For my N scale layout, I keep adding cars until I think the train is long enough. I really don't know how many cars that is however. It may be six or seven feet long. Six feet will have more 40 foot box cars in it than cylindrical hoppers. I've never given it much thought to tell you the truth. I try to limit the number of locos to two or three. The only time I run four is an ABBA lash up of F units. Passenger train length is a different issue where I think seven or eight coaches is long enough.
-Paul
Works for me! ;-)
BOB H - Clarion, PA
JonMN cmrproducts: why not add a few more cars and have the engine act as its own PUSHER ! ;-) BOB H - Clarion, PA Throw mid train helper in there too. Keeps the kids busy trying to figure what is pushing, pulling, front and back of train
cmrproducts: why not add a few more cars and have the engine act as its own PUSHER ! ;-) BOB H - Clarion, PA
why not add a few more cars and have the engine act as its own PUSHER ! ;-)
Throw mid train helper in there too. Keeps the kids busy trying to figure what is pushing, pulling, front and back of train
cmrproducts why not add a few more cars and have the engine act as its own PUSHER ! ;-) BOB H - Clarion, PA
On my current home layout I have 1000' of mainline.
Now I suppose I should be running 100 car trains from my understanding of the previous posts here only because the more mainline the longer the trains should be! ;-)
I run 12 car locals !
It layout is prototypically based on the CR Lowgrade line and it had a lot of towns and curves.
The main industry in the area was coal so the real trains were in the 100 car range.
Unfortunately running trains that long just just looks stupid as the train would be in 3 or 4 towns at once.
I don't have the luxury of stretching out the distance between towns like I would like BUT
There have been Posts on here about layout owners having aircraft hangers they were going to build a layout in and run prototypical distances between town and the responders began to CRY it would be boring!
Yet this thread gives me the impression that everyone wants LONG trains
Having some 300 feet between towns - would give the Dispatcher time to actually get his orders out before the engineer he was talking to reached the next town!
Now as for switching with a long train - one of my regular operators tries this all of the time - running 3 MP15s and trying to switch the local industries - he keeps derailing the 3 unit engine set and is fustrated that it keeps doing this - YET
He keeps trying each OPs session! - YOU would think he would learn - NOPE !
Or if he is on the mainline with a local - it is a 3 unit set again derailing when backing through complicated turnout arrangements - YET he continues in frustration rerailing the engines - and I come through with a single engine local - switch the industries and am on my way while he is still rerailing his power!
Now if I were modeling a 4 track MAINLINE running instead of a backwoods bridge route and only running from staging to staging then 3 or 4 engines would be no problems as you would never be backing through a bunch of turnouts!
So it is down to what the layout is supposed to represent - a shortline atmosphere or a high speed mainline - would determine the length of trains and the length of the passing sidings.
But only having a few feet between the caboose and the engine - is kind of silly and more on the order of a Christmas Tree layout - why not add a few more cars and have the engine act as its own PUSHER ! ;-)
The question is do you want to model the exception or the rule. I think most of us want to model what is common and what we typically see. The picture featured is probably at power set either making up a train or a movement between yards or something, perhaps a "long distance local".
I think we all do like to model interesting scenarios based on prototype. For example, I enjoy late era private passenger trains in the mid-late 1960's and there was often a good deal of varation from the cookie cutter standard consists. Pool power or power sharing lets you mix in "foreign" road power in trains. But I digress.
Rio Grande. The Action Road - Focus 1977-1983
tatans Short trains are not prototypical, if they are too short , your layout is too small, the first item in staging a layout is size of the layout to train size.
Short trains are not prototypical, if they are too short , your layout is too small, the first item in staging a layout is size of the layout to train size.
This is probably most often true, but rules it seems are made to be broken. This is my personal favorite exception to the rule about train length:
http://www.flickr.com/photos/look4trains/3429592333/
Jim
Visit look4trains.com
As a sidenote, my Santa Fe is three decks in a room 29ft by 33ft. A major part of the line is a branch that was called the Enid dist. between Enid OK and Guthrie. This line was originally a short line that Santa Fe bought. I model the area with a date of 1989. Enid also is/was a major grain terminal port so many unit grain trains go in and out. The distance of the Enid district then was 45 miles between Guthrie and Enid and had 4 four towns with passing tracks.
However, the passing tracks were too short for unit grain trains, so trains moving from Guthrie to Enid were held while the trains from Enid to Guthrie ran, then the Guthrie trains were let go to run to Enid. I built my passing tracks at my models of the towns too short also. Makes an interesting twist in operating sessions. The sidings are used then to store grain cars waiting to be used at various places. Sadly after the BN fiasco, the Enid district was abandoned for the BN line to Tulsa which crosses the old Santa Fe at Perry.
It does make op sessions more interesting.
Bob
shayfan84325 Wouldn’t it be most logical to establish train length like the 1:1 railroads do? In Cache Valley, Utah there is a UP branch that serves about 20 (estimated) industrial facilities. The daily local freight is usually about 5 cars pulled by a small (8 wheel) diesel. The train is made up of the right number of cars to satisfy the orders for that day; they don’t add extra cars just to make it look better On my layout the railroad has 5 customers; to me it makes sense to assemble a train that has a few empty cars of the types that each customer requires, and a few that are loaded with materials that those plants have ordered, then venture around the layout spotting cars and picking up the loaded ones. At the end of the run, the entire train is different from the one that departed, and it is probably slightly longer or shorter. I also run a few log-laden cars - those are more like unit trains; they run from the logging camp to the sawmill, carrying nothing but logs. The length of these is determined by the amount of weight that a small shay can pull up a grade. My passenger trains are determined by ticket sales (when ticket sales are really slow I run the railbus). When it comes to things like passing sidings, aren’t their lengths determined more by terrain and easements than by length of train? It seems to me that a 1:1 railroad would make them as long as possible and then limit the length of trains to fit. On my layout, that’s the approach that I’ve taken – my sidings are as long as I can make fit in the space, then I limit my trains to that length. Interestingly, they are just right, considering the small number of customers that my railroad serves.
Logical. When I build the Madison Railroad, the "ideal" layout, sidings will be handled ina a similar way, for the local train. Which will usually be the train. However, come time for them to move the stored fleet of Autoracks to Detroit, that train will far and away surpass the passing sidings, because it won't be the one stopping, except to leave one end of the siding open for the switch train. UP does something similar. Not all Transcon trains fit into a passing siding. The ones that don't wait for the ones that do to get into the hole. Is the +6000ft train common? Not yet. But UP is fascinated by the concept. For now, they play with short trains cause that's their Given, and work until the infrastructure will support their Druther: the ever longer train.
-Morgan
wm3798 But Flashwave, if you're working in a fairly typical model railroad environment, say a 12 x 14 room (which is pretty generous if you live with other human beings) your 16' long train would never have the opportunity to stretch out on a tangent, nor would it ever be able to reach track speed owing to the fact that the engine is passing one town while the rear end is still in the yard. And God forbid you are one of those poor souls struggling to express yourself on a 4x8...
But Flashwave, if you're working in a fairly typical model railroad environment, say a 12 x 14 room (which is pretty generous if you live with other human beings) your 16' long train would never have the opportunity to stretch out on a tangent, nor would it ever be able to reach track speed owing to the fact that the engine is passing one town while the rear end is still in the yard. And God forbid you are one of those poor souls struggling to express yourself on a 4x8...
Right, but even on something pushing 40', there's only a few seconds between stations. Unless you're running a Shay, in which case the Dispatcher will hate you when you hit single track... Not that I've been there... I;ve posted this before, but I actually prefer the 12" view of the layout versus the 12' foot. Yes, that train is in three towns on the layout. Do you see them in the first picture? The second? Granted, it's on the club layout (45'x60') and not the common size home layout, but the princupal is the same. I see not the above, and it's acieveable on any layout of any size. Maybe not 16 on a 4x8, but a 10 is comperable. For long trains, the tail only needs to drop out of sight before the engine re-appears. Headlight glow can be cropped out of video, or ignored. I've even put my chin on the layout just to watch a train go by. Not on someone else's mind you, but the club's for sure. If I'm doing an ops session, I suppose I can see where a 16ft train is a pain, but I don't see me operating with a 16ft train on anything. Trains of those sizes are generally reserved for looks. (See Garry's postabove) At this point, we enter the "run trains to run trains, or run trains to operate" argument. The coal train above was assembled because a freeind and I wanted to see two Big Boys and a Challenger do what they are meant to do. Not because I needed to move 70 hopper cars from the mine to the powerplant. That job is handled by 10-12 car trains on the club layout, for no other reaosn than that's the maximum accepting capacity of the power plant sidings and too many more begins to be problematic for the grade up into the mine. Can I do 70 cars with a Mid Train Challenger on a 4x8? No, but I can probably do 30-40 depending on the trackplan. Yes, that's HO. figuring 6" a car, and 18" for a locomotive. Practical for a 4x8 operation? No. Reasonably long enough to actually look like a Cheyanne-bound Big Boy pulling it's weight, not using a lion to hunt a mouse? Sure is. And even a basic oval, I don't see it kissing it's tail. An over-under could even go 45 cars. Which is why I want to do the Ringling Circus Train, and the full American Freedom Train. They run for looks, very rarely will I be breaking them down for shows. One, because there isn't a yard local to me big enough to handle it, two because that's just not what they are for.
wm3798 But Flashwave, if you're working in a fairly typical model railroad environment, say a 12 x 14 room (which is pretty generous if you live with other human beings) your 16' long train would never have the opportunity to stretch out on a tangent, nor would it ever be able to reach track speed owing to the fact that the engine is passing one town while the rear end is still in the yard. 7 ..... it's no crime to abbreviate the length of your trains. Lee
But Flashwave, if you're working in a fairly typical model railroad environment, say a 12 x 14 room (which is pretty generous if you live with other human beings) your 16' long train would never have the opportunity to stretch out on a tangent, nor would it ever be able to reach track speed owing to the fact that the engine is passing one town while the rear end is still in the yard. 7
.....
it's no crime to abbreviate the length of your trains.
Lee
My last layout was 14 x 24' and it had minimum passing length of 18' feet long trains designed in. That included a hidden 10 track staging yard with storage capacity between 18-24', one siding of 18' and a yard with 18-20' passing capacity. There wasn't a lot of open single track line but one stretch between the yard and the siding, and a small stretch after. Not necessarly a lot of room to stretch legs but certainly capacity to run stage and run 8 to 10 long trains.
No one has to be ashamed of a short train. Whatever you like to run, and if you enjoy it, great!
Personally, I think that the Length of the train should be somewhere near half the average length of all the passing spurs. Hope it helps!
Disclaimer: I AM A TEENAGER!!!
Me
I have either "operating sessions" or "running sessions".
Running sessions allow a variety of train lengths and various equipment.
Operating sessions are intended to simulate a real railroad transportation company. Freight moving from industries in city A must be delivered to other industries in City B.
A road train stops in CIty A and sets off cars for its industries and picks up cars for City B and beyond. Prior to arrival in CIty A, a local train has switched cars at industries in town. After the road train departs City A, the local train is busy delivering inbound cars.
As the road train moves into City B, similar work is performed with its local train.
My branch line has short trains which typically have only two boxcars.
The local trains are small trains, and the road train is much longer. The actual length depends on the work to be performed. However, this is a model railroad, and train sizes are restricted to manageable sizes.
My passenger trains tend to be longer in my running sessions than in the operating sessions. I will run a transcontinental train during a running session. I will run a smaller local passenger train in operating session, and I will assume it connects with transcontinental trains in the terminating city.
GARRY
HEARTLAND DIVISION, CB&Q RR
EVERYWHERE LOST; WE HUSTLE OUR CABOOSE FOR YOU
There's no question that train length in the real world is predicated on the business of the thing, but us modelers are generally striving for a less literal interpretation. Just as our yards have to be smaller, our industries more compact, and our towns closer together, it's no crime to abbreviate the length of your trains.
Route of the Alpha Jets www.wmrywesternlines.net
Train length is determined by the needs of the customer and the railroads.
On my layout, the customers receive between 2-6 cars inbound and 1-3 cars outbound on a bi-weekly schedule (The railroad operates on Monday and Thursday).That means 1-3 cars inbound and 1-2 cars outbound for the day that it's operated.
It's a former branch line with interchange only on one end. That said, It's a ludicrous statement to say that a train with less than 16 cars is not prototypical. The Baltimore and Annapolis Railroad often ran their trains with just an engine and a box car (a box car load of paper for a newspaper plant) on an as needed basis.
http://www.rrpicturearchives.net/showPicture.aspx?id=1967702
The beauty of railroading, both real and model is the rainbow of operations and equipment. That said, if you don't like short trains, don't run them or look at them. If you don't like long trains, don't run them or look at them. Freedom is beautiful baby.
__________________________________________________________________
Mike Kieran
Port Able Railway
I just do what the majority of the voices in my head vote on.
Wouldn’t it be most logical to establish train length like the 1:1 railroads do? In Cache Valley, Utah there is a UP branch that serves about 20 (estimated) industrial facilities. The daily local freight is usually about 5 cars pulled by a small (8 wheel) diesel. The train is made up of the right number of cars to satisfy the orders for that day; they don’t add extra cars just to make it look better
On my layout the railroad has 5 customers; to me it makes sense to assemble a train that has a few empty cars of the types that each customer requires, and a few that are loaded with materials that those plants have ordered, then venture around the layout spotting cars and picking up the loaded ones. At the end of the run, the entire train is different from the one that departed, and it is probably slightly longer or shorter.
I also run a few log-laden cars - those are more like unit trains; they run from the logging camp to the sawmill, carrying nothing but logs. The length of these is determined by the amount of weight that a small shay can pull up a grade. My passenger trains are determined by ticket sales (when ticket sales are really slow I run the railbus).
When it comes to things like passing sidings, aren’t their lengths determined more by terrain and easements than by length of train? It seems to me that a 1:1 railroad would make them as long as possible and then limit the length of trains to fit. On my layout, that’s the approach that I’ve taken – my sidings are as long as I can make fit in the space, then I limit my trains to that length. Interestingly, they are just right, considering the small number of customers that my railroad serves.
Phil, I'm not a rocket scientist; they are my students.
The length of my freight trains is dictated by the length of my sidings.
The length of my passenger trains is dictated by the length of my station platforms.
Rich
Alton Junction
wm3798 sandusky: BATMAN: sandusky: I decided one day that I didn't think you could really get the feel of mainline operations with trains of less than 16 feet in length (loco/locos and caboose not included). I use this for both HO and O scale 2r. As I don't have a layout up yet, I've yet to test this theory out. Mike Okay I'll bite. What's the thinking behind the theory? I think that it's long enough to have sense of beginning, middle and end; that is, the entire train can't be seen at once. Of course, having a bunch of visual barriers would help, but I think I want to have a similar experience to watching a train pass at a grade crossing. It's also a reaction to all the Lionel sets that came with 4-5 cars. MS Brent Really? Train length of 16 feet? And you don't have a layout yet? Good luck with that. I agree that longer trains are fun to watch, but the practical reality most of us face is that there's only so much space to work with. I could theoretically run a train of 16' on my N scale layout, but it would look stupid doing it. It would also quickly succumb to the laws of gravity and other physics, string-lining and dumping cars all over the ground. Rather than assign an arbitrary train length, I think it's important to work with the idea of proportion. But then again, since I'm working in N scale, I can easily run a train that's about 8 feet long, about the same number of cars as your 16' long HO train... so really, it's another clear demonstration that N scale is superior when it comes to showing a train in a more realistic setting, even in a limited layout space! There, I've run circles round you logically! Lee
sandusky: BATMAN: sandusky: I decided one day that I didn't think you could really get the feel of mainline operations with trains of less than 16 feet in length (loco/locos and caboose not included). I use this for both HO and O scale 2r. As I don't have a layout up yet, I've yet to test this theory out. Mike Okay I'll bite. What's the thinking behind the theory? I think that it's long enough to have sense of beginning, middle and end; that is, the entire train can't be seen at once. Of course, having a bunch of visual barriers would help, but I think I want to have a similar experience to watching a train pass at a grade crossing. It's also a reaction to all the Lionel sets that came with 4-5 cars. MS Brent
BATMAN: sandusky: I decided one day that I didn't think you could really get the feel of mainline operations with trains of less than 16 feet in length (loco/locos and caboose not included). I use this for both HO and O scale 2r. As I don't have a layout up yet, I've yet to test this theory out. Mike Okay I'll bite. What's the thinking behind the theory? I think that it's long enough to have sense of beginning, middle and end; that is, the entire train can't be seen at once. Of course, having a bunch of visual barriers would help, but I think I want to have a similar experience to watching a train pass at a grade crossing. It's also a reaction to all the Lionel sets that came with 4-5 cars. MS Brent
sandusky: I decided one day that I didn't think you could really get the feel of mainline operations with trains of less than 16 feet in length (loco/locos and caboose not included). I use this for both HO and O scale 2r. As I don't have a layout up yet, I've yet to test this theory out. Mike
I decided one day that I didn't think you could really get the feel of mainline operations with trains of less than 16 feet in length (loco/locos and caboose not included). I use this for both HO and O scale 2r. As I don't have a layout up yet, I've yet to test this theory out.
Mike
Okay I'll bite. What's the thinking behind the theory?
I think that it's long enough to have sense of beginning, middle and end; that is, the entire train can't be seen at once. Of course, having a bunch of visual barriers would help, but I think I want to have a similar experience to watching a train pass at a grade crossing. It's also a reaction to all the Lionel sets that came with 4-5 cars.
MS
Brent
Really? Train length of 16 feet? And you don't have a layout yet? Good luck with that. I agree that longer trains are fun to watch, but the practical reality most of us face is that there's only so much space to work with.
I could theoretically run a train of 16' on my N scale layout, but it would look stupid doing it. It would also quickly succumb to the laws of gravity and other physics, string-lining and dumping cars all over the ground.
Rather than assign an arbitrary train length, I think it's important to work with the idea of proportion.
But then again, since I'm working in N scale, I can easily run a train that's about 8 feet long, about the same number of cars as your 16' long HO train... so really, it's another clear demonstration that N scale is superior when it comes to showing a train in a more realistic setting, even in a limited layout space!
There, I've run circles round you logically!
Lee, you're right in N regarding train lengths vs actual feet, but in HO, 16ft is only 15 coaches (one Walthers/Rapido car being 83ft, rounded up to 1 ft long, plus a locomotive), or in the neighborhood of 25-thirty cars, which is not THAT unreasonable for a larger but fairly achieveable layout. Now the $64 question, is a thirty car freight train accurate for your setting. A 15car passenger train seems like it could be large, but I can kill ten cars with an NYC train that never went beyond Indiana borders except to hit CHI and Cinncy Union Stations, on it's OFF season.
And you wanna talk logic? Okay, in HO, I can put more weight and your model into my trains to reduce stringlining, plus a larger motor that can overcome drag more effectively. Also, HO feels more right to me. One objective, and one subjective argument make me winner.
Gold Star, Sandusky... I'm glad someone around here is paying attention!
Now it's time to end this thread, as it's become unnecessarily silly!
tatansShort trains are not prototypical, if they are too short , your layout is too small, the first item in staging a layout is size of the layout to train size.
With (limited) apologies to Monty P.
Oh, intercourse the layout!!!!
sandusky BATMAN: sandusky: I decided one day that I didn't think you could really get the feel of mainline operations with trains of less than 16 feet in length (loco/locos and caboose not included). I use this for both HO and O scale 2r. As I don't have a layout up yet, I've yet to test this theory out. Mike Okay I'll bite. What's the thinking behind the theory? I think that it's long enough to have sense of beginning, middle and end; that is, the entire train can't be seen at once. Of course, having a bunch of visual barriers would help, but I think I want to have a similar experience to watching a train pass at a grade crossing. It's also a reaction to all the Lionel sets that came with 4-5 cars. MS Brent
gandydancer19 How do you decide how long your average trains should be? So, how do you determine the average length of the trains that you run on your layout?
How do you decide how long your average trains should be?
So, how do you determine the average length of the trains that you run on your layout?
To me, normal longest train length is one of the most important inputs into a layout design. Normal longest train length directly impacts:
In each of these instances, one has to think long and hard about making any of these less than a normal longest train long. There are instances where any of these can be less than the longest train length, but it must be considered whether or not an operational bottleneck has been created by making the track too short. Conversely, the shortest of these in an existing layout is usually going to drive the most common longest train length to avoid bottle-necking operations.
As has been pointed out, planning for longer trains also requires looking at raising the minimum radius to avoid stringlining, and keeping grades reasonable.
As far as prototype goes, I find it very enlightening to study the tonnage ratings for various locomotives, and the normal trains that were operated over segments of the narrow gauge D&RG and Northwest Pacific. Very seldom did I see more than 18 car trains as possible or practical with the locomotives on hand in 1900 - and often trains that long would be double-headed.
In model railroading, switching more than a 10 car train tends to be more time-consuming and less reliable than one might desires. Running around a 20+ car train for a switching move at scale speeds might be prototypical, but can be a little boring operating on a model layout.
With the very small layouts I have been blessed with, there are a couple of other rules of thumbs to apply. On an oval on a rectangular layout, passing sidings and train length generally work best when limited to the remaining long side distance after the turnback curves are subtracted. In other words, on a 4x8 with 18" radius curves, the longest useful train is 96"-40"=56". In transition era HO, that means 7 cars, caboose, and smallish engine. Even then, the train will probably be too big to fit on the passing sidings. But the siding can be used for runaround moves.
Using the turnback curves as passing sidings or runarounds on 4x8s or smaller is problemmatic for switching. Uncoupling and coupling on 18" radius curves with any common method just isn't all that reliable, in my experience. But there are plenty of track plans drawn that way!
On a shelf layout, Iain Rice postulates maximum train length should be between 1/4 and 1/3 the length of the shelf. I find keeping to 1/4 or less of the shelf length to be much more satisfying, visually and operationally. The 1/3 shelf length feels more like a switching puzzle.
These 2 rules of thumb can lead to very short trains. On my HOn3 logging line, which is a U shelf layout (total 20ft of shelf divided into 4 scenes), the switchback tails are 20" long. This is 3 cars plus the engine, as long as the engine is a tank or short geared lokie. Which ties in well with the 6% average grade. The sawmill scene has 2 staging tracks which are 2 cars each, to serve as a staged version of "loads in, empties out". The log landing and log dump are similarly set up for 2 car cuts. The marshalling area and runaround track for the sawmill is also based on handling a train with 2 car cuts. The lumber loading dock can hold 5 cars, but the lumber schooner is only 90ft. Again, a 2 car cut can be handled at the dock and load simultaneously. Result is I get a 5 car train - 2 cuts plus caboose - on the main. This has to be doubled to use the switchback down to or up from the dock.
Bottom line is an operational scheme, train lengths, and track lengths that all work together.
just my thoughts and choices
Fred W
I would be willing to bet that my short trains are more prototypical than your long trains.
Please tell us what era and train size you are running on your layout.
Dave H. Painted side goes up. My website : wnbranch.com
My trains tend to fall between a protoypical consist, and whether a fair number of engines can pull it. Case in point, I;m modelling the James Whitcomb RIley circe 53, but with only 4 coaches. the RIley can go up to 8, plus a diner, through sleeper from the Southern, a combine, and an observation, as well as a fair mix of Head-end. But, the pics I;ve seen have been a signle Hudson, whixh is happy enough with the train it's got. Now, I know RIley did doublehead with some frequency, so someday I just might.
Otherwise, I tend to veiw my trains from a pictguresque angle. Usually, from a modelled street or overpass, so I cannot see the whole train to realize it fills the entire layout. Rarely do I find myself adiring the layout from an overview. I may see it that way a lot, but then I'm usually not watching the trains themselves.
That being said, most of my modle layout will be shortline trains. I've grown their buisness from one train to up to three, but that's on a good day. Per a 1954 PRR timetable, CMPA trains operating the Madison Incline may be no longer than 12 cars coming UP, or 10 cars coming DOWN. Reasoning, the downhill grade is steep enough that any more is pushing the braking capacity of the lead engine, and besides that, any more cars i just stupid on a 5.89% grade. Ironically, the Madison Hill has another rule. A MINIMUM length of two cars to assure proper braking capability fo the train. Upgrade, train length wil be whatever gets the job done. Most of the time, it won't peak past 15, but CMPA will be doing fleet storage work, so it's possible come New Car season, ~25 cars of Autoracks or Grain Hoppers or what-have you will be snaking their way up to the interchange with Chessie to go to work. I fiigure it should be two to three times the length of a normal CMPA train to give the impression of a monster train, at least for a 20mi shortline.
Yes, I'm also a weirdo who wants a large basement so he doesn't have to selectively compress the CMPA as much, and do a staging loop for Chessie trains to interfere with traffic, and take cars away to staging. Chessie traisn through there will likely be in the 20car range.
Doughless I need to find a one-horse short line support group somewhere.
I need to find a one-horse short line support group somewhere.
I'll be there as well. My inspiration was the branch where I grew up. Trains were a single switcher (RS-1 typically) with 3-4 cars and caboose. Went through a period when I had the room and was modeling n-scale where I was running 3-4 SD-40s and 30 car trains. Over time I found that boring and after a move switched back to HO, even though I had less room. My current layout can handle about a 5 car train without appearing overloaded.
For the heck of it, I have put a couple of SDs and a 20 car train on the line just to see what happened. Even without considering that it was too large for my sidings, it visually over powered the layout. It actually made the towns and layout look smaller. Partly I think it's a trick of the mind. My little 5 car train leaves point A and goes to point B. In between it covers some distance. When looking you see open country, the train, then more country so it seems like it's going somewhere. But when I put the larger train on there, instead of looking like it was covering some miles between towns, it looked like I was standing in one spot train watching as a long train rolled by. So I do think that is part of what drives our choices in train lenght -- what is the effect we're after?
News flash: Not everyone has a giant basement to fill with trains. We have to live within our means. The size of the space available constrains the length of train. Then operational considerations and visual appeal on that layout constrain it even more. Every layout is different and what looks good on one won't on another.
When running ops at the club trains can be anywhere from a few cars long to a couple of dozen. If running a local you really don't want the train too long.
Springfield PA
I like to run one of my Articulated's sometimes with a double header, Mike in behind pulling 50-60 empty ore cars.
http://www.trainboard.com/railimages/showgallery.php?cat=500&ppuser=5959
If one could roll back the hands of time... They would be waiting for the next train into the future. A. H. Francey 1921-2007