Ok, I know the current, conventional wisdom for laying out industrial areas is to try to have all the turnouts face the same way -- easier on the switch crews. Recently I've been skimming Sanborn maps of various small towns and many of them look right out of a track plan book rather than the way we often talk about them. Perhaps the disparity is due to the location/type of railroading (midwest vs southeast & northeast) or era.
What got me thinking about this was trying to fit my hometown into a layout. No one would believe it on a model, but the main industries were a cotton mill and power plant served by switchback, where the spur to the mill was the lead to the power plant. As i began looking at other small town maps, I found lots of switchbacks, in some cases leading to whole other industrial areas, wyes with industries along the tail and in the middle, steam era towns at the end of branches with no turning capability within miles, and little regard for facing vs trailing point. I've even come across the classic spurs crossing each other. This is in a sample of about six towns spread over SC, NC, VT, & NH.
It seems these were more common than typically discussed. It also seems to me if space is limited, the classic shelf like plans, such as the Timesaver and others, may be a better representation schematically of small town railroading than we thought.
jmbjmbI know the current, conventional wisdom for laying out industrial areas is to try to have all the turnouts face the same way -- easier on the switch crews.
For a real railroad or for a model railroad ?
jmbjmbAs i began looking at other small town maps, I found lots of switchbacks, in some cases leading to whole other industrial areas, wyes with industries along the tail and in the middle, steam era towns at the end of branches with no turning capability within miles, and little regard for facing vs trailing point. I've even come across the classic spurs crossing each other. This is in a sample of about six towns spread over SC, NC, VT, & NH.
Biggest difference is length of spurs. A switchback is not a problem as such if the spur used for switching lead for the switchback move is long enough to pull or set out the necessary number of cars at the industry at the switchback, without having to remove cars which are in the process of being loaded or unloaded at the neighboring industry, and you have some sensible way to get the engine on the right side of the cars for the switchback moves - ie that you have an adequate runaround, or switch cars on facing spurs from a train going in the opposite direction.
jmbjmb the classic shelf like plans, such as the Timesaver and others, may be a better representation schematically of small town railroading than we thought.
The timesaver is a a game. It was constructed to be a game, not to be a realistic representation of a small town. It was quite deliberately designed with too short spurs, so the challenge in the game is to move the empty slot around as quickly as possible.
More on the Timesaver and it's relative the Inglenook on Adrian Wymann's Shunting Puzzles web page: http://www.wymann.info/ShuntingPuzzles/
Smile, Stein
From my studies, the vast majority of industrial switchbacks function as switching leads to trailing point turnouts that serve industries. Most any industry on the switchback/lead is sufficiently distant to not interfere with switching moves.
Play games or play railroad. You have a choice.
You've seen my Aeolis trackplan, so you know where I stand.
Mark
As John Armstrong said in one of his design books (and as the accompanying illustrations proved) in the loose-car pre-intermodal world a railroad would do whatever was necessary to put a track next to a customer's loading door. Tracks in the street, spurs switchbacked off other spurs, a track diverging right, looping around to cross the lead at a sharp angle and disappearing into an alley... Anything to get the loads on the balance sheet and the empties where they were needed.
It is often hard to realize just how BIG even a small rail-served industrial development is. Here in the northeast corner of Sin City, one fairly straightforward complex would be about forty feet long if modeled in HO. And yes, there are switches facing in both directions, as well as a long runaround and a switchback.
On the opposite side of the UP (LA&SL) main, there's a classic 'tree' switching lead, built within the present decade (rails dated 2000.) Every turnout faces in the same direction, points toward the mainline connection. Probably dead easy to switch. As a model, it would be as boring as cold oatmeal...
Chuck (Modeling Central Japan in September, 1964)
tomikawaTTAs John Armstrong said in one of his design books (and as the accompanying illustrations proved) in the loose-car pre-intermodal world a railroad would do whatever was necessary to put a track next to a customer's loading door. Tracks in the street, spurs switchbacked off other spurs, a track diverging right, looping around to cross the lead at a sharp angle and disappearing into an alley... Anything to get the loads on the balance sheet and the empties where they were needed. It is often hard to realize just how BIG even a small rail-served industrial development is. Here in the northeast corner of Sin City, one fairly straightforward complex would be about forty feet long if modeled in HO. And yes, there are switches facing in both directions, as well as a long runaround and a switchback.
Chuck makes a couple of excellent points. Of course there were plenty of examples of pretty convoluted track work.
But of of the classic bad design for model railroads is where people create a situation where you have one industry on a switchback spur and another industry on the already way too short switching lead, because people feel that they cannot "waste" the space on the switching lead - that there "should" be an industry there too.
Like this:
Problems here are: 1) To switch industry B, you need the engine on the left of the car or cars, 2) To get anything in or out of industry A, you will have to make repeated shuttle moves, since the switching lead is too short. Even if you first remove the car from industry B, so you can move two cars at a time, it will be tiresome to get stuff from industry A or to industry A.
A better configuration here would maybe be something like this:
You still would have to remove cars from industry B to get at Industry A, but at least that is not just a tiresome repetition of a movement, but a normal switching puzzle.
You could also maybe do something like this:
There are quite a few ways to organize industries on switchbacks on a model railroad in such a way that you don't have to shuttle one and one car in or out through a ridiculously short switchback lead with an industry on the switchback lead that has to be emptied before you can switch the trailing industries.
As for the classic "one industry in each direction" type of situation, there is also quite a few different ways of doing it, including e.g. these:
A is where you switch the trailing industry (to the left) from a train going one way (right), and the facing industry (to the right) by another train going left, or the same train returning leftwards after running around it's cars somewhere else.
B is a variant of the same - you can get the industries closer together, at the expense of increasing the distance between the main and the spurs a little to get in a crossover.
C is having the industries branch off a double ended siding, with minimum enough room on the main that you can run around as many cars as you may need to pull from or push into the facing industry (on the right).
D is a variant - taking advantage of the fact that what you usually run out of first in a model railroad is length, not width.
Lots of other ways to do it too.
It certainly is possible to switchbacks and crossovers and all kinds of fun. Just try to avoid the classical "short switchback lead with industry on it" situation :-)
A lot depends on what your goal is. Since most of us don't have the space needed, we do selective compression. To fit in as much as we want, our industries usually have to be closer together and tracks shorter. As a result our switching is also "compressed". But as long as the scene is pleasing and the switching is interesting and challenging but not frustrating, then I think you're okay.
My second layout which was 6x6.5' in HO incorporated a timesaver arrangement for my main town. For me, it worked out very well.
Enjoy
Paul
tomikawaTTOn the opposite side of the UP (LA&SL) main, there's a classic 'tree' switching lead, built within the present decade (rails dated 2000.) Every turnout faces in the same direction, points toward the mainline connection. Probably dead easy to switch. As a model, it would be as boring as cold oatmeal...
Rail dates don't really tell you when the track was built. A lot of the transcontinenlal mainline has rail dated in the 2000's but was built in then 1860's.
An old track can be relaid with newer rail, and a newer track can be laid with 2nd hand older rail.
Dave H.
Dave H. Painted side goes up. My website : wnbranch.com
hi all,
And you can design a switchback in a way you can have it all. A long spur is going to a fruit (apples) packer. Long enough to be used as a lead for two short spurs facing the other direction.
Next to the fruitpacker is a cold storage building for apple juice, made in the harvest season.
Off season only two boxes are in front of the complex, being loaded with juice. The switch crew enjoys the proper length of the lead.
In season the track is packed with reefers however, waiting being loaded with fresh fruit; changing the area in a switchman's nightmare. If a track is nearby, where the reefers can be parked temporarely, it would help. An other industry closed its gates last year (1956), so that spur is empty. It will take the local some extra moves, but keeps switching doable.
Depending on your mood, or your crew, you can set your date off season, or in 1955 when the industry was still in full operation or in '57 for less hectic.
You can play with the number of cars too; as well as using dedicated bay's.
IRONROOSTERBut as long as the scene is pleasing and the switching is interesting and challenging but not frustrating, then I think you're okay.
So it is not the proper switchback, it is the way we design or use it.
The other Paul
The difference between the prototype's use of complex trackwork and the model's often boils down to the length of tracks and the frequency of service.
The overused model railroading gimmick of industries on both "wings" of the switchback is unrealistic because one industry must be emptied of cars so that the other may be switched. In real life, freight cars aren't unloaded or loaded instantaneously. Moving a partially-loaded car to switch an adjoining unrelated industry would be an unwelcome hassle for the railroad and the industry alike. And these can even become tedious on the model.
The use of "sure spots" and other engaging operating elements adds just as much interest without detracting from realism.
A common related model railroad error is setting up a switch lead so short that only one locomotive and car may be handled at a time for an industry spur that holds several cars. A real railroad would never set up spurs that way.
One of the things to recognize also is that activity on a model railroad is much more intense than on the real thing in many places. Some real-life industries might have only received one car per week (or per month). An awkward placement of that industry along a wye wouldn't have been as much of a problem for the real railroad, since it wasn't dealt with very often.
But on the model, we want to switch industries much more often than every seventh (or 31st) session. When you consider the frequency that we switch our modeled industries, it makes sense that a real railroad would choose efficiency over artificially increasing the number of moves.
It's possible to find real-life exceptions somewhere to every best practice. But there's a reason that they are the rare exceptions and not the norm. If you want puzzle value, add the tricky trackwork. But you may find that it grows tedious over time. If you want realism, think like the prototype -- and choose "typicality" over quirky.
Layout Design GalleryLayout Design Special Interest Group
One other caution regarding Sanborn maps: they weren't always (or even often) accurate for the configuration of the railroad tracks. Sanborn maps were done for insurance companies to judge the risk to properties beign insured. Railroads were typically self-insured and in the days before Haz-Mat, tracks weren't seen as much of a risk.
So the Sanborn mapmakers often ignored or simplified trackage because it was of little or no concern to their customers. I've had many design situations where comparing Sanborn maps to railroad Valuation maps, track charts, or CLIC/SPINS books showed major differences. Of course, era enters in here, too.
As far as spurs crossing, that was not uncommon in congested areas, since it put no extra demands on the crews and didn't require the maintenance of the moving switch elements.
Remember also that real-life crews (at least in the steam era) could "kick", "pole", and "dutch drop" cars into spurs. This isn't practical on the model (in most scales), so we sometimes need more runarounds than would the real-life railroad.
Even today, one can observe real-life crews using a convenient runaround to set up cars on both ends of the locomotive in order to be ready to switch an industrial location miles away.
There's one difference more!
How often does the prototype serve such an industry at a switch back?
And how many cars are switched?
We have at a model railroad much more traffic than the prototype. Usually.
Wolfgang
Pueblo & Salt Lake RR
Come to us http://www.westportterminal.de my videos my blog
It is common for the switchback track to be the switching lead to serve industry(ies). This is exemplified by the common end-of-branchline cement plant situation of the Southern Pacific. Below is the railroad's SPINS track schematic for the cement plant at Creal. It is at the end of the short (9.2 mile) Oak Creek branch running west of Mojave.
The track from Mojave enters the map at the bottom center. The switchback switching lead is at the extreme left. This plant, however, is a bit unusual by having a separate double switchback, here located at the upper right. It serves the coal unloading facility, and I suspect/guess those switchbacks are on a grade to permit gravity to move coal cars by themselves over the coal dump.
dehusman tomikawaTTOn the opposite side of the UP (LA&SL) main, there's a classic 'tree' switching lead, built within the present decade (rails dated 2000.) Rail dates don't really tell you when the track was built. A lot of the transcontinenlal mainline has rail dated in the 2000's but was built in then 1860's. An old track can be relaid with newer rail, and a newer track can be laid with 2nd hand older rail. Dave H.
tomikawaTTOn the opposite side of the UP (LA&SL) main, there's a classic 'tree' switching lead, built within the present decade (rails dated 2000.)
True, Dave, but there's another piece of data that's pertinent. Ten years ago the whole area was unreclaimed desert. The buildings being served were built about the same time that the rails were being rolled.
In checking Mapquest I found still another interesting anomaly - a switching district that is basically one big reverse loop. The center of the loop is an intermodal yard, and there are a couple of bulk material-handling facilities as well. It's just north of the lead to the 'traditional' switching area.
Anyone who would like to do a satellite exploration of the area can enter I-15 and East Craig Road, North Las Vegas, NV, in their mapping program of choice. Just follow the LA&SL tracks to the northeast and you'll shortly hit the turnouts to the various areas I've mentioned.
EDIT: Just as a matter of information, this is the same route featured in the January MR project layout. Anyone who isn't happy with the artificial industrialization of Caliente can simply substitute the real industrial complexes of North Las Vegas.
Interesting firestorm. Please note, I didn't say the Timesaver wasn't a game, nor did I say it was a foot by foot representation of a location. I did say it was a schematic representation. Take that schematic and stretch it a few feet and it becomes a reasonable representation of a small switching area. For example, many of Dave Barrows domino designs are really schematic representations, but on a large scale for the space he has. Not everyone has 50 feet to do an exact replica of an area, so we compromise. We selectively compress. Yes of course the tail tracks need to be more than one car long. But we don't have to limit ourselves to a one way design. Those of us who are trying to fit something along one wall or in a spare room can give ourselves permission to adapt knowing the prototype did it too, and far more often we would tend to believe.
jmbjmbPlease note, I didn't say the Timesaver wasn't a game, nor did I say it was a foot by foot representation of a location. I did say it was a schematic representation. Take that schematic and stretch it a few feet and it becomes a reasonable representation of a small switching area. <...> Yes of course the tail tracks need to be more than one car long.
<...>
Yes of course the tail tracks need to be more than one car long.
Nobody is saying that you cannot build a timesaver game into your layout if you want to - of course you can.
I think one of our regular posters (Lee - wm3798) with a very nice N scale Western Maryland layout has a pretty nice lengthened timesaver built into his layout as industrial trackage. Can be seen in this thread: http://cs.trains.com/trccs/forums/p/140932/1569506.aspx
Likewise, nobody is saying that you cannot (or should not) have facing spurs or switchbacks on a switching layout or in a switching area of a town, or that all industries must be on trailing spurs.
Both facing spurs and switchbacks of course are perfectly fine to use. As with any other design element, they probably are best used in moderation, with forethought.
Btw - I'd recommend trying to go a little easier on words and phrases like "conventional wisdom" and "firestorm" - when you use charged words like that (or like Mark's "play games or play railroad"), it tends to make people annoyed instead of supporting an exchange of ideas.
(edited)
Smile,Stein
I like the discussion.
My take on this is that yes, the prototype does have switchback industries, but those track arrangements should be studied & researched more closely, as the prototype may have the "switching puzzle" inefficiencies worked out in some way. Keeping those inefficiencies in the layout is up to the layout designer, of course.
When commenting on someone's plan and I see a switchback industry arrangement, I've come to call it out as "make sure you understand the switching moves you're gonna get" as opposed to "thou shalt not have a switchback industry".
And as far as Armstrong goes, I don't have a copy of the book in front of me so I can't give an exact quote, but I do remember reading his suggestion to make all turnounts in a switching area face in the same direction except one. He said something like - you want switching to be easy, but not too easy.
Of course, if you extend the track lengths, is it really a Timesaver anymore? Part of what makes a Timesaver to unrealistically difficult to switch (and an ideal game paltform) are the restricted siding lengths. If you lengthen the tracks so you can run around 6 cars at a time instead of 1, you're making it lean more towards a prototype switching area than a limited puzzle.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Perhaps we're getting caught up in the name "Timesaver" and not the concept. I picked "Timesaver" since everyone is familar with the schematic in use and it often gets used as an example in these type discussions. And I wasn't speaking specifically about it, only using it as a general example of the concept. The point is far too often we get caught up in discussions that become almost an "always" and "never" type discussion. For newcomers and those will limited space these type discussions can be limiting.
So now I try to spread the concept that it is ok to not follow all the rules; that even the prototype didn't follow all the rules. Rather it did what it had to do in the situation it was in. I'm not saying people should or should not do things a certain way. Only that the alternatives are legitimate examples of prototype operations too.