Real railroads use three methods of turning engines or trains, wyes, loops and turntables. By far the most common method to turn an engine was the turntable, an example being found in almost any engine service facility. Wyes were the next most common method of turning engines or whole trains as they were more space efficient than loops. There are relatively few examples of loops in prototype practice, but they were and still are used in some instances. There is a loop in the UP (ex SP) Colton Yards in California. The Colton yard makes up only eastbound trains. The few west bounders are made up as east bounders and then run around a tight loop to go westbound. This loop can be seen from Interstate 10 in Colton. The automated Black Mesa Railroad is a 78 mile loop to loop railroad in NE Arizona.
In model practice, it works just the other way, reversing loops are most common as they allow "continuous" running, wyes come next and turntables are fairly uncommon. With two rail track powered model railroads, each of these methods requires some special power routing to prevent short circuits which will occur with any reversing scheme. I won't be discussing turntable methods here as I don't have one and have never worked the methods involved but it appears that they can be treated much the same as loops or wyes.
You can avoid reversing problems altogether by going to battery power, however, this requires that you install batteries and radio controllers in each engine and disconnect the motors from the power pickups so that you don't backfeed power back to the rails.
[ Top ]
A wye is a very convenient way to turn engines, or if the tail track is long enough, entire trains. However, without some special care, it is a hard wired short circuit as shown in the diagram. Current runs up one leg of the wye, back down the other and right to the other rail.
Fortunately this is really easy to fix if you are using LGB turnouts and motors. There is an accessory switch for the LGB 1210 motor, part number LGB 1203, which snaps right on the end of the switch motor. It provides the heavy duty DPDT (double pole, double throw) contacts necessary to accommodate automatic polarity control of the tail track. All the wiring is local to the wye, you don't have to run any wires back to your control panel, other than those required to operate the turnout motor if you use remote turnout control.
Wire the circuit as shown in the drawing. It should be set up so that the polarity of the tail track is correct for whichever way the train is coming from. If your engine stalls when it crosses the insulated joiners the first time, simply reverse the wires to the tail track. Then all you need to do is to make sure that the turnout at the tail track is aligned the correct way before the train crosses the insulated joiners. When your train is all the way on the tail track, flip the turnout and your train will back out without you even having to change the direction of your power pack.
This method of wye polarity control is sometimes called an "X-section" in model railroad wiring books. The method is as old as 2 rail electric trains and has withstood the test of time.
If you don't use LGB motors, then you can use an Atlas "Snap Relay" (about $7) to control the polarity. With two diodes, you can make this twin coil latching relay respond properly to the bipolar drive signals used by LGB and Aristo turnout motors.
[ Top ]
Loops present the same problems as wyes, there is a direct path from one rail right around the loop to the other rail.
There is a common method of wiring reversing loops that uses a bridge rectifier to set the polarity of the loop so that trains can run around the loop in one direction only. This is the method used by the LGB 1015U/1015T reversing kit and is described in detail on page 67 of the LGB Track Planning and Technical Guide. A spring switch routes the train in the correct direction each time. You are required to stop the train when its fully in the loop and reverse your power pack. The bridge rectifier will keep the train running in the same direction. When the train exits the loop, the track polarity outside the loop has changed to match the rails at the exit of the loop.
This method works OK, but it has a couple of operational drawbacks. First, you can't reverse a train once its in the loop so don't put any sidings in there that need to be serviced. Also, you have to remember to reverse your power pack EVERY TIME your train goes through the loop. If you have metal wheels, or you wire power pickup down the length of the train, the WHOLE TRAIN must fit into the loop (between the insulated rail joiners) and pass by the insulated rail gap at the entrance of the loop before you reverse the polarity.
The loop polarity must be initially wired to match the polarity at the approach to the loop. Do this by reversing the + and - leads of the bridge rectifier that feeds each loop so that the train can enter the loop without causing a short.
Also, it is important that every loco that is to run on the loops runs in the SAME direction with DC on the track. It doesn't matter if the "large scale" standard or the NMRA standard is used, but all locos must be the same. If a loco is "backwards", it will short out at the entrances to the loops.
The spring switch is a special problem as not all rolling stock will clear a stock spring switch without derailing. LGB turnouts with a manual control box act as spring switches but the spring is a little stiff. The points have to push all the way over for the turnout to work and some cars and engines just won't do it. Aristo turnouts work better as spring switches as there is a second spring on the curved point rail. The wheel flange on the curved stock rail only has to split the point. This pushes the other point rail part of the way over. Then the other wheel extends the extra spring to push the curved point rail the rest of the way. This only works when the turnout is set to a straight through path. Even this may not work on all rolling stock.
I've taken to ripping off the stock control boxes entirely and using a light coil spring instead. By pegging the other end of the spring exactly where I want it, I can set the tension on the spring so that any engine or car, even the pilot truck of a Big Hauler, will reliably clear the turnout. I use a 1/4" or so light coil spring that I found on a rack at my local hardware store. For outdoors use, you might want to find a stainless steel or brass spring to prevent it from rusting away.
[ Top ]
I found it a bother to switch the polarity every time a train went through a loop. It can be done with magnetically operated track switches alone as described on page 130 of the LGB Track Planning and Technical Guide. This method avoids the troublesome spring switch, but it is prone to getting confused. You have to install magnets under all your engines, and then you can't back trains around because the circuit tends to expect the magnet at the front of a train. Also if you stop your train in the loop and shut the loop off to work with another train on the layout, the system may not work properly when you restart the train again.
I devised an intelligent polarity controller to do it for me. My system uses an infrared beam to detect the presence of a train that is just leaving the loop. When the beam is tripped, the circuit checks the main track polarity and instantly determines if a reversal is required or not. Any kind of an occupancy detector can be used instead of the infrared beam. A reed switch, mechanical switch, commercial between-the-rails detector, or a track current detector will work as well just as long as it provides a logic high when a train is detected. It doesn't matter if the signal is noisy or has contact bounce, the circuit deals with multiple triggers automatically.
I used the IR train detectors because both of my loops exit from tunnels where its fairly dark all the time. I mounted the phototransistor inside a small 2" long cardboard tube that was pointed across the track to the LED. Since this whole arrangement is indoors, weatherproofing was not a problem.
The circuit is designed to handle two loops in a loop to loop arrangement. The two optoisolators detect track polarity and the first D flipflop latches whatever polarity was last detected. This signal is compared to the outputs of the loop occupancy detectors to determine if the polarity of the main line needs to be switched for a train exiting either loop. The other D flipflop is a latch that either holds the reversing relay open or closed. If the power is removed to the circuit, the relay relaxes so that the main line polarity achieves a known state so that manual control of the reversing process is also available. The main line is reversed by the reversing relay, the loops retain constant polarity with a bridge rectifier as shown in the basic loop reversing circuit above.
This is a medium complexity electronic project so I am assuming that those that will try it have sufficient skill to work from a schematic. I don't have records of what kind of IR LED's or photo transistors that I used (probably because I used unmarked parts anyway) so you'll have to use something appropriate. The LED should be an ultrabright type, as it's biased at only about 4 mA. The gates are TTL NAND gates (SN7400) and the flipflops are type D (SN7474) although an RS flipflop will work as well.
The circuit is somewhat sensitive to transients so that the wiring to the loop sensors should be twisted pairs and routed away from other wires that may contain large current transients, such as turnout motor circuits. This is to prevent the circuit from triggering at inappropriate times. It is preferable that the power source for the circuit be isolated from other circuits as well. I use a small plug in DC power supply or battery charger that came with some defunct toy that my kids busted.
Setting up the initial polarity is similar to the manual method, but there is an additional complexity. Since the main line polarity is switched, it must also be set to match the loop polarity at the entrance to each loop. Once the loop polarities are initially set up to allow the train to run around the loop in the "right" direction, the main line polarity must be set (by reversing the feed to the main line) to allow the train to approach the loop. This should initially be done with the loop controller turned off to allow the polarity relay to relax. Check it at both loops by using the power pack's reversing switch to allow the train to approach each loop.
[ Top ]
On the GIRR Mountain Division there is a trolley line that runs around the town and then off to a distant station. The line is really a loop to loop arrangement with a loop in the town with another loop hidden under the scenery. The trolley makes two short timed stops in the town and then make a third long stop for about 3 minutes under the scenery. At each stop the trolley decelerates and accelerates smoothly. The stops are set with reed switches set in the track. In the town, the two stop trigger switches are simply wired in parallel and use the same logic.
I custom built this controller and it is not a project for the faint hearted electronic tinkerer. Not all the part numbers are shown as you'll probably use what you can find anyway. Also, some of the support wiring to the ICs (such as power and ground connections) are not shown. Pinouts for the ICs are not shown either. You'll need to look them up for the parts that you use.
I initially had some difficulty getting the circuit to work reliably due to false triggering of the timers. Model railroads are a very electrically noisy environment so that special care must be taken in isolating grounds between the controller and the rest of the layout. Also the leads to the reed switches must be well twisted to cancel pickup of stray pulses that might upset the circuit.
This diagram is the top level interconnection diagram for the controller. It is built on two boards to keep the track power circuits away from the controller circuit. I used two surplus Bachmann power packs to power the circuit, one for track power and the other to power the controller. The circuit is set up to allow the trolley line to be controlled manually via the power pack, or automatically via the controller. The logic power pack is a set and forget arrangement, it is never touched. The track power pack must be set with a particular polarity for the automatic controls to work. The diode leading to the power control board prevents damage in case the track power pack is set to the wrong polarity.
In the manual mode, the trolly is run into each loop and then the track power pack is reversed to drive the trolly out of the loop. The trolly will not change direction in the loop, as assured by the bridge rectifier, but the polarity change will be there on the bidirectional track so that the polarity will match when the trolly exits the loops. In the automatic mode, the speed of the trolley should be set with the 5K pot on the power controller board. The track power pack will just be run to the max in the one direction that works.
The large capacitor on input to the power control board is to filter the output from the Bachmann power pack. If your power source is cleaner, the capacitor will be unnecessary. The diode between the power pack and the controller board is necessary to assure the correct polarity is applied to the regulator on the power control board.
Reversing control is provided by the controller board. The speed up and slow down ramps are provided by the power control board.
This is the power controller board. It consists of an adjustable regulator IC with the reference voltage ramped up and down with resistor capacitor time constants. The controls for the charging and discharging of the control voltage are optically isolated so that the logic circuitry can run from an isolated power source. The 5K pot set the maximum output voltage to control the maximum trolly speed. The 160 uF capacitor is the main time constant device. It is either charged or discharged through one of the optical isolators. The PNP transistor is an emitter follower to prevent the reference circuit of the regulator from upsetting the time constants.
This is the trolley controller circuit. The track reed switches trip timers that determine the delay at a station stop. The two pots set the station delays. When either timer is triggered, a stop signal is issued and the power controller board begins to ramp down the track voltage. When either timer times out, a start signal is issued and the power control board ramps up the track voltage. If more than one stop is desired with a particular delay, additional reed switches can be wired in parallel with the existing one. The station stops also reverse the polarity on the section of the line outside the loops so that the polarity will match when the trolley exits either loop.
The circuit could also be adapted to a point to point line but some modification of the logic will be required. In the configuration shown, stops are not allowed on the bidirectional section of track as the polarity is reversed as soon as a reed switch is tripped. This would cause the trolley to immediately reverse and then slow down.
[ Top ]
Digital Command Control has its own special reversing characteristics. Since DCC uses track power, short circuits will still occur on wyes, loops and turntables without some method implemented to correct the track polarity.
Some boosters have an autoreversing capability. To make the system work, you either need a regular booster AND an autoreversing booster OR a reversing module. An autoreversing booster is a regular booster switched or programmed to operate to drive a reversing section. Depending on the booster, reconfiguring it may require a wire jumper, the flip of a switch or a change to some electronically set option within the booster. An autoreversing module is a chunk of electronics that uses some power from the rails to feed the reversing section. Autoreversing modules are less expensive and easier to set up provided that the additional power provided by an autoreversing booster is not necessary.
In the case of a reversing loop, you still need to isolate both rails and the whole train (at least the part that picks up power from the rails) needs to be inside the isolated section. The main line is connected to the regular booster. The isolated section is connected to the autoreversing booster or reversing module. Several different reverse loops can be connected to the same autoreversing booster or module as long as only one train at a time is entering or leaving an isolated section.
When an engine enters the isolated section, the polarity may be matched or mismatched. If the polarity is matched, nothing special happens. If the polarity is mismatched, both boosters think that they are shorted but the regular booster keeps providing power. The autoreversing booster guesses that the problem might have been a reversing episode and instantly reverses polarity. Now the boosters match and the train proceeds without even a flicker of the headlights. If it was a real short circuit, the boosters will continue to apply power for a short time and then shut down again until the short clears. The autoreversing booster will be switching polarity every time it retries.
The same thing will happen as the train leaves the isolated section. If there are multiple reversing loops run from the same autoreversing booster the system will still work with multiple trains in the isolated sections as long as only one train at a time is crossing the boundary. The DCC decoders do not care if the track polarity is switched on the fly. However, if there is an analog loco at address zero running on the system and the polarity is reversed due to another event while that engine is inside a loop, it will reverse immediately. Analog locos will handle reverse sections properly if they go all the way through the reversing section before another loco enters or leaves another isolated section hooked to that same autoreversing booster.
This page has been accessed times since 27 Dec 1997.
© 1997-2009 George Schreyer
Created Dec 27, 1997
Last Updated September 19, 2009