Loco Detection (2)

[HornbyRailMasterSupport]

We advise that you do not run other programs while running RailMaster. We do not insist on it. This is simply because some programs use a lot more resources than others. Obviously, reading a PDF guide has no effect on RailMaster, however, running a full-blown

 

CAD system at the same time does. Common sense is needed.

 

Not running other programs while a sophisticated system like RailMaster is running simply makes sense.

 

RailMaster does not crash when communicating with a DCC controller. If it does on

 

your system then you need to look at several things that could be affecting the program: Internet security software, unstable operating system and so on.

 

 

[PJ_model_trains]

Michael_A said:

I'm assuming this will need to utilise another USB port on the computer, as I cant see how else they would get the loco detection data back into 'the system'.

Given Hornby say you should't run RailMatser with any other

programmes running (Bit of a nuisance if you want to read the User Manual at the same time!) I wouldn't be suprised if there also needs to be a significant improvemnt to the communication part of the RailMaster software to stop it crashing while reading detection

data and trying to communicate at the same time with the Elite or e-link!

But, let's remain optimistic, I sooooooo want loco detection, it will transform RailMaster and release the potential of it's programming abilities.

From what I

understand so far, yes, loco detection will run on a sepatate usb port but, from the pc point of view, they are designed to work together. How it works and when it is available we can only wait for confirmation.

[Fishmanoz]

This is one area where we are not drawing a long speculation bow. The loco detection tab in RM Setup window clearly needs a com port selected for the LD controller. Given USBs are now universal as they say and the only way of connecting to most Windows

 

based machines, we can assume that The com port will be for a USB. And pretty clear the communication protocols of the 2 parts of the system will have to be designed to work reliably together. Given it is just one system, that shouldn't be a problem.

[Graskie]

I only have two USB ports on the notebook I use for my railway, one for RM/Elite and the other for my mouse. Perhaps a good job I'm miles away from loco detection.

[PJ_model_trains]

Graskie said:

I only have two USB ports on the notebook I use for my railway, one for RM/Elite and the other for my mouse. Perhaps a good job I'm miles away from loco detection.

Miles away from loco detection Graskie

Does

that mean you are only half a mile away from a dcc bus ;-)

[PJ_model_trains]

walkingthedog said:

DCC uncouplers are already available.

Hi WTD

Do you have the code for this, or the link?

Thanks

[Fishmanoz]

To put all of our loco detection eggs in the one thread basket for convenience, here is the latest info PJ posted from the DCC supplement in the Hornby magazine:

 

LOCO DETECTION UPDATE

 

Hornby magazine May 2014

 

DCC Handbook - page5

 

Hornby

 

loco detection system is made up of 3 compondnts, a loco detection module, track sensors and locomotive detection tags.

 

Each LDM is designed to plug into a USB port on a PC and can read up to 48 track sensors. Additional LDM's can be connected to read

 

more sensors.

 

The track sensors require a 1mm hole to be drilled through a sleeper for installation.

 

The detector tags can be fitted to the underside of the locomotives, and the system will work with up to 106 encoded tags at any one time. As

 

a tag passes a sensor it passes data to RM about which locomotive. If you have more than 106 locos you can duplicate the detection tags on other models not in use at that time.

 

Loco detection is expected to be available later this year. Prices are

 

yet too be confirmed, but it is anticipated track sensors 10 GBP, detection tags 30 GBP (for all 106), loco detection module 80 GBP

 

Loco detection will be able to switch signals as a train passes the sensor, stop or alter train speed according to signal

 

state etc.

 

This is the latest information, source as above.

 

I wonder if the Eureka moment was the 1mm hole through the sleeper, sounds neat though, it would not have been aesthetically pleasing to see tags all over the place. It sounds pretty

 

good, no isolated blocks to set up just work out blocks/zones to match longest rake and plan from there.

 

Now if you go back to page 3 of this rather long thread, you will find all of the previous info we had on price and capability summarised. That

 

should keep us informed on where we are to date.

 

[Gregd99]

Phul007 said:

I hope also that they add in RM the option to simulate the layout so you can test out the sensors with fake trains and see the actions the sensors make. It would save money to see a simulated crash that a real one.

When

you get beyond very simple operations - eg make sound when loco passes - then simulation is very helpful. There are all sorts of combinations of events that can happen that cause surprises on the layout.

[Gregd99]

There was some discussion on another thread re the cost of the system being a bit over-the-top...

 

In my (smallish) layout with a triple loop with about 20m of track i found I only used 6 sensors. I built the system myself and dimensioned the hardware

 

for 16 sensors but found I could auto-run the layout with 3 trains with just six detectors.

 

My IR detectors detect train presence so the one detector can tell you when a train arrives at a station (IR interrupted) as well as when it leaves (IR restored).

 

When I have points I put the detector at the throat end so that a single detector tells you when either block is occupied depending on the point position (this requires knowledge of the current track topology which RM might/might not have).

 

I might

 

fit some more one day but for now 6 is enough.

 

A similar approach for RM might be to fit a label on the bottom of the loco and the last wagon in a train. that would allow different actions to be taken when the train arrived/departed.

[PJ_model_trains]

Phul007 said:

I have put my layout on go slow as it is mostly finished and was always designed for loco detection as it is way too complicated to run manually. I spend more time cleaning the track that using it these days.
I have been

looking at RM and the commands available for detection sensor in the layout mode to understand and plan how I will implement the system. I think people will quickly see how powerful it is but may get too adventurous like me in having so many commands per sensor.

Looking at RM they are a limit on lines you can add, BUT it is too early to make any comment and the PRO Pack may give you more options.
I think the most exciting part is that we are finally hearing officially news about this and not by third parties, BUT

still hurry up!

My biggest fun will be trying to drill 1 MM holes through 4 to 6 inches of material, I have 3mm drills that are 8 inches long for running wires.

Hi Phul007

I also thought OMG about the 1mm holes, I will

have to go through, sleeper, cork, 50mm cavity wall insulation and 9mm plywood. Not as deep as yours but a bit of work. I visualise drilling on an angle, between the sleepers, then the 1mm hole through the sleeper and passing wires through to the underside.

(Hoping wires are long enough to do this or there could be more work)

We now know the 3 components for loco detection, I think it would be good to discuss how we would install them. I guess the installation of these will be easy, working then through

RM with the signals and points is where we make is easy, over complicated, etc.

I will share what I am thinking here, hoping we won't get critised and told to wait until the system is available. I stress these are only mt thoughts and welcome those

of others as a means of planning and prepaing how we will do it on our layouts, bearing in mind all are different.

So far my thoughts are, firstly work out the longest length of train and carriages and take this as the minimum size for a controlled

block/zone. Then work the layout out into blocks, loop by loop, length by length. It was said in a previous discussion we may need to put detectors on either end of the train/carriages, I don't think we need this (correct me if I am wrong). This is how I see

it so far, a train approches a signal, it receives instructions through RM to slow down or stop subject to status of signal. Signals on green train procedes into next block, as train passes signal changes to stop another train entering the block. Once the

train meets the other end of the new block it receives instructions via sensors and RM to slow down, stop, etc as programmed but, also atthis point sends a signal to RM to chabnge the previous points if necessary as this loco has passed the points and is fully

in the new zone.

I welcome thoughts of other members regarding how to install/use loco detection.

Regarding signalling we know Hornby are including multi aspect signals in RM Pro but at this stage do not know which, 2, 3, 4 probably. Feather

we will just have to wait and see.

[PJ_model_trains]

Gregd99 said:

Phul007 said:

I hope also that they add in RM the option to simulate the layout so you can test out the sensors with fake trains and see the actions the sensors make. It would save money to see a simulated crash that

a real one.When you get beyond very simple operations - eg make sound when loco passes - then simulation is very helpful. There are all sorts of combinations of events that can happen that cause surprises on the layout.

Hi gregd99, you are so

right, it will be fun learning though.

I think I will start by setting up say 1 or up to 3 locos, just basic set up. Speed, sound, stopping and starting etc. I think the main control comes when we set up blocks and control them so that a train has to

stop in the block it is in because another train is in the next block. All good fun, the more complex a layout the more mind boggling it will be but step by step not over complicating things and testing locos and blocks has to be the safest way.

[Viaduct]

I'm a bit puzzled by the 106 encoded tags bit. 6 bit binary would only be 64, 7 bits 128. Maybe there are 22 'other' tags for things like coaches and trucks which can be used multiple times. And thus you could define the end of a train when no more tags

 

were read. Maybe this was the 'eureka' moment? Can't think of 22 uses though!

 

V

[37lover]

Viaduct,

Wow; that's impressive and shows up my blinked view. I thought it meant 106 tags that were encoded. Are you saying each tag has 106 codes? I don't understand your reference to binary otherwise. Be gentle with your reply, I use computers but

 

don't understand them!

[PJ_model_trains]

Viaduct said:

I'm a bit puzzled by the 106 encoded tags bit. 6 bit binary would only be 64, 7 bits 128. Maybe there are 22 'other' tags for things like coaches and trucks which can be used multiple times. And thus you could define the end

of a train when no more tags were read. Maybe this was the 'eureka' moment? Can't think of 22 uses though!

V

Word for word the details in the DCC Handbook, Free with Hornby magazine state...

Hornby's loco detection system is made

up of 3 elements: a loco detection module (LDM), track sensors and locomotive detection tags.

Each LDM is designed to plug into a USB port on a PC and can read up to 48 track sensors. Additional LDM's can be connected to read more sensors.

The

track sensors require a 1mm hole to be drilled through a sleeper for installation.

The detector tags can be fitted to the underside of the locomotives, and the system will work with up to 106 encoded tags at any one time. As a tag passes a sensor

it passes data to RM about which locomotive. If you have more than 106 locos you can duplicate the detection tags on other models not in use at that time.

Loco detection is expected to be available later this year. Prices are yet too be confirmed,

but it is anticipated track sensors 10 GBP, detection tags 30 GBP (for all 106), loco detection module 80 GBP

This is what it says... detection tags 30 GBP (for all 106) but magazines have been known to print things incorrectly. We also do not know

if this... 'it is anticipated track sensors 10 GBP' is the cost of just 1 sensor we will have to wait and see

Loco detection will be able to switch signals as a train passes the sensor, stop or alter train speed according to signal state etc.

The

reason for adding the information from the booklet here is to confirm what has been published, not so much the prices which they state 'have yet to be confirmed' but the system. At last we know the elements for the system and how the track sensor will be fitted.

This helps us think about loco detection and how we will install it in our layout.

[Viaduct]

37lover said:

Viaduct,
Wow; that's impressive and shows up my blinked view. I thought it meant 106 tags that were encoded. Are you saying each tag has 106 codes? I don't understand your reference to binary otherwise. Be gentle with your

reply, I use computers but don't understand them!

Well, I'm only guessing, but with a detector that only needs a 1mm hole in a sleeper, I'm assuming that it is an infrared LED and receiver in a tiny package. I can't see any other technology (eg

RFID) being this small at the current time. Therefore the tags are likely to be reflective barcode stickers for the underside of the locos.

I'm assuming that the detector will work in binary ie reading just 0's and 1's from the barcode. Although I

don't know enough about barcode technology to be certain. So each tag is likely to be a series of lines. There'll be more than just the barcode for the loco ID though. Probably a series of lines to signify a start of a tag and end of a tag (I think someone

posted that the speed and direction of travel will be readable, so you need a fixed start and end code to be able to time this and the direction).

In binary, numbers are a series of zeros and ones, so 0000000 is ID 0, 0000001 is ID 1, 0000010 is ID

2, 0000011 is ID 3 etc until 1111111 is ID 127.

I suppose some of the unique 'numbers' or sequences might be reserved for the start and stop sequences, not sure.

I'm only guessing and really not sure how barcodes really work. I just thought

106 was an odd number to have available (only divisible by 2 and 53).

I'm sure we will find out more soon!

V

[Fishmanoz]

If I could just make some comment on what I think is slightly fuzzy thinking over the last 6 posts starting with PJ's at 8.34 on the previous page.

 

The 3 components making up the system have been known for a long time, we've learnt nothing new

 

there. We have learnt a little about the capability or capacity of them and prices though.

 

This is not a block detection system as such, put simply it is a loco ID tag detection system. However, you can think of it in blocks being the length of track

 

between 2 sensors, and that may not just be a single length if there are points in the middle of it.

 

The LD system will not be able to switch points or signals as such. All it will be able to do is report the detection of an ID tag at a particular

 

sensor. I would also be very simple for it to report the direction of travel of the tag if appropriately encoded. It may, and I repeat may, be able to report something about speed. However, given that the tag must be short to fit easily under a loco, the system

 

will have to measure quite accurately the time the tag took to pass over a sensor before that speed reading will have any degree of accuracy. And reading that time at all will be quite different to detecting ID numbers or direction setting codes.

 

So

 

if that's all it can do, how will an LD system be useful? The utility will be in your skill at writing programs that take actions with locos, points or signals etc conditional on the detections being reported by the system. Part of this will also mean that

 

RM will have to remember and have available for use in programs the conditions last set for all of those things - loco last speed and direction, point set direction left or right, and signal state etc.

 

Having a tag at the front and back of the train

 

will give you an advantage in knowing whether the entire train has passed a single detector or not. Otherwise, you will have to rely only on the loco arriving st the next detector to prove that it has fully passed the previous one. If it stops somewhere in

 

the middle, you won't know whether it has fully passed the previous detector or not.

 

All that the maximum of 106 tags in the system at any one time tells you is that there are at least 7 bits in the address system. We may never know what the missing

 

possible IDs are used for. The system must know more than just ID stuff too though and so have at least 8 or more bits being read. If it is to know which direction a tag is traveling, it will need to know which end of the tag it detects first, and it may also

 

need to know this so it can tell whether the address bits it is reading run from least significant to most significant or vice versa (unless the address is encoded twice on the tag, once for each direction of travel).

 

Finally,the loco itself will not

 

be able to report anything back to RM based on anything to do with the LD system, there simply isn't capability in the NMRA spec for its decoder to allow it to do so. Put another way, the loco will be connected to the LD system via its tag, but the loco can't

 

read the tag information, only the LD system and RM can do that. So while RM will know which detector the loco last passed, the loco itself will have no idea. All it knows is what its controller last told it to do with speed, direction and function status.

 

 

 

Does that help, or just make it more confusing.

[Fishmanoz]

Oops, that's last 7 posts as Phil posted while I was typing.

[Fishmanoz]

And for goodness sake, don't hit reply on my long post, just put your content in the Post Reply box at the bottom of the page, otherwise we will have 25 pages in this thread with multiple copies of what I said, along with previous multiple copies of what

 

people have said to get us to 13 pages in the first place - sorry, a pet hate of mine.

[Gregd99]

Phul007 said:

On my Calculation I will need 29 detectors as it is a 3 loop figure eight that intersects each other also I have a 6 track yard so each one would need a detector to stop.

With the caveat that everyone's layout is different

and people can use detectors in all sorts of different ways......

One way of using detectors is to determine which block is allocated to which train. I am not sure if RM will let us do that but that is what I do with JMRI. With block allocation implemented

all sorts of fun things with signaling and/or auto-running are possible!

Typically for location sensors (as opposed to block sensors) you will have a sensor at the entry and exit of each block. A block needs to be long enough to hold your largest train

with something to spare... so if you trains are one metre long (say) then a block size of 1.5-2m is a reasonable size.

Note that to stop a train in a 2m block you will have needed to set the correct signal in the previous block to slow the train down.

This braking becomes the limiting factor.... as in real railways.

In my 20m layout I use 6 sensors. Did cheat and put sensors at the throat of points so a single sensor actually detects entry into to either of the two blocks after the points. Without

this trick I might have had 10ish sensors.

Using these rules of thumb 29 sensors would need a layout of around 60m/200 feet!

Don't forget that you can use a sensor at the entry to a siding to detect the loco (and play a sound) and then to detect

the last carriage (and stop the train).

It is all fun and I found I learned heaps about signaling.

[Gregd99]

That's a very impressive layout.... mine has tools and electronics stacked on parts of it:-(

[St1ngr4y]

Hi Fishy,

I don't think the LD system needs to report anything about speed, because when a detection occurs, RM knows already what speed that loco should be doing.

Ray

[Gregd99]

St1ngr4y said:
RM knows already what speed that loco should be doing.
Ray

This is not actually the case.... RM know what speed it has told the loco to run at.

The actual loco speed will be somewhere in the range from

the "commanded speed" to the "previous commanded speed" as the loco speeds up or slows down. RM can estimate where the loco will be in that range.

Often this will be irrelevant but imagine the case where you say:-
1. set loco to max speed
2. stop

loco
3. if locoSpeed = 0 do something
Does RM use the estimated speed or the "commanded speed" in step 3?

[RDS]

Fishmanoz said:

.. sorry, a pet hate of mine.

and mine!

[PJ_model_trains]

Phul007 said:

If you look at
"https://www.flickr.com/photos/110505779@N08/11178933124/"

you will see my layout map. As it was many points that will require the train to stop if the next block is busy or full yeah 29 is needed.
I

agreed stopping speed will affect things especially in the yard , I think my plan will be put one sensor at the entry and then set speed to 1 and let it run into the end and after a set period of time then stop.

If this gets people more excited about

their layout that is great and hopefully we wont have to wait too long.

Phill

Hi Phill

A very impressive layout, I have seen part before as you mentioned how you control the lifting to turn it on it's side, also very impressive.

You

have a long layout so cannot see it all on the screen, I notice in the image you have removed the locos to show more of the layout. Another layout image that shows just how much room those large icon buttons take up.

Well done with your layout.

[St1ngr4y]

Phul007 said:

NON Hornby trains that RM doesn't know about?

Hi Phil,
I'm not sure what you mean by this. Lots of my trains are NON Hornby, but surely RM needs to know about all trains on the layout.
Ray