Large Scale Central

Building a Heavy Duty Spring switch.

I’m moving this here so as to not take over Rick Marty’s Spring Switch thread in the general forums. For continuity here I’ll re-post a few photos from that thread.

Early Spring 2020 clean up discovered my Aristo stainless #6 switches need some cleaning and work on the throws. I bought a bunch of Bachmann throws from RLD at ECLSTS 2019 with a plan to replace all the Aristo throws outdoors with the Bachmann throw as I find they have a much more positive closure. The problem with that thought is that three of my 5 #6’s outdoors are set up as spring switches so I can negotiate a wye without throwing any switches. Unlike the Aristo, the Bachmann throw has no spring option. So; I thought of trying to duplicate this ride-on scale spring switch…

I thought I might be able to adapt a Bachmann throw to stand-in for that red handled throw. My first step in the experiment was to set up some pen springs on a threaded rod to calculate cut sizes and help me visualize how it could work. I’m using an old Aristo R1 as a test platform since it is much easier to handle than a #6

To connect the rod to the inner workings of the Bachmann throw; I purchased some Dubro Kwik-Links at the LHS, who was having a going out of business sale ! :grin: I really wanted the much smaller Sullivan Gold-N-Clevis, but he didn’t have them. I found them on-line, but shipping lead time is 10 days and I’m impatient. To get the much larger Dubro to work, I needed to trim it down a lot so it didn’t bind as the throw gear rotates…

I was concerned that it would not fit inside the machine, but is does…

And the cover closes just fine. It rotates counter clockwise from this view with no binding. I need to trim the clevis just a bit more as it binds when rotating clockwise. I didn’t take a picture of the brass extension of the throw bar I fabricated from brass flat bar stock. So on to assembly of Version 0.01…

There are plenty of reasons why this version was a fail. The brass extension bar is too long, and the hole drilled too high. When the threaded rod tries to move through the clearance hole in the brass, the threads bind due to the less than perfect line. I took some better measurements and took stuff apart. Version 0.02 features a shorter extension with a shorter rise and a larger hole for the rod to allow for some slop…

BTW, that’s one of Devon’s cast metal trucks. It’s quite heavy, but won’t track cleanly without some extra weight. I’m hoping some of that is due to the sharp R1 points.

One of the problems is that the Bachmann throw is that it moves through an arc as it throws. This system would work much better if the rod stayed at a clean right angle to the rail. This version works, but there is some interference with the Aristo throw bar. You can see it in the photo above. Here it is thrown the other way…

It still needs lots of work. I’m thinking of moving the machine to be parallel with the track, then use a bell crank. This might give me the straight line action I’m looking for.

I’m open to suggestions if anyone has a brilliant idea on how to improve this. It is still a work in progress.

Version 0.02 solved some issues, but also pointed out others. The two most problematic being the angle of the throw bar extension to the rod from the switch machine and the the rod threads creating drag while passing through the hole. A little head scratching resulted in some ideas to try and resolve these issues. First up is addressing the angle of attack. The rod wants to be up against the side of the ties when unencumbered, so an offset extension is needed. I ran through a couple of ideas on how to make one when I realized I had some wide bar stock, I cut it to match the length and bent it to match of the previous version, I then drilled the hole offset to the left, rather than center. Version 0.02 on the left, 0.03 on the right…

Next up was the thread interference. Ideally I would use rod with only threaded ends, but lacking that on hand, I thought a bushing might work. What could I make a simple bushing from? How about heat shrink…

Here is how they look together as a dry fit test…

Time to put on the springs and run some truck tests…

At this point I am happy enough with how it works to move it to the #6 and maybe some operational testing on the layout.

1 Like

Version Beta-1 goes out on the railroad in the morning for testing. I’ll only be able to test the straight route for a foot or so until the rest of the switches go back. I’d really like to paint them first. I made one more improvement to the last version as I installed it on the #6. A pin was added to support the throw bar extension…

If functional testing passes, then I need to replace the springs, nuts, and washers with stainless. The rod is already stainless and the brass will be OK outdoors. The Beta unit may stay in place for a few months before I implement it at the other two positions on the wye.

Real-world testing of Version Beta-1 began today. Several trips through both routes, sprung and un-sprung were successful with diesel power (Bachmann 45 Ton), but as expected, the Bachmann C-19’s lead truck will not push the spring over. I have experimented in the past with adding some weight to that truck with a beefier spring. It’s a delicate balance that I haven’t found yet. Too much spring unloads the front driver so the loco looses tractive effort. Too little and it doesn’t load the truck enough to push the spring over. I’ll keep trying to find the magic spring. I forgot to take a picture out on teh RR. Will grab one now and add it here later.

Photo of Beta-1 in place for operational testing…

Paint went on the test version yesterday. I have decided not to upgrade the hardware until this design has proven itself with a month or so of real world testing. I also decided not to implement this quite yet at the other two wye switches. They will keep their Aristo throws that work as springs by default.

Sorry for the out-of focus photo - I only took one at this stage…

Once the weather improves I will get this re-installed out on the layout.

On my Bachmann 2-4-2, I had to put a sleeve on the pivot mount for the lead truck, so that it was not able to lift up far enough to derail when pushing through my spring switches. In other words, I limited its vertical travel. Since weighting it wasn’t an option, and neither was having it bear some of the weight of the locomotive.

Interesting idea David. The C-19 pilot is already sprung, but very lightly. I’ll take a close look and see if a sleeve might be an option.

Over on the East Broad Top Railroad Facebook page, Randy Hall posted these pictures of a spring switch on the Coal Bunker lead just South of the shops complex. Yet another design to consider!

1 Like

Randy Hall posted these pictures of a spring switch on the Coal Bunker lead just South of the shops complex

Jon, I seem to recall that most switches had a spring incorporated in the linkage from the lever. It forced the point blades to snug up to the stock rails, and didn’t need any adjustment. It also allowed the switch to work as a spring in case someone screwed up and went through it when pointed the wrong way.

Thanks Pete - I photographed a few of them on the wye years back, thinking of fabricating models of the stands from brass strap stock. I probably paid little to no attention to the throw mechanics.

One issue making a model of this, is that in the prototype, the ability for the train to stay on the track and push the switch away is different from our model.

In general our models need to work with much less force, as evidenced by people commenting that a spring switch could not be pushed over by the pilot truck.

I’m not sure, but I think it is the weight on the wheels, on our steam locos, the weight on the pilot and trailing trucks are way below scale weight.

On rolling stock, normally our weight is under even the prototype unloaded weight.

Hard to make the mechanism friction free outdoors, which would allow “gentler” springs which is what you want for our “lighter” locos.

I saw someone used a small “bell crank” in a vertical orientation with a weight at one end, the other tied to the throwbar, seemed a very nice solution and easy to add/remove weight.

Greg

Today I pulled just the switch machines off the other two #6’s on the wye. Last evening I refurbished (cleaned and lubed) two of the machines I replaced previously with the Bachmann throws. I was quite surprised that after many years out in the weather they still function well and the spring is still intact, albeit, a bit weak.

Since the Aristo springs have always worked well for me I decided to keep these two in place and let the new design ride for the summer to see how well it plays.

Greg Elmassian said:

One issue making a model of this, is that in the prototype, the ability for the train to stay on the track and push the switch away is different from our model.

In general our models need to work with much less force, as evidenced by people commenting that a spring switch could not be pushed over by the pilot truck.

I’m not sure, but I think it is the weight on the wheels, on our steam locos, the weight on the pilot and trailing trucks are way below scale weight.

On rolling stock, normally our weight is under even the prototype unloaded weight.

Hard to make the mechanism friction free outdoors, which would allow “gentler” springs which is what you want for our “lighter” locos.

I saw someone used a small “bell crank” in a vertical orientation with a weight at one end, the other tied to the throwbar, seemed a very nice solution and easy to add/remove weight.

Greg

I like to make models !

Just read through this and have used something similar with retractable tail wheel steering on my 1/8-1/9 RC warbirds. To solve some of the issues you experienced I would use a 2/56 control rod threaded just on the end for the clevis Instead of the fully threaded rod. I would then use two adjustable wheel collars (again from RC planes) on each end In place of the nuts for the spring buffers. The collars will allow for easy adjustment and the smooth round rod will slip easily through the hole in the plate. Experimenting with spring size and collar location would allow for fine tuning of the mechanism. Some light grease will keep the rod from rusting or a 2/56 brass thimble could be silver soldered to a piece of SS round wire.

Thanks Ted. I was planning on using rod that was just threaded on the ends, but what I had on hand was too long, so I just went with the threaded rod cut to exact length and covered it with heat shrink tubing. I like the idea of the round adjustment collars. I will look into that.

So far this switch has been out in the weather since mid March and it is still working.

Ted Brito said:

Just read through this and have used something similar with retractable tail wheel steering on my 1/8-1/9 RC warbirds. To solve some of the issues you experienced I would use a 2/56 control rod threaded just on the end for the clevis Instead of the fully threaded rod. I would then use two adjustable wheel collars (again from RC planes) on each end In place of the nuts for the spring buffers. The collars will allow for easy adjustment and the smooth round rod will slip easily through the hole in the plate. Experimenting with spring size and collar location would allow for fine tuning of the mechanism. Some light grease will keep the rod from rusting or a 2/56 brass thimble could be silver soldered to a piece of SS round wire.

Reminds me of building Utility

Jon Radder said:

Thanks Ted. I was planning on using rod that was just threaded on the ends, but what I had on hand was too long, so I just went with the threaded rod cut to exact length and covered it with heat shrink tubing. I like the idea of the round adjustment collars. I will look into that.

So far this switch has been out in the weather since mid March and it is still working.

Good to hear the mechanism is holding up to the elements Jon. My own outdoor layout is stIll in my head and not even drawn out on paper so I’m not sure if I’ll have a need for a spring switch or two yet but your thread is filed in my head for future reference if I do. My plan is for a mostly automatic layout with at least three different trains feeding the outdoor loop from inside the house. I currently have two switches where the trains push the points over when they come out of the yard but they don’t have to spring back because of the sequence the trains operate at and I rarely reverse through them. But if I do, they are both electric motored as they are way at the back edge of my layout.

Update: As I fix picture links I thought it would be good to note that the original pen springs are still in place and functional after 19 months outdoors in New England weather.

1 Like

[quote=“JRad, post:19, topic:79838”]the original pen springs are still in place and functional after 19 months outdoors in New England weather.[/quote] That is impressive. What science-fiction material did they make those from?