i'm very familiar with C/MRI and the benefits using I/O nodes and rs-485 communication to control a layout and minimize wiring.
but how many dispatcher panels, requiring even partial block detection and partial turnout control throughout the layout are directly wired?
by large I'm suggesting 40x25
i assume 4 pair (cat something) cables saves ones sanity
greg - Philadelphia & Reading / Reading
For layouts that are of more recent construction, since all that sort of technology became readily available and inexpsenvie? I'd say not many would have hundreds of wires all going back to a central point. For a layout that's been around 50 years - unless all rewired at some point, there's a good chance that it is all centralized.There are probably still clubs that run most of their trains from fixed engineer cab stations overlooking the whole layout. But I would also think such things are becoming more and more rare.
--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'm helping with a layout and the owner doesn't want computer control (and i'm not sure it's needed)
i'd really like to here from modelers who have a dispatcher panel and don't use computer control.
Does that include a self contained microcontroller like an Arduino? Because it's not terribly difficult to put an Arduino in the panel and connect to others in the field with RS485 or similar. No Windows or Linux PC of any sort in sight. I have to find the link again, I posted it in one of my other threads, but the details are there for making a full CTC panel and controlling remote signal and detection units and it is entirely based on microcontrollers, no PC.
i think the majority of modelers are not tech savy (probably don't need it) and I don't blame them for not using tech that they don't understand and can't maintain themselves.
i'll also guess that larger layouts without such tech, don't have dispatcher's with CTC-ish panels, but i'm hoping someone will prove me wrong
The Pacific Southern Railway is very sophisticated with over 20 nodes having as many as 32 I/O bits. But of the 100+ members, I believe there are at most 5 that can understand the hardware and/or software and I think this is representative of the typical modeler population.
gregc i'm helping with a layout and the owner doesn't want computer control (and i'm not sure it's needed) i'd really like to here from modelers who have a dispatcher panel and don't use computer control.
OK, well I don't have one built right now, and I use DC, but I'm going to build another hard wired dispatchers panel for my new layout.
It depends a lot on the desired goals, and there is plenty of info in older issues of MR.
CAT5 is my control cable of choice, 4 twisted pairs.
It only takes two wires per "route" to control and indicate turnout positions.
It only takes two wires per signal block to clear signals for trains to proceed with DCC control.
It only takes five wires per every other signal block for me to assign and clear routes with four DC throttles.
It only takes one wire (and a common) per block to indicate occupancy.
All of this assumes the presence of the control power feed and common.
The relay control logic boards for turnouts, cab selection and signaling are decentralized, being located near the signal block or interlocking in question.
Only the necessary dispatcher panel control wires go to the dispatcher panel.
So count the number of signal blocks, count the number of turnouts, that will give you a fair idea of the number of wires.
So, my new layout:
24 primary blocks
24 X section blocks (interlockings)
60 turnout routes (about 100 turnouts)
So that means about 300 wires comming into the dispatchers panel, or about 40 CAT5 cables. The panel will be about 10 feet long and 18" tall, so the entry of cable into the panel area can be spread out based on location on the "map".
The layout will fill most of an L shaped room 38' x 43' at its longest dimensions, about 1400 sq ft total.
But here are the questions, what are goals for this panel?
Is it just turnout control and/or detection?
Signaling? What kind? How detailed? What kind of compromises are acceptable?
Does it need to provide actual CTC protocal functions or can they be simplified?
So a little more about my system.
Every element of the dispatchers panel is duplicated on local tower panels at each interlocking on the visable parts of the layout. Staging is controlled on the dispatchers panel and in some cases on local panels.
Signaling is simplified, using only interlocking signals and approach signals.
And again, all this hard wired logic is decentralized, reducing the length and conjestion of the wiring.
It is complex, fully intergrating signaling, detection, DC cab selection, and turnout control into one system.
But its not as much wire as you think, because most of the wiring is on relay panels. So even with computer control digital communication bus, you still need the hard inputs and outputs at some point, unless you use a computer screen input, but then you still need the hard outputs.
In my case a relay panel is right below a local tower panel. Four of five CAT5 cables connect the them. Wires leave the relay panel to operate the switch machines of that interlocoking, direct track power, and relay info and duplicate inputs back at the dispatchers panel.
Here is the thing, any system like this is designed to fit. You can use a lot of stock circuits like those I have posted before, but the actual implimentation takes planning.
My reasons for doing this old tech include:
Not wanting computer screens in the train room.
I prefer the tactile feel of traditional buttons and lights located on a panel, in the track diagram.
I want to minimize dependancy on any proprietary products that would be hard to replace down the road, requiring some major rework.
I am not interested in the learning curve as I am only partly up to speed on much of this new stuff.
My life is pretty busy right now, but I would be happy to help if I can.
Sheldon
About a decade ago, I designed and installed a version of my radio throttle advanced cab control system on the layout of a friend.
Here are some photos of the control panels and the relay panels:
This layout uses ground thrown turnouts, and does not have signals, but uses Aristo wireless radio throttles for walk around advanced cab control.
Cabs are assigned to blocks with the pushbuttons shown on the panels. Other blocks are powered or killed, automaticly based on turnout postion.
Operators do not have to go back and "turn off" anything, there are redundant controls as you progress around the layout.
The KRISTENVILLE AND MICHELLEVILLE Railroad.
When I see things like that, it makes me so happy I created the STRATTON AND GILLETTE long before I met my wife.
It could have become the KEVINHUBBY AND CUTEYWIFE.
She still expects me to number all the freight cars with numbers that are significant to our family/relationship history.
If I number something like "6379" she will ask me why that number was chosen. The answer always is "because the number jumble on the decal sheet let me do it with only two decal pieces", but I need to come up with something like "The first used car I bought after I met you was a 63 Mercury Breezeway that we bought on the seventh and kissed in it nine times the first week we owned it.
Then she is happy.
-Kevin
Living the dream.
rrinkerFor a layout that's been around 50 years - unless all rewired at some point, there's a good chance that it is all centralized.There are probably still clubs that run most of their trains from fixed engineer cab stations overlooking the whole layout. But I would also think such things are becoming more and more rare.
I have not seen a lyout like this since I went to college in Nashville in 1985.
Every layout I have visited since then larger than 50 square feet has had hand held throttles.
ATLANTIC CENTRALOK, well I don't have one built right now, and I use DC, but I'm going to build another hard wired dispatchers panel for my new layout.
any chance you have a photo or can provide a description of you dispatcher panel?
SeeYou190 The KRISTENVILLE AND MICHELLEVILLE Railroad. When I see things like that, it makes me so happy I created the STRATTON AND GILLETTE long before I met my wife. It could have become the KEVINHUBBY AND CUTEYWIFE. She still expects me to number all the freight cars with numbers that are significant to our family/relationship history. If I number something like "6379" she will ask me why that number was chosen. The answer always is "because the number jumble on the decal sheet let me do it with only two decal pieces", but I need to come up with something like "The first used car I bought after I met you was a 63 Mercury Breezeway that we bought on the seventh and kissed in it nine times the first week we owned it. Then she is happy. -Kevin
The railroad is named for his two daughters. Unfortunately he passed away a few years ago, but the family has preserved the layout. His name was Alan Anderson, he was a craftsman of the highest order. The structures and scenery are of the highest order, equal to Howard Zane, better than George Sellios in my view.
The layout was featured on the layout tours during the 2006 NMRA convention in Phily. I was part of the layout hosting team.
More later, off to work.
gregc ATLANTIC CENTRAL OK, well I don't have one built right now, and I use DC, but I'm going to build another hard wired dispatchers panel for my new layout. any chance you have a photo or can provide a description of you dispatcher panel?
ATLANTIC CENTRAL OK, well I don't have one built right now, and I use DC, but I'm going to build another hard wired dispatchers panel for my new layout.
Unfortunately no pictures, I was never good at remembering to do that years ago.
But description, sure.
Much like a CTC machine, linear track diagram of the layout, actually two linear track diagrams. Upper one for visable mainline, lower one for hidden staging.
Located in the diagram, red LED lighted pushbuttons control turnouts. Example, a simple crossover on double track would have three buttons. One on each straight thru route, one for the crossover. The LED's indicate the selected route.
Side note about route control, with two crossovers, left and right, in the same interlocking, only one would be allowed to "crossover" at a time, the other would automatically return to straight thru. No illogical routes are allowed by the controls.
White LED's, in the track diagram indicate detection.
The next part relates to my cab control system. The blocks between interlockings are called primary blocks. Also located in the diagram, for each primary block, are four more lighted pushbuttons, and a reset button.
These buttons assign one of four radio throttles to that primary block. Pushing one button, releases any previously selected throttle, pushing reset disconnects the block from all throttles.
The interlocking territory is called an X section block. It gets its power from the adjacent primary blocks based on the route selected, so there are no cab section buttons for this trackage, it is automated by the combination of the route selection and primary block cab selection.
Only after the route is selected, and the correct primary blocks are assigned to the same throttle on opposite sides of the interlocking, is there track power thru the interlocking route and signals go clear/proceed (also subject to occupancy).
So, in short, red lighted pushbuttons on the diagram show the selected and assigned route, white lights show where the trains are.
More later,
Like mentionned by some, before, I'm not sure about the full necessity to have a computer control.
Second, as a remark, lot of us didn't know a lot about the use and the making of of Arduino, CMRI and other soft stuff able to control and run a layout.
Anyway I'm sure these soft are great and help a lot to control a big layout, unfortunately for me, I don't know a lot about them, like other soft, they need a quiet long learning curve, time I prefer to use on other aspect of my layout construction.
But even in 2020 we don't need a computirized layout to run it.
Because of the extension of my layout and my poor ( to not say very poor) knowledge of these softs, I go to the old way.
But some modern electronic features will be used to control the layout but in a way I'm able to control them, maintain them and understand them.
My basic design is a walk around line with local switching because of industry or small towns.
All the mainline turnouts are controled with stall motors by Hankscraft display motors, local are manual but probably will be motorized in the future.
I will use main line divided in block occupancy and a signal control;meaning the signal , like the real thing permit to run your train on the next block or not.
The conductor of a train need to check the aspect of the signals along the track he is using.
This mean a central control panel which has action on all the mainline turnouts and signal and this way all local are locked, block occupancy detector will repeat on this panel if the track is free or not and also the position of signals.
All the local industry and local switching will have their own panel and control, but these panel will need to be unlocked if they need to run on the main; this is done following the signal aspect by the operators.
Like a friend layout any conductor need to control the signals on the track they use, I model in N scale these are tiny, so I will repeat the signals aspect on the fascia with a bigger logo and bigger leds which show the position of the signals.
The signals aspect also open the next block electricaly but also unlock a local switch area, this mean, the block were a train need to make local switching using the main need to have his main line block free from any coming train.
Relays can easily do the job; some switch control are easy to double from a main panel and block detector affect the aspect of the signals which are repeated on the main panel and on the fascia for more precise view.
The main panel act like a CTC panel but not as complicated, all the trains are running on a schedule; each local switching using the main have a dedicated time to make the switching, after, the main block is open to main train again.
This general view need to be precised and I work on it with a small booklet full of remarks and new thinking about it.
But it will no include a computirized control for sure
Marc_MagnusI'm not sure about the full necessity to have a computer control.
it's not a question about computer control, but communication and wiring. i'm wondering about logistics and practicality.
the layout is triple deck in two joined basements.
the dispatch panel will have ~250 connections between toggles and LEDs on the panels, and block detectors, Tortoise machine motors and contacts and signals.
C/MRI nodes could reduce the connections between the dispatcher panels and layout to a single pair of wires. Each node on the layout could provide 16, 32, ... I/O points. But one or more nodes supporting the dispatcher panels would need to provide the ~250 connections to the panels which still seems daunting and node software is still required
replacing analog dispatch panels with computer graphics displays and mouse clicks would eliminate dispatcher panels and wiring but requires software. I guess this is computer control, but it's simply replacing toggle switches and LEDs
JMRI is an option. not sure if JMRI supports interlocks and signaling.
yes. I'm wondering how practical a non-software approach is on such a layout
placing analog dispatch panels with computer graphics displays and mouse clicks would eliminate dispatcher panels and wiring but requires software. I guess this is computer control, but it's simply replacing toggle switches and LEDs
I can only agree with you Greg, if i had the knowledge to do it sure I will use it
The miles of wire and the number of them can give you gray hairs quickly and ask a lot of time to be put together.
I'm a ferm believer of the KIS method and for sure without any comment possible, wiring is an absolute KIS necessity, and course, JMRI or alike are for sure a enormous step foward.
But like I mentionned, I'm open to the computer approach, but I can use it and maintain it no matter about the reason.
Like I have done on the existing part of my layout I have simplified as far as possible most of the circuit, many works directly from switch motors contacts and more recently I have used some DCC decoder or DCC modules to handle and help to control things with more simplicity than a analog approach.
This is a learning curves and probably I will go deeper in this way including some approach with Arduino; you have published some good approach about it on other forum and they are very entertaining.
So I think my approach will be an hybrid electronic-analog lightly computirized control
In Belgium where I lived before we say " Something affordable by the youngers between 7 and 77 years old".....
Greg,
16-17 years ago, when planning my previous layout, I considered DCC and I considered solid state/software solutions to all these issues. And I considered DC computerized block control.
If we assume basic knowledge but not "experienced" with either approach, my feeling is software solutions have a higher learning curve.
17 years ago we did not have some of the software related products of today. We did have Bruce Chubb's work.
But we also had the work he did with relays in the 60's and 70's. My system is partly based on that, and on work done by Ed Ravenscroft and Paul Mallory. Search them up in the archive.
For me, total time, effort and cost did not seem like it would be much different, just a different set of tasks.
Also, once I rejected DCC, software based solutions for signaling lost a lot of their advantage.
Signaling requires blocks, DC requires blocks, my walk around radio cab control requires power level switching. Software or solid state signal solutions had no real advantage at that point.
I actually have features that would be complex and expensive to implement with DCC, but they are largely a free benefit of what I have done.
I rejected DCC for the following reasons: no interest in onboard sound, poor ergonomics of most throttles, limited selection of wireless systems at that time, existing fleet of 100 locos without decoders, and lastly cost/decoder install time vs added added features.
Control panels...... I dislike computer screens for that sort of thing. That too is a big factor in which way is better. If the user is ok with computer screens, software does have at least some sort of savings/edge.
Even with computer screen interfaces, you still have the same output connections, turnouts, signals etc.
This is why I said before, it depends a lot on the goals of the operator. It sounds like Marc is doing something very similar to me signal wise......
my neighbor growing up was a mechanical engineer from England. On a visit from England his mom, knowing I was interested in engineering, said she told her sons to be mechanical engineers because you can see the workings of the machine, unlike electrical and software.
compared to software, electrical circuitry so much more visible with lamps, meters and scopes than software.
but with software, you can completely change your approach and simply reprogram the device. After getting basic control and display in place, you can add functionality (e.g. interlocks, signaling, fleeting) without changing any hardware.
while I think software keeps it simple, it's relative. I'm trying to work within the knowledge of the owner of the layout. So i'm interested in understanding all the options
Greg, I will agree about the versatility of software. However, I personally am not comfortable with the "design while you build" approach.
Building a large model railroad is a project that requires focus and dedication, and figuring it out as you go may work for some people, but I don't see it as the best approach any more than I would restore a house for one of my clients that way.
Sure, not every question needs to be answered in advance, but the goals and basic platform of methods should be.
If you know what you want, it seems more straight forward to just plan it and build it, no matter the exact tech chosen.
And while software lets you change stuff, to add real world stuff you still often have to add output hardware.
The system I developed is based primarily on MLZ control by Ed Ravenscroft, with the addition of wireless radio throttles and using a theory of pushbutton cab selection proposed by Paul Mallery.
Paul, I hope your watching from above, I demonstrated that your idea of a pushbutton walk around cab selection system works, and works well...... Even the DCC guys in our local round Robin understood it and liked it.
There are any number of ways to get the job done, the first thing is to define the goals.
What does the owner want the panel to do? What info and functions are important?
How close to actual CTC does he want, or can it be highly simplified?
How does he feel about a big CTC like panel vs computer screens?
Does he want or need local controls as well?
I assume we are talking DCC.
All of these things will influence the best choice.
With some goals defined, I could and would make some more concrete suggestions.
ATLANTIC CENTRALI personally am not comfortable with the "design while you build" approach.
your talking about the layout and trackwork, not the electrical stuff, right?
ATLANTIC CENTRALAnd while software lets you change stuff, to add real world stuff you still often have to add output hardware.
yes, but it's just another one of what you've already used.
ATLANTIC CENTRALWhat does the owner want the panel to do? What info and functions are important?
see where trains are, see turnout positions and control turnouts. The dispatcher is the interlock.
None of which requires a computer to control, but a graphic computer screen and mouse would save a lot of work wiring and allow easy corrections, exceptions and enhancements.
ATLANTIC CENTRALDoes he want or need local controls as well?
plans on using latching relays and momentary toggles on both the dispatch and local panel
and just to be clear, i'm not against hardware. i'm in favor of using hardware for what it does best and software for what it does best to minimize the amount of both.
No, I'm not comfortable with designing the control system after the fact, or while building either. Sure, there may be minor adjustments as you go along, that is different from not having a plan in the first place. Or even going back and filling in missing features? With signaling or DC, track needs gaps, I would rather install them while laying track, not later.
Agreed, it is more of the same, just like all my relay circuits are adaptations of the same basic circuits.
Well that is really simple, with no actual prototype CTC functions, l can't see any reason to get too high tech, unless he likes the computer screen thing. What about signals? They need not be shown on the board, they where not shown on prototype CTC machines back in the day. But will there be signals?
Rather than latching relays, why not push buttons and lights. It is a much better approach for controlling the same turnout from multiple locations. If you use monetary toggles, you still need lights. Lights can be in the push buttons, that's what I do. It simplifies the hardware and the panel design. Push buttons don't require latching relays, holding circuit works just as well.
I agree, as I explained, I considered full blown computerized block control, and various other software based and/or more solid state based methods. Because I rejected the computer screen approach, I saw no reason to convert inputs into solid state or digital logic, just to convert them back at the other end.
The thing is, with your control method, the control and signaling are all part of the same system. When you move to other methods of train control, the signal and detection system becomes completely independent of anythign done to direct power.
WHen it's all one system - it's fairly critical to have it all work together from the start, designed that way, and implemented as you go. When the track power is completely independent from the control system (control here meaning dispatching, interlocking, etc - rules of the road for the train, not how the loco is physically controlled), then it is a diffrent story. The rulebook control is layered on top of some independent physical control system.
I have no desire to have the computer be my panel. It's way out of character for my era. If I modeled modern era - I wouldn;t mind so much, and I'd probbaly set up something like the club has, which mimics what some of the memebers dea with in the DS office of the real railroad many of them work for. Yes, any logic you cna do in code, you can do with relays. But even knowing just what relay terminals I need to connect together, I can STILL type up a conditional statement in several different programming languages than I could ever possibly wire the relays. And I still think no matter how much planning you do, some things are subject to change. You may think you have everything worked out for good operation, but after a few sessions, it is suggested that a track may be added to solve a serious bottleneck. With relay logic, now anything affected by that new track has to be rewired to adjust that logic. If the logic is computer driven, cutting in the new track physically is the hardest part of the job. Maybe this has never ever happened - that's either some amazing layout design skills or else everyone is putting up with a bottleneck that on the prototype would have been resolved.
It is possible to do CMRI panel to field with no computer, it just means putting the logic in the mico used for the panel. This is the one I was talking about:
https://forum.aorailroad.com/t/ctc-machine-build/169
JMRI can do full logic, and even mix and max multiple systems. I also just found out that JMRI is flexible enough that my idea of making custom CMRI nodes and making them all transmit and receive the same number of bits regardless of how many are actually connected to anything is completely doable. JMRI does not need the node to support the (I)nitialize operation, but the code running on the node has to know which bytes are inputs and which are outputs (which - why wouldn't it, if you write the code?). Getting 16 bits back in response to a poll when only 8 of them are actually used on that particular node means nothing to JMRI.
When using a virtual panel, you can even have virtual elements intereact witht he logic of physical elements - long ago I did some testing where I made a local power lever on a panel, the state of which was used in logic to accept input from a trackside pushbutton to operate a turnout. This was with Loconet, I was using a LocoIO board, but the same can apply to any supported system. If the trackside Reverse button was pressed, the logic in the JMRI panel combined that Loconet input with the status of the virtual local power level to send a command to the turnout address.
I've incorporated the same thing in the logic of the code on my turnout controller. There is an input bit that if pulled low, locks out the pushbuttons. Allowed to go high, the buttons set the position of the turnout. I made it this way so the layout can be fully operated with the computer off. I could have put the buttons as inputs on a CMRI node, and not as part of the servo driver board, but then the computer would have to be running to see the local button presses and activate the associated turnout - and there would need to be an option to set all mainline turnouts to local control for running without a disptacher sitting at the console. My way, you simply do not power up the computer and CTC panel, the pullups on the servo boards keep the inputs high, and the buttons work.
Randy, completely agreed.
And, my signal system and turnout control can be one of those stand alone systems that are layered over DCC.
I made sure of that just in case I ever changed my mind about DCC.
Because the relay logic is decentralized, changes are not really that hard. Things are well lableled, and there is room for extra relays.
I use solid state detectors...... that come with a relay right on the board....... the detectors and the interlocking contacts are the whole signal system, and I can mix incandesent and LED signals.
There are no permissive block signals, only interlocking and approach signals. The longest distance between two interlockings is 75 feet, only three times the average train, we surely don't need trains that close together. There is only one control block between interlockings in most cases.
Big panels are are best built in sections, which allow changes by just modifing or replacing that section, just like the prototype did.
And yes, one of the key strengths of my system is that the same two relays that threw the crossover turnouts, also powered the frogs, provided the signal interlocking logic, and directed track power to the rails in the X section. And the holding circuit that keeps them on powers the indicator lights.
And the X section power scheme provides free automatic train control at every interlocking - route not aligned? blocks not correctly assigned? no power thru the interlocking - train stops.
The other thing is this, I have not written any kind of code in decades, my work took me away from that stuff a while back. I have not kept up, I had no reason to, and no real interest honestly.
I looked into it 17 years ago as mentioned above. I understood what people were doing, but I was just not modivated to learn the details or get back into that stuff.
The benefits, if there are any over what I am doing, did not justify the learning curve. And it may be different now, but 17 years ago the costs of the various software and solid state solutions I looked at were much higher than what my system costs.
It is the same with DCC. I don't like onboard sound, the other benefits are so small they do not justify the costs or the work of installing 140 decoders, or the infrastructure of bus wiring which I don't need with my system. I have a throttle bus, 10 of them, #12 wire, each connected to the 5 amp power supply feeding each Aristo throttle base. Which are then connected via 10 amp relay contacts to each 50 to 75 feet of track......
ATLANTIC CENTRALRather than latching relays, why not push buttons and lights.
can you explain how to wire this?
with a 2 coil latching relays, a momentary switch powers a coil to switch the relay. the relay provide + or -12V to the switch machine. Two momentary switches are needed on each panel, one for each relay position.
Tortoise switch contacts are used to indicate turnout position on the distance panel.
Like this:
You can use the tortoise machine contacts for the lights, or you can put them in the circuit as shown. Or you can wire them as shown on this diagram and eliminate a lot of wire.
The same wire that powers the coil stays hot, so it can power the LED.
You can stack as many pushbuttons as you want for as many locations as you need.
Yes, this does require two relays for a single turnout, but it also only requires two relays for a crossover, or three relays for a wye. So on most layouts you will average just a little over one relay per turnout in CTC territory.
It also provides lots of spare contacts for powering frogs, interlocking signal logic, etc.
The relays I use have four sets of form C contacts, 4P2T
Looking at the top diagram, when the system powers up, relay #1 (R1) is energized and stays energized. If the button LED is wired directly it stays on with the relay or is powered thru the Tortoise.
Relay #2 is locked out, the Tortoise machine receives 12 volt positive polarity from the common neutral power supply.
When the button for the diverging route is pushed, powering relay #2, the seal in contact set flips to relay #2, it seals in and relay #1 drops out.
The feed to the Tortoise switches from positive to negative, the Tortoise changes.
And so on.......
The relays should be located near the turnout, then it only takes two wires per turnout, plus the control power feed, to go to the remote panel, and a parallel set of two wires to the local panel.
Greg, no offense intended, but I get the feeling you are having a hard time visualizing the physical layout of this wiring, as the ladder logic diagram only reflects the electrical function, not the physical layout.
I don't have any drawings already made showing the physical layout.
But as an example, if you had an interlocking that required control of two crossovers and a diverging turnout on a double track mainline, you would need the following:
5 relays
7 lighted pushbuttons
Only one eight conductor CAT5 cable would be needed between the remote panel and the relay board, and another between the local panel and the relay board.
This would provide one button control of the crossover routes, control of the diverging route, light the LED's in the buttons, and control the Tortoise machines.
If a dual voltage power supply like mine is used, the Tortoise machines can connect directly to terminal strips and relay contacts on the relay board, as can frog power.
The control diagram for the wye, also works as the control diagram for a pair of crossovers, or a double crossover, as well as other track arrangements with three possible routes.
i believe i understand the (top) diagram -- 2 relays. I believe a single relay could be used if pushbutton switches are wired in series with the latching contact and coil.
here's what we're planning on doing: remote/local and local-only cases. the remote case requires 1 relay. the contacts only control the Tortoise voltage. (yes, we can add bicolor LEDs in series with the motor).
gregc i believe i understand the (top) diagram -- 2 relays. I believe a single relay could be used if pushbutton switches are wired in series with the latching contact and coil. here's what we're planning on doing: remote/local and local-only cases. the remote case requires 1 relay. the contacts only control the Tortoise voltage. (yes, we can add bicolor LEDs in series with the motor).
Your way requires twice as many wires between the relay and the control panels. You have four wires between the relay/switch machine and the panels, I only have two? You seemed concerned about wire couts earlier?????????
Your way no makes no provisions to simplify route control of complex routes by automaticly returning other turnouts in the group to their default position.
Your way uses a more complex, and possibly more expensive relay.
Why are you so into these latching relays? When I designed relay circuits we were taught to avoid them if possible.
Think of two crossovers, one left and one right, on a double track mainline. Only one should be set to crossover at a time. My three relay circuit only allows one crossover to be set to the diverging route at a time, automaticly returning one or both to the main with the touch of one button, or in other words, selecting one and returning the other to the main automaticly.
The latching relay provides no advantage, and on most layouts, will not noticeably reduce the relay count, as I have demonstrated with my circuits shown above.
Not sure why you don't see the advantages here?
Ok, I have a question, are you somehow bothered by the relay staying energized? Relay circuits have been working that way for a 100 years, including the ones that ran the signals on the railroads,
It is that fact that saves those extra wires.
I'm sorry, I don't have an easy or fast way to draw additional drawings to help explain.
you asked "why not push buttons and lights" and I asked for an explaination to understand. I'm interested in understanding different approaches
but there is no one best solution
ATLANTIC CENTRALI don't have an easy or fast way to draw additional drawings to help explain.
no diagrams are needed to explain latching relays to less tech-savy modelers
ATLANTIC CENTRALYour way requires twice as many wires between the relay and the control panels. You have four wires between the relay/switch machine and the panels, I only have two?
2 are for switches and 2 are for LEDs (1 if we return +/- 12V). Don't your LEDs in series with the motor require wires?
ATLANTIC CENTRALYour way no makes no provisions to simplify route control of complex routes by automaticly returning other turnouts in the group to their default position.
multiple tortoise machines can be wired in parallel to control a cross-over. Diodes can be used in more complicated situations to do single button route selection.
ATLANTIC CENTRALYour way uses a more complex, and possibly more expensive relay.
there's just one relay and the EC2-12TNU is < $2. Cost vs wiring effort should be considered. The owner is not that concerned about cost.
ATLANTIC CENTRALWhy are you so into these latching relays? When I designed relay circuits we were taught to avoid them if possible.
because they're economical and easier to wire and understand
i believe part of the complication of a relay design is an asymmetry that forces a particular state when powered up (R2's NC contact energizes R1).
ATLANTIC CENTRALThink of two crossovers, one left and one right, on a double track mainline. Only one should be set to crossover at a time. My three relay circuit only allows one crossover to be set to the diverging route at a time, automaticly returning one or both to the main with the touch of one button
i hadn't thought about this case. Now I have and that's why these discussions are useful.
I think using 2 latching relays and 3 momentary buttons can be used to select either cross-over route or switch all turnouts to their normal mainline routes. The mom-switch to select a diverging route on one cross-over also selects the normal route on the other cross-over
on this layout, there is a double mainline with a center passing track with 3 turnouts. The owner is not concerned about an interlock. But I think a similar approach using 2 latching relays and 3 mom-switches can implement an interlock where 1 relay controls one mainline and the passing track turnout.
ATLANTIC CENTRALRelay circuits have been working that way for a 100 years, including the ones that ran the signals on the railroads, It is that fact that saves those extra wires.
your diagram using 2 relays uses 7 relay contacts compared to just 2 contacts in my diagram. Don't understand how that saves extra wires
ATLANTIC CENTRALNot sure why you don't see the advantages here?
i think i'm dealing with less complicated problem for a less-tech savy modeler.
While software and RS-485 communication would be less complicated for me, I'm interested in understanding and using "lower-tech" to satisfy this owner's needs on his layout.
I appreciate your explanations and thoughts.