LD (Loco Detection) v BD (Block Detection)

[PJ_model_trains]

While you are all thinking about how LD works, can I just reiterate some points from previous threads that may assist this thinking.

 

1.  Most important point - as it is currently implemented, RM has no idea where anything is, only you do.  If you were to remove all the straights  and curves from your RM schematic (not including the ones that contain LD sensors), it would not change by one wit what RM knows about your layout.  When LD is added, this remains true with an important unless ............ unless HRMS make a fundamental change to the way RM works, and I think this unlikely anytime soon.

 

It could be said that our schematic layout using RM, (or other similar software) is similar to the control of the main railway system, it is a visual representation of our layout, . Not same in design, not as complicated but they are both just visual representations of the railway system being controlled. Whether real life or our RM system they are has as been said many times just bits and pieces on a screen that only reallly make sense to the person setting up or controlling from them.

 

Even if at a later date Hornby allow more advanced features such as the example mentioned yesterday, for track pieces to be a different colour for occupied blocks, they will still be just pieces on a screen to provide advanced help for the user as the system grows. This example is far on the horizon if ever it shall come but, I think it is good to look ahead, if Bill Gates hadn't looked out of the Window the humble computer may not be what it is today.

 

It is not by the computer knowing what a piece of track on the screen is for, or were a point or signal is, that makes the system good, it is by us simulating our layout and using it, making these bits do things on their own and together, that we can do incredible things and also think of other things to make it even better. I think.

[PJ_model_trains]

2.  Given 1, RM cannot make any sense out of having read a loco tag forwards or backwards.  Knowing a direction would mean it knows where things are and what sequence they occur in along a track.  And it doesn't know either.  The only direction it can know when a detection is made is whether the throttle for that loco is set to Forward or Reverse.  And that tells you nothing about where the loco has come from or is going to.

 

LD will send back sensor code and tag code, to the LD controller and then to RM to process.

 

RM already knows the loco speed and direction, I scratch my head a little as to why we would expect LD which is a simple system to need to know what RM already knows? Just my thoughts.

 

RM will remain the system that processes the data, all LD is going to do is send two small items of data back for every tag read. This loco/tag that sensor, that loco/tag another sensor, this train/tag a different sensor, that goods train/tag a different sensor, (last carriage/tag another sensor), etc.

 

LD is a very basic system that needs sensors, tags, a bus wire, a controller to talk to RM through our PC and a USB cable and mains lead. The final cost, that some may say is to expensive, will be based on all the above plus the research and testing prior to launch, production costs, and bringing us the new system which also includes technical support.

 

[PJ_model_trains]

4.  PJ has talked about having tags not just on locos but also on tenders and last carriages/wagons in a train.  The importance of these is knowing when the entire train has passed a sensor so he can use that information for his signal operation, and others will want to know this too.  If you look in RM loco setup, you will see where you can enter the tag number for the loco, which is how RM can know which loco has been detected.  But there is nowhere you can set up a tender or wagon this way.  To achieve PJ's desired result at present, you would have to set up tenders and wagons etc as dummy locos and then include the tag ID in that setup.  And that really puts the kybosh on throttle setting for determining speed as you won't be using throttles on dummies.

 

When setting uop locos in RM I have, as I guess many have, set them up in categories

 

10-30 steam

30-50 diesel

50-60 dmu's and the like

etc

 

When I set up LD I will do that in a similar way, grouping each type between the numbers alllowed. This means therefore I may allocate...

 

70-80 last carriages

80-90 last wagons

90-100 steam loco tenders or front and rear of DMU's

100-106 guards vans

 

Naturally this will be subject to space under the locos. etc this may be one of the biggest problems as  finding one or two suitably flat areas may not be that easy. I am thinking!

 

By having all locos, dmu's etc using lower tag numbers it is easy to remember when programming that the back end items have the back end umbers. We will all find our own ways these are just my considerations at this time.

 

All the back end items will be set at locos in the list but will only be used for changing signals and/or points if necessary after the last item has passed a signal/sensor point.

 

[PJ_model_trains]

My point Metman is that direction is meaningless given you don't know where anything is, even if you know it's set forward or reverse.  It only becomes meaningful to you as you can see where everything is on the schematic and the layout, but RM can't. 

 

So this means that RM will not be able to fully automate a layout as it stands?

 

Hi hosh

 

RM is ongoing, HRMS have said there is 'lots more to come' but, they know they have to clear the outstanding signal and other issues, they are currently working on increasing the third party signals and decoders within RM and they still say there is lots more to come.

 

Many of us started with RM when it was quite basic, we have seen it grow and improve, signals was a big step forward, LD will be a greater step forward, but they still say 'there is lots more to come' and encourage us to discuss what we would like to see in the future.

 

As for loco direction and loco speed RM already knows this, LD will be an addition to provide detection of locos and other items as we choose and carry out instructions we have preprogrammed into RM for them. With signals and with LD we will be able to do so much more than we have ever done before.

 

In the end automating layouts will be available to us all but, we will probably all have different views of what automation means. For some it may be just one or two locos on different tracks, for others it may be ten or more locos running at once and changing tracks. Some may not want signals others definately do. It willl all be down to us and what we want to do with our systems, do we just run trains, do we run routes, have we a full timetable. Automating our layout will mean something different to us all, but it should be fun learning.

 

For me it is exciting, for me it is also frustrating, (I guess many are in the same boat).

Waiting for signals to work correctly, waiting for LD to come to market, but I am sure they will both come, HRMS have promised to work on past signal issues as a matter of urgency (accepting Christmas is a busy time for them)

 

We wait to find out when LD is coming, all we know at this moment in time is it is...  'coming soon' or 2017?

 

[PJ_model_trains]

@fishmanoz

 

... yes LD will allow full automation as there are If Thens ...

 

Now that is powerful, I did not realise IF / THEN was included.

 

Hi Fishy, for the benefit of members would you like to confirm/list these please.

 

My layout is being lifted for me today as we have visitors coming to stay with us, they will also be with us over Christmas holiday period so I will not have access to my layout until after the New Year. ;-(

 

 

[RDS]

 as we have visitors coming to stay with us, they will also be with us over Christmas holiday period so I will not have access to my layout  ...

 

Hi PJ, You should have bought some 'sleeper trains' problem solved! 

 

[PJ_model_trains]

 as we have visitors coming to stay with us, they will also be with us over Christmas holiday period so I will not have access to my layout  ...

 

Hi PJ, You should have bought some 'sleeper trains' problem solved! 

 

Hi RDS

 

Trouble is, my space is limited (like everyone elses). We find our little niche, a space where we can put our treasured model trains. Unlike Rod Steward where space (and money LOL) appears unlimited and he takes his N gauge (second layout) with him and sets it up in a suite he has taken for the period of the show he is at.

 

For me my space is the spare bedrom, which is only spare when we have no visitors. It sits over a three foot divan bed that pulls out to make two beds at different heights. My wife usually gives me a count down and that depends on when she can get help to lift the layout, I got my warning yesterday afternoon! So... my trains have become their own sleepers until the new year and I can wake them again, wish them happy new year (but don't tell them when LD is coming... cos we don't now)

 

When I started with my layout I was given two options, the one I chose or the garage.

 

The garage! In the North of England! Say no more ;o)

 

[Yelrow]

PJ, when i was young, my mother dried our clothes on an airer that lowered from the ceilng. It had space for 6 rows of clothes, and 2 pulleys were screwed to ceiling joists. She lowered it in sashcord, with a cleat. The rest of the week, it was tight against the ceiling. It was 7 foot long. When i lived in Kidderminster, owing to the need for garaging the car, my layout, was built on exactly the same system.  Just a thought. This way, it does not collect dust, under the bed, and can be lowered or raised in seconds. john

[PJ_model_trains]

I know the ones you mean, they used to hang from the ceiling above the wash tub!!!   LOL

 

It has been considered but the layout is quite heavy, it would need strengtheners above the ceiling joists and more.

 

I have even got to the point of scrapping everything and starting again, a three sided layout 3m by 1.7m in a C shape along 3 walls hinged so that it folds upwards but that won't work either due to the divan bed. We don't want to damage the bungalow in case we ever move.

 

These things we have to live with but for me the most frustrating is having to have my wife, who has rheumatoid and osteo-arthritis get a friend to lift the existing layout. I know it is only once/twice a year but, she shouldn't be doing this for me, I should be doing manual type jobs for her, how long she can do it would be another consideration.

 

Maybe this would be a good idea as a lift for your downstairs layout to your upstairs one.  ;o)

 

I smiled at your comments for the helix you are creating, sounds like a coil idea... you will have to make sure it doesn't over heat though.

 

Great idea though John

 

[RDS]

... on an airer that lowered from the ceilng ...

 

We still use one in the Utility Room at our house.  Well, my wife thinks it is a Utility Room!  

She has one half for her washing and I have the other half as my indoor workshop with my Model Makers Lathe and workbench.

[Yelrow]

PJ, 7 by 7 foot of wet sheets jumpers etc, i belive would be heavier than your layout. As to strengthening the joists, i dont think that would be necessary, as my brother in worthing has just had ceiling hoist to lift him put in, and engineers said not required .Bear in mind, you would be over about 4 joists, with 7-8 foot.. All you would have in your ceiling, would be 2 holes for screw in pullys.  2 minute polyfiller job, if you move.  If i had your problem, i would certainly be asking the question. john

[PJ_model_trains]

Thanks for your suggestions guys, I have a couple of weeks to consider them and also whether my wife could raise the layout by the ropes, I couldn't. Her grip is not good, she can hold the 3x1 side of the layout and tilt the layout until one end reaches the floor then raise into a vertical position (with help), I am not sure she could manage the ropes.

 

I also think it would need 4 ropes to wheels above (one at each corner) and probably additional wheels near the side. Plenty to think about the next comment no doubt would be a motor to winch it up. But I have another concern I think, my daughter and one of the grandchildren may not be happy to sleep under it! Ermmm. Plenty to think about and then talk about I think.

 

I had another assessment recently, they were pleased we had already had installed low steps, grab rails etc and that I had chair raisers, my Lazboy electric chair and the recent electric adjustable bed. They said I would have had to have a rail at side of bed and if that wasn't adequate possibly a ceiling fitting. The electric bed is a blessing for me and for my wife in the middle of the night. Thank goodness I can still play with my trains ;o)    (not in the middle of the night LOL)

[hosh]

It seems there are 2 types of detection systems - Block Detection (BD) via the track or beam tech. Beams can go across the train (and be broken) or up and be reflected and then detected.

 

So it seems Hornby LD will do the latter due to it being more discrete (prototypical - much less above the track than cross beaming) and require the least amount of disruption to existing layouts as would be the case with BD.

 

With the Hornby LD system it becomes a necessity to have ID instead of simply detecting anything above the sensor, since otherwise you have the problem of the gaps between locos/tenders/carriages and not knowing the difference between that and the next train. So this is not done out of desire but out of necessity.

 

To end up with worthwhile and complete automation, any of these detection techniques will require the appropriate software to render the entire exercise worthwhile.

 

Speed is completely unneeded in any case.

 

Direction (not simply forward or reverse but heading on the layout) is unlikely to be provided by Hornby LD and will thus need to be known by the software (RM). Generally, if the software knows where the train is and was, then it knows where it's headed.

 

Point being, at the end of the day, the only thing that will dictate how good Hornby's version of detection is will be the post LD version of RM. Ultimately, there will be no advantage from a technical point of view of any system over any other. The hardware basically detects - the quality of the overall system will be entirely dictated by the software!

[Yelrow]

Pj, my brother has a hospital bed, and this new hoist, but he needs 2 to operate it.   My mothers clothes airer was in our dining room, across the width of the table, and no one ever complained. Ceiling was higher than yours, as old victorian house. At least, you are thinking, outside the box, as if the time comes when your wife cannot lift, you would be playing trains on the floor. You would only need two sashcords, fixed to to 4 corners , and double pullys, but look on the net. There are many gizmos, these days, for such a problem.  With my mums, she had cleats, on the wall, and raised one side, cleated, then the other side, cleated, so it went up in stages.  Lazy Boy, way to go. Mine are great, especially with the in chair heating. john

[PJ_model_trains]

Hi John

 

Yes I am thinking outside the box, I have always been a constant assessor in life, work and otherwise. But, I couldn't play trains on the floor, if I get down, if, I won't get up.

 

First I will consider the options discussed in more detail, my feasibility and risk assessment. Then I will drop feeder comments to the wife (as I do) and then we can chat about it. Having said that she always sees through me and says  'what are you after now' :o)  Me!

 

No rush now as Christmas is coming. I am hoping for a better year coming. Except for the pre-ordered Class 47 TTS I have not had any trains at all this year. After 4 months of signals not working (Laast April)  I made the resolution that I will not buy another Hornby item until the signals are working. I have missed some great offers but I will wait until the signals are working properly. I do have confidence, HRMS will as promised, now get them to work correctly.

 

No trains, carriages, etc this Christmas ;o(

 

[PJ_model_trains]

and require the least amount of disruption to existing layouts as would be the case with BD.

 

Least amount of disruption to the layout is a key consideration. Less disruption the BD I would say

 

With the Hornby LD system it becomes a necessity to have ID instead of simply detecting anything above the sensor. So this is not done out of desire but out of necessity.

 

I feel ID's are not a necessity for the system itself but are a requirment for the additional instructions we will be able to program as a result of including them.

 

To end up with worthwhile and complete automation, any of these detection techniques will require the appropriate software to render the entire exercise worthwhile.

 

RM is capable of doing what is required, as discussed on this thread and others commands are already built in to the software program.

 

Speed is completely unneeded in any case.

 

RM knows the speed and direction of our locos already, we set the speed from from throttles or programs in RM

 

Direction (not simply forward or reverse but heading on the layout) is unlikely to be provided by Hornby LD and will thus need to be known by the software (RM). Generally, if the software knows where the train is and was, then it knows where it's headed.

 

The software doesn't know where the train/loco is hosh. It will know with LD the loco tag ID, the sensor ID as discussed. RM will know the data and programs we have placed into programmable icons which can be 'anywhere' on the screen, not necessarily on the track. RM knows if a loco is going forward or revers but doesn't know where or which track. Only we know by placing items on the RM Schematic layout. It is our visual representation of or layout.

 

Point being, at the end of the day, the only thing that will dictate how good Hornby's version of detection is will be the post LD version of RM. Ultimately, there will be no advantage from a technical point of view of any system over any other. The hardware basically detects - the quality of the overall system will be entirely dictated by the software!

 

Every system will have advantages and disadvantages and nmost of these will depend on whether the system does what we want out to do or not.

 

I think the proof of the pudding for the success of LD will be down to a few items

 

• keeping sensors clean at all times
• positioning tags at the right distance from sensors so they do not catch
• the system being launched and customers having minimal problems

 

You got it hosh, LD is the detection system, RM is the data processsing software.  ;o)

 

[96RAF]

I googled electric hoist systems for model railways and one of the hits suggested a counterweight system if an electric hoist wasn't adopted.

plenty of room for thought there For a full table lift or to assist with lifting a hinged table.

[PJ_model_trains]

I googled electric hoist systems for model railways and one of the hits suggested a counterweight system if an electric hoist wasn't adopted.

plenty of room for thought there For a full table lift or to assist with lifting a hinged table.

Thanks RAF

 

I really did consider starting again! The 3 side idea hinged on the long wall. Seemed good although after so much work to the existing layout it would be a shame. (I would need to design another fish pond). The trouble would be the divan bed would have to be on edge, as I have to it down all the time it would be to high for comfortable operation.

 

As you say 'plenty of room for thought'.

 

May have to change the title of the thread LD vs BD or helping PJ!   [;-D

 

[hosh]

 

The software doesn't know where the train/loco is hosh. It will know with LD the loco tag ID, the sensor ID as discussed. RM will know the data and programs we have placed into programmable icons which can be 'anywhere' on the screen, not necessarily on the track. RM knows if a loco is going forward or revers but doesn't know where or which track. Only we know by placing items on the RM Schematic layout. It is our visual representation of or layout. 

 

Posted by you -

 

Loco Detection

On signal green

On signal red

For Locos;

For train type(s)

For forward direction

For reverse direction

For any direction

Stop loco

Stop loco on signal

Resume loco on signal

Reduce loco speed to

Increase loco speed to

Set to cruise speed

Switch right point

Switch left point

Set clear signal

Set danger signal

Activate loco function

Play sound file

Run program file

 

So if the signals can be set/read when a sensor is tripped, then this is the same as zone control. Only layouts that run trains east/west on the same piece of track would struggle.

 

e.g.

Train A has just tripped sensor A at signal A (which was red) and so train A stops. At this point RM sets the signal in zone B (behind A) to Red. Assuming there is no train in B yet, then the signal at zone C (behind B) is set to green.

 

 

Agreed?

[Howbi]

Hosh.........you keep talking about 'zones'..........Rm does not know about zones...........it only knows that a certain signal has a certain decoder address............RM does not know where it is on the layout or that a particular signal is in front or behind the next signal, only you know that..............with LD you can program a sensor to set the signals as required when a loco passes over it and then reset the same signals when the same loco passes over another sensor further down the line...........All this can happen without RM knowing where each loco is on the layout.............events only happen when each loco passes over a sensor...............the track does not report back to RM, only the sensors...............well that's my take on the modus operandi of LD.      Please correct me if I've got it wrong.  HB :-)

[Fishmanoz]

Perfect HB, that's exactly how LD works.  However, as PJ has told us on more than one occasion, such a system can be used to simulate blocks with the ends of the blocks defined by where sensors have been placed.  RM still doesn't know where these blocks are or that a number have been defined in sequence along a track but we do.  Consequently, we can program signal operations to control train operation in those blocks.

 

hosh, inclusion of the LD sensor commands some 18 months ago (they were already there when that Hornby Magazine article was published) means that from a software perspective, we are already in the "post LD" era.  We can place sensors on our layouts, program them in conjunction with programming both points and signals, even program them to initiate complex programs we are already able to write and test.  All we are waiting for is the hardware to be up and running.

 

Just on knowing loco speed, because locos don't change speed instantly, rather accelerate and decelerate over defined periods, we may not know actual speed on detection.  RM will know what speed step the throttle has been set to, and what actual speed that step means but if the loco is still accelerating or decelerating from the last speed instruction given, its actual speed at the sensor won't be known.  I'm not sure this will cause a problem with automating layouts, but it will need to be recognised.

 

And for those wondering about not seeing any IFs or THENs in the instruction set, a command starting FOR is the same as an IF.  Specifically, the command FOR loco No NN has exactly the same logic as writing IF loco No NN is detected.  The THEN commands are those commands that don't start FOR and will follow the FOR commands when you program the sensor.

[hosh]

Hosh.........you keep talking about 'zones'..........Rm does not know about zones...........it only knows that a certain signal has a certain decoder address............RM does not know where it is on the layout or that a particular signal is in front or behind the next signal, only you know that..............with LD you can program a sensor to set the signals as required when a loco passes over it and then reset the same signals when the same loco passes over another sensor further down the line...........All this can happen without RM knowing where each loco is on the layout.............events only happen when each loco passes over a sensor...............the track does not report back to RM, only the sensors...............well that's my take on the modus operandi of LD.      Please correct me if I've got it wrong.  HB :-)

 

This concept of RM "knowing" is amusing! :)

 

The fact is, if you look at my previous post, RM will indeed "know" what is where in the general case where locos only run, say, in the east direction.

 

If a signal is red then it knows the zone ahead is occupied. A sensor is a zone.

 

Whether there's an intuitive, graphical representation of this is irrelevant - RM effectively and literally "knows" what is where, what it's doing and what it's about to do.

[Fishmanoz]

Hosh, literally RM doesn't, but you do.  Because you do, you can program accordingly.

[hosh]

Hosh, literally RM doesn't, but you do.  Because you do, you can program accordingly.

 

So what would you need RM to be or do to refer to it as "knowing" what is where?

 

It actually seems to me that , literally we don't, but RM does.

[Fishmanoz]

RM doesn't need to be or do anything more than it already is, apart from adding the LD hardware.  It is fine as is.

 

And I meant what I said.  RM doesn't know, literally or otherwise, where anything is, but it also doesn't need to know, because we do.  We can look at the layout itself, and the schematic, then program such things as block simulation using only the capability already in RM.  Just as HB, PJ and I have detailed for you above.

 

May I suggest that you "suspend your disbelief" about how you set up yours or any other BD system, and look only at what we've said about how LD will work in RM, and you should be able to see that, without knowing anything about where anything is on the layout, you can simulate block operation using LD/RM.  it seems to me the logic of BD is interfering your understanding of how it will work, but work differently, with LD/RM.