This begs another question -
With these long trains - how many towns is the train in at one time (is the engine chasing the caboose)?
Like what was stated above the layout must be really large to get 2 or 3 train lengths between towns.
And doing switching (block drops or individual cars) must be real fun if it is in more than one town at a time (you could do your whole train switching at one time 
Bob H - Clarion, PA
50 car freights (maximum), 25 car freights, 1 car freights. It depends on whether we are looking at a staging-to-staging run-through, or a local peddler move.
The time/scale/distance equation can be answered many ways. I use scale speed based on a real-time stopwatch, and how much 1/87 distance a loco travels in a given number of seconds. I consider that the best simulation of scale speed. Through freights move at about 40 mph, passenger trains at 60 or more. However, I do not try to cram the operations of an entire railroad or division (or subdivision) into my layout, so I don’t have to worry about concepts of “scale” time or fast clocks, or time tables, which destroy the illusion for me. I would be more interested in time tables on an outdoor railroad, with thousands of feet of main line; but I am more interested in being the engineer than the conductor or brakeman (and I DO NOT want to be the dispatcher or the yard clerk).
On my 4X10 oval–12 coal cars or 7 box/condolas
I usually run about 50-60 cars around the basement sized museum quality layout that I have. Oops sorry, I must have been dreaming. Usually about 10-12 cars (HO) is as large as I go. Anything more and, in my opinion, it’s just too large for my layout. Smaller trains make me feel as if the layout is bigger, or should I say the distance the train travels seems longer.
I have three main lines, the A line normally has 25 to 30 foot train, mainly 50 box cars and pulled by 4 F-3’s or my Big Boy. B line hauls about 25 cars, its passing spur is all so a commercial siding. C line is the short line that hauls coal cars to K-10 mining. 25 3 bay coal cars is about the max I run on it.
Bench is G shaped 19 X 13 X 9.5 feet.
Cuda Ken
Simply put: length of any train = length of passing siding available on layout. You can’t go wrong with that formula.
Too simplistic…Previous posters have made valid points which I won’t repeat.
One factor left out of this discussion is the scale of the models. Most people perceive that a train is long when it more than fills one’s field of vision. Also, car lengths in the smaller scales are shorter than in larger scales, so more cars are needed, for example, in an N-scale train than in HO-scale. Thus, an N-scale layout requires more to satisfy one’s wish for long-looking trains. Another way of looking at it is that, from a want-a-long-train perspective, an N-scale layout of the same plan (taking about one-fourth the space) will be less satisying than its HO-scale layout equivalent. So, while a 20-25 car train might seem long to one in HO, a 40-50 car train would be needed for one modeling in N. (Please, no remarks from the must-be-a-mile-long-train people.)
Mark
Some people have thought about your question on how to “scale” the layout size. One concept that used to be used often, but isn’t as discussed now days, is the “smile” to measure distance. The “smile” was determined by the ratio of the fast time clock to the lenght of a scale mile in whatever scale you were in. For example, in HO a scale mile is about 60 feet so using a 12:1 fast clock creates a “smile” of five feet. Now two towns 20 feet apart became four “smiles” for the timetable.
For me, I like trains of about 15 cars and two locomotives visually, but my current layout space only allows about five cars and one loco. By modeling a branch, the length is at least realistic.
My layout is being designed for locals consisting of 8-12 cars; through freights of about 25-35 cars; and long coal drags of 35+ cars (perhaps 35-50 cars). I don’t plan for the long coal drags to stop or be switched. They will just run through so will only require a passing track long enough to handle them. The yards will be built to handle the locals and through freights.
Good point about the field of view, Mark.
I see the determining factors (given the same scale) as being:
era to be modelled
size of room available
space for layout in said room, which determines:
length of siding areas available on layout
mainline or branch line representation
I chose to modify No.1 and No.5 to fit with No.2 and No.3. Short trains work better in smaller rooms.
Switching scale from HO to N changes the equation completely, obviously.
I still wouldn’t run trains longer than the longest passing siding, but that’s just me.
Of course, there could be the situation where a trains from staging A travels to a median point, drops a cut or two of cars, then travels to yard B, while other trains will distribute the cars to their respective destinations.
Then the limiting factor would be the yard capacity.
In magazine articles we’ve seen layouts designed to viewed as separate scenes. This can be very effective, and trains don’t need to be as long as to be percieved as realistically lengthy.
My point is that one should design the layout with desired train length and operation style in mind from the outset to get maximum enjoyment out of it. All of these factors affect each other to varying degrees.
Yeah, frequent saw-by maneuvers allowing trains longer than the sidings to pass each other can get old. Still, if one avoids meets where both trains exceed siding length, longer trains can be accommodated easily if train orders give extra-long trains superiority and all normal-size trains taking the siding.
Mark
Again, good point.
Another way to do this depends on the pulling ability of your locos. A switcher may only be able to pull 10 cars. some GP’s only pull about a dozen cars. A pair of GP’s may be able to pull more.
The best haulers tend to be first generation cab units. A set of ABBA EMD F’s or Alco FA’s pull a lot.
A pair of PA’s or E’s will pull a bunch of passenger cars or a nice set of reefers. Having that extra weight in the carbody makes a difference compared to the lighter GP or SD type units. I’d have to compare the wheelbase to see if a GP40 would outpull a GP7.
Having Bullfrog Snot on one axle may be the magic stuff that allows better adhesion of model locos to the rail. The trick is to have something holding the rail & still allows plenty of electrical contact.
Having a bit of the metal frame removed to put in a DCC chip appears to cut pulling ability of newer locos.
My train lengths were determined by two somewhat complementary considerations - complementary since the second drove the first:
Length of passing sidings and staging tracks.
Length of observed trains on the prototype railroads I model.
The practical train length is two locomotives and twenty four-wheel cars, including brake van. If the train includes bogie stock, the length remains the same but the car count goes down accordingly.
Everything on the railroad is sized to accept this standard, from yard lead to classification tracks to lengths of the high platforms used by passengers. If somebody who has visited the Upper Kiso Valley ever sees my (to be) completed layout, he’ll feel right at home.
Chuck (Modeling Central Japan in September, 1964)
Agreed. One must compromise the desire for longer train length with all the other competing factors such as space available, number of towns, radius of curves, steepness of grades, ad nauseum. If these factors are not properly weighed for the individual, the resulting layout will not optimize the owner’s enjoyment.
Looks like my next layout will be bedroom sized. It will be around the walls with a “dreaded” crawl under and viewed/operated from inside so no more than half of the layout will be viewable at a given moment (an advantage). I wanted most mainline trains at least 20 cars long and others about half that to satisfy my train-length “need.” For various reasons, the layout is to consist of a continuous mainline and a point-to-point branchline, with the minimum radius of 30-inches-plus to enable operation of locomotives and rolling stock on hand (up to Yellowstone-sized locomotives and full-length passenger cars).
Here is the current schematic of the track plan:
Accomplishing this required some severe but acceptable compromises. The most serious limitation/compromise was to limit the number of towns to two: one on the mainline, plus the branchline terminus.
The only visible mainline is the town. It is 360-degrees “long” (about 30 feet), with the passing siding about 200-degrees (about 17 feet). The remainder of the mainline consists of hidden st
On my home layout I’m restricted by space, and thus for prototypical operations only run between 6 and 10 cars. On the club’s modular set ups I like to run 30-60 car trains. Another restricting fact is curves and grades. A regular poster here has a layout with 22" curves and 4% grades, and we could not get the train over the pass successfully, even with 3 locos up front, 1 in the middle, and 3 at the rear. I think currently the maximum train length is 9 cars, which is governed by the passing siding.
Chuck, when do we get to see some pictures of the layout or close-ups of your custom trackwork? Please don’t wait for it to be completed.
Mark
39 x 70 and it is HO. I have 75% of the trackwork finished. I like big 4 track mainlines and long big mountains with a lot of rock formations. I will be starting the rockwork soon also. Below is a pic of my club layout. I also run long or longer trains there.
Mike
I model a modern shortline, with a pair of units on every road freight…and my sidings can fit 8 cars and two engines! [:I]