Signal Complexity

This forum is dedicated to Riding Scale Railroading with propulsion using other than steam (Hydraulics, diesel engines, gas engines, electric motors, hybrid etc.)

Moderators: Harold_V, WJH

User avatar
johnpenn74
Posts: 386
Joined: Tue May 04, 2010 12:54 pm
Location: Waiting for next assignment!

Re: Signal Complexity

Post by johnpenn74 » Wed Dec 12, 2018 10:16 pm

Sure it is. Most mainline signals are permissive, which definitely makes it within the realm of possibility that two trains could simultaneously be in the same block.

The permissive concept is intended to keep trains moving. Blocks signals are not anything like highway traffic signals and don't tell an engineer when to stop and when to go, the lone exception being an absolute signal displaying a stop aspect (which cannot be passed).
Yeah, sure. In an Absolute Permissive Block system you could have the second train run the intermediate block permissive red (not the head block) and stop in sight of the first train.

The gains from 2 trains sectioning a siding to siding to block is immaterial on our railroads since our runs are short and bydirectional running of track is going to require you to have traffic flow change direction on a regular basis. IE if your direction changes then the number of trains you will see between the headblocks is probably going to be one train a high percentage of the time. Our mainlines are measured in hundreds of feet not miles between headblocks. So our ability and time requirement to change direction is faster and the cycle time to process a train through a block is less relevent. These two items make the need for sectioning pointless on live steam tracks.

And lets not forget the prototype. I'll venture that permissive reds (usually plate on mast indicated) are only in places where uphill grades might stall out the trains, so we have to keep them rolling if possible. I would bet there are few if any permissive blocks in Florida or other flat places in the United States. :-)
If you enforce a one-train-per-block rule your railroad will get constipated on a busy day as engineers stand at red signals waiting on a slow-moving train up ahead to clear the block.
What? that's just flat out wrong. If you look at ANY track that has two way traffic and capture the flag signals (White Creek, Eaglepoint, CStp&P, Mill Creek to name a few of the biggest ones out there) all that is allowed into the block is one train at a time. Sure you can manually section, but that is the exception not the norm. The real congestion is a result of unmanaged traffic flow. People going where they want when they want with out any multiple passing, priority overtaking, or means to coordinate the flow.

Our operational traffic more or less random, we follow no schedule and only on a few railroads do we dispatch the traffic by radio on the main (CIG for Example). I'll solve that problem next year... :-)

JP
John Pennington

Logging meets that actually move logs

Project
2 Mich-Cal Shays
Allen 4-4-0 Narrow Gauge Conversion
Two Reading A5a Camelback 0-4-0
USRA 0-6-0
Clishay
4 Western Wheeled Scraper NG Dump Cars
N&W 4-8-2
ICM 2-10-2
4 Modern Stake Cars
L&N Caboose
4 Big Four Conversion Gondolas

Like I'm actually gonna build all this stuff :-P

User avatar
BigDumbDinosaur
Posts: 807
Joined: Tue Jun 28, 2011 9:19 pm
Location: Midwestern United States

Re: Signal Complexity

Post by BigDumbDinosaur » Thu Dec 13, 2018 12:25 pm

johnpenn74 wrote:
Wed Dec 12, 2018 10:16 pm
I'll venture that permissive reds (usually plate on mast indicated) are only in places where uphill grades might stall out the trains, so we have to keep them rolling if possible. I would bet there are few if any permissive blocks in Florida or other flat places in the United States. :-)
Well, I'm in a "flat place" (Illinois, the "Prairie State") and can tell you the Union Pacific mainline that runs nearby has permissive signals. In fact, on every road on which I have traveled in the course of business (more railroads than most people will traverse in a lifetime), including the heavily-traveled BNSF mainline that runs across the northern part of the country from the Powder River Basin to Lake Superior—miles and miles of level terrain—almost all mainline signals are permissive. As a fairly general rule, the signals between opposing head block signals in APB territory are permissive. Furthermore, in some areas where unrestricted line-of-sight to the downline end of a block is possible, engineers are not required to come to a complete halt before passing a permissive signal displaying stop—they can roll past at restricting speed if there is no immediate obstruction.
BigDumbDinosaur wrote:If you enforce a one-train-per-block rule your railroad will get constipated on a busy day as engineers stand at red signals waiting on a slow-moving train up ahead to clear the block.
What? that's just flat out wrong.
Not based upon my observations at our club and others I have visited. You'd be surprised how many in this hobby don't understand the concept of permissive signalling.
The real congestion is a result of unmanaged traffic flow. People going where they want when they want with out any multiple passing, priority overtaking, or means to coordinate the flow.
ABS is not going to fix that problem. Furthermore, an overly-complicated ABS is going to exacerbate it when visiting engineers who don't understand the intricacies of your ABS make the inevitable mistakes. You best hope for a functional ABS is one that reduces the likelihood of collisions. For the most part, you are not dealing with professional railroaders, just hobbyists who want to pull the throttle.
—————————————————————————————————
I'm an old guy. What's your excuse? ☻

User avatar
Erskine Tramway
Posts: 270
Joined: Wed Sep 10, 2014 4:13 pm
Location: South Dakota
Contact:

Re: Signal Complexity

Post by Erskine Tramway » Thu Dec 13, 2018 1:24 pm

BDD is correct. When I hired out, the BN Alliance Division in the PRB was in the process of converting from 'Dark", i.e. Train Order Operation, to CTC. At first, there were no 'Permissive' Intermediate signals, the automatic signals between the controlled 'OS' signals. It wasn't too long before the Company discovered that some of the locations where Intermediates were sited made a 'Stop and Proceed' signal 'drawbar city' trying to re-start a coal train with four or five motors on the head end. So, then they started changing some of them to 'Permissive', or as we called them, 'Grade' signals, by installing a "G" plate on the mast. Eventually, there were even some on 'downhill' signals, like on the 'multiple main line' (commonly called double track, though signaled for both directions of traffic) West of Dewey, South Dakota, where several trains were sometimes 'stacked up' waiting to get into Edgemont for crew changes. The purpose of that was to prevent having to stop, than pull down a half-mile or so, and stop again. By the time I took my pension, the Company had changed all the Intermediate signals to Permissive, "Pass without stopping, at Restricted Speed". Even at Absolute Signals, the Control Operator (Dispatcher) can 'talk you by' to give you Authority to Occupy the track past the Stop signal. That was also a common way for the Trainmasters to conduct 'flag tests' of trains talked by an Absolute :shock: They weren't always good at it though, a friend of mine ran over their red flag when they'd told the Dispatcher to set up one track for Stop, and they put the flag on the other track, where he was 'running on the Irish' :lol:

Mike
Former Locomotive Engineer and Designer, Sandley Light Railway Equipment Works, Inc. and Riverside & Great Northern Railway 1962-77
BN RR Locomotive Engineer 1977-2014, Retired

User avatar
johnpenn74
Posts: 386
Joined: Tue May 04, 2010 12:54 pm
Location: Waiting for next assignment!

Re: Signal Complexity

Post by johnpenn74 » Thu Dec 13, 2018 11:07 pm

Well, my next question to the pro crews is: If sectioning is scheduling thing (1 logical train, but broken up into 1 or more physical trains), Does dispatch routinely allow more than one train in between the head blocks?

I mean, what would be the point? All you would end up doing is stacking up and stalling out trains on a grade. I understand the point of the permissive, to keep the train rolling to avoid a stall or broken coupler on an uphill start. But unless the block in 20x the length of the train doesn't it invite conditions to make a larger mess... IE does dispatch INTENTIONALLY allow multiple trains between the headblocks? Yes we know the hard wire and relays can handle it, but does it make good operational sense?
Even at Absolute Signals, the Control Operator (Dispatcher) can 'talk you by' to give you Authority to Occupy the track past the Stop signal.
I thought this was some kind of federal violation to over run an absolute red (not to be confused with a tower operated /controlled red).

Yes I agree, APB / ABS will not do anything to remedy flow. I can add in some time delay relays and give one end of the block a timed priority window over the other. But that doesn't fix the problem either. I think only dispatching could make a change there.

JP
John Pennington

Logging meets that actually move logs

Project
2 Mich-Cal Shays
Allen 4-4-0 Narrow Gauge Conversion
Two Reading A5a Camelback 0-4-0
USRA 0-6-0
Clishay
4 Western Wheeled Scraper NG Dump Cars
N&W 4-8-2
ICM 2-10-2
4 Modern Stake Cars
L&N Caboose
4 Big Four Conversion Gondolas

Like I'm actually gonna build all this stuff :-P

User avatar
BigDumbDinosaur
Posts: 807
Joined: Tue Jun 28, 2011 9:19 pm
Location: Midwestern United States

Re: Signal Complexity

Post by BigDumbDinosaur » Fri Dec 14, 2018 12:58 am

johnpenn74 wrote:
Thu Dec 13, 2018 11:07 pm
Well, my next question to the pro crews is: If sectioning is scheduling thing (1 logical train, but broken up into 1 or more physical trains), Does dispatch routinely allow more than one train in between the head blocks?
Actually, it's the road's rules that allow more than one train between head blocks (I assume you're referring to APB territory when you say "head block"). Case in point: prior to the merger with the BN, a Santa Fe engineer's authority to pass a permissive signal displaying stop would have been rule 9.61. As Mike noted, dispatch might talk an engineer past an absolute stop, but doesn't need to do so for a permissive stop (of course, dispatch might order a train to not pass a permissive stop for various reasons, such as a derailment in the block being governed by the signal).
I mean, what would be the point? All you would end up doing is stacking up and stalling out trains on a grade. I understand the point of the permissive, to keep the train rolling to avoid a stall or broken coupler on an uphill start. But unless the block in 20x the length of the train doesn't it invite conditions to make a larger mess... IE does dispatch INTENTIONALLY allow multiple trains between the headblocks? Yes we know the hard wire and relays can handle it, but does it make good operational sense?
The presence of a gradient usually isn't relevant to the placement of permissive signals. Jerked lungs can occur during a start on dead-level terrain—I've been on a number of trains that have had that happen. Roads do all they can to avoid stopping a train once in motion, if for no other reason than to reduce fuel consumption, as well as wear-and-tear on equipment (think of the cost of replacing brake shoes on 100 hoppers, for example).

Time is also a factor: a stationary train is a crew being paid to listen to radio chatter, gaze at the surrounding landscape and discuss sports. Therefore, signalling rules are written in a way that encourages crews to keep their trains moving as much as possible. Don't forget, railroad signals tell the engineer permissible operating speed (also routing), where being stationary is a "speed." In themselves, signals don't give a train authority to occupy the mainline, but once that authority has been granted, signal rules take over and tell the engineer how fast he can run.
Erskine Tramway wrote:Even at Absolute Signals, the Control Operator (Dispatcher) can 'talk you by' to give you Authority to Occupy the track past the Stop signal.
I thought this was some kind of federal violation to over run an absolute red (not to be confused with a tower operated /controlled red).
There is no federal rule that prevents a dispatcher from authorizing an engineer to pass any signal displaying stop. If such were the case and an ABS malfunction were to occur that resulted in an absolute signal displaying a continuous stop, the entire mainline would come to a halt behind the train standing at the absolute stop. I recall just such a situation on the old Rock Island years ago near the Joliet interchange. Dispatch talked each train past the obstreperous signal until the problem was located and corrected.
Yes I agree, APB / ABS will not do anything to remedy flow. I can add in some time delay relays and give one end of the block a timed priority window over the other. But that doesn't fix the problem either. I think only dispatching could make a change there.

The traffic flow problems at a typical club like EPRR or ILS are nothing like those of a full-sized railroad. As I earlier said, the majority of the trains running at our clubs are not being controlled by individuals with real-world railroad experience. Signal complexity will not help them, only confuse them. Dispatching requires that every crew has a radio and knows enough about the operation of the railroad to understand the dispatcher's instructions.

I urge you to stick a basic three-aspect system, with the only exceptions being at facing turnouts, where diverging indications will be useful to tell oncoming engineers how the turnouts are aligned. Having run at the EPRR, I'm not seeing where anything more complex is warranted. Your goal should be to use your signalling to minimize the likelihood of collisions. Complexity will work against that goal, in my opinion.
—————————————————————————————————
I'm an old guy. What's your excuse? ☻

User avatar
Erskine Tramway
Posts: 270
Joined: Wed Sep 10, 2014 4:13 pm
Location: South Dakota
Contact:

Re: Signal Complexity

Post by Erskine Tramway » Fri Dec 14, 2018 11:47 am

johnpenn74 wrote:
Thu Dec 13, 2018 11:07 pm
Well, my next question to the pro crews is: If sectioning is scheduling thing (1 logical train, but broken up into 1 or more physical trains), Does dispatch routinely allow more than one train in between the head blocks?

I mean, what would be the point? All you would end up doing is stacking up and stalling out trains on a grade. I understand the point of the permissive, to keep the train rolling to avoid a stall or broken coupler on an uphill start. But unless the block in 20x the length of the train doesn't it invite conditions to make a larger mess... IE does dispatch INTENTIONALLY allow multiple trains between the headblocks? Yes we know the hard wire and relays can handle it, but does it make good operational sense?
Even at Absolute Signals, the Control Operator (Dispatcher) can 'talk you by' to give you Authority to Occupy the track past the Stop signal.
I thought this was some kind of federal violation to over run an absolute red (not to be confused with a tower operated /controlled red).

Yes I agree, APB / ABS will not do anything to remedy flow. I can add in some time delay relays and give one end of the block a timed priority window over the other. But that doesn't fix the problem either. I think only dispatching could make a change there.

JP
Hi, JP....


On the BN coal main out of the PRB where I ran, it is common practice for the Dispatcher to set up the CTC machine on 'Fleet', and then go to the bathroom or have a cup of coffee. Since the 'Control Points', or 'OS' (On Sheet, left over from paper Train Sheet entries) sections are generally 8-10 miles apart, several trains will fit in the blocks in between. And, it is 'worth your license' (being 'decertified') for at least a period of time, to get any part of your train or engine past a controlled Red signal. In this situation, any signal manually controlled, either by a man in the Tower or the Dispatcher, is an 'Absolute' signal.

As BDD says, the idea is to keep trains moving. The problems come when you get to a 'choke point', like a crew change, or 'second out' at the mine, then trains start 'stacking up'. Our most common situation was when they would hold trains for trackwork, or lack of available crews, for longer than a crew had left to work on their 'hours of service'. Then, they would start parking trains at the ends of 'double track'. Our blocks were about 1.8 miles long, and a coal train is about 1.25 miles long. If you only put one train in a block, say between the Absolutes at Clifton, Wyo. and East Dewey, S.D., you could only get four trains parked there. By pulling up tight behind the train ahead, they could stuff at least one more train in there, usually by 'talking' the last one in. When the double track at Moorcroft, Wyo. ended at Kara, and the first train was stopped at Kara, the following trains would normally 'hold back' of the Belle Fourche River at the bottom of Moorcroft hill, to avoid having to re-start on the hill. That practice pretty much went away when they started using the slave motors on the rear end, mostly account the two on the head end don't have enough power to pull a train in two :lol:

It was very rare for two moving trains to be in the same block. It wasn't against the rules, but it wasn't something the trailing train wanted to do either. You never knew when the guy in front would unexpectedly stop for whatever reason, and leave you having to stop with crossings blocked, or in a bad starting spot. I always tried to avoid getting 'inside' a Yellow (Approach) block anyway, because once you got past the Yellow signal, you couldn't increase your speed above 35MPH until you passed the next signal, even if it had just changed, and you could see it was Green from a mile away. There were several situations which would cause that, the first train crossing over to the other main, or clearing a Junction OS, for instance.

The bottom line, as BDD says, is that Live Steam clubs aren't 'populated' with folks who make their living doing this kind of stuff. On the big railroad, it can be a matter of life or death that you obey the signal indications. So, how the big railroad does it can just confuse people for whom it isn't second nature. The best signal system is the simplest that will indicate the condition of the block ahead, and/or the route you are taking at the next signal.

I hope this hasn't gotten too 'deep in the weeds' on the subject.

Mike
Former Locomotive Engineer and Designer, Sandley Light Railway Equipment Works, Inc. and Riverside & Great Northern Railway 1962-77
BN RR Locomotive Engineer 1977-2014, Retired

rkcarguy
Posts: 1203
Joined: Tue Aug 22, 2017 10:33 am
Location: Wa State

Re: Signal Complexity

Post by rkcarguy » Fri Dec 14, 2018 5:22 pm

John, could you enlighten us as to what relays you are using?
I ended up using two SPDT relays to properly control the R/Y/G aspect of one signal in a normally green/clear default.
It was a little tricky to figure out actually not being an electrical engineer. At rest relay #1 displays green, when energized displays yellow, and then relay #2 when energized displays red AND drops power to relay #1(power to relay #1 feeds from at rest relay #2), so that yellow or green cannot display. The downfall of this is that if there is any kind of failure in the detection circuits the signals will still show green.

If you want to do it the other way so the system is permissive/red default, it requires a diode be used between the "block sensor" wires for green and yellow otherwise they will both illuminate at the same time while a train is making the connection in both blocks as its crossing the isolators. Keep in mind that the signal will display green until the ENTIRE train has crossed into the "yellow block", so a train length must be compensated for in the placement of the isolators(downside of this setup). The plus side of this, is that the default is all red if the detection circuits fail.

Permissive would require that the rail isolators be a bit ahead of the signal, so that as the train rolled up to a red signal and crossed into the next block it would go green if the bock was clear. For absolute, the isolators need to be behind the signal, such that AFTER the locomotive/engineer passes it and enters the next block it changes to the signal to red.

User avatar
Erskine Tramway
Posts: 270
Joined: Wed Sep 10, 2014 4:13 pm
Location: South Dakota
Contact:

Re: Signal Complexity

Post by Erskine Tramway » Fri Dec 14, 2018 6:51 pm

On the big ones. the 'fail' position is to red. The block is powered from the 'leaving' end, and as long as there is a circuit, the green/yellow signal relay is 'picked up'. As soon as anything interrupts the circuit, equipment occupying the block, a broken rail, or a Trainmaster with his 'shunt', the relay 'drops out' and displays the red indication.

The insulated joints at the signal are always 'offset', one on each side of the signal, even in continuous welded rail territory. That way, a short fast train, can't clear both rails of a block at once. As soon as both joints are passed with the leading set of wheels, the signal drops to red.

In your situation, you have to have a way to tell the guy running whether a signal is an Absolute, or a Permissive, either by rule or by a marker on the signal mast. The only Absolutes we had on the BN were signals controlling moves over switches or crossovers.

Mike
Former Locomotive Engineer and Designer, Sandley Light Railway Equipment Works, Inc. and Riverside & Great Northern Railway 1962-77
BN RR Locomotive Engineer 1977-2014, Retired

rkcarguy
Posts: 1203
Joined: Tue Aug 22, 2017 10:33 am
Location: Wa State

Re: Signal Complexity

Post by rkcarguy » Fri Dec 14, 2018 10:49 pm

Ok that makes sense Mike. Because I'm not shunting across the rails I can't do it exactly like that, but with a permissive setup because it relies on the train to complete the ground through the rails in the detector circuit it would stay red if that ground failed. I think because my RR is going to be pretty simple, at best case a wye or reversing loop at one end and a passing siding in the middle, any "meet" is going to have to happen on the siding/main section so I think signaling with a green/clear default would suit me the best in reality. I think the signal changing from red to green right as one is rolling up to it would be confusing to any guests. I'm not sure if the Southern Pacific's track through Oregon was signaled in Absolute or Permissive, which could sway me to go one way or the other:)

User avatar
BigDumbDinosaur
Posts: 807
Joined: Tue Jun 28, 2011 9:19 pm
Location: Midwestern United States

Re: Signal Complexity

Post by BigDumbDinosaur » Sat Dec 15, 2018 12:19 am

rkcarguy wrote:
Fri Dec 14, 2018 5:22 pm
John, could you enlighten us as to what relays you are using?
I ended up using two SPDT relays to properly control the R/Y/G aspect of one signal in a normally green/clear default.
It was a little tricky to figure out actually not being an electrical engineer. At rest relay #1 displays green, when energized displays yellow, and then relay #2 when energized displays red AND drops power to relay #1(power to relay #1 feeds from at rest relay #2), so that yellow or green cannot display. The downfall of this is that if there is any kind of failure in the detection circuits the signals will still show green.

If you want to do it the other way so the system is permissive/red default, it requires a diode be used between the "block sensor" wires for green and yellow otherwise they will both illuminate at the same time while a train is making the connection in both blocks as its crossing the isolators. Keep in mind that the signal will display green until the ENTIRE train has crossed into the "yellow block", so a train length must be compensated for in the placement of the isolators(downside of this setup). The plus side of this, is that the default is all red if the detection circuits fail.

Permissive would require that the rail isolators be a bit ahead of the signal, so that as the train rolled up to a red signal and crossed into the next block it would go green if the bock was clear. For absolute, the isolators need to be behind the signal, such that AFTER the locomotive/engineer passes it and enters the next block it changes to the signal to red.
Pardon me for being blunt and undiplomatic, but the above tells me you don't understand how railroad signalling works.
—————————————————————————————————
I'm an old guy. What's your excuse? ☻

rkcarguy
Posts: 1203
Joined: Tue Aug 22, 2017 10:33 am
Location: Wa State

Re: Signal Complexity

Post by rkcarguy » Sat Dec 15, 2018 12:21 am

greensig.jpg
ylwsig.jpg
redsig.jpg
redylwsig.jpg
So here is what I've got. The white wires on the relays are the trigger wires which would be tripped by occupancy of the adjacent blocks.
In the last picture, I've "simulated" a turnout switch which indicates the lower yellow, and a "diode" jump wire to relay #2 drops the upper head to red. You'll notice there is no red in the lower mast, this was an SP specific thing used on some of SP's routes in Oregon, and a single yellow was indicated below red if the train was to take the diverging route. Otherwise the lower head was dark and the upper head displayed the track occupancy ahead. Not the best, but that's what I'm modeling so that is what it is.

The relay trigger wires require 1amp, 6 volts minimum as these are large heavy duty automotive relays. The nice thing is I can use whatever timed relay circuit I feel like for detection(you want a few seconds delay so the lights don't flicker due to bad wheel connections across the rails), so my method of detection is still wide open as long as it supplies these relays with the trigger amps/volts above.

rkcarguy
Posts: 1203
Joined: Tue Aug 22, 2017 10:33 am
Location: Wa State

Re: Signal Complexity

Post by rkcarguy » Sat Dec 15, 2018 12:30 am

BigDumbDinosaur wrote:
Sat Dec 15, 2018 12:19 am
rkcarguy wrote:
Fri Dec 14, 2018 5:22 pm
John, could you enlighten us as to what relays you are using?
I ended up using two SPDT relays to properly control the R/Y/G aspect of one signal in a normally green/clear default.
It was a little tricky to figure out actually not being an electrical engineer. At rest relay #1 displays green, when energized displays yellow, and then relay #2 when energized displays red AND drops power to relay #1(power to relay #1 feeds from at rest relay #2), so that yellow or green cannot display. The downfall of this is that if there is any kind of failure in the detection circuits the signals will still show green.

If you want to do it the other way so the system is permissive/red default, it requires a diode be used between the "block sensor" wires for green and yellow otherwise they will both illuminate at the same time while a train is making the connection in both blocks as its crossing the isolators. Keep in mind that the signal will display green until the ENTIRE train has crossed into the "yellow block", so a train length must be compensated for in the placement of the isolators(downside of this setup). The plus side of this, is that the default is all red if the detection circuits fail.

Permissive would require that the rail isolators be a bit ahead of the signal, so that as the train rolled up to a red signal and crossed into the next block it would go green if the bock was clear. For absolute, the isolators need to be behind the signal, such that AFTER the locomotive/engineer passes it and enters the next block it changes to the signal to red.
Pardon me for being blunt and undiplomatic, but the above tells me you don't understand how railroad signalling works.
What did I miss? I could have got permissive and absolute mixed up? I'm at a bit of a loss of the definition of each type of system and how they work relatively. From what I have seen in real life operations, viewing a signal (as a southbound train would), I would see these aspects displayed on the signal as traffic came north into town. The interesting thing is, the signals on the UP sub along the Deschutes river in Oregon would stay green all night long even void of traffic. The ones in my hometown, would go dark unless there was activity in the area. As a kid this good and bad, bad because if the signal was dark nothing was happening, but good because when the signals did come on I knew there was going to be some train action.

Post Reply