DCC++ modified to produce Railcom cutout?

LMSFan72 Aug 19, 2018

  1. LMSFan72

    LMSFan72 TrainBoard Member

    118
    18
    9
    Hi All

    I wonder if anyone has already, or can, modify the DCC++ software to be able to use Railcom. I did find an old github branch but it appears 2 years old and still only alpha! I would greatly appreciate any input on both the software and changes required to the motor shield (I have both the std and Pololu so for wither) to enable broadcast and receive of the Railcom packets. I have a Mega.

    Many Thanks

    Andrew
     
  2. Quax

    Quax TrainBoard Member

    27
    5
    8
    Hi!

    I guess you refer to the approach of evili. Looks like his code creates a cutout in the DCC signal using an additional timer (2). I find this really interesting but have not tested his code so far. The author states that the brake signals are involved to signal the cutout (shorten the output), so you should not cut the brake traces on the motor shield. If you want a global detector inside the base station, then some circuitry (for detecting the data from the track/locos) would be needed. In addition, the data (RS232 coded, 250 kHz) must be read and interpreted by the software. Reading this data could be done via serial read or using timer/interrupt features. This may put more load on the processor but then, on the other hand, a separate programming track is no longer needed.

    Christian
     
  3. LMSFan72

    LMSFan72 TrainBoard Member

    118
    18
    9
    Hi Christian, yes, that's the one. Evili does include hardware schematics in kicad files in the repository. I am trying to work out those too! I took a look at the code, but, I haven't tried it either. I was trying to figure out if it was complete or not!
     
  4. Pieter

    Pieter TrainBoard Member

    152
    46
    10
  5. LMSFan72

    LMSFan72 TrainBoard Member

    118
    18
    9
    Thank you, that looks really good. There are both sketch and PCB designs available if you follow some of the construction links. Now that I have been looking harder at this I discovered that many of my decoders are not Railcom compatible - a lot of my Digitrax ones use their own transporting system. Therefore, I am also now going to look at Arduino based RFID to see what it there too! Thanks again
     
  6. Pieter

    Pieter TrainBoard Member

    152
    46
    10
    For decoder report back there are 2 flavors: Railcom (basically European decoders) & Loconet (USA decoders). For feedback (modules) you have to keep in mind your DCC system bus protocol. I don't use RFID but saw some feedback where speed and in some cases the scale used, caused problems. If you want to see what main feedbacks and its options are available have a look on Digikeijs website.
     
  7. Pieter

    Pieter TrainBoard Member

    152
    46
    10
  8. LMSFan72

    LMSFan72 TrainBoard Member

    118
    18
    9
    Hi, the barrier to railcom for me is that i use 8 different brands of decoder. Some have no transponding, and others have different types, not just railcom. I figured RFID was at least a common system. I'm less worried about bus/comms/interfaces as I believe the hardware is diy-able if it has to be, although the Digikeijs system lets you integrate them all as I read the spec.... I do see cost implications though with the RFID - readers are reasonably pricey and I'd need a few. More research...!
     
  9. Pieter

    Pieter TrainBoard Member

    152
    46
    10
    Paco has on his site a diy transponding module.
     
  10. Pieter

    Pieter TrainBoard Member

    152
    46
    10
    Digitrax sell 2 function (FL4 & 1) decoders that include transponding. Other make should also work the same. You should also be able to slave (lock) it to the decoder installed on your engines.
     
  11. LMSFan72

    LMSFan72 TrainBoard Member

    118
    18
    9
    Thanks Pieter. The challenge is finding the space to fit additional electronics inside some of the locos and then the cost of doing so. A series of reliable RFID readers and low cost tags may well be cheaper, more flexible and easier to fit..... I feel some experimenting coming on
     
  12. vasilis

    vasilis TrainBoard Member

    110
    39
    10
    Hi, is there any news? :)
     
  13. LMSFan72

    LMSFan72 TrainBoard Member

    118
    18
    9
    Still to start experimenting - track re-laying taken precedence with me!
     
    vasilis likes this.
  14. vasilis

    vasilis TrainBoard Member

    110
    39
    10
    Both approaches are interesting. I didn't know about the railcom cutout and the transponding. My first interest was in cab signaling and the rfid seems good for this, but the expensive implementations, the interventions in the track and speed limits discourages. For car's classification is the only way and i'm sure can be done in affordable cost.
    Among my searches i found this http://www.silogic.com/trains/RFID.html
    On the other hand the transponding in cutout is more attractive for use in cab signaling. I would be grateful for any sources about cutout, transponding, example with evili's code.
    I wish you succeed and i will be glad to see your results.
     
  15. RCMan

    RCMan TrainBoard Member

    271
    132
    12
    I like Railcom, it has many features, now there is Railcom+ that will identify the Loco when you put it on the track and down load the setup of the decoder.

    The real part of Railcom is the Engine ID number is transmitted back to the Command station and also the direction the engine or train is traveling, this includes going in reverse.

    Railcom works with both short and long address.

    You will need to have block detectors that can pass the information to the command station. I have a 16 block detector board and DCC++ and can do some testing for anyone's sketches.

    Biggest part will be the train software pages being configured for DCC++ to route the information correctly to the software.

    JMRI, ITrain, Rocrail, and Train Controller can all handle Railcom.
     
    vasilis likes this.

Share This Page