Has anybody ever build TMCC command base out of Arduino?
Original Post
Replies sorted oldest to newest
I'm working on a system that with an Arduino that will send commands to Legacy when I throw a switch from the panel. The command sent is to actually control the switch. I'm going to to use DZ-2500 switch machines.
I was thinking of trying to build an accessory controller out of an Arduino that responds to TMCC signals. I don't know if it could be done or not. I was thinking of using the Z-stuff data wire driver to get the signal to the Arduino. I'm unsure if a circuit would be need to for the data driver before the commands go into the Arduino or if the Arduino would be able to read the commands from the data wire using the Serial library. Also, the documentation shows a hot and common off the data wire driver and I'm not sure if this for power for the data driver or what. Can anyone answer this? I don't have one and planning on using DZ-2500 and with the data driver anyway. I know other people have made there own TMCC controllers like Millhouse River Studios for their turntables and transfer tables. They use the data wire for their system.
Z-stuff Data Wire Documentation
The Arduino would have to store TMCC ID and such in EEPROM and also would have to have a switch to put into program or run.
I think it would require a lot of work to build an actually base with Arduino. I know they have done it with DCC, but I think you would only be hurting yourself trying to develop your own system. Especially when it comes to LCS. However, I think an accessory control is possible instead of using an Electric RR board. This way you could setup your own commands for your bought or custom accessory.
I hope this discussion goes somewhere, since I have been thinking about this for awhile.
Chris
Pennsylvania & Ohio rr conway yard posted:Has anybody ever build TMCC command base out of Arduino?
I think it is impractical to "build a command base". Perhaps what you mean is, use Arduino to issue commands into the serial port on the base. That is very possible, and people have done that. You could also use a full blown computer for this task.
I saw this morning that you put a like on an old topic of mine, a good way to mark it so you can find it again easily. That was a good conversation, and a bunch of the guys involved are well versed in various aspects of this stuff.
The Data Wire Driver is simply an opto-coupled isolator. You can easily drive the Arduino directly from the Base 9-pin signals. You may want to opto-isolate the data output to the switches to avoid grounding conflicts.
Dale Manquen posted:The Data Wire Driver is simply an opto-coupled isolator. You can easily drive the Arduino directly from the Base 9-pin signals. You may want to opto-isolate the data output to the switches to avoid grounding conflicts.
Yes, one could control the Arduino from the 9-pin serial port on the base, but what if you want the Arduino near the accessory, which is not near the base. Would the data wire from the Z-stuff send the command or no?
When you say onto-isolate the switches? You mean track switches?
Chris
The signal coming out of the Data Wire Driver is identical to the signal going in.
The opto isolation provides a separation of the Base ground and whatever ground you attach to the Data Wire Driver for operating the switches.
Dale Manquen posted:The signal coming out of the Data Wire Driver is identical to the signal going in.
The opto isolation provides a separation of the Base ground and whatever ground you attach to the Data Wire Driver for operating the switches.
Dale,
I apologize, but I'm confused. Are you saying that the data wire drive doesn't send commands out on the data, but just sends commands to the base?
Chris
The Data Wire Driver is unidirectional. It only buffers/resends the serial commands coming out of the Base to its own data wire. Nothing goes into the Base.
Dale Manquen posted:The Data Wire Driver is unidirectional. It only buffers/resends the serial commands coming out of the Base to its own data wire. Nothing goes into the Base.
Okay, that is what I thought. So one could use it to connect to an Arduino if they wanted to get the commands coming out of the base. All your saying is in order to connect the data wire to the arduino you need to put the data wire through the opto-coupler because of different grounds. Correct?
Chris
Do a search for "Arduino" and look at:
Arduino Switch Controller for less than $16.
and
New Life for a ECU-1 Train Control Transmitter.
Using an arduino to read signals from a TMCC base, or to inject commands into the base is pretty easy, assuming you are familiar with both Arduino and the TMCC base. I've done this for my LionChief/TMCC bridge project, and once I got past some stupid mistakes I had solid communication back and forth.
I also built my own Accessory/switch controller using Arduino and inexpensive relay module boards which works quite well, and costs almost nothing compared to the real thing.
As for actually scratch building a TMCC base, I don't know that it would be worth the trouble, though I'm sure it is possible. I think it would be a much better use of time to replace the track signal with 99 cent 2.4GHz transceivers. I've put some thought into building a replacement for the R2LC board that uses these cheap transceivers and could plug right in to any normal TMCC engine, but have not yet actually done any physical experiments.
JGL
David Nissen posted:Do a search for "Arduino" and look at:
Arduino Switch Controller for less than $16.
and
New Life for a ECU-1 Train Control Transmitter.
Dave thank for sharing your posts with me. I really like your ECU-1 control transmitter.
JohnGaltLine posted:Using an arduino to read signals from a TMCC base, or to inject commands into the base is pretty easy, assuming you are familiar with both Arduino and the TMCC base. I've done this for my LionChief/TMCC bridge project, and once I got past some stupid mistakes I had solid communication back and forth.
I also built my own Accessory/switch controller using Arduino and inexpensive relay module boards which works quite well, and costs almost nothing compared to the real thing.
John ~ I'm aware that the Arduino can read and inject TMCC signals pretty easy. I'm very familiar with Arduino and TMCC/Legacy. I have my switch controller figured out. I'm going to use TMCC controlled switches, but besides throwing a switch from a panel. I want to be able to have panels along the layout, so I can also push a button to throw the switch. The Arduino will read the button and send out a command to a "Master" arduino over wireless serial and then the master will determine what TMCC command to inject into the Legacy base. The wireless serial module I plan on using is an HC-12. I have heard a lot go good things about the HC-12.
http://www.banggood.com/HC-12-...ntenna-p-973522.html
I know I can build my own accessory controller, so that Arduino can read the TMCC commands. What I'm having a hard time figuring out is getting the commands to the Arduino from the base. If I was going to do this. I would want to put the Arduino near the actual accessory and not the Legacy base. I was going to try to use the data wire from Z-stuff for this, but it might be better to have an another Arduino board read the signals and then send over wireless serial communications utilizing an HC-12 to the other boards. It would just act as a receiver for the TMCC signal.
I guess my point is I'm trying to DIY an Electric RR board with Arduino. The Arduino may be controlling motors and such, so I would like to have operate exactly like an Electric RR board if I could. I think it would pretty hard to tape into the TMCC rail signal with an antenna (I could be wrong), so I was thinking the Data Wire Drive would help. The whole point of creating my own Accessory controller is in order to use the commnds I want on the CAB-2 remote like brake, boost, aux1, and so on.
Chris
Gary E did some things along these lines some years ago. He programed a logic chip to operate switches from a panel without the nest of wire, etc. One of his projects. Here is a link to some of his activity.
My point is that you don't need a Data Wire Driver to interface with the Base. My SwitchMaster massive switch controller that I built several years ago using a PIC just interfaces with the Base serial output directly. http://www.trainfacts.com/trainfacts/?p=193
" I have heard a lot go good things about the HC-12. " I don't have any of these, but most of what I've heard is not so good, especially with the shorter range versions. I like the nRF24L01+, which are equally inexpensive, and are very reliable. they also send two way digital signals so they can be set up for self error correction, to make sure the proper message is always received.
As for an implementation, What I have planned for my own layout is to use a raspberry Pi at the main control station, which will talk to the TMCC/Legacy base, keeping track of any data sent from the Cab1/2 remotes, as well as any input from buttons or switched placed throughout the layout. I intend to use the RPi to provide a graphical interface for the layout as well as master control, then to use Arduinos scattered about to talk to it.
I think this is a good way to go where the switches are thrown vial relay boards at one central location, but manual throw buttons could be places anywhere on the layout that use radio to speak to the main base station.
I think it is more cost effective to use larger relay boards in one place, but you could use smaller ones with their own radios to have a wireless set-up.
JGL
Dale Manquen posted:My point is that you don't need a Data Wire Driver to interface with the Base. My SwitchMaster massive switch controller that I built several years ago using a PIC just interfaces with the Base serial output directly. http://www.trainfacts.com/trainfacts/?p=193
Dale,
Yes, I realize that. My point was that if the microcontroller is not near the base. It needs away to talk to the base. I can't foresee running an RS-232 line 20 feet or whatever from the base. Hence, why I was asking about the data wire and if that would work.
Chris
JohnGaltLine posted:" I have heard a lot go good things about the HC-12. " I don't have any of these, but most of what I've heard is not so good, especially with the shorter range versions. I like the nRF24L01+, which are equally inexpensive, and are very reliable. they also send two way digital signals so they can be set up for self error correction, to make sure the proper message is always received.
As for an implementation, What I have planned for my own layout is to use a raspberry Pi at the main control station, which will talk to the TMCC/Legacy base, keeping track of any data sent from the Cab1/2 remotes, as well as any input from buttons or switched placed throughout the layout. I intend to use the RPi to provide a graphical interface for the layout as well as master control, then to use Arduinos scattered about to talk to it.
I think this is a good way to go where the switches are thrown vial relay boards at one central location, but manual throw buttons could be places anywhere on the layout that use radio to speak to the main base station.
I think it is more cost effective to use larger relay boards in one place, but you could use smaller ones with their own radios to have a wireless set-up.
JGL
John,
The HC-12 link I posted in one of my previous posts and has a decent range to it and will work with the Software Serial library in Arduino. I don't know where you heard that they aren't good, but Dave Bodnar (trainelectronics.com) has used them. Here is a link http://trainelectronics.com/Ar.../HC-12-Serial_Radio/
I do agree that nRF24L01+ are good too and can place in "node" networks.
My switch system that I'm designing doesn't use any relay boards at all. I planned on using TMCC switches (DZ-2500) and the control panels around the layout will send to a master that will inject the command into the base.
Chris
A number of people have long serial daisy chains to drive multiple TPCs and other controllers. The RS232 signal from the Base isn't very fast.
Pick your poison. Either direct or buffered output will probably work just fine, but the Data Wire won't get you back into the Base.
Dale Manquen posted:A number of people have long serial daisy chains to drive multiple TPCs and other controllers. The RS232 signal from the Base isn't very fast.
Pick your poison. Either direct or buffered output will probably work just fine, but the Data Wire won't get you back into the Base.
Are there any examples of these daisy chains?
This is a really interesting thread and I am always impressed with the knowledge and talent on the Forum. As I read through this thread it seemed to me the LCS does everything that is needed. I have had mine for 2 1/2 months now and the LCS works flawlessly. I decided to have no manual controls of any kind and rely completely on iPads, iPhones and Cab 2's. I now have no regrets. Just touch the turnout on the iPad layout mimic to throw it. The complete layout track plan is spread out over 7 iPad screens (each loop and yard has its own screen.) Manual control switches can be installed on the layout if desired because the STM2's will report the turnout route setting back to the base and all other devices. The layout has both Tortoise and DZ 2500 turnout motors.
Since the layout is 2 rail AC there are a lot of detectors and relays to switch the power source of the turnout frogs and for the reverse loops. These would not be needed in 3 rail. The LCS also controls all the building lights, street and yard lights, signs, siding power, roundhouse tracks power, uncouplers and operating tracks.
The LCS eliminated a whole lot of control wiring and components. A visitor can pick up an iPad and an iPhone and operate the layout with just a couple minutes of orientation. It is totally intuitive. The layout is relatively complex with 680' of track, 45 turnouts, a turntable and 4 reverse loops.
I just built an accessory controller from an Arduino Pro Mini; it reads serial commands coming out of a SER2 module (and buttons from a control panel) and controls one of those ubiquitous 8-relay boards.
I wasn't sure if optical isolation was necessary - the Arduino is powered by an eBay AC switching module connected to layout common - but in case it was, I used a 6N139 as the line receiver.
The HC-12 modules look interesting. Can multiple modules can receive the same signal from a single transmitter? If so, you could have one transmit serial commands from the Base, and modules in rolling stock receiving the commands. Couple the module with Pro Mini and you have a mobile, easily programmable TMCC receiver for about $10.
I've ordered a few and will report if multi drop communication works.
Professor Chaos posted:I just built an accessory controller from an Arduino Pro Mini; it reads serial commands coming out of a SER2 module (and buttons from a control panel) and controls one of those ubiquitous 8-relay boards.
I wasn't sure if optical isolation was necessary - the Arduino is powered by an eBay AC switching module connected to layout common - but in case it was, I used a 6N139 as the line receiver.
The HC-12 modules look interesting. Can multiple modules can receive the same signal from a single transmitter? If so, you could have one transmit serial commands from the Base, and modules in rolling stock receiving the commands. Couple the module with Pro Mini and you have a mobile, easily programmable TMCC receiver for about $10.
I've ordered a few and will report if multi drop communication works.
Professor Chaos:
From what I have read the HC-12 can have more than two modules. The Arduino board would just need to confirm if the data is for that board, since the HC-12 doesn't have a unique id. Which is easy enough to do with the TMCC id.
The HC-12's have to be on the same frequency for this to work.
AmFlyer posted:This is a really interesting thread and I am always impressed with the knowledge and talent on the Forum. As I read through this thread it seemed to me the LCS does everything that is needed. I have had mine for 2 1/2 months now and the LCS works flawlessly. I decided to have no manual controls of any kind and rely completely on iPads, iPhones and Cab 2's. I now have no regrets. Just touch the turnout on the iPad layout mimic to throw it. The complete layout track plan is spread out over 7 iPad screens (each loop and yard has its own screen.) Manual control switches can be installed on the layout if desired because the STM2's will report the turnout route setting back to the base and all other devices. The layout has both Tortoise and DZ 2500 turnout motors.
Since the layout is 2 rail AC there are a lot of detectors and relays to switch the power source of the turnout frogs and for the reverse loops. These would not be needed in 3 rail. The LCS also controls all the building lights, street and yard lights, signs, siding power, roundhouse tracks power, uncouplers and operating tracks.
The LCS eliminated a whole lot of control wiring and components. A visitor can pick up an iPad and an iPhone and operate the layout with just a couple minutes of orientation. It is totally intuitive. The layout is relatively complex with 680' of track, 45 turnouts, a turntable and 4 reverse loops.
Tom,
Do you have the STM2 hooked up to the DZ-2500?
FYI, it isn't too much more difficult to write Arduino code to send and receive Legacy commands as well. My autonomous train control is very similar to yours, Professor Chaos, except that I wanted to be able to use Legacy commands for better control over announcements, effects, momentum, speed control, etc. You need the Legacy spec from Lionel, which requires signing up for the LCS Partner program - and for that reason I'm not sure I would be free to share that part of my code - but it's easy to sign up and get the spec for Legacy. It isn't as trivial as TMCC codes, but it's very logical and straightforward once you get your head around it.
To anyone reading this thread who is interested, it's super easy to build inexpensive Arduino-controlled devices that control 16-relay boards (less than $1/relay on eBay) that can throw switches based on CAB-2 or control panel commands. My control panel is so tight that I didn't have room for toggle switches, so I used cheap pushbuttons, just ONE per turnout, that toggle the state of the turnout (the panel has 3mm LED indicators to show the current orientation of each turnout.)
I use a bunch of shift-register boards called Centipedes from Mace Tech so even a $5 Arduino clone can control and/or read up to 128 devices (some read pushbuttons and relay contacts, and others control relays that throw turnouts or control accessories etc.) My project uses seven Arduino megas, about seven 64-bit (input and/or output) shift register boards, roughly 50 relays that are tripped by occupancy sensors, and about 80 relays for controlling turnouts and accessories. My control panel has well over 100 LEDs (white, green, and RGB) that indicate turnout orientation, sensor occupancy, block occupancy and reservations, etc. Each Arduino also has its own 20x4 LCD display that shows what each Arduino is doing, and several other devices that add to the fun and confusion.
So far I've run everything in "proof of concept" mode, including identifying/reserving routes, throwing all turnouts for a particular route, and real-time table-driven train control where my application constantly populates a "to do" table with instructions that are triggered by time delays or when a certain event happens (i.e in 30 seconds, accelerate train 4 at momentum 3 to speed 40, or blow the whistle of train 17 for 3 seconds when sensor 47 is tripped, or turn on accessory 5 indefinitely, or make an announcement, etc.) -- but have not yet put all the pieces together so the trains become autonomous. The number of trains that can run simultaneously is limited mostly by the number of sidings on the layout - and the layout itself is stored in a table that can be easily changed, so there's lots of potential for future expansion. I hope to run about six or seven trains at once on my current 15' x 30' layout -- more than that and I fear that the trains will just get bogged down -- but I won't know until I try it!
The two biggest challenges at this point are 1) Avoiding deadlocks where a train can't move, and 2) Avoiding Gomez Addams-style fiery crashes due to software bugs or hardware malfunctions.
Chris, yes I do. It requires a DZ1008 relay and a +5VDC power supply. I have posted the wiring diagrams on the forum. Look in my profile under activity, there are 3 wiring diagrams.
Randy P. posted:FYI, it isn't too much more difficult to write Arduino code to send and receive Legacy commands as well. My autonomous train control is very similar to yours, Professor Chaos, except that I wanted to be able to use Legacy commands for better control over announcements, effects, momentum, speed control, etc. You need the Legacy spec from Lionel, which requires signing up for the LCS Partner program - and for that reason I'm not sure I would be free to share that part of my code - but it's easy to sign up and get the spec for Legacy. It isn't as trivial as TMCC codes, but it's very logical and straightforward once you get your head around it.
To anyone reading this thread who is interested, it's super easy to build inexpensive Arduino-controlled devices that control 16-relay boards (less than $1/relay on eBay) that can throw switches based on CAB-2 or control panel commands. My control panel is so tight that I didn't have room for toggle switches, so I used cheap pushbuttons, just ONE per turnout, that toggle the state of the turnout (the panel has 3mm LED indicators to show the current orientation of each turnout.)
I use a bunch of shift-register boards called Centipedes from Mace Tech so even a $5 Arduino clone can control and/or read up to 128 devices (some read pushbuttons and relay contacts, and others control relays that throw turnouts or control accessories etc.) My project uses seven Arduino megas, about seven 64-bit (input and/or output) shift register boards, roughly 50 relays that are tripped by occupancy sensors, and about 80 relays for controlling turnouts and accessories. My control panel has well over 100 LEDs (white, green, and RGB) that indicate turnout orientation, sensor occupancy, block occupancy and reservations, etc. Each Arduino also has its own 20x4 LCD display that shows what each Arduino is doing, and several other devices that add to the fun and confusion.
So far I've run everything in "proof of concept" mode, including identifying/reserving routes, throwing all turnouts for a particular route, and real-time table-driven train control where my application constantly populates a "to do" table with instructions that are triggered by time delays or when a certain event happens (i.e in 30 seconds, accelerate train 4 at momentum 3 to speed 40, or blow the whistle of train 17 for 3 seconds when sensor 47 is tripped, or turn on accessory 5 indefinitely, or make an announcement, etc.) -- but have not yet put all the pieces together so the trains become autonomous. The number of trains that can run simultaneously is limited mostly by the number of sidings on the layout - and the layout itself is stored in a table that can be easily changed, so there's lots of potential for future expansion. I hope to run about six or seven trains at once on my current 15' x 30' layout -- more than that and I fear that the trains will just get bogged down -- but I won't know until I try it!
The two biggest challenges at this point are 1) Avoiding deadlocks where a train can't move, and 2) Avoiding Gomez Addams-style fiery crashes due to software bugs or hardware malfunctions.
Randy,
Sounds like a really cool project . Keep us posted.
AmFlyer posted:Chris, yes I do. It requires a DZ1008 relay and a +5VDC power supply. I have posted the wiring diagrams on the forum. Look in my profile under activity, there are 3 wiring diagrams.
Tom,
Thanks for the info.
Chris
Was doing some research tonight. Here is an RS485 shield for Arduino and use the software Serial library. It will work with 30 devices over 4,000 ft. cable length. Seems like a good option for hardwiring.
https://www.sparkfun.com/products/12965
So an Arduino would be connected to a SER2 and would then send the commands out over the RS-485 line.
Chris, how many receivers and what distance are you thinking of?
At 9600 baud, TMCC serial is not very demanding. If you have a Legacy SER2, then it may be enough to just rely on the native RS-232 communication without resorting to RS-485 or wireless.
You would need a receiver circuit at the Arduino to convert the RS-232 signal to TTL polarity and levels. This can be done with a transistor, optocoupler, or a MAX232 chip.
Professor Chaos posted:Chris, how many receivers and what distance are you thinking of?
At 9600 baud, TMCC serial is not very demanding. If you have a Legacy SER2, then it may be enough to just rely on the native RS-232 communication without resorting to RS-485 or wireless.
You would need a receiver circuit at the Arduino to convert the RS-232 signal to TTL polarity and levels. This can be done with a transistor, optocoupler, or a MAX232 chip.
Professor Chaos,
I can't answer that yet, since I'm still planning the layout. I think for a wire connection the RS485 seems the way to go or an I2c connection with buffers to increase the length capability of the I2C bus. Which typically uses cat5 wire. In the wireless arena I would use HC-12 or nrf24L01+.
I know there are several ways to do this, but original point was if there was away to use the zstuff data driver to talk to the Arduino. The answer that I got is basically no, so one needs to place a "translator" with a SER2 and that send the commands to the custom accessory controls over wired or wireless connection.
Randy P. posted:FYI, it isn't too much more difficult to write Arduino code to send and receive Legacy commands as well. My autonomous train control is very similar to yours, Professor Chaos, except that I wanted to be able to use Legacy commands for better control over announcements, effects, momentum, speed control, etc. You need the Legacy spec from Lionel, which requires signing up for the LCS Partner program - and for that reason I'm not sure I would be free to share that part of my code - but it's easy to sign up and get the spec for Legacy. It isn't as trivial as TMCC codes, but it's very logical and straightforward once you get your head around it.
To anyone reading this thread who is interested, it's super easy to build inexpensive Arduino-controlled devices that control 16-relay boards (less than $1/relay on eBay) that can throw switches based on CAB-2 or control panel commands. My control panel is so tight that I didn't have room for toggle switches, so I used cheap pushbuttons, just ONE per turnout, that toggle the state of the turnout (the panel has 3mm LED indicators to show the current orientation of each turnout.)
I use a bunch of shift-register boards called Centipedes from Mace Tech so even a $5 Arduino clone can control and/or read up to 128 devices (some read pushbuttons and relay contacts, and others control relays that throw turnouts or control accessories etc.) My project uses seven Arduino megas, about seven 64-bit (input and/or output) shift register boards, roughly 50 relays that are tripped by occupancy sensors, and about 80 relays for controlling turnouts and accessories. My control panel has well over 100 LEDs (white, green, and RGB) that indicate turnout orientation, sensor occupancy, block occupancy and reservations, etc. Each Arduino also has its own 20x4 LCD display that shows what each Arduino is doing, and several other devices that add to the fun and confusion.
So far I've run everything in "proof of concept" mode, including identifying/reserving routes, throwing all turnouts for a particular route, and real-time table-driven train control where my application constantly populates a "to do" table with instructions that are triggered by time delays or when a certain event happens (i.e in 30 seconds, accelerate train 4 at momentum 3 to speed 40, or blow the whistle of train 17 for 3 seconds when sensor 47 is tripped, or turn on accessory 5 indefinitely, or make an announcement, etc.) -- but have not yet put all the pieces together so the trains become autonomous. The number of trains that can run simultaneously is limited mostly by the number of sidings on the layout - and the layout itself is stored in a table that can be easily changed, so there's lots of potential for future expansion. I hope to run about six or seven trains at once on my current 15' x 30' layout -- more than that and I fear that the trains will just get bogged down -- but I won't know until I try it!
The two biggest challenges at this point are 1) Avoiding deadlocks where a train can't move, and 2) Avoiding Gomez Addams-style fiery crashes due to software bugs or hardware malfunctions.
Will it work with a cab-1 remote as well? I'm trying to modernize a club layout that as more on the floor control rather a control tower feel.
christhetrainguy09 posted:Randy P. posted:FYI, it isn't too much more difficult to write Arduino code to send and receive Legacy commands as well. My autonomous train control is very similar to yours, Professor Chaos, except that I wanted to be able to use Legacy commands for better control over announcements, effects, momentum, speed control, etc. You need the Legacy spec from Lionel, which requires signing up for the LCS Partner program - and for that reason I'm not sure I would be free to share that part of my code - but it's easy to sign up and get the spec for Legacy. It isn't as trivial as TMCC codes, but it's very logical and straightforward once you get your head around it.
To anyone reading this thread who is interested, it's super easy to build inexpensive Arduino-controlled devices that control 16-relay boards (less than $1/relay on eBay) that can throw switches based on CAB-2 or control panel commands. My control panel is so tight that I didn't have room for toggle switches, so I used cheap pushbuttons, just ONE per turnout, that toggle the state of the turnout (the panel has 3mm LED indicators to show the current orientation of each turnout.)
I use a bunch of shift-register boards called Centipedes from Mace Tech so even a $5 Arduino clone can control and/or read up to 128 devices (some read pushbuttons and relay contacts, and others control relays that throw turnouts or control accessories etc.) My project uses seven Arduino megas, about seven 64-bit (input and/or output) shift register boards, roughly 50 relays that are tripped by occupancy sensors, and about 80 relays for controlling turnouts and accessories. My control panel has well over 100 LEDs (white, green, and RGB) that indicate turnout orientation, sensor occupancy, block occupancy and reservations, etc. Each Arduino also has its own 20x4 LCD display that shows what each Arduino is doing, and several other devices that add to the fun and confusion.
So far I've run everything in "proof of concept" mode, including identifying/reserving routes, throwing all turnouts for a particular route, and real-time table-driven train control where my application constantly populates a "to do" table with instructions that are triggered by time delays or when a certain event happens (i.e in 30 seconds, accelerate train 4 at momentum 3 to speed 40, or blow the whistle of train 17 for 3 seconds when sensor 47 is tripped, or turn on accessory 5 indefinitely, or make an announcement, etc.) -- but have not yet put all the pieces together so the trains become autonomous. The number of trains that can run simultaneously is limited mostly by the number of sidings on the layout - and the layout itself is stored in a table that can be easily changed, so there's lots of potential for future expansion. I hope to run about six or seven trains at once on my current 15' x 30' layout -- more than that and I fear that the trains will just get bogged down -- but I won't know until I try it!
The two biggest challenges at this point are 1) Avoiding deadlocks where a train can't move, and 2) Avoiding Gomez Addams-style fiery crashes due to software bugs or hardware malfunctions.
Will it work with a cab-1 remote as well? I'm trying to modernize a club layout that as more on the floor control rather a control tower feel.
Chris,
Yes, Randy's system would work, but will only work with TMCC commands and not Legacy Commands. The cab-1L can decode Legacy commands from what I read. If you are using LCS you need a SER-2 module.
Chris
