Signaling Crossing Tracks and Double Track Junctions

Transcription

1 Signaling Crossing Tracks and Double Track Junctions Welcome. In this tutorial, we ll discuss tracks that cross each other and how to keep trains from colliding when they reach the crossing at the same time. The tutorial is divided into four sections: Crossing tracks Double crossovers Integrating the Invisible Speed Signal with crossovers Junctions for double track main lines Dealing with tracks that cross each other A crossing is different situation from the usual switch from a mainline to two diverging branches. Have a look at the drawing below: The upper part of the drawing is the switch-to-branches arrangement, while in the lower part of the drawing, the tracks cross. The Trainz AI understands how to deal with branches, and if you place signals in the usual configuration, at A, B and C, then Trainz generally works fine. The crossing situation is different, because if you place a signal at D, it does not know about the other track, and will not signal an approaching train to stop if the crossing track is occupied. Fortunately, we have some third-party rules created by Trainz users that can help solve the problem. My favorite is the Trigger Multiple Signals rule, <kuid2:116387:26:1> by Maggs. Since this is a rule created by a Trainz user, you ll have to get it from the Download Station; it is not part of the standard installation. The rule allows you to change one or more signals to stop if a track (or tracks) is occupied. Consider the following drawing:

2 Above we have four signals, labeled N, S, E and W. We want to change all of these signals to stop if any train hits the crossing. We do that by placing triggers at the crossing point. If any train hits either of the triggers, all four signals go to stop. (And since we absolutely, positively, want all trains to stop, we use an 04 signal, which does not allow a user-controlled train to pass, even after a full stop. In the drawing, the signals are Signal USA 04. ) Why do we need two triggers? Because we have two tracks, and neither of them is aware of the other. If we were to place a single trigger right where the tracks cross, that wouldn t do the job, because the trigger only works for the track it s attached to. If the trigger is on the east-west track, it will not detect a train on the north-south track, even though the train might appear to drive right through it. In the drawing, I ve shown the triggers separated a bit for clarity, but in practice you can place them near each other, just make certain they stay on the correct track. Now let s set up the rule. Assuming you ve downloaded and installed it on your system, click on the Edit Session button on the toolbar at the top of the Surveyor screen. Then click Add and then select Trigger Multiple Signals from the pull down menu. You should end up with something like this:

3 Click on Trigger Multiple Signals, then click Edit to bring up the rule edit screen: The dialog box allows you to specify which signals will be controlled and which triggers will control them. When you click on either Add Signal or Add Trigger it brings up a pull down menu where you can make your selections. Above we ve added the four signals and the two triggers for our junction. In effect, we ve said that if a train passes either NS or EW, then all four signals should go red.

4 Notice one other thing. I named the rule Junction 1. If you then return to the Edit Session screen, you ll Junction 1 listed there, instead of Trigger Multiple Signals. This is particularly handy when you have more than one instance of the rule, so you ll know which rule does what without have to open it. If you want to protect more than one crossing or junction, having multiple copies of the rule seems to work just fine. Later in this tutorial, I ll show you where I m actually using six copies of the rule at the same time. Now before we get to the actual operation, there s one more concept we need to discuss: Detection radius. For the purposes of this rule, detection radius is how close to the trigger the train has to be before it s detected. That distance appears to be about 100 feet, or about 31 meters for those of you who speak metric. What we re saying is that if any train comes within the detection radius (100 /31m) of either of the triggers, all four signals will go red. And that raises an important restriction on this rule. If we are going to change all the signals when a train gets to within 100 feet of the trigger, then the signal must be placed so the train passes the signal before the trigger detects it otherwise the train will trigger it s own signal to go red! Did you get that? What we re saying is if the signal is not placed correctly, the train will cause itself to stop. Keeping that principle in mind, let s return to our track drawing. Notice there are a few differences from what we discussed earlier: Each of the four signals is now placed 100 feet from crossing. What this does is allow a locomotive to pass the signal on a green indication, and by the time the trigger sees it and reacts, the train will not be stopped by the signal when it turns red. This technique works very nicely for slow speed traffic and when trains do not reach the signal at the same time. Places such as infrequently used, low-speed crossings would be fine. This method does not work very well for high-speed traffic, where the stopping distance between the signal and the crossover would be too short. It also has a problem if two trains reach the crossing at the same time. In that case, if both pass the

5 signals at the same time, neither would get a stop indication, treating you to the sight of the trains passing through each other. So, it would seem we need to make the stopping distances longer, and we need to find a way to deal with trains that reach the crossing at about the same time. In preparing this tutorial, I ve spend a lot of time examining these issues and have come up with what I think is a possible solution. As the old saying goes, situations alter cases, so once you ve seen the method I m about to demonstrate, you ll have to experiment on your own route to find a solution that works for you. For example, if you typically run trains of 70 cars loaded with iron ore, your stopping distances will be much longer than on my Midwest Central, where the average train is about 20 cars. Now let s look at my proposed solution: Principle 1: Give one track a higher priority As you ll see in minute, we want to give one of the crossing tracks a higher priority. We do that by putting its trigger further from the crossing, so an on-coming train will trigger a stop on the second track well in advance. It doesn t matter which track you choose, but it might be your main line track as opposed to one that crosses it. Let s look at the track layout drawing below: Wow, big changes, right? There are now four triggers, NS, NS2, EW, and EW2. The signals are the same, N, S, E, W, but they are in different positions. In this arrangement, notice the position of the triggers and signals are reversed between the two tracks. The triggers on the east-west track are farther from the crossing than the signals. The opposite is true for the north-south track. In this case, I ve made the east-west track the priority, meaning that a train reaching EW or EW2 will trigger a stop indication for N and S. Conversely, the north-south train has to get almost to the crossing before it will trigger a stop for the east-west trains.

6 Principle 2: Use a separate rule to control each track As we said earlier, when the triggers controlled all four signals, it was possible that neither train stopped, or that higher speed trains couldn t stop in time. I suggest using a separate rule for each track, so we can more closely control how we want the crossing handled, and we are not limited to putting triggers directly on the crossing. First, we need to add another copy of the rule. We then edit each rule to control it s own track, as follows: Compare the drawing below with the two rules above. The two EW triggers control the north-south track signals, and the NS triggers control the east-west track signals.

7 Here s an actual example based on the drawing above. Let s say we have a train approaching from the west. Once it reaches the detection radius of EW, signal S (and N, of course) goes red. S is far enough from the crossing that even if a train is right at the signal when it changes, it will still have time to stop. Now check the drawing below: In the drawing above, the red trucks located at C and D are adjacent to where triggers are placed on the track. (We re in Driver mode right now, so you can t see the actual triggers.) In this case, train A has just reached detection radius of the EW trigger at point C. That changes signal F to danger, forcing train B to stop. Now let s reverse the situation and have the first train approach from the south. In the drawing below, train B has passed signal S and has entered the detection radius of trigger NS2 at D. That changes signal E to danger, forcing train A to stop. Notice how far A is from the crossing. So even if A is a high-speed train, it will have a much better chance of stopping.

8 But what about when two trains arrive at the same time? I tested that specific scenario, and what happens is that both trains stop. Then after a short interval, the north-south train will proceed. Why north-south? Because train B has stopped within the detection radius of the NS2 trigger at D, keeping the east-west signals red until B clears the crossing. Cool. In the drawing above, this is how it looks when both have to stop. Both trains came to a complete stop. After about a ten second pause, train B proceeds, followed by train A once B has cleared the crossing. (Interestingly, when I tested the scenario with just the two locomotives, one always triggered the other. It was only when I added the passenger cars that I saw the both stop situation. Remember I said you ll have to experiment a bit to determine the best placement for the triggers on your particular route? Although the both stop situation is rare, it can occur and you might want to move train B s signal further from the crossing, to give it a bit more stopping room, especially if you run longer and/or higher speed trains. The principles remain the same, though. Keep one trigger further from the crossing than the other and use separate triggers for each track. You can also consider a temporary speed restriction on the approach to the crossing. If your mainline speed is, say, 70, then you might want 45 a quarter mile or.5km before the crossing, then restore the 70 once the train has cleared the crossing. Principle 3: Be careful in the placement of signals and triggers When a train encounters a red signal, it will stop one to two car lengths from the signal. Where it stops is important because if it is in the detection zone of another rule s trigger, you may have an impasse, where two trains force each other to stop and neither can move. Consider the following:

9 In the drawing above, both trains approach the crossing at the same time. Let s say the F40PH on the left hits the detection zone first. It triggers a stop for the P42 on the right. But as the P42 is coasting to a stop, it enters the detection zone of right trigger, causing the F40 to stop as well. You end up with a standoff and neither train can move. The solution is to make sure your trains don t stop within the detection zone of another rule s trigger. Like the picture below: Now we ve moved the signals and triggers further apart. If a train passes T1, it will stop a second train at signal B. Of if a train passes signal B, it will send a stop to the signal at A. Nothing new about that, but here s what I want you to see. Even though signal A is red, a train approaching it from T1 can proceed through the detection zone all the way to signal. When it get there, it stops, but it is now outside the detection zone of T1, so it does not trigger a stop on the track by signal B. So, no standoff. Get the idea? Signaling a double crossover A double crossover is where trains from one parallel track joins the other track in both directions. Like this:

10 We normally signal a crossover with 02 signals, since we have diverging tracks, as in the drawing below: For those of you new to Surveyor, we have two similar, but functionally different signals for a divergence where the tracks separate. If track branches to the right, as at SW2, we use an 02 signal. If the branch goes to the left, as at SW1, we use a L02 signal. Notice at the other end of the crossover the signals are reversed, but even so they are placed correctly, depending on which way the track branches. For the most part, double crossovers work just fine just as you see them above. In order for an oncoming train to proceed, the switches at both ends of the crossover must be aligned correctly or the signal will stay red. In the picture above, a through train on the right track must have both SW1 and SW4 set correctly or the signal at SW1 will remain red. It s also true for the left track: SW2 and SW3 must be correct or a though train will get a red light at SW2. Moreover, a train crossing from one track to the other must also have both switches aligned or the signal stays red. For example, a train on the right track requires both SW1 and SW3 be set for a train to proceed through the crossover. So far, so good, but there is a potential for a collision under the right circumstances. We said earlier when two tracks cross, neither knows about the other. We have the same situation here, and if two trains try to use the crossing at the same time, they will collide. Check out the drawing below:

11 In the drawing, both trains are about to take the crossing. But the AI doesn t know how to handle crossings, so both trains get a clear to proceed. The results are spectacular: This is a great opportunity to use our Trigger Multiple Signals rule. By doing so, we can lock the signals if a train is anywhere in the crossing. Check out the drawing below: This looks pretty complex, but let s take it one step at a time, and you ll see it s really not all that tough. Principle 1 says to give one track a higher priority, so let s establish that first. Since I m in the USA, my trains run on the right side from bottom-to-top on the right track and top-to-bottom on the left track. With that in mind, trains that stay straight have priority over those taking the crossover, and trains that cross right-to-left have higher priority than trains that cross left-to-right. In the picture, you see three triggers. T1 and T3 are for trains taking the crossover from right-to-left on either mainline. They have priority, and so their triggers are placed further away. T2 is for those trains crossing left-to-right and the train has be actually in the crossover before T2 will stop a train crossing right-to-left.

12 In the drawing there are four dwarf signals, Sig1-4. Their placement is important. Since through trains have priority over crossovers, our triggers should only affect trains taking the crossover; through trains should be allowed to proceed uninhibited. What we re saying is that just because a train hits T1, it should not force a through train on the other track to stop. T1 should only stop a train crossing from Sig2 to Sig 4. In the same way, a top-to-bottom train on the left track hitting T3 should not stop a through train on the right track. The triggers should only affect trains actually taking the crossover. Got it? But how can we stop a train taking the crossover without affecting the through route? Good question. We do it by placing the dwarf signal on the crossover track, not the mainline. Let s take a closer look at the signal placement. In the drawing, Sig1 and Sig2 appear to be on the straight track, but they are not; they re on the crossover track. And since they re on the curved track, they only affect trains taking the crossover. Through trains on the straight track are unaffected by Sig1 and Sig2. In the same way, Sig3 and Sig4 at the other end affect only crossover trains. In practice, I found it difficult to place the signals exactly where I wanted them. In playing with it, I discovered an easy way to place the signals close to the switch points. All you do is temporarily delete the straight track, place your signal, then add the straight track back in. In the picture to the left, I deleted the short piece of straight track above the switch. That made it easy to place the signal on the curved section. Then all I did was drag the signal down to a point almost to the spline point. Once that s done, I just added the straight track back in, being careful to keep the track direction the same as it was before.

13 Notice also the trigger right at the crossing tracks. That was also difficult to place exactly, so I used the same technique: Delete one track, add the trigger, put the deleted track back in. What we end up with it the picture to the left. Sig1 and Sig2 look like they re on the through track, but they aren t, and so they affect only trains taking the crossover. That s how we allow through trains to pass, because the dwarf signal doesn t affect them. About now some of you may be saying, Hold up a minute. Are you saying that through trains can pass Sig1 even if it s red? Yep. That s exactly what I m saying. Why? Because the signal has to go on the curved track or it will affect both through trains and crossovers. And because it s on the curved track, it will be totally ignored by trains on the straight track, as if it s not even there. All this, of course, is highly unprototypical, and some of you are saying, No way, man. I m not doing that. In that case, you have three choices. (1) You can just abandon the whole trigger concept and leave the crossover just as it was created. That s fine, and if your traffic at the crossover location is light, you ll probably get away it with 95% of the time. Maybe that s good enough. (2) Your second option is place the signal before the spline point instead of after it. In the drawing, that would mean moving Sig1 from above the switch to below it. That will work, too, but the signal will stop both through trains and those taking the crossover. (3) There is a third option. How about an invisible signal? An invisible signal is one that is placed and acts exactly like a regular signal, but cannot be seen in Driver mode. We could use such signals to replace Sig1-4 in the drawing. That would allow you to use the triggers to stop the crossover trains, and also do away with the through trains passing the red light issue. There are several invisible crossing signals on the download station. My favorite is the Invisible Signal <kuid2:45324:24010:1> by NorfolkSouthern37. (Again, since this is a user created asset, you ll have to get it from the Download Station, since it s not part of the standard installation.)

14 Let s go ahead and the place four of these signals on the route. We end up with this: As we mentioned before, these signals look like they are on the straight section, but they aren t. Instead I followed the placement procedure I used before: Delete the short straight section; place the signal on the curved track section; drag it down near the spline point; replace the straight track. Functionally, these signals work just the dwarf units we used before, but cannot be seen in driver. One final thought before we leave this subject. Principle 3 says not to place a signal so that when it stops a train, the train is in the detection zone of a trigger. Let s check our setup: The trigger in the middle of the crossover has a detection zone of about 100 /31m, so when one of our invisible signals stops a train, it has to stop outside the zone. And since the train should stop 1-2 car lengths from the signal, we re safe. I ntegrating the Invisible Speed Signal with crossovers Those of you familiar with this tutorial series will remember the Invisible Speed Signal (V2) by BPanther. <kuid2:137715:23002:2>. It s an invisible signal that allows you to set the train speed through either (or both) branches of a switch. Check the drawing below:

15 In the left half of the drawing above, the ISS has been set up so when switch SW1 points into the crossover, the train speed is reduced to 20, but when the switch is set straight ahead, train speed is unrestricted very prototypical operation. I won t get into detail on the ISS operation here; you can check out that tutorial for details, but here s a quick look at the set up dialog: The top line allows you to name the control and I use a name similar to the name of the switch, so I can keep track of which ISS affects which switch. Notice that reference junction is set to SW1, left. That tells the control to monitor SW1, and if it faces left, set the speed limit to 20. You can optionally add a second rule to control speed for the other switch direction as well. The ISS is great tool to use, even when there are no triggers involved. In fact, on Midwest Central I have one at the entrance to every sidetrack and crossover more than 90 in all. The rule even has an option to set speed based on train priority, a definite option for those of you who like to run your freight and passenger traffic at different speeds. If you use an ISS, be sure to restore train speed on the other side of the crossover! Otherwise your trains will proceed to their destination at 20mph! Junctions with double track mainlines In this part of the tutorial, we ll discuss how to handle a situation where you have a double track mainline that is joined by a double track branch line. I developed the technique I m about to demonstrate here on my own Midwest Central route.

16 The junction basically consists of a double track Union Pacific north-south mainline, which is joined by the east-west MidCen mainline. The MidCen mainline joins the UP in both directions, so MidCen trains can go either north or south and UP trains from either direction can head east on Midwest Central. The actual junction is too large to show in one piece, but here s a drawing illustrating Westfield Junction. The UP mainline connects to portals located about a mile north and south of the junction. The Midwest Central mainline extends 50 miles to the east, where there is a similar junction with BNSF at East Shelton. Since MidCen is located in the USA, trains run on the right side. A train entering the junction from the east would be on the upper MidCen track. From there, it can head either north or south by taking the appropriate direction. In a similar fashion, UP trains not only stay on their own mainline, but can head east on MidCen trackage as well. All the tracks are single direction only. Now as you can imagine, signaling this junction was a challenge, but, in fact, it was easier than I expected. Let s take it one step at a time, and you ll see how I did it. What I d like you to do is spend a few minutes looking at the drawing above. There are six possible routes through the junction. See if you can figure out what they are. Once you ve done that, see if you can determine where the potential collision points are and where we can let the Trainz AI do the work for us. Stop now and think about it, then continue below when you re ready. *****

17 Did you get them all? Check out the drawing. As noted earlier, MidCen trackage is single direction, so we don t have to worry two-way traffic on the same track. That simplifies matters greatly. Now that we ve got the routes established, let s think about which of those routes would work okay using the Trainz AI, and which require special signaling to deal with crossing tracks. At first glance, routes 1, 4 and 5 could be left to the Trainz AI, because appropriately placed signals could handle traffic flow with no further assistance. Let s take a closer look. In the drawing below, I ve placed 04 signals at A, B, and C. In each case, the switch and the track conditions govern the signal. For example, at signal A, if switch 2 is aligned toward A and the track ahead is clear, the signal is green. If either the track ahead is occupied or the switch is in the other direction, signal A goes red. This works fine and no other action on our part is required. The same principle applies with signals B and C. We do have collisions possible at several places. Check the drawing above and see if you can spot them. Then scroll down to continue.

18 C1, C2 and C3 point to places where we have crossing tracks that might lead to a collision because the AI does not know about the crossing. Let s look at C1. Traffic running from switch 1 to switch 5 has a collision possible with traffic running from switch 6 to switch 4. At C2 the collision paths are 1-to-5 and 3-2. At C3 we have a potential collision from routes 6-4 and 3-2. In those situations, the Trainz AI can still help. Let s take a closer look: I ve added two more signals, at D and E. At each location the signal is controlled primarily by the switch (5 for D and 2 for E) and track conditions ahead. For example, if switch 2 is facing toward switch 6, signal E gets an automatic red. It s only if switch 2 is aligned toward signal E that we have a potential collision. In the same way, if switch 5 is not facing toward signal D, then D gets an automatic red. But if switch 5 is facing toward D, then we have a potential collision with traffic coming from switch 6. Get the idea? It s now time to lay out our crossover rules using Trigger Multiple Signals. The first principle in that process is to decide which tracks have priority at each collision point. Let s talk about the area around switches 5 and 6 first, and let s decide that UP traffic on it s mainline has priority over traffic entering or leaving Midwest Central. We ve said that we place the trigger further from the crossing on the priority track. Check out the drawing on the next page:

19 Things have gotten a bit more complicated, haven t they? But really, it s no more so than what we discussed earlier. Let s break it down. We ve added two triggers, S1 and S2, and one new signal, SigS1 (for south1). We ve also given the old signal D a formal name, SigS2, since we ll need a name for the rule. Trigger S1 gives SigS2 a stop when a train enters S1 s detection zone. In the same way, trigger S2 gives SigS1 a stop when a train passes through the S2 s detection zone in the middle of the crossing. Let s see how we d set up the rule for this junction: In the drawing above, we have two rules in operation, called Trigger S1 and Trigger S2. I picked those names because they relate to the trigger they work with, but you could just as easily call it something else, like UPJuncS1 or something similar. Just make sure you pick a name that makes sense. On the left side of the drawing, you can see how the names of the rules show up in the session edit list. Look back at the picture again and note the location of SigS1. It is on the mainline track past the branch line turnoff. With the signal placed there, it only affects traffic taking the straight path. Trains taking the turnoff into the MidCen route are unaffected by the trigger. That means its quite possible to have a train exiting MidCen toward the bottom of the picture, while at the same time, another train is making the turn into Midwest Central territory. Check the pic on the next page:

20 Our goal in designing the signaling for this junction is not only to prevent collisions at the crossovers, but also to avoid stopping a train unnecessarily. The above picture is a step in the right direction. By using our triggers, we have prevented a collision at the crossover, but we have also allowed the train turning into MidCen to continue. Unfortunately, we can t cover all the potential problems. Study the drawing below. There is a certain alignment of the switches, signals and triggers where we will stop a train when we don t have to. See if you can figure out what the problem is. Once you ve found the answer, then scroll down.

21 The problem arises when a train passes trigger S1 and turns into the MidCen branch. According to our rule setup, any time a train enters the detection zone of S1, it sets S2 to stop. That s what we want. Or is it? We do want through trains from switch 6 to switch 4 to trigger SigS2 to red. But what about trains which turn into the MidCen branch? If a train passes S1 and turns into the branch, it makes SigS2 red--preventing a second train coming from MidCen to enter the southbound UP main. In truth, there is a relatively small chance of this happening only if the MidCen train happens to reach SigS2 just after the UP train passes S1. I think we just say we can live with it, because we ve prevented the collision at the crossover and that s the important thing. Before we bring this tutorial to an end, I want to show you the other two parts of the junction, using actual trackage from Midwest Central. It s laid out a little different, but this is a good real world example of all the things we discussed in this tutorial. Remember we mentioned invisible signals earlier? If you look carefully, you ll see one at position 1. This signal affects only traffic crossing from right, southbound, UP main into the MidCen eastbound main. Three triggers, A, B and C, control signal 1. Signal 2, on the left, northbound, UP main is controlled by trigger D. Why three triggers for signal 1? Because the UP mains are high-speed tracks, and they have priority over trains taking the crossover. By using three triggers, we make sure to have overlapping detection zones all the way from A to the crossover. So as soon a train hits A, we lock the crossover and keep it locked until the northbound train clears it. If we only did one trigger, at A, the train would leave A s detection zone before the train clears the crossover. Using three triggers prevents that. On the other hand, if an eastbound train actually reaches trigger D in the crossover before the northbound hits trigger A s detection zone, the northbound train has plenty of time to stop before it gets to signal 2.

22 This is great example of assigning a priority and longer triggering distance to one track over another. The high-speed train, with its longer stopping distance, rightly deserves the priority over the train taking the 25mph, low speed crossover. But if the low speed train is actually on the crossing, the signaling prevents a collision. Now to the third UP/MidCen junction, shown below. This junction is where the two MidCen mainlines join UP territory. North is to the right and South is to the left. Again we have three triggers, A, B and C controlling invisible signal 1. An incoming UP train reaching A will lock the crossover, and the overlapping detection zones of A, B and C will keep the crossing locked until the UP train has cleared the junction. Conversely, if a MidCen train heading south onto UP territory reaches trigger D before an eastbound train reaches A, it will take control of the crossing. The eastbound will have to stop, but the placement of the triggers will help ensure it has enough room to do so. In setting up Westfield Junction, I used six separate instances of the Trigger Multiple Signals rule, two at each of the three legs of the junction. For example, in the picture above, one rule covered triggers A, B, C and signal 1. A second was needed for trigger D and signal 2. Conclusion This document has grown to some length, but I hope you have a better idea how the proper use of triggers and signals can help prevent collisions where tracks cross. Thanks to Maggs for providing us with a very useful rule that makes it all possible, and to NorfolkSouthern37 for his invisible signal. Here s a review of my principles for where tracks cross: 1. Give one track a higher priority. 2. Use a separate rule for each track of the crossover. 3. Be careful in the placement of signals and triggers so trains don t stop in each other s detection zones. And let s add one more:

23 4. Consider using multiple, overlapping triggers on the priority tracks to extend the effective detection zone further from the crossing. Happy trainzing. Chuck December 2008