I was watching a pgm on The Empire Builder (Chicago to Seattle) & it mentioned that when the line was being pushed through they tried switch backs in a tough part of the Rockies. Eventually they tunneled as avalanches kept taking out the switch back lines.
Have any HO layouts been built that have used switch backs??
I was pondering how the switching & gradient transitions would be engineered.
Tanked
It's been 45 years since I built/had a switchback main track. The plan for my current layout puts a double switchback on the branchline's main track, but I haven't worked out all the details for it.
The gradient transition (vertical curve) will depend on the type of equipment operated. Obviously, a Heisler with 40'-long cars can handle a vertical curve sharper than a 2-10-2 with Pullmans. You need to avoid couplers disconnecting, parts of equipment (like locomotive cowcather) inappropriately touching the track, or wheels lifting off the track. I'd think a 1% grade change for every car (or loco sans tender) length would be more than adequate, but that's just a guess. Also, there should not be a vertical curve in the immediate vicinity of a turnout. One-half-inch-thick plywood can be bent to make a reasonable vertical curve, but three-eights-inch may be necessary for sharper vertical curves. Someone can become a hero by making a "scientific" study of the issue and distributing the results.
I also recommend that the reversing-move track should have a lesser grade than those used in forward movement. Backing up places a greater load on the locomotive so it won't be able to push quite as many cars as it can pull uphill.
In most situations it would be best to have an even number of switchbacks so the train ends up going forward for the rest of the journey.
Mark
The engineering is thusly: Your landings, so to speak, where the train reverses direction for a subsequent grade up or down, should be as close to level as possible. The turnout that permits the diversion from the incoming direction should also be level, as Marks suggests above. So, your level track with the turnout should be supported on sub-roadbed of your choice and construction, but what follows must either be pylons allowing the grade to be followed or cookie-cutter plywood (or appropriate material), or splines. The simplest might be the foam pylons that make a ramp put out by Walthers, but they'll need to rest on a surface or support of some kind themselves. Also, you'll have to fashion the transition vertical curve at each end youself...I think...depends.
For splines, you use risers screwed or glued to joists or cross-bracing as and where you need it. Splines make great transitions easy. Time consuming. and materials heavy.
Cookie cutter technique is relatively easy, simple, and fast, but you still need intermediate supports to keep spans form sagging, and this is particularly acute with thin plywood or luan/doorskin/masonite type materials where you desire a long span with a long transition. But here is how it works. Just after the turnout on the level direction changing track, you anchor a thin, flexible material and then commence a downward curve until you meet your intended grade. At the other end, reverse it until it anchors to the same solid structure on which the next level segment of track and turnout rests. The anchored ends of this thin flexible material are crucial, as you must visualize. It permits the levering downward and the formation of the curve you need for a transition.
Note that the transition curve necessarily encroaches on the distance you may have calculated being necessary for any clearances or height you had intended. The transitions must be figured out and then their average grades figured in the true grade. If you had expected to reach 3.5" of clearance or simple height and had thought you would do that easily in 10 feet, with transitions you can now only achieve it in about 8 feet, and it gets worse as the grade steepens because your transitions become more important and longer.
Save yourself a lot of heartache, as Mike hints, and don't place your turnouts well forward to the extent that you actually have to bend the two routes past the frog almost immediately to start the transtions. It can be done, but it is a lot of work. Just ask me. Far better to have the entrance to the turnout proper as close to level and non-curved as possible for about the last 8"....no super-elevation, either!
-Crandell
My layout has one switchback situation in place, and is intended to have another partway up the canyon on my private coal-hauler.
First - Nonomura, which is located in hidden staging, has two back-in staging yards. An arriving train runs onto a stub-end lead track, then backs onto its designated storage track, which has auto-stop at the clearance point. A departing train runs to the end of the lead, auto-stops, the mainline crossover is set and it proceeds to back out, up a 2% grade on the DOWN main, through a spring switch, onto the bypass, over a summit, down 2% to almost the same level it departed from, through the other bypass turnout to an auto-stop on the UP main. It then proceeds, locomotive forward, up a 2% grade to the surface - hopefully on schedule. Total backing distance is about 1 scale kilometer.
Second - Yamamoto. This is a design I have seen in the prototype, so I know it works. Main track, on a continuous upgrade, crosses a bridge, crosses a double slip switch (!), crosses a diamond crossing, runs through a trailing-point turnout and immediately enters a tunnel. Non-stop trains run straight through. DOWN trains which stop at Yamamoto, either passengers, way freights working the 'three men, three women and a dog' mine or anything that has to clear the way for a unit coal train, take the curved leg of the double slip, then back to either the station platform DOWN spur (through the other straight double slip route) or the freight trackage (across the diamond.) Departing DOWN (actually upgrade) trains take whichever route through the specialwork that will aim them into the tunnel.
UP passenger trains take the diverting route through the tunnel-mouth turnout to a stop at the UP platform spur. To depart, they back across the diamond to the tail track, then run the curved route through the DSS, across the bridge and on to Tomikawa.
It is theoretically possible for four trains to clear the main for a coal unit - passenger trains on both platform spurs, one freight in the freight station-mine area and one freight on the tail track. Getting them in, and out again, is sort of like a sliding-block puzzle - one wrong move and chaos sets in.
Two notes:
Note that Spacemouse's latest track planning venture includes a switchback.
Chuck (Modeling Central Japan in September, 1964 - with switchbacks where necessary)
TankedEngine I was watching a pgm on The Empire Builder (Chicago to Seattle) & it mentioned that when the line was being pushed through they tried switch backs in a tough part of the Rockies. Eventually they tunneled as avalanches kept taking out the switch back lines.
Switchbacks were common on industrial railroads facing steep ascents but usually avoided by common carriers. Nevertheless, they were sometimes used on common carriers, sometimes as a temporary route because of a delay in planned tunneling. The Great Northern crossing the Cascades with six switchbacks is one example. Another example is the Colorado Midland Railroad crossing the Rockies. Some secondary lines had permanent switchbacks. The Oregon, California & Eastern and the McCloud River railroads are two examples, and Sierra Railroad's Angels Camp branchline is a third.
Let me quote an excerpt from Clement C. Williams book published in 1917 entitled The Design of Railway Location, (a former college engineering textbook): "An extreme use of the switchback principle was used in the ascent on the Crown King extension of the Santa Fe R.R. ... The total length of the line is 28 miles, and the ascent of 2436 ft. between Turkey Creek and Crown King is accomplished in a distance of 17 miles. There are ten switchbacks with five backup sections on the line. These sections are from 1500 to 4000 ft. in length, and have slighly easier grades and curves than the go-ahead sections, in order to give the train a little advantange in backing up. The tails of the back-up sections are at present 300 ft. long and are continued on a rising grade of 2.0 per cent beyond the switch stands, the maximum grade approaching the switch-stands being 3.5 per cent. The frogs are No. 6.5 and No. 9."
The texbook gives some principles in switchback construction. Tails must be long enough to accommodate the longest trains that will pass over the line. Tails should be given a (rising) grade in order to conserve the energy of the train in stopping (help stop a downward train) and allow the same to be expended in starting back. (And I might add to help start an upward-bound train). The author also adds that switchbacks usually occur in pairs so that the train need not proceed far while running backwards.
markpierce".......... The texbook gives some principles in switchback construction. Tails must be long enough to accommodate the longest trains that will pass over the line. Tails should be given a (rising) grade in order to conserve the energy of the train in stopping (help stop a downward train) and allow the same to be expended in starting back. (And I might add to help start an upward-bound train). The author also adds that switchbacks usually occur in pairs so that the train need not proceed far while running backwards.Mark
"..........
Obviously technically quite an interesting exercise.
Are there any magazine articles or drawings from layouts that have done this, out there??
I have one on my layout and as others have suggested the turnout and short dead end track are level. The diverging end of the turnout has one track going down and the other going up. It's on my logging branch line as it would drive me crazy on a main line (That's what happened in real life too!).
If you ever come to Australia, there is a Steam operated switchback called the Lithgow Zig Zag operating just west of Sydney.
It was on an original main line, but a tunnel now bypasses the steep grade that the Zig Zag uses. The roadbed has been relaid with 3' 6" track and narrow gauge steam trains run on it.
Read more about it here.
http://www.zigzagrailway.com.au/
There are some impressive stone viaducts still in use.
cheers
Alan Jones in Sunny Queensland (Oz)
In one of his Kalmbach books, John Armstrong had a trackplan featuring the Northern Pacific switchbacks that they used for a few years until they finished the tunnel that replaced it. However outside of logging railroads I don't recall seeing it done on a model railroad.
Alantrains I have one on my layout and as others have suggested the turnout and short dead end track are level. The diverging end of the turnout has one track going down and the other going up. It's on my logging branch line as it would drive me crazy on a main line (That's what happened in real life too!). cheers
If I do one I envisage running Heisler/Shay/Climax crawlers - logging/ore etc.
Is it possible you could post a couple of shots of your switch back so I can get an idea of what they look like?
Thank you
Others have addressed the construction aspects of switchbacks. How will you feel about operating one when you've built it?
Probably the most famous (and most copied) switchback layout is Chuck Yungkurth's Gum Stump & Snowshoe. The track plan is at Carl Arendts's site (http://www.carendt.com/) and in Sept 1963 and April 1966 Model Railroader issues.
If your switchback has no bypass, it means every single time you want to take a train over the route, you must go through the switchback. Some get bored by the switchback operation very quickly. Others find it relaxing. Without a doubt, a switchback without a bypass will interfere with "sit back and watch 'em run" continuous running. Unfortunately, knowing which operational preference you fall into probably won't be known until you have built the switchback. If you have access to a switching layout, I would definitely give it a try before committing to a switchback operation.
That said, I am building my version of the GS&S without knowing for sure I will enjoy the switchback operation.
my thoughts, your choices
Fred W
fwrightOthers have addressed the construction aspects of switchbacks. How will you feel about operating one when you've built it? Probably the most famous (and most copied) switchback layout is Chuck Yungkurth's Gum Stump & Snowshoe. The track plan is at Carl Arendts's site (http://www.carendt.com/) and in Sept 1963 and April 1966 Model Railroader issues. If your switchback has no bypass, it means every single time you want to take a train over the route, you must go through the switchback. Some get bored by the switchback operation very quickly. Others find it relaxing. Without a doubt, a switchback without a bypass will interfere with "sit back and watch 'em run" continuous running. Unfortunately, knowing which operational preference you fall into probably won't be known until you have built the switchback. If you have access to a switching layout, I would definitely give it a try before committing to a switchback operation. That said, I am building my version of the GS&S without knowing for sure I will enjoy the switchback operation. my thoughts, your choices Fred W
An interesting variation of the Gum Stump & Snowshoe is here http://www.nmra.org/beginner/vandalia.html
Enjoy
Paul
I lived in the Andes Mountains of Peru for 9 years. The Central Railroad of Peru uses switchbacks extensively. It is the only way they could maintain a decent grade and still generate revenue. So.....when ya gotta, ya gotta.......!
This was many years ago, so I can't be positive, but my recollection, from having walked along the tracks and onto a switchback reversing section, past the turnout, is that it was level, or very close to it.
A switchman and his family lived across from the turnout at the one I recall exploring. He was a busy feller dropping what he was doing on his property to run over to the switch stand while the train thundered past and came to a halt. This must have happened several times a day....it wasn't the busiest of lines, but traffic had to come up from the coast with supplies of all kinds, and the product from the smelter had to be taken to the docks. So, it was a two-way single main, and reasonably involved.
selector The engineering is thusly: Your landings, so to speak, where the train reverses direction for a subsequent grade up or down, should be as close to level as possible.
The engineering is thusly: Your landings, so to speak, where the train reverses direction for a subsequent grade up or down, should be as close to level as possible.
I have a different opinion. Prototype railroad engineers (designers and operators) prefer for switchback tails ("landings") to be on a rising grade, say about one-half of the grade of the switchback's, to make operation safer and easier to stop and start the train. (See an earlier post of mine.) Say you have a switchback grade of 4%. Gradually reduce the grade to 2% by the time you've reached a car length from the turnout and extend that grade for the length of the tail. On the other turnout leg continuing upward, the grade would initially be going downgrade by 2% and begin transition from -2% to +4% a car length from the turnout.
fwright ... every single time you want to take a train over the route, you must go through the switchback. Some get bored by the switchback operation very quickly. Others find it relaxing. Without a doubt, a switchback without a bypass will interfere with "sit back and watch 'em run" continuous running.
... every single time you want to take a train over the route, you must go through the switchback. Some get bored by the switchback operation very quickly. Others find it relaxing. Without a doubt, a switchback without a bypass will interfere with "sit back and watch 'em run" continuous running.
On my planned layout, the switchbacks are on the branchline which is point-to-point, so it doesn't interfere with non-interactive train running because non-interaction is impossible on a point-to-point railroad unless one has automatic reversing circuits as might be applied to a trolley line, which this isn't. The switchbacks avoid the problem of the branchline circling the same scenes twice to reach the branchline terminal at the top elevation. If I want to see a train chase its tail, I can just let it run on the mainline oval.
markpierceOn my planned layout, the switchbacks are on the branchline which is point-to-point, so it doesn't interfere with non-interactive train running
Using the following parameters, can someone give me a rough idea of how much wall length I would need to put in a switch back:
HO scale
U shape layout.
One wall available up to 20' long.
2 layers - Lower layer is a mainline operation.
Top layer is a small logging/ore operation.
The 'Interlayer Consist' would be say a Shay/Heisler etc & 5 cars max? (Is this too many cars?)
Top layer is say 20 inches above lower layer.
To get the above example consist from the Lower to Upper layer using a switch back, how much overall horizontal wall length [if looking straight at the switch back at its horizontal centre so that it stretches to left & right of where one is standing], would roughly be needed to put in the switch back and how many 'zig zag' legs, or 'transition layers' would be needed to climb the 20"??
(Or putting it another way, assuming it enters the switch back running left to right on the Lower Layer - it goes in & climbs a bit, then stops & backs up & goes forward again via a point? to a higher level. Then stops & backs up etc??] - How many "Forward/Back cycles would there be, approx to get it up a Layer??
Tanked.
Tanked,
I'm too occupied to do your homework. But figure your max. train is 3 feet and a couple inches long, so the tail length including turnout (presuming you'll be using a no. 4 or 5) plus a car length beyond the turnout frogs, youwill need about 4.5 feet for each switchback tail/turnout section. Twenty feet is plenty of length but several switchbacks will probably be necessary to gain 20 inches elevation. This will depend on the steepness of grade you'll accept (I'd recommend less than 5%), and remember the need for vertical curves. If 4% is your grade on the switchback and 2% is on your tail track and assuming a car length of 6 inches, transitions will be 1 foot for transitioning from 4% to 2% for an average of 3%, and 3 feet transitioning from -2% to +4% at an average of 1%. Now you do the math (but draw a sketch first).
Hi Tanked,
here's a few photos
First the Climax heads down to the switchback turnout. The track levels out as it approaches the turnout
The climax draws to a halt in the level switchback track
Now carefully reversing with brakeman watching the tracks ahead
Heading down to the main line
Overview of the branch. The main line is darkly ballasted and runs thru the same cutting as the branck line. The Switchback turnout is just out of the picture in the distance on the left. I hope that brakeman sees the other caboose at the end of the line, it's going to be hard to stop even this short train on this grade!
Alantrains here's a few photos First the Climax heads down to the switchback turnout. The track levels out as it approaches the turnout
Thank you Alan - very helpful & simpler than I was envisaging.
Thank you markpierce for the math despite your busy schedule & thanks to other posters for their insights