FORUM CLINIC: Designing for satisfying operations

[jfugate]

Peter:

80% is the absolute maximum ... 60% is probably more like it for a really fluid op session.

The numbers represent limits. You don't have to do what they say, but it gives you some idea of how far you could pu***he design before you would "break it", if you will.

You could think of it as "stress testing" your design. [;)]

[jrbernier]

Joe,

I think your 60%(fluid) number is right on. I have used a similar calculation system(not quite the same as yours), and found if I run at 66% Maximum Capacity, that is the 'sweet spot' I am comfortable with.
I do allow 'seasonal' car orders(and the extra cars needed are added). Somehow the car orders get removed, but the cars hang around and I start using a yard track to store extra cars. At least until I bill out an extra east/west train to move them 'in block' to staging for removal from the system. Just picking them out after the 'grain rush' just creates havoc for a couple of operating sessions after the 'purge'. At one time I put an orange yard sale sticky dot on the extra cars and ran jobs in sequence to move the cars 'off-line' Now I put the dot on the car card and instruct the yardmaster to park the car on the track assigned for car storage.
I suspect 80% is workable, but I like running a low key secondary/branch line operation and the 66% number works just fine.

Jim Bernier

[SilverSpike]

I am a sponge, and you guys are the water!
Keep the ideas flowing, because these are great nuggets of info!
Hey Joe, thanks for the thread!

- Ryan

[jfugate]

Topic this post: More cool things we can learn from the layout stats

Besides the things discussed in the last post, there are some very useful insights you can learn about your design relating to trains and your layout operation.

NUMBER OF TRAINS: We can divide the number of cars moved by our average train length to arrive at the average number of trains we can expect in a typical operating cycle. Average train length is the smaller of average passing train length or average staging train length.

One operating cycle is defined as running the layout in a realistic manner until the trains you run begin to repeat. Ordinarily this will be one "24 hour" day according to the modeled train schedule.

Depending on your fast clock ratio, the experience of your crew, the reliability of your equipment, the length of a typical run, and the level of detail to which you simulate prototype operating practices, the actual time it takes to complete one cycle could vary from one hour to dozens of hours. Three hours is probably a good typical cycle, however.

DISPATCHING THRESHOLD: Compute as (3 x shortest passing siding + 2 x average passing siding + longest passing siding) / 6.

Two opposing trains of this size or larger will tend to create a dispatching bottleneck because they cannot easily pass each other except at select sidings. If you want to ease the dispatcher's workload, keep the typical train length at or under this size.

If you want the dispatcher to more easily manage longer trains, then lengthen your passing sidings. The best way to increase this threshold is to lengthen your SHORTEST passing sidings first. Of course, you need to keep the length of your staging tracks in sync with passing siding lengths as explained above under the train length stats.

Another less obvious tactic to improve this stat (if your passing sidings are smaller than your staging tracks) is to declare very short passing sidings to be switching runaround tracks only (and thus connecting track instead of passing track), thereby removing them from routine consideration as locations where the dispatcher might arrange meets.

This tactic also has the effect of increasing the number of cars moved since it creates more connecting trackage.

This concludes the series of discussions on layout stats. Now on to other design insights ...

Next topic: Easements and superelevation.

[SilverSpike]

Great topic Joe!
You have enlightened my track planning stage of my layout incredibly. As it stands in the current form of the plan I have two passing sidings (double track sections actually) that are at least 1.5 X the length of my longest staging yard track.

Look forward to the next topic!

Ryan [^][:D][^][:D]

[jfugate]

SilverSpike:

Many an old timer on the Layout Design SIG yahoo forum have said that you should get yourself some string and pretend running trains on your track plan. You'll find many of it's shortcomings early if you do that.

The nice thing with the layout analysis stats is they do the same thing, but are far quicker and less awkard than playing with strings on your plan.

By computing the values for your track plan, you are forced to think through many of the issues according to the insights a savvy layout designer will have already considered.

So you kind of get "instant expertise" in track planning by using these formulas.

By the way, all the formulas and more explanation on doing this sort of track plan analysis can be found on my web site at:
http://siskiyou.railfan.net/model/layoutDesign/layout.html

Glad to help out my fellow modelers in any way I can!

[TurboOne]

Great topic ! Enjoyed all the info, and as I haven't added any yards yet, now I know a lot more. The knowledge is great, but have a simple mind. I like watching trains. Wish I had Big Boys layout, but like my 4 x 8 until benchwork for new is completed.

Tim

[SilverSpike]

Thanks for the tips, Joe!

I have your website saved as one of my favorites. I will review the Layout Design pages again.

Thanks again,

Ryan

[conford]

This is an informative and enjoyable topic. I think the statistical models also encourage the designer (me) to think about a certain cohesion or balance in the layout's design. Perhaps it should be obvious that passing siding lengths should be somewhat consistent along the line, or that the lengths of staging tracks will effect the length of the trains you can run, but putting it in a formula has made it make more sense to me.
Has anyone run the numbers on any of the layouts in the popular press besides the ones on Joe's site? It would be fun to see how some of them measure up.
Thanks
Peter
conford

[jfugate]

I ran some numbers on Rick Rideout's L&N since I'm a big fan of his layout. I don't have the results handy, but I recall it was interesting to see that the layout was clearly designed for 20 car trains.

That was not obvious by reading the article in MR years ago, but it was as plain as the nose on your face by seeing the stats.

It would be interesting to see stats on other well known layouts, or various published track plans. Which is the point of the article on my website, but the way. [:D]

[jfugate]

Time to resurrect this forum clinic since several of the recent questions relate to posts on here.

[:D]

[n2mopac]

Yes, Joe, pleas do. I have been very much enjoying following this clinic and was hoping it would continue.

Ron

[jfugate]

Topic this post: Easements and superelevation

The prototype puts easements in between the straight track (called a tangent) and a curve so the straight track slowly "eases" from straight to the radius of the curve. This prevents cars from "jerking" into the curve from the straight track, and makes the journey easier on passengers and freight.

On the model, easements don't matter because the physical forces on the cars do not scale down, so there is little operational benefit from having easements. However, adding easements adds tremendously to the visual authenticity of your track alignment and makes it look more "right" even to the untrained eye.

As to track planning, adding an easement might offset the location of your curve by 1/2" to an inch (in HO, more in the larger scales, less so in the smaller scales). This is not enough difference to even concern yourself with adding easements to your track plan. Even the process of transferring your track plan to the benchwork could introduce this much error.

There are several ways to model easements -- bent stick, curve templates, spline roadbed -- but that's beyond the scope of this discussion. Suffice it to say that if you use spline roadbed like I do, the easements are formed naturally by the roadbed, so I don't even have to worry about how to add them.

The other track alignment topic related to easements that often comes up is "banking" the curves, or superelevation. The prototype superelevates for the reason you expect: it allows you to take a curve at speed and the banking offsets the outward centrifigal force as the car goes around the curve.

Again, on the model, the physical forces do not scale down, so superelevation is unnecessary. However, it's hard to beat the realism of a model train on a realistically superelevated model curve.

The prototype commonly applies a superevlevation transition in the same place as the easement. While the track is easing from straight to curved, the outer rail is also slowly rising to the superelevation angle that will be maintained throughout the curve. And the reverse happens at the other end of the curve in the transition back to the straight tangent.

Modeling superelevation is beyond the scope of this series, but it's worth pointing out a couple of things. Raising the outside rail more than about a scale 6" (1/16" in HO) is plenty. Also, pay attention to superelevation on the main when you have a non-superelevated passing siding. The superelevation actually reduces the distance between the tracks, so you should increase track spacing to compensate.

Otherwise, you might have cars on the superelevated main hitting the tops of cars on the non-superelevated siding (when the siding is on the inside of the curve).

Ask me how I know. [B)]

This completes the topics for this series, however, I'll also entertain direct questions as long as they are layout design related.

[vsmith]

Question? can anyone here recommend a good source (book, magazine,website) for setting up one person operations systems for a small layout?

I had in mind a simpler system like a dice roll setup but do not know where to find the info for it? I think a full blown card system might be a bit too much for my tiny layout.

BTW heres my layout if anyone is interested...
http://1stclass.mylargescale.com/vsmith/GarageStudyDOWNSIZEnew%20studyA.pdf

Thanks for any info. Vic

[jfugate]

Vic:

Here's a great link:

http://www.gatewaynmra.org/designops.htm

This is an excellent overview of all things operation, includes different options on how to route cars (your question) and a thorough bibliography of further reading.

This link contains some of the best model railroad operation tutorial info of anything I've ever seen, in print or online. Richard Schumacher is to be commended for a top notch piece of work.

[jfugate]

This link is also a good one:

http://www.gatewaynmra.org/articles/essence-of-ops.htm

[vsmith]

Thanks, I'll be reading up on those sites.

[SpaceMouse]

I read this topic a couple weeks ago and reread it today. It pretty much took all day.

I am way too new to this hobby. So far I have been designing my layout with a couple things in mind.

1) I want a lot of track and places to go so my son can run his Hogwarts.

2) I want to see my 19th century steamers cross a tressle bridge.

3) I want to see the steams run through the forest, so I'm planning a logging operation.

4) Yards intregue me, but I've never operated one.

5) I've planned a staging yard beneath my track because I know it will increase the tracks capabilities.

6) I know I need operations because I will get tired of running laps no matter how complex the laps will be. But I have never done anything but run laps. My club is threatening to do some ops but since I've been there all we have done is work on the tracks. I have two new BLI locos that have never made a complete circuit at the club. The sit and idle in the passenger yard I am renovating.

That is all to say that this has been very interesting. Time to re-read Armstrong.

[West Coast S]

The concept i'm exploring is a large branch line operation basd upon the SP in S scale with two focal points for all switching as opposed to placing a switching district ever few feet. Open space will not be devoted to yards or engine terminals instead secondary trackage will serve as makeup/break up tracks and for switching.

Three easily accessible stagging yards-one acting as an interchange when operations call for it will provide storage capability. Provisions for one passing siding which will serve a second function as the interchange lead. This design fits well with an around-the-walls,point-to-point design. So far on paper I like it.

[Anonymous]

QUOTE: Originally posted by jfugate

Topic this post: But watch out ... don't make it TOO high!


It's far better to take an evening mocking things up, than to spend money and time building benchwork only to find it's too low or too high!

[Anonymous]

QUOTE: Originally posted by jfugate

Topic this post: But watch out ... don't make it TOO high!



It's far better to take an evening mocking things up, than to spend money and time building benchwork only to find it's too low or too high!

Yes I wish I had forseen this error .I had to remove all of the riser's I had installed on my
multi-leveled "Beaverton & North Plains Rail Transfer"Freelanced Branchline of the Rivergate Industrial Park in Portland Oregon.It would have saved me nearly 200 bucks,and the time fixing the miscalulation.
But now that I have reparied my work I can now revise my Trackplan.With changing the height of the lower level this has changed the way trains travel onto and off the branch,since the Helix is NON-MOVABLE ( I say this because there is no other place to put this Floor Hog!)in refrence to the height of the lower level the ramp to lodeck is extended two and a half feet in to the lower level.Entrierly changing the track plan.

[nobullchitbids]

I come late to this forum, but a few things do not appear to have been said:

1) It is possible for trains to pass when both are longer than the siding (it is called a double saw-by meet); however, if the trians are much longer than the siding, this can get pretty hairy, and even with short overages, one probably would not want very many of them in a session, since they definitely can tie up a line.

2) People looking for selective compression of prototype facilities might want to check out some of the older USGS maps, which state right on them that the trackage shown in terminal areas is only a rough approximation of the real thing. USGS maps can be found in most university and some public libraries.

3) Freight Terminals and Trains, as well as the companion Passenger Terminals and Trains (Droege), remain the prototyper's bibles for foundations of operation as well as prototype design prior to the World War.

4) One thing to remember about prototype operation is that, at 12 inches to the foot, walking becomes a major concern in terms of how things are done. It takes a lot less time to drop off a car cutter, then pick him up, than it does to oblige him to walk the length of the train as much as twice. Many modelers' "operational" moves would make no sense to a real railroader for this reason.

5) Prototype railroads do not deliberately create switching "puzzles" unless no other solution exists. If you are one to like switching puzzles, then you need to be thinking in terms of situations which make these necessary, such as modeling an access-restricted wharf area where e.g. to switch right, it might be necessary to turn out left, then cross back over the normal route to get to a cramped business location. Even in the common situation of more than one business on a siding, the prototype will (if revenue permits) build a runaround to avoid having to pull and respot cars already switched to be loaded.

6) On the question of single location versus branch, I definitely would opt for the branch, even if the branch were no more than a long approach to servicing a single industry (e.g. a cement plant). One of the best track plans ever incorporating this idea was Charles Small's Cerro Azul (shown in MR's 101 Track Plans). If you do not want to use the switchback format, simply horseshoe the lower connection around to reach the upper level. The branch brings loads to the base terminal, where they then are switched into passing road freights coming and going from staging. Such a plan offers the lone wolf really all the action he could want.

7. I respectfully disagree with the assertion that easements have no more than cosmetic value on the model -- anyone who has run long-wheelbase steam, especially backward, knows this is not true. Easements aid tracking for longer equipment, and if the choice were to keep a minimum radius and sacrifice the easement or keep the easement and sacrifice the minimum radius, I definitely would opt for the latter.

[Roger38]

On the former NP southwestern branch at Lisbon, ND, the house track had three grain elevators, an LP gas distributor, and the Standard Oil Company bulk plant, plus it was also used as the team track. I remember seeing 1955 Chevy's being unloaded from an automobile boxcar, and was asked to drive one of them downtown to the dealership. Multiple industries on a siding or spur are prototypical.

[pocovalley]

This forum for operation is a great idea.

Regarding "passing sidings" and their length, this would depend on the design of your layout. One of the members of our Tuesday night round-robin operating group has a loop to loop design (with staging in each loop). Thus, since you usually use all of the passing sidings sometime during the evening, your shortest siding determines the length of train you can run. Also, don't confuse passing sidings, meant for passing or meeting two or more different trains, with runaround tracks, used for way freights or local freights to run around their train to spot a car at a facing point spur. The latter can be used for runaround, but the other way around wouldn't work.

In our group, the host is always the Dispatcher, since he knows the railroad better than anyone. I get to run my railroad in between sessions, sometimes to fini***he evening's schedule if a certain train or trains didn's get run.

Staging tracks: If you are just designing your railroad, put in as many as you can get. You will NEVER have too many. The Poco Valley runs three different sessions, Morning, Afternoon, and Evening. A train might run westbound in the Morning, and not have its eastbound run until Evening. That train sits in one of the staging yards all Afternoon, tying it up, but not moving.

I find the 60% rule works pretty well (cars versus space). I take cars off the layout frequently. They stay on a shelf for a week or so, and this gives each operation more variety.

Poco Valley

[jfugate]

QUOTE: Originally posted by nobullchitbids

7. I respectfully disagree with the assertion that easements have no more than cosmetic value on the model -- anyone who has run long-wheelbase steam, especially backward, knows this is not true. Easements aid tracking for longer equipment, and if the choice were to keep a minimum radius and sacrifice the easement or keep the easement and sacrifice the minimum radius, I definitely would opt for the latter.

True enough ... since I model the 1980s era I keep forgeting about long wheelbase steam locos. I just know for most of the stuff I run in the modern era, an easement doesn't make *that* much difference in how it tracks.

But you're right, so I stand corrected. For longer wheelbase stuff (locos, and 80 foot cars) an easement can help how it tracks. That said, it's also true that an easement does help your track alignment look more prototypical, and that's the reason most modelers use them ... improving how things track is not as frequent a justification for using easements on the model.

[jfugate]

Once again, resurrecting this clinic because there have been some requests for how to find it.

[SpaceMouse]

I am a total newbee so if this seems elementary ....uh, keep it to yourself.

Do you need passing sidings if you have a two track main?

[nobullchitbids]

None less than Union Pacific says, "Yes." Without them, how will your crack limited get around that local hog loping along at 15 per?

U.P. typically uses high-speed "Y's" to create a third track in the middle which can be used in both directions.

[jfugate]

SpaceMouse:

You need places where faster trains can overtake slower trains, where trains can stop and switch, and so on. You put crossovers every so often between the two mains and consider one main to be passing siding and the other to be mainline. The distance between crossovers tells you "passing siding length" as it were.

--Joe

[SpaceMouse]

QUOTE: Originally posted by jfugate

SpaceMouse:

You need places where faster trains can overtake slower trains, where trains can stop and switch, and so on. You put crossovers every so often between the two mains and consider one main to be passing siding and the other to be mainline. The distance between crossovers tells you "passing siding length" as it were.

--Joe

Dang! Back to the drawing board.