Signal Help

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

southwestern737
Posts: 138
Joined: Sun Nov 12, 2006 8:45 am
Location: Magnolia, TX

Signal Help

Post by southwestern737 » Wed Oct 26, 2016 12:06 pm

I am trying to upgrade the signals on my railroad and am seeking expert electronics help. I need a train sensing circuit that can switch 12vdc by sensing when my track resistance goes from (infinity-40ohms) to 0ohms. That is to say that my track has anywhere from infinity to 40ohms resistance when there is no train present and 0 when train is present. I also need for the circuit to switch immediately when the resistance goes to 0 but hold for 5 seconds when the resistance returns to normal. If this can be built I can use it to trigger the relays for the signals, but I would would like to add on to this and possibly do away with the relays if it is practical. If any of you are willing to help with this please Send me a PM and I will send you my contact info.
Thanks
Brent Courtney
The Southwestern Railroad

User avatar
makinsmoke
Posts: 1789
Joined: Thu Jul 10, 2003 12:56 pm
Location: Texas Hill Country

Re: Signal Help

Post by makinsmoke » Wed Oct 26, 2016 8:21 pm

Hi Brent,
Reach out to the SWLS guys, Richard Day in particular.

This sounds a little complicated for what you are wanting to do, but I may be wrong. To me simpler is always better if possible, especially when exposed to the weather like our stuff is.

Take care,
Brian

southwestern737
Posts: 138
Joined: Sun Nov 12, 2006 8:45 am
Location: Magnolia, TX

Re: Signal Help

Post by southwestern737 » Wed Oct 26, 2016 9:56 pm

Richard is on my short list of people to call.
Brent

User avatar
BigDumbDinosaur
Posts: 606
Joined: Tue Jun 28, 2011 9:19 pm
Location: Midwestern United States

Re: Signal Help

Post by BigDumbDinosaur » Thu Oct 27, 2016 12:41 am

southwestern737 wrote:I am trying to upgrade the signals on my railroad and am seeking expert electronics help. I need a train sensing circuit that can switch 12vdc by sensing when my track resistance goes from (infinity-40ohms) to 0ohms. That is to say that my track has anywhere from infinity to 40ohms resistance when there is no train present and 0 when train is present.
Hmmm...where to start? (Rhetorical question.)

I'll start with one or two questions. What gauge is this track? Are you using wooden, plastic or concrete ties? How many ties per 100 linear feet of track are you using? What is your maximum desired block length? What sort of ballast are you using? How did you come up with the 40 ohm open circuit minimum?

Now for some technical mishmash. Any Robinson failsafe track circuit implementation using on riding scale track will be working with off-the-shelf relays, not the purpose-designed "vital circuit" track relays used by real railroads to annunciate block occupancy. Off-the-shelf relays have considerably more hysteresis that vital relays, hysteresis being the spread between pick-up and drop-out voltages. The hysteresis of most off-the-shelf relays places more restrictive electrical requirements on track circuit design than demanded by the low hysteresis of the track relays found in full sized practice. Here's why.

Robinson track circuits are generally operated at very low voltages in order to minimize the effects of track "leakage," which is the parasitic current flow that passes from one rail to the other via the ties and ballast. Leakage is unavoidable in any installation, even when conditions are bone dry, and has to be accounted for in the design. The loss caused by leakage varies with the square of the track circuit voltage, which explains why voltages are kept very low (3 volts or less in full sized practice).

At the same time, current limiting is necessary in order to prevent component damage when a train occupies the block and effectively shorts out the power source, as well as the track relay. Paradoxically, current limiting is also an unwanted feature when the block is not occupied, as it conspires with leakage to "steal" voltage from the track relay's coil. The designer walks a bit of a fine line between keeping short-circuit current to a safe maximum and not having too much current-limiting resistance in the circuit. If the latter is too high and leakage becomes considerable, as it would when a soaking rain is falling, the open circuit track voltage will not be high enough to pick up the track relay, resulting in "false occupancy." Higher track circuit voltages exacerbate this problem, as correspondingly higher ballast resistor values must be used to limit the short circuit current flow. This causes the leakage problem to become much more difficult to solve.

For the sake of discussion, I'll toss in some numbers so you can better understand the situation. In order to accommodate off-the-shelf relay hysteresis in a Robinson circuit, your block preparation needs to produce a minimum open circuit rail-to-rail resistance of at least 10 times that of the track circuit's current-limiting ballast resistor in order to produce consistent operation. Typically in a riding scale implementation using a 5 volt DC track relay (e.g., an Omron G5V-1-2-DC5 or similar—5 volts is generally the lowest voltage for which suitable relays can be readily obtained), the ballast resistor would be around 10 ohms in order to limit the shunted track circuit current to a level that can be accommodated with reasonable off-the-shelf components (such as the ubiquitous 7805 voltage regulator), but still be able to compensate for a reasonable amount of leakage. The guaranteed pickup voltage of the Omron G5V-1-2-DC5 is 80 percent of nominal at 25 degrees Celsius, which number would be 4 volts—consider that to be a ragged edge number (drop-out for the same relay is 10 percent of nominal, or 0.5 volts). If you put 40 ohms (your proposed minimum open track circuit resistance) across a 5 volt power source ballasted by 10 ohms, the absolute maximum open circuit voltage will never exceed 3.9 volts due to the combined current draw of the relay coil and leakage. In other words, the track relay will never pick up and the block will never clear.
I also need for the circuit to switch immediately when the resistance goes to 0 but hold for 5 seconds when the resistance returns to normal.
That requirement is practical. However, I wonder about your need for a specific delay-on-clear value. What are you trying to accomplish?
If this can be built I can use it to trigger the relays for the signals, but I would would like to add on to this and possibly do away with the relays if it is practical.
Trying to go entirely solid state, as you are implying, is not wise, in my opinion. The outdoors is a very hostile environment for solid state electronics, especially when connected to metal objects (rails) lying on the ground out in the open. Also, the long wiring runs typical of ABS tend to pick up electromagnetic pulses (EMP) from nearby lightning strikes. EMP will definitely knock out most electronics. For these reasons, full sized railroads continue to use relays in signaling due to reliability concerns and have very cautiously started to accept the use radio telemetry in place of wire interconnections between instrument cabinets (the wayside houses that contain the ABS hardware).

I recommend that you stick with relays for track circuits, signal lamp control and instrument cabinet interconnections, and keep the amount of electronic content down to a minimum.
Science makes it known. Engineering makes it work.

southwestern737
Posts: 138
Joined: Sun Nov 12, 2006 8:45 am
Location: Magnolia, TX

Re: Signal Help

Post by southwestern737 » Thu Oct 27, 2016 7:23 am

Ok, I really appreciate the reply. I guess I should have mentioned in my original post that I don't know much about electronics beyond how a relay works. Also to answer your questions, the track is 7 1/2" gauge aluminum rail on concrete ties, two ties per foot, crushed limestone ballast, and my longest block is about 400'. As far as the 40ohm number that is what I measured when the track is wet (it actually was 50 ohms but I was trying give a little margin) much higher when dry. I am trying to make my signals more reliable when wet. If I understand your calculations and comments it is not possible.

southwestern737
Posts: 138
Joined: Sun Nov 12, 2006 8:45 am
Location: Magnolia, TX

Re: Signal Help

Post by southwestern737 » Thu Oct 27, 2016 7:27 am

Sorry I forgot to address the five second delay, ocasionally I have trains run with very rusty wheels and the signals have a difficult time with them, the delay would keep the signal from blinking and also give the trains a little extra time to clear the block.
Thanks
Brent

User avatar
ccvstmr
Posts: 1495
Joined: Sat Jan 04, 2003 10:37 am
Location: New Lenox, IL

Re: Signal Help

Post by ccvstmr » Thu Oct 27, 2016 8:04 am

SW737...

I figured "the Dinosaur" would comment...I wasn't disappointed. Some 25 years ago, the local club started with a simple relay based signal system. Problems cited with that system were varied. One of the more frustrating situations was when it rained...and all the signals went red. Other issues included the use of high lumen output 12 volt lamps that had a short life (even if visible in bright sunlight), broken rail bond jumpers, and more.

When Dino joined the club, he took it upon himself to develop a working signal system. The club members wanted 2 particular sections of track signaled for safety due to blind spots in curved, high traffic areas. Several members threw every problem encountered at Dino...and he continued to refine his signal system design to eliminate those problems. The club made up a short section of track (alum rails on treated pine ties) to experiment with. Word had it, Dino took that home and soaked it in the bath tub to get "wet" rail-to-rail resistance measurements. Keep in mind, once you get to the point where you have a workable control system, then comes the installation...distances, wire size, etc.

The end result WAS a reliable signal system. Signal lamps are now LED (thanx to technology advances). Large LED's are not cheap, but they last longer than filament bulbs...AND...can be seen at a distance in bright daylight. Of course, there are trade-offs. That reliability comes at a price for the controls. Worst of all...any signal system is only as good as the people using it! If engineers aren't paying attention, it's easy to roll by a signal they shouldn't. Here too, signal systems require rules. Rules for the club signal system are...mainline wayside signals are PERMISSIVE. Funny, most people don't seem to get that. To avoid choking the RR with stopped traffic, engineers are allowed to stop at a red mainline signal, and are permitted to proceed past the red signal. Maybe the engineers just want to see the lamps change color. There is one signal on the RR at the station exit that is ABSOLUTE. This signal should not be passed when displaying red unless directed by a Stationmaster.

Not trying to talk you out of a signal system, but as you can tell, there's more involved than meets the eye. Dino won't steer you wrong. Carl B.
Life is like a sewer...what you get out of it depends on what you put into it!
I don't walk on water...I just learned where some of the stepping stones are!
I love mankind...it's some of the people I can't stand!

southwestern737
Posts: 138
Joined: Sun Nov 12, 2006 8:45 am
Location: Magnolia, TX

Re: Signal Help

Post by southwestern737 » Thu Oct 27, 2016 8:26 am

Thanks Carl B for the comments, I guess I need to give a little more info. I already have signals and they work perfectly about 90% of the time, I am attempting to get that to 100%, I had a meet earlier in the year and we had rain each day which made the signals unreliable for about an hour each day. My railroad is designed for by directional running on two sections, the railroad is in dense woods so signals are a must.hope this clears things up.
Brent

User avatar
NP317
Posts: 1229
Joined: Tue Jun 24, 2014 2:57 pm
Location: Northern Oregon

Re: Signal Help

Post by NP317 » Thu Oct 27, 2016 9:39 am

I learned a lot from this signal discussion! Especially Dino's description on specific electrical requirements.
Thanks, all. 'Illustrates the value of this discussion board!!
~RN
Last edited by NP317 on Fri Oct 28, 2016 10:54 am, edited 1 time in total.

User avatar
Atkinson_Railroad
Posts: 166
Joined: Mon Jun 08, 2015 6:27 pm
Location: Michigan
Contact:

Re: Signal Help

Post by Atkinson_Railroad » Thu Oct 27, 2016 12:06 pm

Section 9 of an instruction manual that complimented a signal product I was making in the 1990s dealt with
block signaling circuits and the issues related to track ballast resistance.
Eleven pages from that January, 1996 document can be scrolled through using the Previous and Next buttons
found in the upper right corner of the URL below:

http://www.pbase.com/visual_first/image/164388227

(Probably could have posted the entire document here in a PDF format, but was not sure this site provides for that option.)

One method of dealing with varying track ballast resistance using off-the-shelf parts was solved by driving a relay with a transistor.
This arrangement allowed for track detection during very low resistance track circuit conditions.

Because the manual information is dated, the Radio Shack 275-214 relay mentioned in the literature may be difficult to find today.
A cross reference equivalent to relays made by other manufacturers can be found if need be.
For example, this NTN part number R12-17D3-12 crosses from the Radio Shack part.
Relay specs. PDF reference:
http://www.nteinc.com/relay_web/pdf/R12.pdf

Should also note that the circuit diagrams shown in the manual pages were rendered with a “railroad theme”
using railroad circuitry symbols instead of typical electronic symbols.

Clicking the [Submit] button,
John

southwestern737
Posts: 138
Joined: Sun Nov 12, 2006 8:45 am
Location: Magnolia, TX

Re: Signal Help

Post by southwestern737 » Thu Oct 27, 2016 1:42 pm

John,
Thanks for that info, i will need to get home so I can print it and read in full size but at first glance it will be very helpful.
Brent

User avatar
BigDumbDinosaur
Posts: 606
Joined: Tue Jun 28, 2011 9:19 pm
Location: Midwestern United States

Re: Signal Help

Post by BigDumbDinosaur » Sat Oct 29, 2016 1:59 am

Atkinson_Railroad wrote:For example, this NTN part number R12-17D3-12 crosses from the Radio Shack part.
Relay specs. PDF reference:
http://www.nteinc.com/relay_web/pdf/R12.pdf
In my opinion, that is not a suitable relay for use in block occupancy detection.

A track circuit relay should be a high sensitivity, low current device, operating at low voltage, and fitted with contacts rated for switching other relays. To explain this last, it's important to understand that block occupancy detection is a function that is independent of signal operation. The relays used to operate the signal lamps (or LEDs) are handling very different loads than the relays that implement the control logic. The latter relay types should be fitted with bifurcated, gold-flashed contacts so they will reliably operate at low current levels, often as low as 30 to 40 milliamperes. On the other hand, relays that switch signal loads should have silver-flashed, single button contacts rated for at least 50 percent more current than the heaviest load that is expected. Such contacts will not reliably switch low current logic circuits.

Track circuit voltage should very low to minimize the effects of leakage. About 15 years ago, I undertook a research project to determine the electrical characteristics of 7½ inch gauge track using wooden ties and coarse ballast. The particular railroad where I did my research uses fairly typical track construction and installation methods, with nominally 260 ties per 100 feet of linear track, the ties being cut from 2×6 inch treated lumber split lengthwise to make two ties of 2×3 inch dimensions. It should be noted that treated lumber is conductive, even when dry, with a typical dry DC resistance per tie of around 60,000 to 75,000 ohms—I use the lower number for all calculations. 60 Kohms per tie translates to a theoretical dry open circuit resistance of 230 ohms per 100 feet of track. I say "theoretical" because continuity between the feet of the rails and the ties is not perfect.

Anyhow, the result of that research was the development of a block occupancy detector that operates at a nominal 4.9 volts on the track circuit when the block is unoccupied, with a maximum short circuit current of 0.5 amperes. Real-world testing showed that this detection is marginally reliable with a 450 foot long block—occasional false occupied states would occur when it was raining. At 300 feet, the detection is 100 percent reliable, even in a driving rainstorm and with part of the track under water. In wintertime testing, it was found that a 300 foot block would consistently clear even with the track under wet snow.

The keys to it, of course, are very low track voltage, relatively high short-circuit current when the block is occupied, and use of a track relay that is sensitive and has consistent hysteresis throughout normal temperature ranges. There are also some other design tricks that compensate for the electrical characteristics of riding scale iron or steel wheels rolling on extruded aluminum rail. I am currently analyzing a new block occupancy detector design that will operate the track circuit at 3 volts DC, with a short circuit current of about 0.75 amperes. Based upon the track characteristic data I have, it appears this detector will be reliable when connected to a block approximately 1000 feet long, equivalent to 1.5 miles in scale.
Science makes it known. Engineering makes it work.

Post Reply