are six short blocks to store the locos.  

 

The 'passing contact' method for train detection is probably the leanest in terms of equipment required -  additional hardware, and wiring, will be required for the occupancy detection method.

 

Cheers ... Alan  

[RobinofLoxley]

11 hours ago, Andymsa said:

it is possible to have more detectors than blocks, the term block may mean something different to you. In Traincontroller a block can have more than one feedback sensor, the same applies to I train, the more detectors you can have the better automation will be.

That was the heart of my question really. I was surprised a layout of that size would run automatically with what are effectively 80 sub-divisions for control. So I wasnt really looking for

block

 

3 hours ago, Alan Kettlewell said:

Some blocks have two detectors - ie momentary contacts (reed switches) - particularly useful if a block or section is bi-directional.  On long runs of track there may be just one block to control that whole section, which could be up to 20ft.  Most block sections are a minimum of the longest train length which is about 8 ft. except for the loco shed area where there are six short blocks to store the locos.  

 

 size information as its obvious from the stated dimensions and block numbers that some blocks will be large, more maybe concerning the nature of the layout. I didnt want to ask for a track plan.

[Andymsa]

5 minutes ago, RobinofLoxley said:

That was the heart of my question really. I was surprised a layout of that size would run automatically with what are effectively 80 sub-divisions for control. So I wasnt really looking for

block

 

 

at this time I have around 560 individual detection sections. This may go up or down as I’m in the process of a big alteration.

 

i have started a post in layout topic  called llamberg, I will be going into more detail at some point on the control methods used 

Edited December 13, 2020 by Andymsa

[RobinofLoxley]

Just now, Andymsa said:

 

at this time I have around 560 individual detection sections. This may go up or down as I’m in the process of a big alteration.

Which as why I was curious as to how Alan manages with 100, as I have the impression the layouts are of comparable size.

[Alan Kettlewell]

Hi Robin,

 

Not quite sure why the number of blocks surprises you.  I can provide a schematic track plan from Train Controller if you wish.  The plan can be seen in my H0 layout thread but perhaps hard to find wading through 30+ pages, but happy to oblige if it helps understanding, t'would only take a minute.

 

Cheers ... Alan  

 

 

[RobinofLoxley]

4 minutes ago, Alan Kettlewell said:

Hi Robin,

 

Not quite sure why the number of blocks surprises you.  I can provide a schematic track plan from Train Controller if you wish.  The plan can be seen in my H0 layout thread but perhaps hard to find wading through 30+ pages, but happy to oblige if it helps understanding, t'would only take a minute.

 

Cheers ... Alan  

 

 

If its already on here I will find it, thanks Alan are we talking Hufeisental, thread from 2015 from your link?

Edited December 13, 2020 by RobinofLoxley
Addition

[Andymsa]

3 minutes ago, RobinofLoxley said:

Which as why I was curious as to how Alan manages with 100, as I have the impression the layouts are of comparable size.

my layout is an L shape and length is 65 feet by 12/14 feet

[Alan Kettlewell]

1 minute ago, RobinofLoxley said:

If its already on here I will find it, thanks Alan

 

No worries.  Since those plans in my thread were posted I added a further section for a narrow gauge mountain track - so a few more blocks got added.

 

It might be worth adding that I didn't do any shunting manoeuvres with the H0 layout - it was a 'watch the train go by' type of layout so I only needed blocks at the stations and in the storage yards - plus a sparse number out on the running lines just to keep the trains separated.  This happily provided up to 5 or 6 trains moving at any one time - quite enough for me and I definitely needed a spare pair of eyes to watch it all.

 

Cheers .. Alan     

[RobinofLoxley]

1 minute ago, Alan Kettlewell said:

 

No worries.  Since those plans in my thread were posted I added a further section for a narrow gauge mountain track - so a few more blocks got added.

 

It might be worth adding that I didn't do any shunting manoeuvres with the H0 layout - it was a 'watch the train go by' type of layout so I only needed blocks at the stations and in the storage yards - plus a sparse number out on the running lines just to keep the trains separated.  This happily provided up to 5 or 6 trains moving at any one time - quite enough for me and I definitely needed a spare pair of eyes to watch it all.

 

Cheers .. Alan     

Yes Ive realised that from looking at the track plan.

 

My project is C 1960 using all steam and therefore I have to plan for loco switching in stations. I have planned for three blocks per track on station lines. Similar in most sidings that are dead end, either 2 or 3 blocks depending on siding length and likely use.