New DCC system


Vernon

New Member
I just built a new DCC system for N scale using an HS12 microprocessor that drives a 754410 H bridge and that IC then drives the track. This board is packaged with a Mod 5270 Netburner board to provide both internet and local control.

I attached files showing the track circuit board and the display while in operation. IP means internet control, A is locomotive A, Fwd is the current direction, and 1 is the speed. One will be able to switch to local control and activate a throttle lever and direction switch for both or either loco. It will be possible to have a friend run one loco from a web browser while you control the second locally. I will put in the second line when I get the other locomotive in about a week. The system will run two locomotives addressed as 2 and 3 - I think two at a time is probably enough for most people.

Try it at

http://www.cttestset.com/camtest.htm

I intend to add a couple of switches and the ability to control them ..
 
That's great that you have the technical know how to do that Vernon, and no disrespect intended, but the question is: Are you a model train hobbyist or an electronics hobbyist? I have so many unbuilt kits in the basement that the idea of building my own DCC system is a non starter. Most all the starter sets on the market now do more than you describe for a very modest cost. What was the component cost for your system, and how many hours did it take you to build it?

I have limited time for my hobby and tend to buy what I can ready to go, and save the building for what I can't get any other way, or for special projects that no manufacturer will likely ever do.

Cheers!
 
Are there starter sets with a web interface ? I am not aware of any.

I am essentially an electronics person and just bought the DCC locomotive, track and etc on Ebay for this project. I did not have a prior interest in model RR. I have a real airplane and my hobby is actual flight.

The object was not to exceed the norm in DCC capability - the point was to provide browser control of locomotives and other elements of a model RR layout. I think there could be an advantage in sharing control of multiple locomotives with remote friends. One could simulate the conditions of a large railroad with, say, 16 engineers on realistic runs controlling respective locos from their own home and executing complex logistical missions.

The hardware cost:

I drafted a circuit board and had it prototyped (10 boards) by Advanced Circuits $295
Nano Core 9S12C128 processor module from Technological Arts $75
Parts for Ckt board (approx) $20
Netburner MOD5270 embedded processor board with ethernet $79
2 x 16 display from Jameco (approx) $15
Trendnet IP camera $165

It took maybe 40 hours to design the circuit board, program the HS12 in assembler and the MOD5270 in C.
 
Last edited by a moderator:
Are there starter sets with a web interface ? I am not aware of any.

Computer interface capability yes, web based capability no, but that could be a matter of software couldn't it? I would think that, for this concept to catch on, some sort of interface between existing systems and the web would be more likely to draw interest.

I am essentially an electronics person and just bought the DCC locomotive, track and etc on Ebay for this project. I did not have a prior interest in model RR. I have a real airplane and my hobby is actual flight. The object was not to exceed the norm in DCC capability - the point was to provide browser control of locomotives and other elements of a model RR layout. I think there could be an advantage in sharing control of multiple locomotives with remote friends. One could simulate the conditions of a large railroad with, say, 16 engineers on realistic runs controlling respective locos from their own home and executing complex logistical missions.

That explains alot. Most modelers I know want to be there to see the models they have built run on a real live layout, theirs or someone's. There are programs for virtual railroading like Train Simulator that are set up for group activity via the web. Not to say there would be no interest in what you propose, but I think that those who would be interested in it would not likely be model railroaders. Railfans maybe?

The hardware cost:

I drafted a circuit board and had it prototyped (10 boards) by Advanced Circuits $295
Nano Core 9S12C128 processor module from Technological Arts $75
Parts for Ckt board (approx) $20
Netburner MOD5270 embedded processor board with ethernet $79
2 x 16 display from Jameco (approx) $15
Trendnet IP camera $165

It took maybe 40 hours to design the circuit board, program the HS12 in assembler and the MOD5270 in C.

Well, it will be interesting to see what other comments your post draws!
 
Dang, Vernon, I have no idea what you're talking about but controlling remote layouts over the net? There's already software available to control your own layout by computer and some of the folks here use it regularly for large operating sessions. The problem I see is that model railroaders like to be next to their trains when they're moving, seeing what the actual cars and engines are doing. They also like to look down the track for any trouble ahead so they can get things stopped or change routes before there's a major problem. In addition, coupling and uncoupling cars is by no means automatic and certainly not trouble free. None of this takes into account the occasion derailment, which can be fixed by a human in a few seconds. In a perfect minature world, you idea sounds pretty good. Model railroads are pretty far from a perfect minature world.
 
Are there starter sets with a web interface ? I am not aware of any.

I am essentially an electronics person and just bought the DCC locomotive, track and etc on Ebay for this project. I did not have a prior interest in model RR. I have a real airplane and my hobby is actual flight.

The object was not to exceed the norm in DCC capability - the point was to provide browser control of locomotives and other elements of a model RR layout. I think there could be an advantage in sharing control of multiple locomotives with remote friends. One could simulate the conditions of a large railroad with, say, 16 engineers on realistic runs controlling respective locos from their own home and executing complex logistical missions.

The hardware cost:

I drafted a circuit board and had it prototyped (10 boards) by Advanced Circuits $295
Nano Core 9S12C128 processor module from Technological Arts $75
Parts for Ckt board (approx) $20
Netburner MOD5270 embedded processor board with ethernet $79
2 x 16 display from Jameco (approx) $15
Trendnet IP camera $165

It took maybe 40 hours to design the circuit board, program the HS12 in assembler and the MOD5270 in C.

Neat concept, but I think I would do it differently. Since NCE and other DCC systems can tie into a PC, its probably not that far of a stretch to modify the Decoder Pro software to allow the "virtual throttles" in Decoder Pro to be access via a web browser from a remote site. Decoder Pro is freeware, and this may be something they are already considering for future upgrades.


Lets face it, any modeler with a sizable pike (big enough to allow remote operation) is already going to have an NMRA compliant DCC system (have you looked into that yet? Your system may fall short with the limited addresses).

Oh, and most folks running 2 at a time is enough? Then call me abnormal!! I've run 6 consists at a time with one NCE Pro-Cab.:D

Like I said, neat concept. I can think of more than one layout that I have run on before and would enjoy running on from home if I could. Just think of all the gas I could save!!:D Now if you could integrate a high quality miniature camera in the lead loco, and DCC with sound, all remote controlled in a web browser , now we're talkin!!:)
 
I agree that it is certainly better to be there !

The idea is to share control with someone who, for one reason or another, can't be there. Like we know it is better to visit your cousin but, if you can't, his Facebook stuff might be interesting - as would controlling his Model RR from a couple of states away. My plan was to build 4 x 4 and 4 x 8 N scale tables with a selection of track layouts, integrated DCC control of two engines and switches. You install your scenery, put on the trains, plug it in the wall, plug in an ethernet cable from your router, set up the router and share.
 
All you need is a PC interface, such as Decoder pro, and a computer that can handle multiple online users accessing it...
 
DCC permits 127 addresses so it would be possible to control a large number of locomotives. I just thought that two would be enough for most people - I have one amp available so I could probably do 4 or 5 (N scale) with the existing H bridge and power supply. I believe there are boosters available to increase the power for more locomotives. I could have one web page for each of say 64 locomotives if someone wanted to run that many.

I can certainly relate to the part about having somebody there. I went down to the internet cafe to test it remotely and the cat jumped up on the table, watched for awhile, then knocked the caboose off the track.

The issue with the "Decoder Pro" idea is getting the source code so you could add a TCP/IP server. They probably would not supply it. I use an external server because I don't want the entire world accesssing a port on my PC. There is not much they can do to the Model RR server in the way of viruses or hacking and it works even if the PC is turned off.

I like the on-board camera and may add that.
 
Vernon, you need to look at http://jmri.sourceforge.net/ and read everything that's going on with JMRI and Decoder Pro. It's all open source so you can look under the hood and see what you think. You may be better off trying an add-on to JMRI than trying to reinvent the wheel from scratch.
 
Dear kbkchooch,

CP6027 has just replied to a thread you have subscribed to entitled - New DCC system - in the Wiring, Electronics and DCC forum of ModelRailroadForums.com.

This thread is located at:
http://www.modelrailroadforums.com/forum/showthread.php?t=13983&goto=newpost

Here is the message that has just been posted:
***************
You guys are doing a great job of running this guys ideas and design down. We will be lucky if he makes another post. The knowledge required to do what he is doing is not normally found in model railroading, why don't you try and encourage him instead.

Reminds me of my post on hooking up my Tsunami sound decoder to my home theater speakers. For some reason I had to explain the practical use of the project.

Maybe you have no need for what he is doing but someone else might.
***************

CP6027, what happened to the post????:confused:

I don't think we (Jim, Alan & Myself) were running the ideas down. I even suggested that with the addition of an in-cab camera and sound, it could be even more interesting. What we were suggesting was there were other parts of his concept that bear further thought. DCC standards & Decoder Pro are both ideas that one must bear in mind before fully developing this type of system. Also the real life issues of a derailment (not to mention dirty track of the household pet on the track) have to be taken into cosideration.

I think the responses given were completely appropriate.If Vernon is serious about development of this concept, it will be information that will serve him well as the process continues. I wish him well with the idea! :)
 
Karl, I think maybe that was a PM...or something, not sure. :confused: At any rate, I certainly don't see any of us as trying to chase Vernon off. He's already invested a fair sum in this project. Since he says he a new model railroader, I'd hate to see him spend more money on electronics and programming if there's a way to piggyback on what JMRI and Decoder Pro is already doing. I also think it's good he understands some of the real world issues of model railroaders and how we think. Just as I would want advice from experienced programmers if I was embarking on a new project, I hope Vernon sees us as giving advice from a model railroading perspective. Heck, I barely know how to plug in a USB cable but I'd hate to see him spend more money on electronics if there's another way when he could be buying more engines. :)
 
Karl, I think maybe that was a PM...or something, not sure. :confused: At any rate, I certainly don't see any of us as trying to chase Vernon off. He's already invested a fair sum in this project.

Absolutely not! I just wanted to know what he as going for, which he expalined. I think an interface with already available systems would be the way to go with this, but I'm not the guy doing the work, so my opinion and $2.25 will buy you a coke at Chicago O'Hare! I just didn't want to see him spend so much effort on an entirely new DCC system when so many are available. Would you buy a second system to play trains this way? Know anyone who would? How about a black box that hooks up to your existing system and enables the web capability? I think that would be more likely to generate interest. If Vernon is developing this for commercial reasons, he's going to have to answer questions like this anyway, unless he's independently wealthy and can finance his own developement and get the product to market himself, then finance ther production of enough units to satify demand. If this is a hobby project for him & his friends, then that's another matter. I think the in cab camera or some sort of video feed would be essential. You can't run a train on instruments! At least not at most operating eseeions I've been to. Someone always seems to figure out how to bugger up the works, and if you can't see his mess, you can't avoid it!

Good luck in any case Vernon. I don't know if I'd be interested or not, but some will surely be.
 
DCC permits 127 addresses so it would be possible to control a large number of locomotives...

Actually, DCC permits over 10,000 addresses. If you got your addressing information from NMRA standard 9.2, then you need to look at recommended practice 9.2.1 which allows for extended addressing. Although it is not required(that is why it is a recomended practice, not a standard), every current decoder that I know of allows for extended addressing.

You may initially think that extended addressing is not important since you're not likely going to be running more than 127 engines at a time, but there are a couple of reasons why it is important. First, many modelers have more than 127 engines. Even though they don't run them all at the same time, with only 127 addresses, you are either limited with which engines can be run at the same time because of duplicate addresses, or you have to reprogram some addresses to be able to run certain combinations. The other reason for extended addressing, which applies even if you only have a few engines, is easier addressing. Extended addressing allows for a 4-digit numeric address and most locomotives(in the US anyway) have a 4 or less digit road number, so with extended addressing most people assign the road number as the address.

...I use an external server because I don't want the entire world accesssing a port on my PC. There is not much they can do to the Model RR server in the way of viruses or hacking and it works even if the PC is turned off...

This is the biggest advantage to what you are doing. Something I suspect most people reading this thread don't realize is that your system is standalone, i.e. it does not need to have a computer(other than the one that is accesing the web page).

The people working on JMRI have done some work in this area, the current version actually has a mini webserver built in. Unfortunately, the sample webpages they have are not very useful. Although a programmer by trade, I have never done any web development and I have no idea how much trouble it would be to build on what they have done.

Something else that JMRI has is LoconetOverTCP(which, as far as I know, only works with Digitrax command stations). The computer connected to the layout can run a LoconetOverTCP server. Then you can run JMRI on another computer using LoconetOverTCP as the layout connection and connect to the server over a network or the internet. With this setup, you can do anything on the remote computer that you can on the local one.

Of course, with any JMRI solution, you have to have a computer connected to the layout.

If I was trying to develop something like this, instead of developing a complete command station, I would use a Netburner or something similar to develop an interface to a commercially available command station. That way, in addtion to the web control, you would be able to use commercially available throttles and other accessories. I think it would be really neat to have it interface to Digitrax and not only give you web access but also have a built in LoconetOverTCP server.
 
i don't know, to me seems like tremendous effort to recreate something that already sort off exists. if i had to allow a person to join into running my layout i'd stick to DCC throttle software running on PC and that user logging in remotely. with today's VPN tech security is not an issue.
besides, that PC can also run another tasks like VOIP or better yet- a video call (with movable cam like logitec orbit) so the remote person is much more "present" so to say.

but then again its a hobby. sounds like you enjoy this project so by all means go for it.
 
You should take a look at RocRail for additional ideas.
http://wiki.rocrail.net/

Or perhaps think about integrating your work into that project.

Rocrail includes computer control (client/server), fully automated running of trains and accessories, web browser control (including from palm, blackberry), other mobile platforms using Java client.

The project also includes (but does not require) designs for DIY hardware of all sorts for DCC including DCC control stations with computer interfaces.
 



Back
Top