Scorpius wireless digital train system


Hi,

Oh every point has already been discussed, its all in the thread here:)


Rick

Actually, it hasn't. I have asked two specific questions that haven't been answered. I will ask them again, including more information so you will understand what I'm asking.

1. Different locomotives will run at different speeds, even if made by the same manufacturer. Even the top end speeds are different enough to cause a problem when these locos are run together. How does Scorpius handle matching the speeds of these locomotives so when they are consisted together they will start, run through all ranges of speeds and not "fight" each other. How does Scorpius handle this and where is this information stored.

2. How does Scorpius adjust the starting voltages of different locomotives.
Example; Loco 1 needs 2.5 volts to overcome the friction of its own motor and drive train while Loco 2 needs only 1.75 volts to do the same. Again, how is this handled and where is this information stored in Scorpius.

Note that this information is stored in each decoder in DCC.
 
Last edited by a moderator:
Actually, it hasn't. I have asked two specific questions that haven't been answered. I will ask them again, including more information so you will understand what I'm asking.

1. Different locomotives will run at different speeds, even if made by the same manufacturer. Even the top end speeds are different enough to cause a problem when these locos are run together. How does Scorpius handle matching the speeds of these locomotives so when they are consisted together they will start, run through all ranges of speeds and not "fight" each other. How does Scorpius handle this and where is this information stored.
How does Scorpius adjust the starting voltages of different locomotives.

2. Example; Loco 1 needs 2.5 volts to overcome the friction of its own motor and drive train while Loco 2 needs only 1.75 volts to do the same. Again, how is this handled and where is this information stored in Scorpius.

Note that this information is stored in each decoder in DCC.


Hi.
I answered this in post number 62 :)
Configuring is dead easy, figuring out the differences is the hardest part.

As post 62 says using times each loco takes to get around the track could be used........... I wont repeat it here.
A more complex method is using current sensing on the mobiule decoder and measuring back EMF to automatically configure the locos whilst actually being used together. I figure this would be months of testing alone to get right.

How does Scorpius adjust the starting voltages of different locomotives?

Easily, you can actually graph the throttle curve on the screen (laptop) or by adjusting numeric values on the controller. The laptop is quicker and you get a visualisation of what the throttle curve looks like.

Heres a pic showing screen grab off the laptop during configuring, a minimum 15% PWM, or 15%x12V=1.8V DC.

Once the curve is configured on the screen press send, press menu on controller, scroll to reconfigure page, press button to accept reconfigure, press exit. The throttle now has the latest throttle curve. These

You can see you can set a minimum speed and maximum speed. The X axis is divided into 10 sections, thats throttle movement. The voltage is on the Y axis or % PWM in this case is divided into 100 segments or 0.12V increments.

Capture261209.jpg


Instead of being a function of the controller I would configure the throttle curve to each mobile decoder, that way each loco can have its own independent min and max speed, and every thing in between if required, either a straight linear function, parabolic functions or customise the throttle curve to 1000's of possible combinations.


Rick
 
I am a bit confused. In post #8 you say "Im just appealing to people sitting on the fence or new comers to the hobby." Then you say how easy it is to get out a laptop and page through menus to continually configure or re-configure settings. What I'm not seeing is how do these hobby newcomers simply place an engine on the tracks, select it, and run it. It seems to me that Newcomers needing to page through menu after menu are going to be ebaying their trains and back to video games pretty quickly where things happen fast.
 
I believe, to jump in at long last, that Keith is looking for people with some vision, some interest in possibilities, who are quite receptive to new technologies, and so on. There are many in the hobby like that, but not so many that they comprise a majority. Maybe....15%.

The hobby wil probably draw in new users, and they will be much less averse to having to use a device other than a simple paddle. They will know to scroll through menus to get to where they need to be.

This system has a ton of possibilities. Just as there are many who feel DCC has set them free, there is bound to be a lot of people who eventually look to be set free from the complexities of DCC. It is the usual process...a new set-point of reference that become familiar, and then becomes something from which one would want to distance himself if given the chance.
 
I am a bit confused. In post #8 you say "Im just appealing to people sitting on the fence or new comers to the hobby." Then you say how easy it is to get out a laptop and page through menus to continually configure or re-configure settings. What I'm not seeing is how do these hobby newcomers simply place an engine on the tracks, select it, and run it. It seems to me that Newcomers needing to page through menu after menu are going to be ebaying their trains and back to video games pretty quickly where things happen fast.


Dont assume someone new to MRR cannot use a basic menu. A newbie is just a person, nothing said about his intelligence level. The way I see it everyone with a mobile phone uses a menu system because thats what every phone uses. And almost everyone has one. One has to ask why? :)

However buttons it will be (as well as the menu) :) A combination of the 2.

The menu is so easy any modeller could use it, watch a child use simple menus on Wii or Playstation, a lot of people are growing up with it, its not an issue and you can pack infinite number of functions with as little as 2 buttons and a trim pot if one ever wanted to.

The menus are on the controller only, simple to use programs on the laptop.

It is a top end system for ANYONE. I simply said I dont want to push it to guys who are entrenched and or happy with DCC. Hence it would suit someone who hasnt spents $100's or 1000's on DCC, hence a NEWBIE or someone sitting on the fence ready to jump.

Anyone can get started on Scorpius. A $399 starter kit would give you one controller, one turnout controller, one mobile decoder and 1 dongle. All wireless. Programs are free.

How to transfer a ID from controller to loco is in the thread but here it is again :)

How to set up a loco to run in basic mode.

1)Press menu on controller
2)Scroll to "SET ID' page
3)Scroll to ID number required, press select.
4)Transfer this ID to loco by placing loco on rails and press send on controller.
5)Exit menu

All configuring to loco from controller is done in the same way.

Configuring settings on the controller is done in a similar fashion.

Loco and controller are now on same channel. Ready to run.



To run a new Scorpius powered loco out of the box.

1) Install mobile decoder into loco, default ID number is 1
2) Install batteries in controller, default ID is number 1
3) Connect 12V DC direct to rails, place loco on track.
3) Turn trim pot on controller to make train move

It doesnt get any easier :)

No PC is required for basic operations or basic reconfiguring.

Laptop or PC is always OPTIONAL and recommended.




Rick


PS. Any other questions anyone please just ask, :)
 
Last edited by a moderator:
I have a question, Can the controler run more than one loco at a time? For instance, I have one train running, will it keep running while I am calling up another?


:cool:
 
Some problems with the reversing loop you show:

It will not be acceptable for the locomotive to always stop when passing through the reversing section. The reverser needs to switch the polarity of the power quick enough that the locomotive continues on as if nothing happened.

You need another isolation point on the other leg of the turnout. As shown, when you reverse the polarity of the loop you will still have a mismatch at point B and a short when the locomotive tries to cross that point.

You would need another beacon at the other isolation point for when a train is running through the section in the other direction.

The locomotive picking up a signal from an LED is not an acceptable means of controlling the reversing section polarity. If a loco is pushing cars with metal wheels, the wheels will cause a short before the loco gets there to signal the switch in polarity. I think you need to stick to the DCC method of actually allowing and sensing the momentary short and quickly switching the phase of the circuit.
 
Last edited by a moderator:
Some comments on your "block detection" scheme:

First of all, the addition of a sensor on the bottom of the locomotive complicates the installation of the decoder, believe it or not, greatly in some instances(on some locos you would actually have to mill paths in the frame for the wires to get from the decoder down to the sensor on the bottom, and even locating the sensor would be difficult in some situatios).

What you show is not "block" detection, but "point" detection. You are not sensing when a train is in a block, but when the engine passes a point. This causes some problems. When a layout is first powered up you have no way of knowing which blocks are occupied and which ones are clear. Even if you remember from the last time, if a train was added or removed it would chamge the state of the blocks, which your system would not detect. When a train enters a new block, the block it was in will not be clear until the last car of the train leaves that block, something your system can not determine(this is very important for signaling systems). Also, when pushing cars, your system would not see a block as occupied until the engine reached the block limit.
 
Last edited by a moderator:
Some problems with the reversing loop you show:

It will not be acceptable for the locomotive to always stop when passing through the reversing section. The reverser needs to switch the polarity of the power quick enough that the locomotive continues on as if nothing happened.

Firstly thanks for the heads up, much appreciated.

I can just shut down power for a few milliseconds so loco does not lose momentum and switch polarity, so thats sorted.

You need another isolation point on the other leg of the turnout. As shown, when you reverse the polarity of the loop you will still have a mismatch at point B and a short when the locomotive tries to cross that point.

Yup I just realised I should have moved isolation point A under B. Thats sorted.

You would need another beacon at the other isolation point for when a train is running through the section in the other direction.

Correct.

The locomotive picking up a signal from an LED is not an acceptable means of controlling the reversing section polarity. If a loco is pushing cars with metal wheels, the wheels will cause a short before the loco gets there to signal the switch in polarity. I think you need to stick to the DCC method of actually allowing and sensing the momentary short and quickly switching the phase of the circuit.

The only solution is to make a cheap decoder for the caboose with no motor drive.

One of my concerns is wiring up and isolating all those blocks where one power circuit could do the lot (except reversing loops of course) with the LEDs already attached to the board, just fix in place with hot glue.

You see one one the bonuses here that the LED in track system can talk to all the trains generally and not just trigger an event, it can give any message you program. I need this because for future functions I want to know exactly what loco is where without using transponders.

So the LEDs in track could DIRECTLY trigger locos to slow, slow more, stop, or proceed. Is this useful?

Im assuming block detection is a resistance detecting system? Or?
And can I assume the system knows theres a loco there or any metal wheeled carriage but not which one.....or? Obviously ID detection can be very handy?



Some comments on your "block detection" scheme:

First of all, the addition of a sensor on the bottom of the locomotive complicates the installation of the decoder, believe it or not, greatly in some instances(on some locos you would actually have to mill paths in the frame for the wires to get from the decoder down to the sensor on the bottom, and even locating the sensor would be difficult in some situatios).

What you show is not "block" detection, but "point" detection. You are not sensing when a train is in a block, but when the engine passes a point. This causes some problems. When a layout is first powered up you have no way of knowing which blocks are occupied and which ones are clear. Even if you remember from the last time, if a train was added or removed it would chamge the state of the blocks, which your system would not detect. When a train enters a new block, the block it was in will not be clear until the last car of the train leaves that block, something your system can not determine(this is very important for signaling systems). Also, when pushing cars, your system would not see a block as occupied until the engine reached the block limit.

Ok Im hearing you.
A photo diode at either end of the train
I guess it depends on what percentage of locos need anymore machining other than a 3mm hole. If it means milling or any other work other than a drill press or cordless drill this would be unacceptable. But can I ask what percentage of locos this would be? 5%? More? Preferably the photodiode (looks like a 3mm diameter bulb type (typical) LED) should be as far forward as possible in the loco.

Ok point detection, I get you. I guess the question is what am I trying to achieve?. The obvious reason why you would add block detection is for controlling trains and signalling. As above Id need an extra decoder for those wanting signalling and routing control.

If I decide its too hard I have a very nifty plan B which may appeal to some of you :)


Rick
 
Then the bonus of integrated ID detection which I havent really spoken about.

With the point detection concept, because of of nifty comms, Scorpius already has a great ID system, not just detecting a some train somewhere in block 14.

Using throttle mapping I can actually show in real time on your PC all trains moving around, signal status and programming and of course a routing program.

It saves for a 16 block system I belive the Digitrax RX4 has a msrp of $49.95 and does 4 zones. However it needs to plug into a BDL162, then into possibly into a LNRP, then to PC via a MS100.

For this I would need 4 of RX4, 1 x BDL16, optional LNRP then a MS100 plug, a cost of $350. Add this to the previous tally if you want detection. And I assume its a goody some folk would love to have:)

See all this is standard with Scorpius. Of course as Robert has pointed out we need to get a 50 cent relatively small photodiode into the chassis of the majority of locos.

So theres the trade off. The LED/photodiode combination on top of 2 way comms gives exact locations with throttle mapping and use of the laptop.


From here you can do any signal or routing, including trains being keep on exact timetable by beinf calibrated automatically at every beacon point. Once the motor has warmed up simply do one circuit. The system records the times between points and calibrates the motor at every point to stay on schedule. Good on alarge layouts where you would like a timetable so the trains run in sequence the same all day long.

So some of the things you can achieve are amazing.


Rick
 
...
The only solution is to make a cheap decoder for the caboose with no motor drive.
...

What about trains that don't have a caboose? Putting a decoder in the last car won't work because that will seriously limit what you can have at the end of the train(unless you put a decoder in every car).

...
Im assuming block detection is a resistance detecting system? Or?
And can I assume the system knows theres a loco there or any metal wheeled carriage but not which one.....or? Obviously ID detection can be very handy?
...
Block detection uses current detection to know when a block is occupied. Locomotives and lighted passenger cars will be detected because they naturally draw current. Any additional car can be detected by adding a metal wheel that has a resistor across it. Yes, ID detection could be very handy.
...
If I decide its too hard I have a very nifty plan B which may appeal to some of you :)
...

I 'd like to hear it.

...
It saves for a 16 block system I belive the Digitrax RX4 has a msrp of $49.95 and does 4 zones. However it needs to plug into a BDL162, then into possibly into a LNRP, then to PC via a MS100.

For this I would need 4 of RX4, 1 x BDL16, optional LNRP then a MS100 plug, a cost of $350. Add this to the previous tally if you want detection. And I assume its a goody some folk would love to have:)
.. .

To be clear, the RX4 is not needed for block detection, but it is needed for Transponding. Block detection will tell you if a block is occupied. Trnasponding will tell you who is in each Transponding zone.
 
Scorpius SMP

Ok heres the new system. Ive listened to the concerns of the members here and have addressed the issues accordingly.

It combines wireless with DCC and DC very cleverly using just one magic box for everything, a modular add on system with one box per 4 blocks, but it does everything from powering wireless locos, DCC locos and even DC locos. The magic box also does every other operation for trains, including turnout control, block detection, ID detection (transponding), signal operation, routing and auto reversing.

Add one more universal decoder for each set of 4 blocks.

Each universal decoder has 12V DC power input and 40 outputs or 4 blocks of 10 wires.

TrainslatestMultiStstDecoder.jpg


Each block or zone has the following outputs:
Track power for wireless, DCC or DC and auto reversing.
Block detection
Turnout control
Transponding
Signal outputs.

Each zones wiring is in a ten wire multicore that is split as required to go into your circuit.

Apart from track power, leftover transponder, turnout and signal outputs can be used in any zone.

Only one (big) power supply is needed for any system and no power boosters are required.

TrainlayoutMultiPurposeDecoder1.jpg


Once again the emphasis is on simplicity of hardware. Here just 6 components allowing for 1 mobile decoder and no signals, to do any complex task.


Close up.....showing transponder, signal and turnout connection.

TrainhardwarelayoutJPG2.jpg


* The thick purple wires coming out of universal decoder are multicored (10 core), this keeps things neat.


So how does track drive work?

Essentially each block is a seperate railway system, well actually the block the train is in now, plus the next block(s).

As the loco is about to pass to the next block (one transponder receiver is used with each block and probably adjacent an isolation point), the information is passed to the next block, assuming it is now just gone vacant. The information would include if the loco is wireless, DCC or DC and can run any one type in each block (or pair of blocks).

Interesting, this system can wireless, DCC and DC all on one track, albeit different blocks! I assume this would be a first.
And apart from transponding, lights and a few other functions this system can pretty much make a DC train do what a DCC train can do in terms of motor control.

All this is based on assuming theres a vacant block at any one time between trains.

The Scorpius mobile decoder is now hybrid (multi-protocol) DCC/Scorpius wireless.

DCC
The universal decoder receives throttle info and will send DCC motor drive commands to the block the train is in plus the next block only. The system finds out where the train is when it crosses its first transponder.

So this system is now forward and backward DCC compatible 100% :)

Use any DCC loco on this system. Prior to starting the train is configured as DCC and the track signal (power and data combined) is sent to rails accordingly. Example the block 16 says to block 17, heres comes a train, its ID number 6 and its DCC, so send DCC to rails and get ready to pass the message on to block 18 as soon as the train leaves my block, Ill let you know when........

WIRELESS
Throttle signal goes directly to loco as per traditional Scorpius.
Train is configured to say number 7 and the mode is wireless. In this case it simply sends constant 12V DC to the rails in its block as it moves around the circuit. Simple.

DC

Invite your mate over finally who only has DC trains, and run together on the same track with same controllers. And no your not dreaming.

Trains have this advantage over slot cars because of the block concept.
And with new new "SEQUENTIAL" hopping concept Ive invented you can run DC, DCC and Wireless together albeit not in the same block. (However Scorpius Wireless can nick power from DCC to run)

Even without ID, once a block is designated as DC to start it off initially the voltage output from universal decoder to rails is relative to throttle. 50% throttle = 50%x12V = 6V.

No drilling, no photodiode installation.

Ive made the universal decoders 4 blocks/zones for a reason, to keep track power, signals and transponders wires to a minimum length and keeps things neater. And 40 outputs is enough for one unit.
The transponder receivers could be plugged into the universal decoder so those not wanting them can save money.

So the entire system composes of:
1)Laptop
2)Dongle $99
3)Controller $199
4)Universal stationary decoder (per 4 zones) $199
5)Mobile decoder $39
6)Light board $29
7)Associated programs. (included)
8)Power supply (Allow $100)

Total cost to get started......$665 (assuming you have your own PC or laptop) and a rapid install time.

Total cost on a 12 zone system and 12 mobile decoders.....$1492 (assuming you have your own laptop or PC).

So I have only 5 components to develop which keeps costs down and speeds up development. Less inventory, less shelf space, less postage, less install time.

Yes you can do basic operations without a PC or laptop.

Im going to call this type of system Sequential Multi Protocol or SMP.
Hope you like it. Ive addressed the issues so far raised by the forum members (I hope) and thank everyones input so far.

Rick Field
Sydney
Aust.
 
Last edited by a moderator:
Just use it in Scorpius wireless mode, train will stop, go, etc. In this case it will just cost you for the controller and mobile decoder, $238 all up.


Rick
 
Last edited by a moderator:
. sounds very nice and all, but as you were pointed out you have multiple instances of Baron Münchhausen pulling himself out of the samp by his own hair.
well, since this thread seems to have come to a conclusion i think i can add that all is left to do for you Rick is to build a layout to test/showcase your consept.
 
....all is left to do for you Rick is to build a layout to test/showcase your consept.

I'm assuming that's what he's about to do..... He solicited input, received it in spades and appears to have listened..... [Some may even say he got a lot of free marketing advice :)]

My only comment for now is I think the following is a bad assumption:

All this is based on assuming theres a vacant block at any one time between trains.

That could mean a *lot* of blocks, but adjacent occupancy needs to be "supported" IMHO - Especially in yards etc.


Cheers,
Ian
 
...well, since this thread seems to have come to a conclusion i think i can add that all is left to do for you Rick is to build a layout to test/showcase your consept.


Hmmm, not sure why it has come to an end, Im sure theres much much more to think about before hardware architecture can be set in concrete.

Then theres the software side.


I'm assuming that's what he's about to do..... He solicited input, received it in spades and appears to have listened..... [Some may even say he got a lot of free marketing advice :)]

My only comment for now is I think the following is a bad assumption:



That could mean a *lot* of blocks, but adjacent occupancy needs to be "supported" IMHO - Especially in yards etc.


Cheers,
Ian

Thanks for that.
So a collision could occur if blocks occupied by different trains are adjacent? One is stationary, the othe knows to stop when it enters the block but maybe too late?
This system really needs that safety margin of one block, do real trains work this way on main lines, from memory they do. Please correct me if Im wrong.

And yards would be generally done maually? Or have smaller blocks for auto routed yard work? Additional transponder receivers could be used too to define areas.

Rick
 
...So a collision could occur if blocks occupied by different trains are adjacent?...

Huh?... Sorry, what collision?.... You said:

All this is based on assuming theres a vacant block at any one time between trains.

There are many occasions when there isn't a vacant block between trains..... [And that's OK, but the signaling system better know!]

Cheers,
Ian
 



Back
Top