Large Scale Central

Servo controlled turnouts/points

I saw this project on Trainelectronics by Dave Bodnar http://trainelectronics.com/Servo-simple-controller/index.htm

Having one of the servo testers and some servos I decided to see if I could adapt it to my outside layout.

I have made a mock up of the controller and after a bit of experimenting in the shed with a test bed I put 2 servos out on the layout next to 1 set of LGB points and 1 set of Aristocraft points to see how they will go when I run a train through them.

If successful I will convert all my points to servo control.

I plan to put the controllers into a weatherproof enclosure and put a signal box over the top to disguise it.

My biggest worry was that the distance between the points I want to control and the controllers might be too large and the signal would degrade.

I made a 12 Ft long test cable and successfully changed a set of points over that distance.

As I have 2 signal boxes I will split the 12 sets of points I have into groups of 2 (6 controllers in each signal box).

The control switches will be weatherproof ones and mounted in enclosures with a lid.

Edit;

As the servo tester has outputs to control 3 servos at the same time, I have come up with the idea of using a second servo to move a semaphore signal so I can see which way the points are set at a glance.

I tested the idea on my bench set up and the second servo moved in time with the point driver.

Pictures of my test set up.

I need to check out that article. I’m considering automating a couple of throws that are hard to reach on my indoor. My RailPro system has accessory devices can control up to 4 turnouts. I could use the standard Aristo motorized throws, which I have - but something a little smoother might be nice.

Jon Radder said:

I need to check out that article. I’m considering automating a couple of throws that are hard to reach on my indoor. My RailPro system has accessory devices can control up to 4 turnouts. I could use the standard Aristo motorized throws, which I have - but something a little smoother might be nice.

Following Dave’s video and instructions the wiring of the switch and resistor took less than 5 minutes, I used a small “trim pot” potentiometer instead of a fixed resistor and left it in circuit so now I can fine tune the point blades if anything changes eg servo wear etc.

Picked up a few of the servo testers off Amazon. Paid a little more but got them quick. Exactly the same on the outside, and operationally as the one in Dave’s video. Circuit board is a little different, with less parts. The needed modification looks easy. I have one hooked up and working without modification. I need to figure out linkage, then on to the electronics side.

Great thread, and good luck guys!

FWIW, I’ve used boards from Tam Valley for servo control, here’s one:

http://www.tamvalleydepot.com/products/microsingletservodecoder.html

Unfortunately, the Singlet II board I’ve been designing around is halted in production due to Covid-caused limitations.

Cliff Jennings said:

Great thread, and good luck guys!

FWIW, I’ve used boards from Tam Valley for servo control, here’s one:

http://www.tamvalleydepot.com/products/microsingletservodecoder.html

Unfortunately, the Singlet II board I’ve been designing around is halted in production due to Covid-caused limitations.

I’ve operated the Tam Valley system at Al’s RGS of NH. If I simply wanted electric with fascia or DCC control they would be a good choice. Price is very reasonable. I’m trying to put something together that I can operate from my RailPro hand held. Not that I need that, just looks like something fun to try and make work.

Looking at how Dave Bodnar mounted his servo and the video on the Tam Valley site, I’m not a real fan of how they do the linkage, pivoting through a hole in the roadbed. I’m thinking of trying a more direct link, maybe recessing most of the servo below, leaving the top part above the deck. To me that would make a much more positive connection to the throw bar. I can easily hide the servo top in some line side junk, or an equipment box.

I got Version Zero of the bread board put together this evening. I quickly learned that the servo torque is superior to the holding power of good double sided tape on plywood. I think I need to stop at the RC car store and grab some supplies (https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-cool.gif)

Jon Radder said:

Cliff Jennings said:

Great thread, and good luck guys!

FWIW, I’ve used boards from Tam Valley for servo control, here’s one:

http://www.tamvalleydepot.com/products/microsingletservodecoder.html

Unfortunately, the Singlet II board I’ve been designing around is halted in production due to Covid-caused limitations.

I’ve operated the Tam Valley system at Al’s RGS of NH. If I simply wanted electric with fascia or DCC control they would be a good choice. Price is very reasonable. I’m trying to put something together that I can operate from my RailPro hand held. Not that I need that, juat looks like something fun to try and make work.

Looking at how Dave Bodnar mounted his servo and the video on the Tam Valley site, I’m not a real fan of how they do the linkage, pivoting through a hole in the roadbed. I’m thinking of trying a more direct link, maybe recessing most of the servo below, leaving the top part above the deck. To me that would make a much more positive connection to the throw bar. I can easily hide the servo top in some line side junk, or an equipment box.

I got Version Zero of the bread board put together this evening. I quickly learned that the servo torque is superior to the holding power of good double sided tape on plywood. I think I need to stop at the RC car store and grab some supplies (https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-cool.gif)

I like the idea of mounting the servo with the top protruding through the roadbed, I did see on a UK forum someone used a pile of sleepers to hide a servo and also a lineside equipment cabinet.

I am thinking of cutting a hole slightly wider than the body that will let it pass through but not the mounting lugs, a little roof and gutter sealer with small locating screws should hold it in place.

I can then make dummy control rods out of stiff wire that run back to the signal box (longest one would be about 12Ft long) so it looks like a mechanical set up. I could make the rod guides out of 4mm irrigation micro-sprinkler riser.

I have ordered waterproof servos and I am using model aircraft control rods with a thread on one end onto which is screwed a snap on clevis pin connector the other end is bent in a Z to hold it on the point throw bar.

After all the ideas/suggestions I have seen on various forums I think this is going to turn into a Ben Hur production I can just see it coming.

As for glue I found that hot melt glue (I melt it with a micro blow torch and dab it on) is strong enough to hold the servos and is easily removed with a box cutter knife.

I used this for my Kadee clone remote coupler. Simple standalone servo control https://www.sparkfun.com/products/13118

Thanks Dan. I’ve seen that one too.

GAP - The airplane rod with the clevis is the way to go. I used them on a remote mechanical set up a few years ago. You may want to add a brass stiffener to the throw bar to protect it some…

I could do it this way again, but I want to try something new (https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-money-mouth.gif)

Dan Gilchrist said:

I used this for my Kadee clone remote coupler. Simple standalone servo control https://www.sparkfun.com/products/13118

I looked at this again over on the Sparkfun website. It’s really a nifty product. If I could find some modified firmware for it it would be ideal. As shipped it only holds the servo position as long as the the logic trigger is set - a button push would be a simple trigger. As soon as the input is removed, the servo returns to the starting position - a momentary condition. I’d rather have a latching condition; when the trigger is applied the servo moves and stays until that trigger signal is removed and a second one tells the server to return to original position. Probably a simple code change, if I was a coder!

Alternately, the default (un-triggered) condition could be set to mainline, and when trigger applied it moves to siding. When the control source is disconnected, the switch would default to mainline. Not the ideal situation and I think I would like some feedback as well. Might be a good choice if the $2 part doesn’t cut it.

1 Like

Jon Radder said:

Thanks Dan. I’ve seen that one too.

GAP - The airplane rod with the clevis is the way to go. I used them on a remote mechanical set up a few years ago. You may want to add a brass stiffener to the throw bar to protect it some…

I could do it this way again, but I want to try something new (https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-money-mouth.gif)

John this is a photo of my test rig with a mock up of how I plan to fit the servos, good suggestion about reinforcing the throw bar.

The micro servo on the right is for a signal, as it moves at the same time as the points servo I can add a semaphore to show which way the blades are set, I could even use it to open/close a micro switch to turn on and off a signal light.

Hey GAP - I hope you don’t think I’m stealing your thread, but since we are going down identical paths I thought it was appropriate to post here (https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-innocent.gif)

This is my breadboard…

The electronic device at left is a DC Buck converter to bring my 12V accessory bus down to 5V for the system. A “Server Tester” out of it’s wrapper at right top. My in-stock servo selection is down to this old steering servo from a Black Foot that is probably still in the attic. Somewhere in my youngest’s attic junk collection is probably plenty to choose from, but at the price of new I’m not going to hunt. The linkage was made from a paperclip and a 2mm machine screw.

This works enough to prove concept, but not ready for mainline service. I need better, adjustable linkage, a better servo horn, and a more square approach to attachment. I still need to add the switch and trim pot to the servo tester. Switches I’ve got but might need to get a few trim pots.

EDIT to add: Grabbed a 4 pack of servos off Amazon that will be here tomorrow. Got impulse sold this two-pack of these 16 Ch. logic input / pwm output servodrivers to play with.

They are meant to be driven by an Audrino, but old-school TTL logic is something I can handle (https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-smile.gif)

John,

Nothing wrong with running parallel projects.

My goal is to keep the project as simple as possible, I did consider going down the arduino or Raspberry Pi (I have a Pi) path but out in the garden that just adds points of failure that I do not need. I could use the Pi on my indoor HO layout once I learn Python programming language.

I have re thought this project so many times that it bears no resemblance to the original idea for example the testers are now going to be spread around the layout and not concentrated in one place, I am mounting the servos through the baseboard(elevated layout).

So far I have installed my solar battery array and 12V-5V buck converter to give me the start of a 5V bus to power the servos.

Just waiting for the testers and waterproof servos to arrive.

Solar panel

(https://www.largescalecentral.com/FileSharing/user_3049/GAPs Pictures/Solar Panel Mounting.JPG)

Charge Controller and Buck Converter

(https://www.largescalecentral.com/FileSharing/user_3049/GAPs Pictures/Charge Controller.JPG)

Battery and Charge Controller install

(https://www.largescalecentral.com/FileSharing/user_3049/GAPs Pictures/Battery and Charge Controller.JPG)

And yes I am using irrigation pipe as conduit, mainly because; it is cheaper than electrical conduit, it is waterproof, is readily available, comes in 2 sizes 13mm &18mm and it cuts easily (doesn’t take like chicken though (https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-wink.gif)(https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-wink.gif))

Cool - Thanks GAP.

Well, I guess I should have read a little deeper on those servo controllers. I thought I could use simple TTL logic to trigger servos, but I missed that it use Audrinos 12C communications protocol and data from the Audrino exclusively, so that was a waste of $10. If Anyone can use them I’ll pass them along. Or maybe, just return to Amazon.

I know how US people love napkin drawings but in Aust our pubs do not supply napkins but they do beer coasters in a big way.

So here is a coaster drawing of how I am going to wire the tester and trimpot in an enclosure (plastic food container with sealable lid).

There will be one of these next to each set of points under the baseboard.

(https://www.largescalecentral.com/FileSharing/user_3049/GAPs Pictures/Tester Enclosure.jpg)

I ordered in some 10K trim pots and they arrived today. I decided I did not want the easily disturbed pot on the Servo Tester board, so I removed it and extended all three wires. These 3 are connected to one of the 10K trim pots exactly the same way as the on-board pot was. Then I added the Bodnar modification using a second 10K trim pot. I purchased the type with a tiny screw trimmer that allows for very fine adjustment. I also added the switch using one from from my junk box for testing. My breadboard now actually includes an electronic breadboard. I also swapped out the servo to a new one with a basic horn.

Shown here in the curve position…

And moved to the straight route…

I’m very happy with how easy this is to adjust for a perfect throw at both ends and how well it moves the points. Proof of concept complete! Time to order up that RailPro accessory module so I can work out the interface. Should be pretty simple, but I want to add LED position indication and possibly a fascia switch as well, so having all the components wired as they will be should help me work out the details.

Jon Radder said:

I ordered in some 10K trim pots and they arrived today. I decided I did not want the easily disturbed pot on the Servo Tester board, so I removed it and extended all three wires. These 3 are connected to one of the 10K trim pots exactly the same way as the on-board pot was. Then I added the Bodnar modification using a second 10K trim pot. I purchased the type with a tiny screw trimmer that allows for very fine adjustment. I also added the switch using one from from my junk box for testing. My breadboard now actually includes an electronic breadboard. I also swapped out the servo to a new one with a basic horn.

Shown here in the curve position…

And moved to the straight route…

I’m very happy with how easy this is to adjust for a perfect throw at both ends and how well it moves the points. Proof of concept complete! Time to order up that RailPro accessory module so I can work out the interface. Should be pretty simple, but I want to add LED position indication and possibly a fascia switch as well, so having all the components wired as they will be should help me work out the details.

John,

Nice idea about replacing the Pot with a trimpot never thought of that now you have me thinking of having the adjustments out on the facia along with the switch that would save me from having to crawl under the layout to make adjustments. Looks like Version 2,000.999 coming up (https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-wink.gif)(https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-wink.gif)

For the LED position indicator I did toy with the idea of using a micro servo to make and break a micro switch that switches the LED on and off.

My thinking was that there are 2 spare servo outputs on the tester that move the same as the throw bar servo which could be utilised for what ever I wanted.

I didn’t follow through as I want to minimise the number of electronic parts outside (just more things to fail), I will most likely go with a mechanical semaphore signal operated off the throw bar.

I also found that outside LEDs just did not work in bright sunlight (couldn’t see them)

For the throw bar adjustment I listened to the servo and when it wasn’t buzzing at either end of the throw then I was not over driving it.

I measured the current at both ends of the throw and found 4.7mA was the lowest current that it was drawing, rising up to over 200mA as the blades moved.

Good thought on the extra outputs. Gives me something to play with this evening while I wait for more components to arrive (https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-cool.gif)

Nothing on the extra outputs is usable for me except as an additional place to grab 5 volts. I was interested in looking to see if the data line had any possibilities, but the data burst is so short it doesn’t offer anything I can use.

I’m pretty sure I am going to need a relay to replace the SPST switch in the Bodnar design in order to interface with RailPro, so if I’m going to need a relay I’ll just go with a dual pole and use the second pole for the position LEDs. I had SPDT relays in stock, so I hooked one up and with a single wire connection from RailPro (a lighting output) I’m able to trigger the servo / move the points with my handheld (https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-wink.gif)That breadboard is starting to get busy.

DPDT relays should arrive tomorrow. This is almost ready for prime time!

Jon Radder said:

Nothing on the extra outputs is usable for me except as an additional place to grab 5 volts. I was interested in looking to see if the data line had any possibilities, but the data burst is so short it doesn’t offer anything I can use.

I’m pretty sure I am going to need a relay to replace the SPST switch in the Bodnar design in order to interface with RailPro, so if I’m going to need a relay I’ll just go with a dual pole and use the second pole for the position LEDs. I had SPDT relays in stock, so I hooked one up and with a single wire connection from RailPro (a lighting output) I’m able to trigger the servo / move the points with my handheld (https://www.largescalecentral.com/externals/tinymce/plugins/emoticons/img/smiley-wink.gif)That breadboard is starting to get busy.

DPDT relays should arrive tomorrow. This is almost ready for prime time!

Good to see you have it sorted.

I have been slowed down by rain so I have made no progress outside except for putting in an On/Off switch. I had a 500V 30A waterproof stashed away so I am using that (should handle the 5V 3A ok).

Inside I am converting the testers by adding the trimpot, I am going to put them near the points.