Greetings,
I am an electro-mechanical engineering student at Penn State Berks campus. Currently I have a class on the introduction to PLC basics utilizing the Allen Bradley RsLogix 500 PLC. I am working on the planning stages of a model train controlled by PLC project and am looking for input. I have a basic understanding of writing code but need help in reference to controlling the model train.
The basic overview and background to this project:
The model train will represent how the PLCs are used in industrial settings. I would like to be able to control the movement of an engine with a single car attached. There will be a user interface to chose specific functions or tasks to be completed by the train so I will need to be able to control the movement of the car(the speed does not matter--just movement). Also, I would like to incorporate signals along the trains path, crossing areas, and track switching which would all be automated by the plc. I have the track layout determined but I am not sure how to add it to this post. Basically it is an oval which the train will travel around in the clockwise direction. There will be two chances for the train to complete inner loops and an outer loop. Upon entering each loop I need the train to stop for industrial processes to take place then the train to continue around to complete the loop. I am leaning towards HO scale DC trains. I do not have the track and items in my possecion yet but will next week.
My questions are:
What I/O modules would you reccomend for these controls.
What methods can be utilized to control the trains movement?
Is it possible to communicate with DCC components with the PLC?
Any wiring recomendations for track and signals, etc.
What is PLC?
PLC = Programmable logic controller
You will not be able to put a PLC in the rolling stock (unless it is a very large scale). I think your only option is controlling the the power to the rails and automatically lining the switches.
Once you figure out how you will control the process you can determine what inputs to and outputs from the PLC are needed. Once you know that you can figure out what modules you need.
Think of communications protocols as languages. If two people do not know a common language, they will need a translator or not communicate. It is the same with electronics. I seriously doubt anybody has, or ever will, make a gateway to allow any DCC protocol to communicate with any process control protocol.
"No soup for you!" - Yev Kassem (from Seinfeld)
I was considering wiring each section of the track independantly and then have the PLC sequentually power each section of the track and power down the previous section as forward sections are energized. This will essentually control the engines location be able to stop and start it anywhere on the track. Also I would be able to possible control more than one engine at a time.
The most common module to use would be relay type modules.
Atlas snap switches for turnouts would be the easiest to set up. These are twin coil type machines. Activate one coil momentarily and the turnout sets one route, activate the other coil momentarily and it switches to the other route.
Most other devices will work with relays as well. Grade crossings and Signals, etc. These could be simulated with different color of LED's or small bulbs. Simple relays to control each LED on and off. Grade crossing lights flash alternately. If there are solid state relay modules, they may work as well but individual LED's draw small current, 20ma or so.
Block power following like you have suggested will work as well.
If you want to add some pizzazz, set up a module or two that can control the voltage to the track. Then you can start the train slowly by ramping up the voltage and ramping it down when you want to stop instead of a bang on and bang off.
DCC requires a specific digital pulse train type signal and would require you to purchase a DCC system to use as an interface to the PLC system. I doubt if there are any off the shelf PLC to DCC interface modules available.
Hope this helps.
Elmer.
The above is my opinion, from an active and experienced Model Railroader in N scale and HO since 1961.
(Modeling Freelance, Eastern US, HO scale, in 1962, with NCE DCC for locomotive control and a stand alone LocoNet for block detection and signals.) http://waynes-trains.com/ at home, and N scale at the Club.
Right down the road from me. Interesting project idea - great working a hobby into somethign practical.
I think Elmer pretty much covered it. Simple stop/start for the train is probably easiest. DCC interface would be much more complex - it would be possible to use a D/A module to control the throttle of a DCC system, but the only way you can get that off the shelf without actually cutting intot he DCC system would be a Digitrax Zephyr, and use a jump port - it's designed to hook a standard DC power pack up to add an extra throttle or two, and works with a 0-12 volt input. You'd still have to actually select the loco and activate it. The Bachmann EZ Command is probably cheap enough to not care about voiding warranty - you could tie a contact to one of the buttons, which is how it selects a loco, question is what needs to replace the poentiometer - you could use a servo output and just control it, rather than electrically repalce it. All depends on you level of electronics skill. DC control with simple relays shutting off the power at stop locations is DEFINITELY the easier route.
Turnouts, if you use the solenoid machines, need a momentary pulse output at realtively high current, or you could use Tortoise motors, which take a continuous 9-12V at low current, simply reverse polarity to change position.
One thing, if you use relay outputs to simply chop power to make the train stop at each location - wire the relay outputs to short the track when power is off. IE, when 'Go', track power is connected to the power supply, when relay goes to 'stop' it disconnects from the power suppyl but shorts the track. This will make the stops more precise - shorting the motor outputs will brake it to stop faster than just letting it float.
I love this sort of stuff, but never get to play. I was doing some work at RACC the other week and was in the building where they have the industrial control labs and it makes me want to take a class so I can go play with all the cool toys.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
I agree with the opinion that in this specific case, controlling the train's movement might be easier and less expensive using DC, unless you already have a DCC system laying around.
However, the LocoNet_hackers guys (LocoNet is Digitrax's communication method) seem to have a lot of experience interfacing PLC-type stuff with the LocoNet. You may want to join their Yahoo! Group and pose your question there. They may be able to give you some pointers or suggestions.
http://groups.yahoo.com/group/loconet_hackers/
how do you determine the position of the train?
if you could determine the location of the engine, you can simply control power and polarity of the track.
greg - Philadelphia & Reading / Reading
If we assume the car will always be coupled to the loco, a super simple way would be to use some magnetic sensors and a magnet ont he bottom of the loco and car. Or possibly trip contacts the signal the PLC when the loco gets to a specific point. Unless the object is to simulate a working railroad with multiple trains running, I don;t think there is a need for more traditional block detection.
It sounds to me that this is more like a simulation of an industrial transport system than a railroad system. So perhaps some of the things we'd need to make a realistic railroad function aren't strictly needed for this.
raider49xrWe could use the sensor to power up and power specific blocks of track which might make it possible to control more than one train. And yes the main idea is to model an industrial setting but the more processes we can include the better so it makes sense to use accessories compatible with model trains.
And yes the main idea is to model an industrial setting but the more processes we can include the better so it makes sense to use accessories compatible with model trains.
what do you mean by industrial process? and what accessories do you have in mind?
i write firmware and often use state machines. If the train needed to back into a spur, presumably
presumably the train is now ready for the next movement. Is the an example of part of an industrial process sequence thru a couple states?
If so, it seems that the events triggering the state machine would be the position sensors indicating the presence of the train and possibly a delay, resulting in one or more actions triggered by those events. Is this the way to think about the industrial process sequence(s)?
a TC4423/4/5 may be useful for controlling train (other things) power and polarity. But position sensors seem essential.
Think microcontroller on a macro level. After having some experience with PLCs and more so with machine tool CNC controllers, microcontrollers like the PIC are simply reduced scale low power versions of PLCs.You have inputs and outputs - contact sensors, switches, analog input, analog output - except much larger and capable of handling much higher voltages and currents. Guess it's just me, I find this stuff cool to play around with. I need to settle on a platform and start using it though. I started with PICs, but the whole DIY world seems to have gone Aurdino.