TrainTech new signalling product launch

[PJ_model_trains]

Additional note

 

I notice TT also do Track sensors only, the same as the signal but without the signal stand or head. 

 

From their sheet it states...

Track Sensor The Track sensor is similar to a Sensor Signal except that it has no signal head. It plugs into the track in the same way, works on both DCC and DC layouts and has the same sockets to connect it to other Layout Link products.

 

You can use a Track Sensor to:

• Link to a Mimic Switch or Light to show trains passing or locations

• Link to a Sensor Signal to change the following block section signal status

• Change a Semaphore signal automatically using the SC300 Dapol controller

• Trigger the LC10 Level Crossing lights and sounds when a Train approaches

ST1 

 

Maybe it could have other uses, station platforms, sidings, etc?

[PJ_model_trains]

In reply to my own thoughts two messages back, I see the following on the TT sheet

 

Sensor signals work exactly like the real railway where signals automatically turn red when a train enters a block section to prevent other trains from entering the same block and preceeding signals automatically change accordingly.

 

Bottom of page 4

http://ccgi.dcpmicro.plus.com/traintech/pdf_manuals/traintech_brochure.pdf

 

Interesting.

But no good to me I have over 30 TT signals already on my layout.

[RB51]

Hi PJ - in your post above the "additional note" I have a different understanding. I think the signal light sequence is altered by a detection at a signal sensor, not a loco, or carriages or wagons clearing a signal sensor. So it would, in theory, be possible to have a loco go past four signals but they are so close together that the first (at the train head) is red, the next one back is double yellow, the third on back is single yellow and the last one is green but the train is present in all four gaps (simulated blocks) between signals. As I say, just a theory and I recognise in practise you will have signals as far apart as you can manage. R-

 

EDIT: you posted again as I was typing. I think you are so far advanced with your system PJ that you are a little locked into the arrival of LD. R-

[Howbi]

@ PJ.........as I understand it, this new TT system is to control Block Signalling which will make a layout look more realistic.......it does not control the train movements so when you say that when a signal goes Red it protects that block that is only visually, there is no control over a following train entering that block against the Red...........the operator still has to control the train speeds to avoid collisions.........that's my take on it anyway...... 😀

[52train]

Hi guys it says about a mimic board does this mean that it will not change the signals on the railmaster track plan.

Do we need this board bought sepertly. I dont think TT said that it was train detection unless I missed that section.

Maybe it would be possible to add a system insulated rail joiners and relays to match the signals.

[RB51]

That is my understanding. There is no control link with RM - except as Chrissaf has discovered - to force a Red or resume auto operation. R-

[Chrissaf]

That's my take too HB.....there is no DCC feedback from the signals into the controller to take avoiding action. Also no feedback into RM to synchronise signal display in RM.

.

Since TT has stated that the sensor detects a wheel on an axle (see previous post showing TT response to emailed question) then the sensor cannot in all reason tell the difference between a loco and a piece of rolling stock behind the loco.

.

Thus in Roger's scenario where the signals are so close together (unrealistic) that a train spans across more than one signal. I would expect all downstream signals to get rather confused and may all try to go to Red.

.

The drawing below represents my understanding of what TT would be expecting the layout builder to design when using their SS products, using their 'Layout Link' wiring system.

.

/media/tinymce_upload/414ae227c41a245162afc68cb1686270.jpg

.

 

[RB51]

I think any board you have 52train will be of your own manufacture, but switches etc. can be obtained from TT. R-

[PJ_model_trains]

Hi PJ - in your post above the "additional note" I have a different understanding. I think the signal light sequence is altered by a detection at a signal sensor, not a loco, or carriages or wagons clearing a signal sensor. So it would, in theory, be possible to have a loco go past four signals but they are so close together that the first (at the train head) is red, the next one back is double yellow, the third on back is single yellow and the last one is green but the train is present in all four gaps (simulated blocks) between signals. As I say, just a theory and I recognise in practise you will have signals as far apart as you can manage. R-

 

EDIT: you posted again as I was typing. I think you are so far advanced with your system PJ that you are a little locked into the arrival of LD. R-

Hi Roger

 

No problem, there is quite a bit to take in and it is new. I think also there is excitement here that Hornby should be taking note of. 

 

I agree that it is not the loco that is detected and this system will not know what loco or where it was a train in reverse with carriages etc first.

 

The TT brochure shows the simple loco heading towards block, the common example but in reality a train will span two blocks as the leading end enteers another block. Therefore the block ahead and the block it is still in should both be protected with a red signal, only after a last carriage, wagon or guards van leaves the initial block should signals change 'back down the line'.

 

As with all TT products they are well thought out, good quality, and usually value for money. I did know feathers were being launched by the end of this year but was told in confidence and promised not to say anything, it was when I had lots of talks with them over their signals and adding feather signals, which I eventually went for CR signals ones, which look the same.

 

It will be interesting to watch this I am sure there are many considering it but I think we need to know a little more yet.

 

I have already thought about sensors for my TT level crossing lights but a controller and two sensors is a little costly. Maybe some sensors could work for sidings, being 4 per card would make it better value for money. Hornby no doubt will be looking closely at the system, and it cost compared to what they think they could evenetually launch LD at. Something else to watch for I think.

[PJ_model_trains]

@ PJ.........as I understand it, this new TT system is to control Block Signalling which will make a layout look more realistic.......it does not control the train movements so when you say that when a signal goes Red it protects that block that is only visually, there is no control over a following train entering that block against the Red...........the operator still has to control the train speeds to avoid collisions.........that's my take on it anyway...... 😀

Hi howbiman

 

I agree it is a brilliant step forward and hopefully will make Hornby think about LD, same old story, sit on your larels to long someone else will launch something similar. 

 

I also agree it willl look more realistic as you say (providing sensors are in signals, I think the stand alone sensors could look igly little boxes that need hiding, but fair play to TT, they have done what they have in a signal and given the option for single base sensors.

 

I underestand the sensor picks up the movement of the passing train, loco or carriage, which we are controlling, and changes signals accordingly, my only concern is not changing R,Y,YY,G when part of the train is still in the previous block which would need to be R,R,Y,YY until the last carriage passes the saif signal. Maybe this can be controlled in the controller I don't know yet.

 

 

 

[PJ_model_trains]

@chrissaf

A train can occupy one block or two

 

The blocks in the previous note were small to fit on the page in the forum

 

take a train consisting lof a loco and say just 2 carriages

 

___Sig-5__Loco-carriage-carriage_____Sig-4_______<<<

 

Here signal 4 is Red and Signal 5 is Green - the Train is in the Block (the gap between two signals)

 

The train moves on and we have this...

 

_Loco-carriage_S-5_carriage_______S-4_______<<<

 

Here the train spans two blocks, Sig-5 is Red and Sig-4 remains Red

 

It is only after the last carriages passes Sig-5, and the previous block is clear, and that the Sig-4 can change from stop to proceed according to aspect shown.

 

The TT example only shows the sequence R,Y,YY,G, we do not know yet if it can provide the above method.

 

The details in the brochure make me think possible he wont as it says...

 

Links to other Sensor Signals with a single wire for fully automatic block signalling!

 

and shows it that way in the diagram on page 4.

[RB51]

PJ, as soon as S5 goes to red S4 becomes Yellow, regardless of where the train actually is. There are no sensors to say any part of the train is still in the intervening block, just the sensor on the next signal to say that it has reached the next signal. R-

[Chrissaf]

I totally agree PJ....in real life that would be so.......but for some this is a toy (albeit an expensive one).....would TT be that accurate in modelling real life. There is one possibility, as a train passes Signal 5 (in your example above) the sensor will receive a trigger event as each wheel passes it. There may be a timer built in that has to wait, say for example, 2 seconds to pass without any further triggers before Signal 5 tells Signal 4 to change state from Red to Yellow. Once the trigger events stop, Signal 5 will know that the train has passed into the next upstream block. In electronics terms, such a timer is a relatively simple bit of code to include in the programming. Although that could be a possibility, I would tend to agree with Roger's view that change of state downstream would be immediate.

[Howbi]

@Chrissaf..........Chris, remember your RPR 220 sensor circuit with PNP transistor for LED on when shaded

/media/tinymce_upload/b4b945108a0bd08599bf5819561e0a70.JPG

the TT block signalling design prompted me to experiment with the RPR220 fitted next to the rail like this...

/media/tinymce_upload/b1b3a249cf9ba584791d0d60e1b489ba.jpg

it works very well, as the loco and train move past the LED flashes as the wagons/coaches intermittently shade the sensor until clear when of course the LED is unlit as no further current is induced.........

/media/tinymce_upload/3d6356786e5eb6f6d56dcf29f97b31da.jpg

[PJ_model_trains]

@Chrissaf 

 

I mentioned earlier a time delay seems the only feasible solution, that is if already encoded to do so. We don't know yet.

 

As much as I like TT products (and service) the system, if it works in sequence as we all think at this stage, everything points that way, it is wrong and also wrong to advertise as...

 

Sensor signals work exactly like the real railway where signals automatically turn red when a train enters a block section to prevent other trains from entering the same block and preceeding signals automatically change accordingly.

 

One could say there is a play on words here, the changed to Red as a train passes the signal to enter the next block, preventing other trains entering 'the same block'... preceeding signals changing automatically accordingly. 

 

But the system is not correct. It is saying a previous block is safe to enter, it is clear, when it may not be, in all probability won't be.

 

I stick out and wait for LD where we hope to be able to program signals, and other signals down the line, by tags read from first and last loco/carriage or wagons or guards vans. At least LD will know which loco passed which signal, and hopefully which last carriage, from that we can control our back down the line signals.

 

Another item we don't know yet with TT SS signals, does it automatically change signals back down the line subject aspect types installed. e.g. 2 aspect Red and Green, 3 aspect Red, Yellow, Green, 4 aspect Red, Yellow, Yellow/Yellow, Green. By that example what I am trying to say is it will alter the number due to the different aspects irrespective of how many we want to change.

 

We may only want to change a 4 Aspect from Red back to Yellow or Yellow/Yellow and no further due to a branch further back down the line. With LD  it seems we should be able to do this, with TT SS signals it doesn't.

[PJ_model_trains]

I totally agree PJ....in real life that would be so.......but this is a toy (albeit an expensive one)

 

I agree it is a toy for some, although I have to say for most it is definately much more than a toy.

 

We can thank loco manufacturers, but I think most credit goes to Hornby with RailMaster. Computer control has made much more possible than we could have dreamed of a few years back. This is where the advantage remains, with LD and computer control through RM all the items discussed in the previous messages in this and the LD thread are possible. Providing LD allows us to tag all locos and all last carriages/wagons/guards vans, we will have total control. 

 

I have confidence in Hornby, they have over the last few years shown us that they do listen to us.

 

If we start with all signals to proceed, except the ones occupied which are protected with a red signal, we can be sure (providing we have pre-programmed correctly) every signal/sensor passed will turn a signal red, and every last carriage will then change signals back down the line, to the full sequence discussed or part, back to junctions where we don't want to change further back down the line.

 

With LD we will have the control... when it eventually comes. It will come we just don't know when. Hopefully TT bringing the SS Signals out will help add pressure on Hornby, they now know what TT offer, but what they maybe don't know is who else is near, or testing to release their option.

 

So model trains are a toy to some, I tend to find the ones who consider them toys are mothers who buy a set for a kid for Christmas, for us they are real, everything except the smell of smike and an oil rag, although the latter is easy add.  😎

 

 

[Howbi]

@PJ........see what you've done by mentioning LD so much again......Mystic Meg is back...........

/media/tinymce_upload/207e26be08f5c0a7cdcac540f33eddfc.jpg

[Chrissaf]

HB, how can I forget (re RPR220).........looks like you have been busy today.

[Fishmanoz]

I'm a little puzzled by this discussion.  TT have announced the system works like the real railway, so instead of assuming the system won't work, why not hypothesise how it can.  You already know the system can detect the presence/absence of a whole train via detection of axles, then it must have delays built in so as to not false trigger between the first axle and the last of a loco or a train.  So why would it not be able to do its back down the line triggering on the loco or train clearing the signal, rather than on its first reaching it?

 

Also for closely spaced locos or trains, won't they just re-initiate the sequence each time each signal detects one, from that signal back, without getting confused?

 

And it's clear to me, it is just a clever signalling system using sensors to initiate a signal sequence.  It is not a DCC system, it is separate to any DCC you are running, except as Chris has noted about RM being able to force a red.  It does not replace or interact with RM, with or without LD, and cannot program detection to do other than initiate the signal sequence.  If you want detections to do other things, like the whole list of things possible in the list of LD commands in the locked thread at the top of the RM forum, you will still need LD to do them.

[Howbi]

It's always worth waiting the 12 hours time difference for Fishy's point of view......... 😆..........HB

[Fishmanoz]

Thanks HB, I try.

[PJ_model_trains]

Time delay was mentioned as maybe a consideration they may have used, it was only a consideration at the point in the discussion, we do not know.

 

Fishy is right, if we want the workings of LD we will have to wait a bit longer for it.

 

Picking up on the conversation, it seems feasible then that the sensor that detects the axle movement, keeps detecting the axle movement as the loco and all carriages pass over it. It is only after the last detection, when detection stops, that back down the line signalling would be initiated. This would be a better consideration than time delay, and would work so long as a train does not stop with ppart of the train in the new block and part in the previous block it has not fully cleared.

 

One thing for sure, TT have leaped forward in what is offered for signalling and loco detection, adding to their history of quality products and excellent customer service they will no doubt do well. Hornby, I think, should see this that anyone interested in Loco Detection, that goes for an alternative system, whether as good as LD or not, is in effect, potentially a lost customer.

[Fishmanoz]

Wasn't thinking just a fixed delay PJ, rather a short delay to see if more axles are coming without a false trigger.  That won't allow for your stopped overlapping trains though.  And try stopping one right on the point of axle detection and see what the system does about it, as Murphy will tell that's exactly where they will stop.  But they will have thought of all of that too so I wonder what their solution is?

[PJ_model_trains]

It is good for Train-Tech to fill a gap

 

It is good for LD to further that gap.

 

The biggest gap maybe time.  😉

[PJ_model_trains]

That brings us back to where LD left us after many years waiting and a few forum discussions   😎

 

Patience!

 

Hornby may define LD like a good wine it takes years to get it right.  😉