Bidirectional crossing signal actuation?

1. Automotive relays are designed for DC. Does this mean I can't use insulated rail to actuate it, without dumping DC into my track circuit?

No.All you need is a bridge rectifier. Hookup is shown here

www.jcstudiosinc.com/BlogShowT...=410&categoryId=

For multiple relays you can also use a separate transformer that does not share a track common and 1 bridge rectifier driving multiple relays. The relay circuit shares a common with the track outer rail (the minus of the bridge rectifier for example). There is no DC leakage.

The relay described is 12VDC usually with an 88 ohm coil and would work with a 10VAC source for the coil,a bridge and capacitor. However it is SPDT.

2. Assuming the relay works, will the circuit work as I describe?

Your approach I believe is wrong. No need for a timer as gate activation is a function of distance,not time,at least in model railroading.

3. Is there a simpler way to do this without spending more than $20?

I dont have a circuit off the top of my head but I think it could be done with 2 or 3  multi contact relays, I would have to sketch it out to be sure. You would need 3 outside insulated rails adjacent to each other on the crossing approach. Such relays are $8 each

No need for IR detectors in 3 rail. Expensive and not that reliable.

MTH ITADs are junk. If you go with ITADs use the Lionel ones.

Dale H

A

Above is a hand drawn diagram of a crossing approach,the crossing being placed by IR2 (outside insulated rail 2) using 3 relays and 5 insulated outside rails. You can click on it to enlarge.  Relay A is DPDT,the other 2 are SPDT. A separate transformer is used to  power the relay coils which does not share a common with the track. Instead the minus of the bridge rectifier does. If the relay power  transformer shares a common you could use a small bridge on each relay,but this way only 1 is needed. The 470 uf capacitor is connected across the + and - of the bridge in proper polarity. Diodes across the relay coils remove voltage spikes. 

Outside insulated rails IR A and IR B are at least 1 train length from insulated rails 1 and 3 respectively in the loop on approach to the crossing.   Relay A is self latching in series with its own NO (normally open) contacts and through Relay B NC (normally closed) contacts.

A train moving clockwise crosses IR A,the circuit is completed by the train wheels and Relay A latches energized. The second set of contacts on relay A connects IR1 and IR2 together and disconnects IR2 from IR3. The crossing gate relay connected to IR 2 is then energized by  occupancy of IR1 or IR2 by the train wheels. The contacts of relay 2 switches the crossing gate (not drawn here for clarity but shown on my previous link).  When the train clears block 2 the gates go up as relay 2 is de energized.

A train moving counter clockwise crosses IR B, de energizing relay A if it is electrically latched. IR2 and IR3 are connected and the gates are down when IR2 or IR3 are occupied,the circuit again being completed by the train wheels.

You are looking at about $25 in parts, plus shipping to do it.

For 12 VDC coils 10VAC input will work.  For 24VDC coils 18VAC input will work.

IR A and IR B need only be 1 track section long each.  IR1,2,and 3 can be made to suit the operating characteristics desired.

Dale H

You the man, Dale. I need to sit down with a pencil and retrace your steps.

It's funny, after posting my think-out-loud idea, I figured that good ol' Peter Riddle must have already addressed this. But no, his Greenberg wiring book proposes the use of a manual toggle switch to make a crossing-signal circuit bidirectional.

Hi Alan

All a relay is, is a remote electromechanical toggle switch. So if you see a toggle switch in a schematic it can be done with a relay if it needs automation or there is some other advantage. In addition, extra contacts of multi contact relays can be used for switching logic. Of course complicated logic is now done with PC chips but it used to be accomplished with relays,the contacts used as switches. 

I used to service pinball machines and the old ones were electromechanical marvels with mechanical latching relay banks,step switches,motorized switches, etc. No longer done that way but for simple stuff relays are still a viable solution.

With another relay and timer module, the crossing bell module can be shut off after a set time. I do that with mine because the bell gets annoying after a while.

See my blog for timing circuits using modules if interested

www.jcstudiosinc.com/BlogCategoryMain?catId=426

Dale H

Don Grabski   Daylight Don used to sell a small board that was a real easy connect for dual track crossings   Haven't seen him on the forum for awhile but I think he still has them  www.daylightdon.com

Thanks for these detailed replies. I should point out that I'm approaching this with a scrounger's mentality, and pretty limited electrical skills. That said, I believe I've come up with a way to do this with an Atlas Snap Relay, two short trigger rails, and two longer control rails, on opposite sides of the crossing.

Here is my thinking. It's easy to make an insulated rail actuate a signal for a train that only approaches from one side of a crossing. It's also easy to put an insulated rail on the other side and use a center-off toggle to disable one or the other control rails. This prevents the right-side control rail from keeping the signal lit long after a "right-bound" train has left the crossing, and vice versa.

So at the most basic level, the problem I'm trying to solve is the automation of that toggling effect. I want direction of travel to dictate which of the two opposing control rails actually provides a ground to activate the signal.

Here goes:

1. A train approaches from the right. It contacts a short trigger rail - one that is a train-length away from the crossing - that completes a circuit energizing the coil of the Snap Relay.

2. This connects the power source of the signal to the right-side insulated control rail AND cuts off the path from the power source to the left-side control rail.

3. When the train reaches the right-side control rail, it powers up the signal and continues to do so until it leaves the control rail on the other side of crossing.

4. The left-side control rail remains unable to complete the signal circuit, because of the latching action of the Snap Relay, UNTIL the train reaches the left-side trigger rail. Now you see why the trigger rails need to be a train-length away from the crossing, to ensure that the train doesn't electrically bridge the control rails and defeat the toggling effect.

5. Whether the layout is a loop or a line with reverse loops on the ends, the train will always encounter the correctly enabled control rail for its direction of travel. 

Now it seems to me like the only drawback to this is that the solenoid of the Snap Relay is going to be energized whenever two wheels are on either trigger rail. On a modest-sized layout, that could be most of the time that a train is moving, which will cook the relay in a hurry. I've addressed that, hopefully, by purchasing two of these gizmos:

http://www.frys.com/product/6232010?source=googleps&gclid=CP-M4f_IzLMCFYuZ4AodIEUAaQ

It's an interval timer. It responds to the energizing of a circuit by permitting current to flow for a pre-set length of time, and then cutting off power for a pre-set interval. I intend to put one of these between each trigger rail and the Snap Relay, set to permit about a half-second's worth of juice before shutting off.

Since I already have a Snap Relay, my total investment in the project is $12: the price of the interval timers, inlcuding shipping. Pretty sure I have a little regulated DC power source to power the timers.

If you've made it this far, thanks for bearing with me! When my timers arrive I'll see if this works and report back.

    About 10 years ago I built bi-directional detector for my MTH motorized gates. I have a road crossing next to the staion. I wanted the gates to rise as soon as the last car cleared the road, no matter what the speed of the train was. Picture the last car stopping just past roadway and then the gates go up.

    I used a 10" insulated section across the roadway and 40" insulated sections on each side of the center section. The logic circuitry is 7400 chips. For reliable operation to detect the wheels, I used 40VDC feeding opto-isolaters to the logic chips. Think 40 volts at extremly low current. Wheels don't conduct as good as you would think at low current. We are not talking gold wheels on gold track here.

     The system has worked flawlessly for the last 10 years including the gates.

          No,  I  don't  have any  cats.