Station siding electrical question

Described here. Further relays and a timer can provide soft stop and start and time delay between trains.

http://www.jcstudiosinc.com/BlogShowThread?id=856&categoryId=426

 The electronic part is easy to do. The above link is harder to explain than to actually do. Do not be intimidated it is simple switching. The more serious problem with automating a siding is the reliability of the switches. I purchased a pair of the new of Ross O tubular at York to mate with my K-Line Shadow rail. Steve assured me his switches were reliable and derailment was not an issue.. I will be doing the above circuit on the layout I am building with added soft start/stop circuits.  I am still laying track for 6 loops. Since the layout is about 1400 sq feet it will be a while.

Dale H

Hi Dale

I'd love to implement this feature into my passenger stations, but the track is already in place and not easy at all to add an insulated track section. Can the Lionel 153IR or Z1011 trigger the relays when the train enters the station? Or must it be insulated rails only?

ralph

Ralph

With 3 rail the insulated rail method is the cheapest and most reliable. I would not use IR detectors for collision avoidance due to poor reliability. However having said that if they were made reliable I think the link circuit could be made to work,even with less relays assuming the detector circuit had sufficient hysteresis and would hold in as the train left. (and not go on and off between cars as it passes the detector)

If it is just a stop circuit and not collision avoidance the consequence of a malfunction is less of a problem.

A reed switch and magnet integrated with a latching circuit either mechanical or electrical might be a better choice. I would have to think about it and sketch it out. That could be installed without tearing up track.

Blocks can be cut with a dremel maybe without picking up track.

There are a lot of ways to do things but since we put up with the 3rd rail we may as well use it as it is a natural block detection advantage over 2 rail.

The basic circuit linked before would cost about $35 to build. Maybe $70 with a time delay and crude soft starting. Cost depends on the amount of realism we want.  Real trains do not jack rabbit out in a split second of each other.  In conventional this can be overcome with additional circuitry.

Dale H

Hi Dale

Thanks for the reply. I use tubular track, so making the blocks is a bit more difficult, since the metal ties mess things up.

If you have the time and could sketch out the 'reed - latching circuit' and email me directly, maybe I can induce you to build it for me? The relay solution for lighting has you sent me in the past been an immense help.

Larry: The request you posted sounds deceivingly simple, but it's not. I've been a model railroader for over 25 years and have authored 25 published articles on advanced electronic/electrical control systems for trains. The closest to what you asked for was an article I wrote that was in Classic Toy Trains October 2007 issue. This article described running two trains simultaneously on the same loop of track. Oddly enough, I was in our local train shop about a week ago and a customer voiced the same request as yours. My first reaction was: no problem. But when I got back home to my lab and started to work out the details, I realized it wasn't all that easy!

To do what you and that other guy wanted requires several key steps:

1. You need a detection scheme to know where the trains are.

2. You need to switch the two turnouts which feed the siding at the appropriate time.

3. You need to control the power to the correct blocks at the right time.

I am convinced that this can be accomplished and am working on the approach now. When I finish, I will post the the circuit details.

Bob H. Walker 

Ralph

I would have to think a bit and draw it out. Off the top of my head it could be done with 2 reed switches for sure and IR detectors maybe. Basically we are making a circuit for 2 rail.  The problem with the IR detector is that when the train is leaving the block the gap between cars makes and breaks the detector as the train moves. There is a timing circuit in the Lionel IR detector I think but if the train stalls in that position it may not be reliable. There is a difference between a block signal not working right and trains colliding if it does not work right so we have to be careful here. If the reed switch failed we would not have a collision.

Anyway I will sketch it out and look in my parts box. Most useful would be a twin coil mechanical latching relay .  This would have memory so the trains would not have to be parked on the siding when shut down. I need to know what kind of turnouts you are using. They have to be thrown in tandem.

Even with this since 2 trains are running on a loop the center rail of the 2 siding have to be isolated if they are not now.

For parts we are looking at I think 3 relays,a latch relay and a few rectifiers and 2 reed switches and capacitors. For pause between trains a timer and another relay. Again give me a bit of time. The best laid plans of mice and men sometimes go astray.

Dale H

I have a bunch of Atlas Snap-Relays, they're a nice twin coil DPDT latching relay.  That's probably what you're looking at here.

I agree with the problem of IR detectors dropping out at the space between adjacent cars, but this is only true with reflective detectors. If you use transmissive detectors with the emitter on one side and the detector on the opposite side and place them at 30 degrees relative to the track, this problem goes away. However, using a time delay circuit with a reflective detector will also work as well.

Bob Walker

If the train stops at the gap with a reflective detector, the time delay doesn't solve the problem, right?

Hi all

The station siding solution that Dale H has IDd is attractive to me.  Let me explain.

I constructed the track layout and placed all the accessories, etc before I understood about the need/use of isolated rails - like I really know now, Ha Ha. Actually, to do what I want, most of the layout would have to be isolated rails. So I opted for no isolated rails and jumped into the IR cart. I use IRs to operate signals and some accessories.

I have quite a number of them from different manufacturers. Zstuff 1011, Lionel 153IR MTH ItADs and one called MrMax. I'm sure there are other manufacturers as well. These all tout the ability to accomplish the things I want to do. They seem to perform well, though my layout is not 100% operational in terms of running trains to all levels and track sections - more like 45%.

Zstuff includes application drawings on their site to do wiring, using their own brand of relay for a siding and collision avoidance applications. The lionel includes its IR and relay as a unit, but doesn't include instructions for applications other than operating accessories - basically you are left on your own. MrMax suggests several on their website. MTH's is similar to Lionels.

The Zstuff 1011 relay comes wired, by color. One of the big problems is this wiring, since the unit is proprietary, it isn't clear what wire does what. Eg, the yellow wire never seems to be used in any application, the blue wire is labeled COM, but half the time in the application drawings it gets connected to power, Red/Green always are joined and go to power or common depending which accessory it is trying to operate. Some of the drawings are missing components referred to in the instructions and notes. It seems like this relay has to be customized to each application, with the email advice from Dennis - and sometimes he suggests a trial and error solution.  

The lionel product doesn't have any applications similar to the Zstuff listed in its manual. Mr.Max will talk thru them on the telephone, but his products are pricey - they actually use a two component send/receive IR and even under track sensor item. 

I'll probably implement the Zstuff collision avoidance solution on some crossovers in my layout. The station siding solution that is being discussed here will need time to percolate, but it may be the way I go.

Ralph

Hi Arthur

Thanks for your reply on the 'insulated rail'. A good how to for making one.

I've tried it and after my un-handy work the revised track never gets back the way I had it before I removed it. I thought of removing a track and replacing it with a true insulated rail, but there is the issue of location. The area where I want to do this has some long stretching and placing the IR next to the tracks with relays under the table top is most convenient for me than working on the tracks proper.

Ralph

Ralph

I am drawing up a circuit for the sidings,there are several ways to do it.. I need to know the type of turnouts used and if you want a time delay between trains.

Crossover protection is very easy and only requires 2 relays if the insulated outside rail is used. 2 additional relays will eliminate roller jumping on longer trains if needed.

www.jcstudiosinc.com/BlogShowT...4&categoryId=426

Dale H

To Ralph: I understood your comments on the various IR detectors. I have used every one you mentioned in your post with different results. The Z-Stuff units seem to be the best of the refelective types. I must add that the transmissive units I have used (Pioneer Valley #021) have been outstanding performers so far.

Bob Walker

Numerous ways to circuit the passing siding. The reed switch offers an advantage over the insulated rail method in that it can select trains which will react. Only those with a magnet underneath will trigger the switch. So for a stop circuit a passenger train can stop at a station while a freight train is ignored. However we will assume that only 2 trains are on the loop in this application. The trains on the passing siding can go in the opposing or same direction.

The center rail on each siding is blocked with fiber pins. Reed switches are placed where the train is to stop allowing for stopping distance and train length. The reed switches could be made to switch power to a relay coil which could latch through one of its contacts. However I would instead trigger a timer with each reed switch,shown is an Infetec single shot,described here.

www.jcstudiosinc.com/BlogShowT...8&categoryId=426

Triggering the timer would require much less current and time than a relay coil and increase reliability IMO of the reed switch operation.. A homemade 555 circuit could also be used instead.

The power supply for the circuit is a 10 VAC tap off a transformer. The bridge rectifier and capacitor boost voltage to about 14 VDC for the 12 volt timer and relays.

The timers in turn will power 2, DPDT relay coils (coils 1 and 2 left and center). If reed switch 1 or 2 is made by the train magnet,the timer is triggered, coil 1 or 2 will energize for the time set by its timing resistor ( a 1 meg pot from a few to 180 seconds),then de-energize.

Power from the transformer is routed through the left sets of relays 1 and 2 so that if either relay is on power is cut to the track center rail. This allows for individual set timed periods between train arrival-departure.

Relay 3 is 4PDT (use 2 , DPDT relays in parallel).  When relay 1 is energized, coil 3 is energized by relay 1 (right contact set) and electrically latched through its second set in series with the right set (common- Normally closed contacts) of relay 2. When relay 2 is energized the latch is released and coil 3 is de-energized.

The above can also be accomplished with a twin coiled latch relay which would have memory at shutdown. The above would have to be reset manually (Latch restored on relay 3) if shut down while block 2 is vacant. Some twin coiled latch relays do not take constant voltage so the coils would be thrown by capacitor discharge by the right set of contacts of relays 1 and 2.

If relay 3 coil is de-energized track power is on block 1. If relay coil 3 is energized track power is on to block 2. This is accomplished by contact set 1 of relay 3. Both blocks then can not be energized at once.

The third set of contacts on relay 3 would throw the turnouts ,wired in tandem in the proper direction. Atlas switches could be thrown with capacitor discharge.

The optional 4th set of contacts on relay 3 could be used for block signals,wired opposite of each other.

In operation, train 1 would pull into the siding tripping the reed switch,after the set time the opposite train would pull outand run the loop,when it returns  train 1 would run the loop. It would not matter if the train stopped dead on the reed switch.

Further relays and voltage dropping diodes could add a soft start circuit. The parts for the above circuit would cost about $55 to make.

Dale H

Dale,

You have mail

Ralph

Hi Larry,
 
    What you want is in chapter 22 of "Model Railroading, Prepared by the Staff of the Lionel Corporation" from the 1958 Bantam book, chapter titled "Wring for Effect" pages 353-355, from the 5th edition.  No relays or special circuits, just insulated rail sections and switches (turnouts) wired together.  Clever and simple!
 
    Take care, Joe.

Joe - I'm interested in knowing this circuitry, but I lack access to the book. Would it be possible for you to post (or e-mail me) a schematic or a description of how it's accomplished?

Pete

This worked for me.  MTH RealTrak 42 switches.  Lock engines in direction and run slow, it is possible to coast through the Stop Blocks  

I have ideas on how to run two trains in oposite directions and three trains in the same direction on this track plan, but these are untested

Hi Pete,
    Here are the 2 diagrams from the Lionel book.  If it violates the copyright, I guess the moderator will remove it.  
        
    Take care, Joe.

Joe - Thanks a million!  Was already aware of the joy of O-22 non-derailing switch wiring tricks, just couldn't figure out how to activate the stopped train(s) without a relay.

Pete

Hi Pete,
 
    It was my pleasure!  No need to complicate a simple thing.  These techniques were done by boys in their attics and basements 60 years ago.  You don't have to be an electrical guru to wire trains!
 
    Take care, Joe.

Joe

The diagram is not a single siding which is what I think the original poster wanted,it is rather of 2 sidings on a loop,not one and simply runs 2 trains point to point using 3 points and relying on the anti derail feature of the turnouts.. The problem with this arrangement,although it works is the lack of realism,as it will perform very mechanical. As soon as one train pulls in,the other one will jackrabbit out at full speed. Also the wheels carry current for the other train and operation can be erratic. The advantage of relays is that it can be made to perform smoothly with a time delay on one single passing siding,the trains need not run point to point, rather when 2 trains share a single siding,the one that was there first will leave.. A soft start circuit can be made with additional relays where after one train stops the other,after a time delay pulls out slowly and accelerates gradually..

Not knocking the diagrams they work as drawn but this method will not work for 2 points on only one siding. It would need 3 points and be tedious in operation.

Dale H

All that may be true, but it's fascinating to me that such a feat can be accomplished by using only the "technology" of the Lionel system.

Pete

I agree with the Dale H. analysis of the suggested postings from the Lionel book. The original request from Larry S. is not an easy thing to do when you really look at what is required. I have worked out what I hope is a simple solution and when I get it checked out, I will make the circuit available.

Inspired by MP147's single non-latching relay method, here's a potential starting point that does not require insulated rails. Using ideas from rrman's thread on grade crossing detection, I cobbled together a circuit from spare parts and tried it on a simple Realtrax loop.  Reed switches detect a magnet placed under the forward truck in each engine.  Using a strong Nd magnet only 1/4" in diameter, I was able to place the reed switch under the track.  Rrman describes how the heavy O-gauge steel wheels from a derailed truck slice through a reed switch casing placed top-side.  The reed switch is off-centered on the left side of the approaching engine; likewise, the magnet is placed on the left side of the truck.  This allows upgrading to bi-directional operation though that's a separate discusssion.

This requires electronic component assembly so I realize this does not meet the OP's criteria but the parts cost is under $5.  To be clear, this does not have all the neat features Dale's method provides and I offer what I discovered in the spirit of a forum discussion. 

A 555 chip is configured as a bi-stable latch so a cheaper non-latching relay can be used. The latch is set and reset by reed switches (one on siding, one on main). The set and reset switch signals are edge-triggered which means the stopping engine can sit over, or coast past, the switch once it triggers the changeover. I believe this edge-triggering feature allows easy adaption to ITADs. On the DPDT relay, one pole applies Track AC to the center-rail of the siding or main. The other pole configures both switches to straight or curve.

I only had a 5V DPDT (on the right) in the spare parts bin so the circuit runs on 5V DC provided by a 7805 regulator chip on the left of the board. To keep things simple everything including the switches is powered by track voltage.

The short video below shows two MTH PS2 engines locked in Forward. PS2 engines wait about 1-2 sec after power is applied to start moving. Considering this pause is “free”, it might relax the need for an additional timers. Also, PS2 engines have some built-in acceleration ramping in conventional so they do not jack-rabbit on start. Again, considering this ramping is “free”, it might relax the need for diode blocks (or whatever) to ramp up the speed.  It also seems for engines with sound, it might be nice to lower (rather than remove) the voltage to keep sounds running while stopping the engine but that's yet another discussion.

Pete 

If you look at old Lionel books they actually use relay block circuits for "advanced" layouts,though some of the circuits are drawn wrong. My first circuit could have been made in 1955, although selenium rectifiers, wirewound resistors and mechanical timers would be used instead of diodes and solid state timers. Relays,capacitors certainly existed then and even binary switching circuits using vacuum tubes. A lot of electronics technology was learned from WW2 development and engineers were eager to try it out on consumer products. Lionel had a remote control set way back and very bright engineers. The reason for not implementing more advanced circuits was cost. That little whistle relay in a tender is a fine piece of engineering and there are many other examples. A little over a decade later men were put on the moon with similar technology (slide rules and vacuum tubes) not that much advanced from 1955. The limitations were back then the time,money and effort put into a system rather than lack of technology,though it is certainly easier to do today..  Having said that I do not think even the most optimistic person in 1955 would conceive of the technology we have today. Really a great era to live in.

Dale H

My hat goes off to Stan2004 for his clever solution to the automated siding request. The 555 IC timer is a great device and as a EE, I am blown away by how many applications that chip can handle. Again, my congrats to Stan. Now, for a 2nd offering.

I devised a circuit (attached) which accomplishes essentially the same task. It uses eight components (Stan used 15). The track setup is the same, an isolated center rail siding and an isolated center rail section between two switches. Two magnetic sensors are alongside each track section and are triggered by a magnet mounted on the train loco or one of the cars. The isolated track sections can be any length depending on available layout size. The benefit to my approach is that it is simpler and perhaps easier to implement for a model railroader. I calculated the total cost of the parts to be around $42.00 (the latching relay cost $5.20). Also, the DPDT relay provides two sets of double throw contacts which might be necessary with some switches and can be useful for other signalling capabilities.

Bob

What happens if the train stops dead on the reed switch and it remains closed? Some latch relays can not handle continuous current and are only designed to be pulsed. Also the e unit would not release when the other train arrives. A solution could be capacitor discharge. Also the reed switch would have to be closed long enough as the train passes over it to make the E unit coil. This may not happen if the train is moving fast. That is why I opted in my design for the timers, all a reed switches needs to do is pulse the timer trigger for a few milliseconds and carry a few milliamps. This could also be done with a 555 circuit rather than my stock timers. Twin coil latch relays multiple with 10 amp contacts get a bit expensive,the ones I have seen with these capabilities cost about $25. Please link a part # for your latch relay.

Dale H

Stan2004:

Thank you, thank you, thank you!

Your circuit is exactly what I have been trying to gome up with.

Fast and easy to build.

I will be using it on a display layout that belongs to the Edgerton Model Railroad Club, which used to be known as PAL Model Railroad Heaven. 60 years old and still going...

The circuit I posted functions properly with any reasonable train speed past the magnetic sensors, including stopping dead center on the sensor. (The sensors are Cherry Electric MP201801 which are almost perfect for layout use). The latching relay is a Panasonic DS2E-SL2-DC5V. This is a PCB mount DPDT with two 5V 139ohm coils. Putting a 120ohm resistor in series with each coil keeps the current well within a tolerable range for continous duty. I have used this relay on other projects and it holds up well. 

Stan2004:

I have some questions about your station sideing circuit. Could you contact me off line? My email is in my profile. Thanks.

Jay Edgerton

Jay, I suggest just asking your questions here. I think the topic is of general enough interest that others might find it interesting or better yet have additional insight or experience to contribute.  I'm sure if things get too technical someone will pop in and tell us to get a room take it off-line.

Stan;

I built up your circuit to be used on a display layout that a club I belong to takes to train shows to attract new members and promote the hobby. I have some problems:

When the circuit powers up the relay is energised without contact on the pin 2 circuit.

Contact on the pin 4 side won't release the relay. I have a led and 1k resistor between pin 3 and ground. It shows output on pin 3. When the off contact is closed the led brightness drops to about 1/3 to 1/2. But not enough to turn the relay off. I have switched the 555 chip with the same results. Any ideas?

Thanks.

Jay Edgerton

Do you have a meter?

As you know, the pin 3 output is digital so it should be either high or low.  The LED should be either on or off.  My guess is the power supply is sagging for whatever reason.  What is your source of DC power to the circuit?  Are you sure it's stable?

What I'd do next is leave the LED and 1K resistor connected to pin 3, but remove the 10K connection between pin 3 and the transistor-relay.  The circuit will still toggle upon application of alternating switch closures except you'll only see the LED go on and off.  You'll also be able to confirm the edge-triggering behavior of the the switches.  That is, it's the initial switch closure that toggles the 555 output.  Either or both switches can remain closed, or can release from closed-to-open without incident.  Once you get the circuit working without the relay, it should be easy sailing...

As for the initial condition, the 0.01uF caps are initially discharged when power is first applied meaning the voltages at pins 2 and 4 are both at 0V simulating a momentary switch closure.  In your case pin 2 is "winning" the race.  If you absolutely must have a guaranteed known starting condition, there are simple workarounds but this is a separate issue.

The power supply is a 12 volt, 1 amp supply with a 7812c regulator. It has a 4700 uf filter bypassed with .2 uf. I've used it for years with no problems. I'll let you know what happens without the relay circuit as son as I can check it out.

Thanks.

I removed the 10k resistor to the relay circuit.

The same problem is there. Pin 4 circuit doesn't turn the output off.

I'm using the same type of breadboard that you show in the picture of your circuit. I got to thinking about the contacts in the board. Sure enough, when I moved the circuit over on the board it now works. Must be a bad contact in the board somewhere. Time to replace it.

Anyway, your circuit now works as intended. Now to make a pc board for it and put it in a case.

Thanks for your help.

JAy Edgerton

Since we discussed IR detectors extensively in this forum, I thought I'd mention another available product. Bouldercreekengineering.com offers a pair of packaged detectors called nightscope at $19.95 each. One has a built in delay and one does not. These are designed to be mounted in the roadbed between the rails looking up at the train. They can drive most low drain accessories directly, but require relays for heavy power like a loco. The website has a lot of info including a 12 page applications manual. I plan to check this product out and report on it.

Dale if you are still watching this thread, I would like to try your solution, but need some assistance. I have all the pieces needed except the timers. Could you please identify exactly what they are. Radio Shack?? Some other online distributer?? Catalog # esp if radio shack would be very helpful. Also I have atlas #200 snap relays. Are those ok to use with your setup

Thanks

Larry S