Throttles

Bruce

From Figure 1.  It looks like you still have to have throttle network for each Brand of Throttle and the network will then connect to the LLC module.    You will not be able to plug a NCE throttle into Digitrax UP5 or I am wrong?

Thanks

Doug W

Throttles

As I see it, you plug your throttle into a node that is appropriate for the brand of throttle. The node speaks Lenz, for example, on one side and LCC on the other. So, if you want distributed throttles for all three brands, then, yes, you would need a XpressNet for Lenz and a Cab Bus for NCE and a LocoNet for Digitrax. Not very practical.

However, I've been in several cases where folks have purchased one brand of DCC system and wanted to keep their throttle investment and change the basic system. So, for example, you keep your LocoNet for throttles and use LCC to transition to Cab Bus to operated an NCE system.

I found the info very

I found the info very informative and easy to follow thanks to your writing. I also have discovered that it will be something not needed for a layout I am likely to build ever. Even if I hit the lottery and build a barn filler the systems I am likely to use would not need a system such as that. Think ground throws very few electrically controlled turnouts etc. In working on our club layout I have discovered things I find very enjoyable and things I would happily do with out.

I have also discovered that any layout I build for my self will be large but rather simple and involve long runs simple track and the least amount of electrical gadgets possible. I have thought of ways to eliminate much of the electronic gadgetry that has become common place and to eliminate the up keep associated with such things.

Even though it is something I can say I will not use knowing what it is and what it does will allow me to make a very informed decision regarding such a device. For doing that I thank you tremendously as now I know about a great piece of equipment and that it has great potential and the ability to do many things even though I won't need it to do those things for me.

Throttles

Hi Bruce/Doug,

   The picture Stephen showed is of my development nodes at the Cleveland National last year.  I have been thinking of bringing it to the NMRA meet in Tucson in a few weeks if I can find the time.  What is not mentioned is not only is there NCE/Digitrax/Xpressnet throttles there is a software throttle on a computer, pure LCC throttle on a a touchscreen and a WiFi throttle on a mobile device operating that train.

"From Figure 1.  It looks like you still have to have throttle network for each Brand of Throttle and the network will then connect to the LLC module.    You will not be able to plug a NCE throttle into Digitrax UP5 or I am wrong?"

In the photo no you can't.  It would need another adapter node to convert the LCC stream to Digitrax at the UP5.  I never owned a commercial DCC system (always designed my own command station) so I did not prototype up one of those.

"As I see it, you plug your throttle into a node that is appropriate for the brand of throttle. The node speaks Lenz, for example, on one side and LCC on the other. So, if you want distributed throttles for all three brands, then, yes, you would need a XpressNet for Lenz and a Cab Bus for NCE and a LocoNet for Digitrax. Not very practical."

Not necessarily, these nodes where generic prototyping boards I built.  The only reason there is 3 of them for the 3 busses is that was all I could do with the hardware I had.  There is no reason a single higher end microcontroller could not manage the LCC bus and generate the at least 2 if not all 3 of the busses.  Or a single node could have mulitple low cost microcontrollers and talk between themselves with SPI or I2C in a single node.

Bruce one thing that everyone is getting mistaken on is that LCC = CAN.  LCC does NOT = CAN.  We had to have a CAN document to define the CAN interface into LCC because there is not a common physical layer standard that exists (connectors etc).  LCC is a PROTOCOL that can run on ANY physical medium.  I have abandon CAN and now have everything working on Ethernet and WiFi (the physical interface, connector, etc exists from the IEEE)

Jim

Rob

You hit on what I was trying to do and I'm so glad it resonated.

I've watched the NMRA-net, LCC development along the way. I'm very thankful for the public exposure at the Portland convention and felt that it was time to "go public" here.

Isn't it nice to have something that you do NOT have to monitor and think about?

Let's enjoy our dark territory together.

Jim

THANK YOU for your comments. I have been trying to nail down who did that demo, to no avail, until now.

I'll be in Tucson, please look me up if you come. I'd love to chat.

Apparently, I didn't do justice to your efforts in the text. I understood that what was shown was just a, "see what we can do!" moment. Like you, I see no reason that there couldn't be an (almost) universal throttle connection node. But you are closer to the hardware and software than I am.

Again, I think this "universal throttle" solution is becoming possible through LCC, but at a time where the smart phones are taking over that space. They run on any DCC system and are wireless. The ergonomics are very good, given the form factor of the phones. They still lack the KNOB, that I like.

Great Article

Thanks Bruce. While I was at Portland I had to choose between other higher priority clinics running at the same time as the LCC ones. My stop by the LCC stall gave me the impression that while LCC could become very useful, there are as yet no commercial products available.  Your article was a very good intro and I hope you and/or MRH keep us appraised as things develop.

I am presuming that LCC wil require another Bus but I have no idea what the possible current draw and ergo the wire gauge should be.

Thank you too for your great tip on clearing slots. Everywhere I go this always seems to be a problem and until now I hadn't realised that an easy solution could be found in Decoder Pro.

 "but at a time where the

Hi Bruce, Have you done any articles on setting up smart phone control for digitraxx? I'd like to know more  to consider it for a new layout I'm working on and I find your explanations always easy to follow. Regards, Dave Branum 

Peter

Yes, an LCC bus. Which is nothing more than Ethernet cables - available off the shelf.

Dave

No need to wait. Works for both Digitrax and NCE that I've done.

Get your layout running on JMRI through an appropriate interface.

Have your computer connected to a WiFi network. There are lots of ways to do this with hubs and routers and all sorts of stuff - way beyond even a column.

Make sure the smart phone to be used is on the same WiFi network and has the appropriate app running: WiThrottle for iPhones or Engine Driver for Androids.

In JMRI click on ACTIONS> START WITHROTTLE SERVER

The smart phone may go through a handshaking window. Then it should show the roster from JMRI and allow you to pick a loco.

Knob - off topic

You mean a knob on your phone like this:

http://www.slashgear.com/diy-physical-controls-for-iphone-touchscreen-video-1385538/

...

This all looks like a very important step in the development of DCC.  If you think LCC is unnecessary on your layotu but you have Fascia panels and such, you are already very close to having an LCC second bus on your layout; my UP-5s are all connected together to each other with RJ-25, for instance, so I already Had an LCC bus without knowing it, minus the LCC node that ties it all in to a common computer operating system without a command station in play.

I am a bit amused or depressed at the sight of the automobile connector in this setup.  Right now the computer world has been driving on Cat-5E and USB for the last twenty years and is quickly moving over to HDMI.  I would think the most logical interface here would be Ethernet Cat-5e at the minimum [and I'm very happy to hear that the last round of devices was all done with Cat-5e!] because that is what computer systems use for communications busses!

It all looks good, but here i'm hoping it's not Too Little Too Late.  And I say this as we look at where the smart phone revolution is going to take train control in the next ten years, where communication busses become meaningless, by looking at where Computer Control is taking us now [do we have an acronym for Computer Control yet???? We'll need one for both Central Computing control and one for Indpendent Computing control - I'll explain further down!].  We haven't seen the arrival of the blue tooth control from Bachmann yet, but if that allows direct interfacing control without any busses at all, I do believe that will be the nail in the coffin for dedicated DCC throttles and with it, DCC altogether.

I dare see Independent computer control displacing even Central Computer Control here in a decade.  Geoff has shown us how to make a 17 function decoder for $5, now imagine what we could do if we started with a USB thumb drive and that decoder and worked towards a decoder that is in essence a standalone computer like every other cell phone, camera and electronic device is nowadays. What independent computer control would mean is that you'd end up with a locomotive that can carry it's own copy of JMRI, take pictures, play music, and coordinate the movements in the yard or on the mainline without any user input whatsoever.

It would be in essence a standalone computer, and before anyone cracks the old Blue Screen of Death jokes about OS failure, take a real good look at the OSs that have been rolled out on cell phones.  Everywhere you look, there's an OS running the electronic device.  And it's not hard, if you consider each output as not a function but a peripheral device...

Exciting times for all of us, even as many are running for the hills and for hand thrown ground throws!!

Benny

Automobile connectors?  The spec (and this setup) is RJ12 connectors if you are running with a CAN bus.  If running Ethernet then you just use the IEEE defined RJ45 connector and isolation transformer.  LCC is just a protocol and can be run over any transport medium.  We have defined the packet structure for CAN and Ethernet/WiFi (TCP) to date.  Running it over Zigbee, BlueTooth, USB, etc is just a matter of defining the the packet structure details to carry the protocol.   There has been discussion on these various transports but no one to date has created a test-bed and proven out the packet structure.  The general rule is code and hardware wins for a spec and no one has worked those other transports yet.

Thank you for the insight

As always clearly and well written. So much new stuff (note the techie term) comes along and many of us feel lost and left out. I use NCE dcc because it was simple enough for even an oldie like me to learn and you clear up, for me anyway, the next generation of electronics and how to decide if I need it. Just reading the comments and all the terms that seem so obvious to the writers that it seems to me they are speaking in acronyms...I guess they are so familiar with the technology that the acronyms are the way to go for them. This is not meat as a derogatory to the other writers, but more as a compliment to you Bruce for the way you explain things that is clear but not condescending...so thanks for another AWESOME column... it is pretty much the first thing I  read each new issue.

Dispatching Digitrax UT4

Bruce, Thank you for your column which I always find interesting.

You also explained how to put JMRI DecoderPro to good use for realising slots with a Digitrax DCC system. I do also clear slots before an opsession, and sometimes during as I had a Zephyr with only 10 slots, Now I have upgraded to the Zephyr Xtra which has  more slots.

There is an easier way to dispatch, at least with the UT4D, and that is to just change the position of one of the rotary selctors. From the manual:

I did thought that was possible with the normal UT4 and UT4R, but the manual does not tell about it. I have to check next time if that still is possible, as I have used that procedure to get rid of an address and dialing a new one. 

PS

By reading the quote above again, I see that it seems that the turning of one ofthe rotary adress selectors just relase the adres from the certain UT4, and do not Dispatch it.

I really doubt that we will

I really doubt that we will see the end of dcc anytime soon. The capabilities of this "new" system are interesting but also involve lots of additional complexity. Complexity also involves maintenance of equipment and connections to that equipment.

There are now ground throws that can power turn out frogs. Connections can be from the track to the switch to the frog. No electronics, no software, likely very little maintenance, and likely to still be working many years after it is installed. Computers seem to go obsolete rather quickly. There are huge amounts of data that can not be found that were stored electronically. In contrast the dead sea scrolls can still be read, as can the hieroglyphics on the pyramids.

If you have a way to use this technology and make it work for you that is great, I suspect a lot of other things might happen such as the technology will find a way to use you instead. In some other threads there was a lot of discussion regarding making things simpler and easier to maintain. If you have a strong background in computer use and electronics these things maybe just what the doctor ordered for you. If on the other hand you appreciate the way a single use device like a railroad throttle fits in your hand and uses a format designed for one purpose and only one you might feel as I do about it.

To steal a line from a movie and possibly paraphrase: "My daddy always said when the object is to put a nail into a piece of wood don't try and make it complicated just get a hammer and hit it till it's in there."

I have seen lots of electronic parts fail after a short time of being used. I have also found that the spare parts are often in short supply or unavailable and that cheap piece of electronics renders a much more expensive system inoperable.

Our club has a refrigerator we use that is likely 50 or more years old. It still works very well. I have seen lots of them made since this one that have failed and were unusable. Sometimes the methods used by the Luddites were a good idea.

For me the things that this new technology offers are not things I want to add at this time. Maybe it will offer things I want later and then I will be jumping to get some. Right now I need this like I need slot cars on the layout.

Throttles

There's some very mixed messaging here.  The NMRA FAQ suggests that LCC is going to "unload" your DCC bus, but DCC is on the rails, most of the traffic on any DCC (on the rails bus) is cab -> loco traffic as speed and direction change often and those messages are repeated.  Aux decoder commands for turnouts (kind of an expensive way to go if you have more than a few) are only sent once and use an insignificant amount of bandwidth, so why bother moving them?  Now they're suggesting that you ought to put your cab bus (not the track) on LCC, so is the point to unload DCC (seems unnecessary) or to replace your cab bus?  My NCE cab bus works just fine.

LCC ad WiThrottle

why would you use LCC to implement WiThrottle when it has worked for years with just JMRI and a connection to your favorite DCC command station (with a computer interface)?

...

LCC basically gives you the backbone for a common utility Command and control input line, so you can use any manufacturer's throttle with your layout regardless of either the manufacturer of the throttle or the maker of the command station, so long as there is a module to allow the interface to occur.  If you have Digitrax, this is basically a universal Loconet.

LCC in Texas

Rob, you guys do need LCC, or at least you needed it 20 years ago.

LCC can replace all of this with a two wire buss.

Every signal, detector, switch machine, animation and so on will be connected to the buss. When you install such a thing, you will wire it via a chip similar in form to a loco decoder, but potentially much cheaper-- after all they don't need sound (some might, like crossing signals) or complex speed functions. I would expect the cost of a single aux power control to match the price of the toggle switch it replaces. Still want toggle switches? A chip for those will be even cheaper, and you will be able to have as many as you want, anywhere you can reach with the buss.

All of this stuff will require programming, so you still need someone like Bob T around, but I expect it to be a JMRI function.

Just how many man-hours did it cost to loose that piece of paper?

pqe

Seth

I'm not sure if your rhetorical questions were aimed at me or elsewhere.

I'll try to reply. I wasn't saying anything about LCC as a support for WiThrottle. I was saying that Jim Kueneman was able to get LCC to use different brands of throttles on the same DCC system a year ago. Folks have always lamented the fact that brand A throttles won't work on brand B system. Here is a solution. Like so many solutions, it comes at a time when folks are using WiThrottle and smart phones to run any system with a "free" radio throttle.

I know several places (the Scottsdale HO club, for example) where they have two busses: NCE Cab Bus to run trains and Digitrax LocoNet to handle their signaling. I was saying that, given the development of LCC hardware in the next little while, LCC could become a competitor to LocoNet or CMRI for that second bus.

@Pelsa

Let me see if I follow your logic. And I will add I am not as electronically talented as you are.

Every signal, decoder, switch machine, detector will be connected to the buss via some sort of chip. The buss will connect to a computer and all the chips will need to be programmed and connected to a computer that will need to be working to manage all of this.

Sounds like the same thing only different. When Bob T. designed and built the system he did so with what was available at the time. All the circuits were designed and built by him and were put in this dispatchers panel that was designed for our layout.  Should something like this be undertaken for the future a computer and LCC thing might be warranted for an application like the club layout. Looks like a different route to the same thing, no disagreement there.

As to hours used up 4 guys three hours tops to document wiring. I actually spent more time under the layout labeling the legs and photographing the wiring and components on each leg so we could write the manual. You see when the layout was built no one documented any of the wiring except Bob T. who had some sketches of the circuits he designed and built. I suggested that we prepare something that was the equivalent of a shop manual for our layout and Bob T. embraced the idea whole heartedly. Until the manual all the information was in Bob's head and covered things he had figured out by trouble shooting what was done by others ahead of him. The rest of us had some things committed to memory but if something needed attention we might spend hours solving it or get Bob and he would have it corrected in minutes.

Now as to my comment as to not needing LCC. I was not referring to the club layout. I was considering the needs of any layout I might build and how it would be operated. I am not a fan of lots of little chips placed all over the layout if an alternative exists.

Ground throws can power frogs and do not need a buss or taxi, a simple set of jumpers from the turnout itself to the switch and to the frog. No motor, no power supply.

Signals with out electronic detection units do not need a buss or taxi just some wire from a selector from the tower to change the aspects and a power supply. For any layout I am planning I do not anticipate more than a dozen signal heads. Detection will be handled via the mark II eyeball method employed by tower operators for decades. A compact closed circuit tv camera and monitor system will give the individual a view from all six towers likely to be employed and allow visual detection and subsequent changing of the signals. I saw an example of this in a magazine and one camera can cover multiple tracks on a layout and needs to not be connected to anything relating to track power.

The tower positions will deal with a few mainline crossovers and nothing more so I do not believe more than two dozen tortoise machines will be used and likely it would be closer to half that number. Does not seem to be a need for a lighted dispatchers panel, a train sheet should suffice.

In my case I will be building something that is as uncomplicated as possible and reflects the transition era of the late 50s.

As to your statement thinking the club layout would benefit from that system it might be a great addition if it was being built today but the club layout was originally built as a dc layout and moved up to dcc. We still do both when ever the need arises. In another couple of threads discussions took place on size and complexity of layouts. For me eliminating as much wire, circuits, computer connections etc. is a great way to simplify the layout and reduce maintenance. Sorry for the confusion my original post generated it was meant to be only dealing with my future home layout. 

One thing I have found though is when an electronic component fails it seems to take a great deal of time to correct what ever the little gadget was supposed to do as that particular part is no longer available, wire on the other hand generally is close to as reliable as gravity and when it does fail is easily fixable.

That might make me a Luddite.

re - Throttles to Benny

Benny, I think you are assuming facts not in evidence: has anyone actually made a throttle translator -> LCC module for any manufacturer's throttle with the intention of selling it?  Is there any demand for such a thing other than as a "science project" to show it can be done (as the other poster did)?  Realistically people select their DCC system for two reasons: 1) because all their neighbors use it (note how there are clusters of Digitrax, NCE and CVP) and 2) because they like the throttles.  So it seems unlikely I'd go out an establish another bus (I've already got NCE and CMRI) just to be able to run (say) Digitrax cabs.  By the time I got done with the CAN BUS interface to my PC and the translator module I could have bought a couple Cab06s.  It's more likely I'd use WiThrottle/Engine Driver (although I'm not a fan of touch screens for operating layouts) or maybe just make my own "Hard" WiThrottle with an Arduino, a pot and and a WiFi shield, this would communicate through my WiFi Router, to JMRI without ever touching my existing Layout Control Bus (CMRI) or LCC.

Bruce - Separate buses

So are you arguing that there is a benefit in having a single bus for both cab control and layout control?  If so why did the NMRA LCC FAQ in the NMRA Magazine lead with the benefit of "unloading" your DCC bus by removing layout control functions from it? 

It seems to me (I'm not a Digitrax user, so I'm weak on the details) that a single bus was the idea behind Loconet.  My impression is that it works pretty well for all but the largest and most complex layouts.  I doubt the guy with a garage layout, or even the Pebble Creek Club would ever run out of capacity with Loconet.  I personally use NCE for cabs and CMRI for signals, detection, CTC and staging control.  My system also runs well. 

I know you've been around DCC for at least as long as I have and I distinctly remember in at the Portland Convention in 1994 that the reason the cab bus was not standardized (resulting in cab incompatibility) was that the manufacturers would not invest in DCC systems if they didn't have a way to differentiate themselves.  Do you think that has changed and NCE, Digitrax etc are now going to develop to LCC as a cab bus?  Does LCC open a large enough market for new manufacturers to enter in hope displacing the incumbents? What happens to all of the investment in cabs that NMRA members have made?