Turnout control methods?

Model Railroader Electronics and DCC
1

Hello everyone:

One of the next choices that our club has to make in building our new layout is how to control the turnouts. I would like to hear what you recommend as well as what to avoid.

As far as how to actually move the throw bars, we are already looking at Tortoises, Blue Point manual controllers and Caboose Industries ground throws. I would like to hear your opinions of those options as well as any others that are out there. We have already decided that we are not going to use twin coil ‘snap’ switch motors, and the frogs will be powered.

In addition to the types of switch motors I would like your advice on how to control yard ladders. In our case the main yard has six tracks and is double ended. I found one article in the August 2011 MR (pgs. 58 - 61).

Of course, the other issue is cost. Although most of the electronic solutions seem to be pretty slick, I don’t think the club is prepared to spend $15+ per turnout just for control. We will have about 75 turnouts.

Thanks,

Dave

2

Hi there. Probably outside budgets but I use DCC Concepts Cobalt IP Digital motors. These are Tortoise type motors but with a DCC accessory decoder and frog polarity management built in. I connect the frog polarity connection to power two colour LEDs (red and green) to show switch settings.

You can read about them on my blog - recently HERE

3

Hey Dave-

The original plan for my current layout build was to have manually conrolled Peco turnouts. We’ve used them on a club layout, and they have withstood at least 10 or 20 years of use (and abuse . . . club layout). The other original plan was to install a rudimentary signal system “sometime in the near future”, and that might require block detection and automated turnout control to be installed in the near future as well.

I figured that while everything was clean and open, maybe I could install Tortoise machines on four turnouts now. Just as a test. That got me to thinking that I could install signals for the four plants now. And while I was at it, I might install the Digitrax BDL168 and SE8C cards now as well. And the wiring and so forth. So, I did.

The first four turnout motors were pretty easy to install and they worked perfectly well, and I thought, “Hey, neat!” So I added a few more. One nice feature of the SE8C is that in addition to signal control, it serves as a stationary decoder allowing remote operation of slow-motion turnout motors (such as the Tortoises) from the hand held throttle. It also allows control of turnouts via momentary push buttons mounted on the fascia or control panel. In my case, I was able to easily wire up the push bottons to control a set of crossovers from either side on the middle peninsula.

One thing led to another . . . There are 42 turnouts on the upper level, 24 of which have Tortoise motors. There will be a lot on the lower level (not sure how many), but only the benchwork and the decks are constructed down there right now. The trackage and whatnot will come sometime in the near future.

Regarding costs . . . The Tortoises cost about $15 each purchased in 6-packs. The SE8Cs cost about $90 each and control 8 turnouts; that works out to about $12 per turnout. The momentary contact push buttons (you might not use them) are pretty cheap on Amazon-dot-com. Wiring was free, stripped from a surplus 1000-foot sp

4

And after pricing this stuff out you can see why I abandoned Tortoises and went with servos. The servo AND electronics are less than $15 per turnout.

If anyone had done the slightest bit of experimentation with Arduino - Geoff Bunza over at MRH has several articles now that show the hookup AND give yoou the software, one Arduini os less than $15, and can control 12 or more servos, that cost about $1.50 each. Or fewer servos in trade for pushbuttons to operate them, instead of using DCC control (or along WITH DCC control). There are several of us here who could modify one of Geoff’s programs to set it up so you have one button per yard track plus an indicator LED to shoow which one is selected - in other words, you want to use track 5, you push the button for track 5, and all turnouts will be lined to route the train to track 5, and the light for track 5 will go on.

It sounds like a lot of complex wizardry but really it boils down to a bunch of If This, Then That - literally the code looks like:

If button 5 is pressed

Set servo 1 to reverse

Set servo 2 to normal

Set servo 3 to normal

Set servo 4 to normal

Set servo 5 to reverse

Turn LED 5 on

The “set servo to reverse” part is handled by canned code that you don’t have to write.

If you have anyone who is a logical thinker, who may have programmed in BASIC before, or some other language, they can pick this up pretty quick. Maintainence is pretty much a non-issue, with various Arduinos being more like in the $5 category, you doon’t repair one when it fails (though I might - except my one attempt thus far was a major failure), you load the code on a new one and swap it in (all plugs in, no soldering). You would of course have to modify the program if you added or remove tracks.

Commercial options do exist, just no

5

One thing Randy didn’t mention about the Servo driver is the Arduino Relay modules, 8 SPDT 10 amp conact relays per module for under a buck per relay off eBay (optically coupled). The Arduino can drive the relays to perform other external functions with the servos.

EDIT:

6

Randy…

Two requests…first how are you issuing commands to the servo modules, and second, can you post some of your simpler code so the learning curve to use the modules is less?

7

Google Geoff Bunza, there are links to almost all of his projects. They link to the web site of an MR competitor so I won’t post the direct links here.

The circuits are pretty much all the same, a servo has 3 pins, power, ground, and signal. Power and ground go to, well poower and ground, and the signal goes to the Arduino. Most of the 14 digital pins are fair game, in the software you define which servo is which pin, so that when the program says to move servo2, it know that means the one on pin 4.

Geoff’s examples include the schematics and the software. Some of it is free downloads that you must be a subscriber to download, but it’s free.

My design I will be sharing the hardware and code for anyone that wants to duplicate it. I will NOT be making or selling anything, but if someone wants to duplicate what I’ve done, go for it. My devices are not so high density (though I may make something different for my yard), each board will control just 2 turnouts, but that includes 2 buttons for each one, plus 2 indicator LEDs foor each one, and a relay for frog power on each. The other pins are for connection to a CMRI based controller board for remote control and remote lockout (so the buttons don’t work unless the dispatcher enables them). Preliminary schematic here:

http://www.readingeastpenn.com/images/TurnoutController2.pdf

–Randy

–Randy

8

Question for Randy, what kind of servos and where are you getting them?

For me, I designed my layout so that turnouts that would be maunal on the prototype would be close enough to the layout edge for ground throw operation, but I still need electrical switching for my wiring scheme, signals and powered frogs.

So I use sub minature slide switches, set right in the roadbed at the throw bar, with little wire springs I make connecting the handle to the throwbar. The small screwdrivers I use for uncoupling also operate the manual turnouts.

So yards, industries, etc, are all ground throws, I don’t worry about the appearance, it is just a very small little “blob” where the switch stand would be.

If it is a yard that needs dwarf signals, the switches provide the info to light them.

The mainline turnouts are CTC controlled and powered with Tortoise machines. Those machines are controlled by relays so that I have lots of contacts for signal logic. The relays also allow for easy matrix setups for one button route control.

The tower control panels have lighted pushbuttons that show the route selected. Push one button and all the necessary turnouts move to the correct route.

Because of the typical motor control push button station wiring I use, any given route or turnout can be controlled on as many different control panels as you want. In my case a local panel and the CTC panel. Both panels show the route selected and the CTC dispatcher can disable the local panel controls which will still illuminate to show the current route.

The ice cube relays I use only cost me a few dollars each, and can still be had surplus or new for $5 or less. A complex interlocking will require about one relay per turnout

9

Sheldon

I use Tower Pro SG90 servos, I buy the 10 pack for $15 off eBay. They work very good with one exception. They advertise that they have 180° movement but that is a stretch, 9 out of 10 vary from about 120° to 170°. That much swing isn’t needed for turnout control so no problem there. Using the supplied crank with 6 holes the max being .6” from the pivot point about 30° works great with my Atlas Custom Line turnouts.

Mel

Modeling the early to mid 1950s SP in HO scale since 1951

My Model Railroad
http://melvineperry.blogspot.com/

Bakersfield, California

10

I use the little 9G servos as well, plenty of power to move even a Peco with the spring still in it, or a Fast Tracks style all-rail turnout (no hinge). They are all over the place on ebay from any number of Chinese sources, usually in packs of 6 for about $10 shipped. I have yet to have one go bad, and yet to have one DOA.

I like to mount thm so they have an over-center action like a Tortoise - so when fully in position it’s really the springiness of the music wire holding the points closed, since servos that can;t reach the commanded travel point will buzz and overheat. The latest one I am trying doesn’t even need the travel range to be set, you just run the servo from minimum to maximum. As such, I wasn;t including any settings for this in my controller, outside of actually editing the program. I do have a pair of jumpers to set the srvo to the center position to make it easier to physically attach the servo and mount under the table and line it up with the throwbar hole.

I did once try a smaller one, they’re either 4.5G or 4G. The idea was to mount it directly under the turnout like a Peco motor - it works very well, but the downside is having to carve out the space before laying the turnout. The advantage is the distance is so short you cna use rather fine wire and still impart enouhg force to securely hold the points. Plus they won’t stick down below even 1/2" plywood let alone plywood with some other roadbed like cork or homasote. In the end I just stuck with the 9G type and mounted them Tortoise style under the layout - they are however a fraction of the size of a Tortoise.

–Randy

11

Randy, I like the idea of the small size of the servos, but that seems like just as much wiring for two turnouts as I do using simple relay logic?

I can have unlimited control stations

I can have a master location lock out other control stations

My lighted pushbuttons show the route

The relays provide the signal logic and frog power directly with extra contacts

Here are a few examples:

The diagram with the wye shows how three relays provide all the logic needed with contacts left over for other stuff.

Sheldon

12

Randy

The 3.7g servos work great for animation.

Mel

Modeling the early to mid 1950s SP in HO scale since 1951

My Model Railroad
http://melvineperry.blogspot.com/

Bakersfield, California

I’m beginning to realize that aging is not for wimps.

13

Robert, Randy, Mel and Sheldon:

Thanks for the suggestions, and the interesting discussion back and forth.

One thing that I should have mentioned in the first post is that whatever system we use has to be understandable for current and future members of the club. I love the Arduino concept but if I were to say the word “Arduino” to the club members most of them would respond with a blank stare.

Sheldon, your wiring system does sound like something we should consider. However, I don’t understand some of the components/symbols on your schematics. For example, you have a component labelled ‘R1’ and then you have a component labelled ‘R1’ with a circle around it. There are also several symbols with two vertical lines with an angled line across them. On some of these symbols the angled line is from upper left to lower right but on others it is from upper right to lower left. I don’t recognize the symbol at all, but then again, I’m no electrical wizard. Could you post a key explaining what each of the symbols mean?

Thanks,

Dave

14

Dave, these are all standard symbols for electrical controls, going back to the 1920’s or before.

The circles with letter or number inside are the relay coil.

The parallel lines are the relay contacts, if it has a diagonal across it, it is a normally closed contact. So a contact labled “R1” is controlled by relay coil “R1”, etc.

When you see a normally open contact, and a normally closed contact labled the same with a common connection between them, that is/can be a “form C” contact, common on relays, like a single pole, double throw switch.

This is “old tech” if you have anyone in your group with industrial motor control experiance they will understand the drawing.

And I can send more info, that was just what I had scanned a

15

For your purposes, yoour system may be better than servos - but mainly because route selection is part of your MZL control system. When turnout control is completely isolated from any power control, it becomes a different story.

The servo drivers can and should be considered a black box. In that case, the wiring couldn’t be simpler. Two wires in for power, and the servos all connect with a 3 pin plug and header.

Dave, if “most” of the club would glaze over at the word “arduino”, that tells me there are some who DO understand them. The key to longevity is documentation - as time goes on, the number of people who understand the device will only go up, but without documentation, if those who created the configuration are no longer there, it will be tough to figure out. At the same time, I see no need to tell members at large the details of what’s under the layout - what they need to know is that if they want to go on a certain track, they push the button for that track. When the light turns green, that means they can go. What happens behind the scenes to make that happen - the wiring and technical committees should know. We have the same thing going on with various RR-CirKits block detectors and signal drivers. We have some members that after nearly 10 years of the layout being DCC still can;t figure out how to punch in the number on the side of their loco, but the majority have no problems running trains and following signals, despite not having the slightest clue about how DCC works, or how the signals get changed. They just follow the lights. I should probbaly get more involved, the original DCC guy recently passed away unexpectedly - no, it wasn’t me because I didn’t have the time to put in to make every meeting and every work session, let alone every exhibition. The previous chairman has also stepped down and he was the one who made the entire JMRI PanelPro setup for the dispatcher, so I’m not sure who’s handling things lately. The

16

Well said Randy!

It’s worth a shot Dave, plus a bunch cheaper in the long run . . . . $4.50 per turnout.

Mel

Modeling the early to mid 1950s SP in HO scale since 1951

My Model Railroad
http://melvineperry.blogspot.com/

Bakersfield, California

I’m beginning to realize that aging is not for wimps.

17

Randy,

And that is what it comes down to, hardware vs software.

Because I use hardwire logic, I don’t need the redudant relays for the frog, I have spare 5A contacts.

I don’t need the information buss.

I build the relay panels on the bench, just like you write code at a desk, but again, looking at your diagram, you have just as much wire as me.

I don’t need a “pin” for each addtional pushbutton station, they just daisy chain in parallel.

I have wired and programed PLC’s, going all the way back to their inception in the early 80’s. Ardunio is just the latest generation PLC, yes finally very afordable.

Yes, some of my turnout relays also direct track power, and they power the frogs. So regarding powered frogs, I do with three wires what take you a whole sub system and addtional board.

My CTC, detection and signaling is exactly like Chubb or Ravenscroft, series logic from the detector, thru the turnout and dispatcher choices, to the signals. Simple versions of the same circuits used by the prototype for 90 years…

Yes, to change things I have to rewire…I’m a good planner, I don’t make many changes…

I have written code, I would rather not. I have built relay panels since age 20, I don’t mind it one bit.

My costs are similar per turnout, $2 for the average relay I use, you still need the same buttons, LED’s, wire, etc. I need less resistors, caps, etc. My power supplies are simpler, two 12V transformers and a couple rectifiers create my split 24V/12V power for control circuits, switch motor power, and signal power.

Sheldon

18

On my home layout, I tried a number of different techniques. For twin-coils, I used simple SPDT toggles on a control panel. I built a number of separate panels around the benchwork to be close enough to the turnouts to see them, since my SPDT toggles were left in the center position and did not indicate turnout position. For my subways where the turnouts were not visible, I used latching relays and LEDs.

Later, I switched to Tortoise machines, and controlled them with DPDT toggles which did give me position indication at the panel. That’s a big advantage.

The last phase of my layout was a separate physical bench, connected only by a liftoff section. Since I no longer had a turnout power bus there, I decided to go with DCC control. I put LEDs in to a control panel so I could see how the turnouts were set. My mistake, though, was using one of the simpler, cheaper DCC turnout decoders that did not allow for both throttle and panel control. After running this scheme for a while, I realized that I don’t personally like using DCC for my turnouts and really would prefer a panel.

19

Keep in mind that is an electronic schematic and most of the lines seen on that drawing will be PC board traces, not wires. Wires for this part of the system will consist of an RJ45 cable from the control panel to the board for each turnout handling the pushbuttons adn the indicator lights for each turnout. No, that does not need 8 wires, but by using a twisted pair for each signal it aids in noise immunity, and RJ45 cables are common off the shelf items. The board will have screw terminals for the frog power. The servos plug in via their 3 pin plugs. Power to run the whole thing comes in with a pair of screw terminals. ANd one other RJ45 cable connects this board to the main controller.

Also - this device is for mainline turnouts in a given control point. Other places that will never need dispatcher control, like my yard, will have a much simplified circuit (read that as, higher density). Since the logic for the master control are not needed, I can control more servos with the same chip. I MIGHT also go with the method of attaching a microswitch to the servo for frog power, rather than the relay. In its simplest form, as seen in Geoff Bunza’s designs, it is possible to get one Arduino (the chip in my circuit is the Atmel ATMega328, same as used in the Arduino Uno, Nano, and Micro) to drive at least 9 servos with buttons, sacrificing indicator LEDs, although it’s easy and cheap enough to glue 2 microswitches to each servo, which effectively makes it a Tortoise with 2 sets of SPDT contacts. Since a prebuilt Arduino Nano is about $3, and the servos are about $2 each, you can see the per turnout cost is greatley reduced with servos.

Also note that since there are no relays in my track drive, being DCC, there are none I can just piggyback off of for the turnouts or frog power. The simplified “just moves the servos” circuit could be substituted for the Tortoise in your system.

&n

20

[quote user=“rrinker”]

Keep in mind that is an electronic schematic and most of the lines seen on that drawing will be PC board traces, not wires. Wires for this part of the system will consist of an RJ45 cable from the control panel to the board for each turnout handling the pushbuttons adn the indicator lights for each turnout. No, that does not need 8 wires, but by using a twisted pair for each signal it aids in noise immunity, and RJ45 cables are common off the shelf items. The board will have screw terminals for the frog power. The servos plug in via their 3 pin plugs. Power to run the whole thing comes in with a pair of screw terminals. ANd one other RJ45 cable connects this board to the main controller.

Also - this device is for mainline turnouts in a given control point. Other places that will never need dispatcher control, like my yard, will have a much simplified circuit (read that as, higher density). Since the logic for the master control are not needed, I can control more servos with the same chip. I MIGHT also go with the method of attaching a microswitch to the servo for frog power, rather than the relay. In its simplest form, as seen in Geoff Bunza’s designs, it is possible to get one Arduino (the chip in my circuit is the Atmel ATMega328, same as used in the Arduino Uno, Nano, and Micro) to drive at least 9 servos with buttons, sacrificing indicator LEDs, although it’s easy and cheap enough to glue 2 microswitches to each servo, which effectively makes it a Tortoise with 2 sets of SPDT contacts. Since a prebuilt Arduino Nano is about $3, and the servos are about $2 each, you can see the per turnout cost is greatley reduced with servos.

Also note that since there are no relays in my track drive, being DCC, there are none I can just piggyback off of for the turnouts or frog power. The simplified “just moves the servos” circuit could be substituted for the Tortoise in your system.

&nb