DCC and Automation

[shephsj]

I have the Lenz 100 system

 

I am building a round and round exhibition layout and would like to have automatic operation.

 

How would I go about it or does anyone have a good source of information.

 

I have the PC interface, what software would you recommend?

[Lady_Ava_Hay]

I have the Lenz 100 system

 

I am building a round and round exhibition layout and would like to have automatic operation.

 

How would I go about it or does anyone have a good source of information.

 

I have the PC interface, what software would you recommend?

 

I think that Railroad&Co is the industry leader, mainly because it is well written and well supported. There are others and Roco has had some positive comment on here. Then again there is the JMRI suite which is free and therefore requires a good deal of knowledge of the open sourcing that goes into the programme.

 

You can therefore guess which is the most expensive but for near enough load and use, RR&Co is the best.

 

There have been some threads on here and recently I think and Two Tone Green of this parish is a keen fan I believe.

 

If you require a shuttle type operation.then try CML electronics.

[Nigelcliffe]

The Shuttle from CML requires a Digitrax command station, so no use on a Lenz system. On Digitrax kit, its excellent.

 

 

 

The biggest issue with automation is accurate train detection (note detection is different to identification, discussed below); work the detection methods carefully. There are numerous detection methods, track current drawn in sections, IR beam breakers, magnets and switches in track, etc.. The detection needs to get back to the computer through some sort of bus system. I don't know much about Lenz' bus (Xpressnet), but have a feeling that some users adopt S88 rather than Xpressnet for feedback on Lenz systems (so another bus standard which will require another computer interface).

Thereafter its software. If the software knows the starting position of trains then detection should be enough and you won't need train identification.

 

Train identification gets you dangerously balanced on the ragged edge; Railcom (promised as the next great thing in DCC for an awfully long time, and still very little to actually buy), RFID, Barcodes under trains, Lissy diode transmisions, Digitrax's proprietry transponders, etc.., etc...

[ian]

XpressNet is just for throttles, boosters and the computer interface. The feedback bus for Lenz is called RS and is only supported by Lenz and LDT - there is no information available to build your own modules.

 

Feedback can get expensive very quickly for automation - you often need more detectors than blocks so that the computer knows when to start decelerating and when to stop. RR&Co can work the train's position out within a block - provided the loco has been profiled within the program, thus saving on the cost of extra detectors and offsetting the cost of the program.

JMRI is free but needs a big investment in time and effort if you go down the automation route.

 

Fortunately most of the computer programs have free downloads that you can play with - try a few to see what they can, and can't, do and what they need to do it.

[greslet]

Must admit I don't follow the argument in favour of RR & Co, just looked at the price list and I'm alarmed. It all looks very very expensive compared to JMRI.

 

Yes, I'll agree JMRI does need some investment in terms of set up time, but JMRI offers very rich functionality and is infinitely adaptable through scripting to do just about anything you want. Dont be fooled into paying out good money for something with a few whizzy looking screenshots

[Edwin_m]

If only there was a JMRI equivalent to the RR&Co manual, describing the key features in a clear manner starting with how to install it and get it running! I get the impression that by the time users have worked out how to use it they don't find it worthwhile writing any documentation for others to follow. The documentation seems to assume the reader is a long way up the learning curve already. On a quick glance the same applies to RocRail.

 

At least that was my impression from looking at it a few months ago - if anything new has appeared, or it was there all along but I didn't find it, then I'd be interested to hear about it. I had no major difficulties making RR&Co do roughly what I wanted in signalling terms, but as explained somewhere else on this site I'm not convinced its method of train tracking is sensible for what I want to do with it, and there doesn't seem to be any way of changing it. I would also have to upgrade not only the software but also my PC to get it working properly.

[ian]

It's horses for courses, Greslet. Using a computer to automate some or all of your layout isn't an easy task. You need to be patient, methodical and willing to spend money on things that don't add to the visual appeal of your layout, like block detectors.

 

Getting to grips with Railroad & Co is a difficult task. It is a complex computer program and you are configuring it to operate a complex system (your model railway). However it is within many people's reach. Trying to achieve the same (remember we are talking about automation here, not using the computer as a control panel) with JMRI is a much bigger mountain to climb. You need to be happy tinkering under the bonnet of Java code to do it.

 

I have created and run simulated Minories to fiddle yard layouts in both RR&Co and JMRI and can honestly say, as a computer programmer in a past life, that the only reason I persevered with JMRI was out of sheer bloody-mindedness. I got them both to work as simulations (eventually), but if they were connected to a real world layout I'd bet a pound to a penny that the real-world tweaking necessary to get things to stop in just the right place would be a lot harder in JMRI.

 

It may be free, but to be honest I don't want to spend more of my hobby time wrestling a program into submission than I have to.

 

For use as a control panel, JMRI PanelPro is great. For use as a loco programmer JMRI DecoderPro is unequalled. For automating some or all of your layout it is very hard work.

[Nigelcliffe]

Must admit I don't follow the argument in favour of RR & Co, just looked at the price list and I'm alarmed. It all looks very very expensive compared to JMRI.

 

The software cost is pretty much irrelevant compared to the cost of train detectors for all but the simplest of model railways.

 

If you want automatic running, the computer needs to know where the trains are. That requires sensors all over the place. If bought assembled, it can quickly be £10+ per sensor.

 

JMRI is a good tool, I am one of those contributing to its development. JMRI is a big package, doing dozens of different things for different types of user. Some features are more developed than others. There are some JMRI automation tools around (eg. Georgio's robot throttle) which take some of the pain out of automation.

But for automated running, RR&Co has a clear advantage in being a packaged aimed at automated running, its got a user manual, etc.. Unless you're interested in programming, use RR&Co.

 

- Nigel

[RFS]

The software cost is pretty much irrelevant compared to the cost of train detectors for all but the simplest of model railways.

 

If you want automatic running, the computer needs to know where the trains are. That requires sensors all over the place. If bought assembled, it can quickly be £10+ per sensor.

 

JMRI is a good tool, I am one of those contributing to its development. JMRI is a big package, doing dozens of different things for different types of user. Some features are more developed than others. There are some JMRI automation tools around (eg. Georgio's robot throttle) which take some of the pain out of automation.

But for automated running, RR&Co has a clear advantage in being a packaged aimed at automated running, its got a user manual, etc.. Unless you're interested in programming, use RR&Co.

 

- Nigel

 

I entirely agree. To get JMRI up and running will take a lot of time and effort, and no doubt most rewarding for those who like that sort of thing. I evaluated both JMRI and RR&Co before deciding that my real interest was getting the trains running sooner rather than later. Although RR&Co had its price tag, for me I decided it was affordable. I'd just retired and thought life was too short for the JMRI solution. I had RR&Co up and running within 15 minutes although only operating points and trains at the time. Automatic running is the next stage for me but I'm glad all the functionality is there ready to use (and fully documented on how to do it) - I've just got to install the detectors.

[Guest Digital]

I have used RR&CO for several years now and for me it is the best program currently on the Market for automation. I agree it is quite expensive but if you want computer control and don't want to get involved with programming then this is the way to go. You can even download the software for free to evaluate and see what it can do before committing yourself to buy (time limited version). Excellent forum support as well as detailed instructions. Yes it is a very sophisticated piece of software but things soon fall into place and with a bit of time you soon get the hang of things. There are several threads on RMWEB about the software and a few experts here who can help two tone green being one of them as said before.

Yes there is expense with hardware such as occupancy detectors and other things (Which are needed for any automation software to work) but in my mind all worth the effort. It reminds me of when I first moved over to DCC and have not looked back since.

Regards

John