Thoughts on Bear Creek Yard

Have you considered running the main at the rear of the yard?  1 to 1 hands interfering with a run through freight or passenger train.  It could also add interesting possibilities as locomotives go to and from the engine terminal to the A/D tracks

Don

Small?

Good article with lots of thoghts when designing a yard. However - is the title really correct? Building a layout in N-scale measuring about 9' x 9' i thought a "small" division point might be something for me. Are all of you guys building so large layouts that you consider this division point to be small?

The reference to a "small" yard

I think the reference to a "small" yard is from the perspective of the prototype, not the model. Charlie's South Jackson yard is a small yard in prototype terms.

Everything we build is small!

Probably. But if you used that kind of argument, all yards, cities and sceneries we model should be described as small. Rod Stewart's layout isn't described as a layout with a small downtown area in MR, and I consider that correct.

Have you considered running

Don,

Thanks for your comment. I thought about running the main at the rear of the layout. However, as you point out, motive power would need to cross the main(s) to go between the A/D tracks and the service facility. On the model I was concerned that the amount of traffic through the yard would create delays if lite engine moves were blocking the main.

Also, in South Jackson yard, the main in front has proved to work well. Crews don't seem to snag equipment with sleeves and the better access makes it easier for the yard and road crews to interact with each other. So I don't think this will be much of a problem.

Charlie

Small?

Hi eriwe,

Sorry for any confusion the title caused. Yes, this is a very large yard for a home layout. But for a division point yard it's quite compact. A prototype divison yard is often huge and couild easily have doiuble or triple (or more) the classification tracks and the tracks could be twice or three times longer than they are in the Bear Creek design.

I guess the name seemed appropriate when I was writing the column, but looking at it from your point of view I can see how it might seem misleading.

However, I think the issues I addressed still apply even to a yard on a much more compact layout where there might be a main and 3 tracks.

Best regards,

Charlie

Kind of small!

Thanks for your reply, Charlie!

When reading my own comment, I realise that I sound a bit upset... I was not, and most of all I enjoyed reading the very thoroughly written article with all aspects you possibly need for the design. Size is always a matter, unfortunately. Your yard is larger than many model railroad yards that are made to depict much larger prototypes. Since you have choosen a smaller prototype, the compression doesn't have to be that cruel.

Thanks for writing in MRH!

Best

Erik Wejryd, Linkoping, Sweden

General comments on overall BC&SI track plan

Is this the forum where we can ask about the general layout? Or make observations? Is there some back article that lays out the track plan's purpose and conditions for each area?

I was wondering if the section between Siskiyou Jct and Albany Jct (past the main yard) is intended to be run as double-track main. If so, which orientation (LH or RH running)?

I've toyed with the idea of

I've toyed with the idea of making the track between Bear Creek yard and Browning double track main. Whether or not to do so will be an interesting question.  The final decision will depend on the actual traffic that passes through Bear Creek, Junction City, and Browning. It's really hard to exactly predict op session dynamics in advance of actually having op sessions! It's also possible that the aforementioned area of the layout might get signaled as another to expedite traffic though it's pretty unlikely I'll take the step of going to CTC. I generally prefer single track mainlines though. And a benefit of keeping train density a bit lower is not needing as many trains, engines, cars, and cabeese. I've got about 265 freight cars on the layout now (with no peninsula). I've got about 180 car kits waiting to be assembled. I figure I'll likely be needing another 150 cars when the monster is finished. At today's prices that's enought to get my attention!

It will probably be quite a while before the benchwork for the peninsula gets finished and then there's tons of track work to install and wire.  If I follow my previous pattern I'll probably mock up Junction City and Browning with atlas flex and see how they operate before putting down the final ME flex (ME if fragile enough to not recycle terribly well if it's been glued down).

As far as RH or LH running, I'll look in my 1952 SP rule book and see how they did it.

Cheers,

Charlie

Traffic density with yard and Junction City industrial area

Since you have a branch coming off both ends of the "double-track" section that's why I thought it was there.  And with a major industrial area and a yard along the same stretch ... having two-line capacity seemed reasonable.

However, if that does become a busy section of the line you might want to think  about putting long spurs on both sides of the double main, or maybe sidings, that all the industries branch off. This would keep the switcher out of the way while the through trains zip by unimpeded. it would also give a separate runaround for each side of the mains, so the switcher doesn't have to keep crossing back and forth over the mains to position cars the right way. That is a LOT of industry to switch, and switches facing both ways. Real time-killer.

Making two long sidings with industries branching off also makes two more passing tracks in case things get really tight and you need to get 3 trains through there while the switcher works. You have a lot of the track already in place to do so, with room to add the rest, by adding some turnouts.

Since I don't know your traffic density nor your schedule, I don't know if you need more than one mainline access point to/from the west end of the main yard. As it is now, there is only one, so any trains trying to get iinto or out of the yard will have to wait if there's another one doing so.

What is the scale of the grid on the layout drawing?

Traffic Densities

First, the grid is 24" squares. It's a rather big yard for a non-club model.

Traffic density is the real issue isn't it?  Since I'm not modeling a particular prototype, I'm free to set the density to what ever works. A good question is, how much traffic is required on the track(s) from Bear Creek heading west to Junction City, Albany Jct, and Browning to balance out what the yard crews are capable of handling? This leads to other questions:

How long does an experienced yard crew take to deal with a train that terminates at the yard? Blocked and hodge podge numbers (I block trains leaving staging, but cars coming back on locals/turns can be a mess).  Dealing with a train means sticking the caboose on the nearest caboose track, sorting the cars onto the class tracks and mt storage tracks, and sending the motive power over to the engine service area.  I'm hoping that with two crews, one on each end of the yard, plus a hostler, this could be handled in around 15 minutes (wall clock time)  assuming the cars are well blocked, or 20 to 25 minutes if they're not.

How long does an experienced yard crew take to build a departing train on an A/D track and add power and caboose to it?  This should be faster than breaking down a terminating train since there is no classification involved (assuming the YM isn't a moron and the body tracks are kept well classified).  I'm guessing 15 minutes or less (wall clock time).

How long does an experienced yard crew take to swap blocks with a through train?  This will be highly dependent on where the cars in the arriving block go and where the outbound cars come from. Experience on what's built so far in the train room (using South Jackson yard) suggests somewhere between 5 minutes to 20 minutes depending on whether a helper needs to be called and added. Most trains should be in the 10 to 15 minute range.

Next question: How long does it take a train to emerge from staging at Browning or Albany Jct and run to Bear Creek?  Estimating the distance from the staging access hole at Browning to Bear Creek yard at 43 squares x 2' is aproximately 86' or close to 1.5 miles.  At a scale 30 mph this is 3 minutes assuming the train doesn't need to stop. At a scale 20 mph this is almost 5 minutes assuming no stopping.  But is no stopping realistic?

If the yard can handle two trains every 15 minutes - one through and one that originates or terminates at the yard, that suggests a 7.5 minute train spacing. In practice, this is quite agressive for track warrants in dark territory. It's not exactly inconsequential even if everything was under CTC control.  It also means the yard crews will be hopping!

I think it also means that only half of the trips from Browning to Bear Creek will be stop free. Maybe less than half.

This suggests that if I try to run the trackage from Bear Creek to Browning I may be clogging the mainline.  I don't have a string diagram prepared, but that would be the next step and would reveal how many meets and their consequent delays would occur.

Let's try another way of looking at it.

Right now I have two miles of mainline with South Jackson, Mill Bend, Deschute Jct, and Oakhill on it between Salem and Pocatello staging (not including the length of the staging areas).  The Deschutes branch line is about 1 mile in length from Deschutes Jct to but not including Descutes staging passing through the town of Redland.

I run a 4x fast clock for a 12 fast hour session working out to 3 real hours in an 'ideal' session. In this time we run 15 or 16 trains.  Assuming we keep the same trains/hour density, how many trains would run on the 5 mile + 1.5 miles of branch full layout?  Without going for rigorous mathematical proof I'm guessing

6.5 miles / 3 miles = 2.25 * 16 trains is 36 trains! Yow! That's a lot of trains!  How many would pass through, arrive or terminate at Bear Creek?  Perhaps 80% of them or roughly 28 trains. 28 trains in 3 hours is 9 trains per hour passing or interacting with Bear Creek yard. Assuming that 10% of those are passenger trains that don't do much at the yard and another 10% are freights that pass through without any consist changes that drops us down another 5 or 6 trains to 22 trains or 7 trains per hour. Obviously something has to give!

It turns out that while I can run a 15 train session in 12 fast hours with a 4x fast clock, the dispatcher *really* needs to know what they're doing. Issuing inefficient warrants, taking too long to figure which train to move next, or failing to coordinate with the South Jackson YM and sending too many trains into that cramped yard for it to handle all will cause trains to be delayed. Once the delays start, the line up is full/tight enough that it's really hard to catch up.

My crew and I have considered that a 25% reduction in train density reduces the trains through Bear Creek to 5 per hour. Now things are starting to sound more reasonable.  However, as I mentioned before, I haven't done a string diagram for a 'full' session on a 'full' railroad. It might not be possible to have that many trains moving all at once. And the more simultaneous trains, the longer they wait for warrants.  Perhaps another 5% or 10% reduction in train density would be good.

If, as I'm hoping, I make a switch to TT&TO operations then it's likely that train density may still be too high.  I think that only experimentation will tell.

In any event, I think the capacity of the single track main is probably fairly well balanced with the ability of the Bear Creek yard crew to deal with the various trains swirling around and through their yard.

There is another wild card though. The design of the west end of the yard (at the blob) is such that two switch crews could operate there with one crew on the siding and the other taking the switch lead.  I think this would be a desperate measure however as the space next to the ladder is already a bit shrunken and having two switch crews there, plus the ever present hosteler hovering about, plus other train crews either arriving/departing, or passing through the yard.

Best regards,

Charlie

Train frequency, type, consist make-up, and direction

I think I followed your explanation to the end result of 4 trains an hour entering and being worked at the main yard. That's a good first step in planning. However, as  you said, the specifics of each train are really the determinants of how much time it will take to receive, service and dispatch them.

Yes, returning locals are a mess when it comes to blocking. Plus also the wrong cars are sometimes picked up or not delivered at all, which entails special handling. Sorting large inbound locals is a huge time-eater for the YM, Looks like you would have only two locals, plus maybe a interchange run. If you can schedule the locals 12 hours apart you'll be okay.

Will the 4 trains coming into the yard in that hour be alternating, east-west, east-west? Or, like 3 eastbounds and one west?  Nothing like having one YM sitting  around drinking pop while the other one is tearing his hair out at the jam-up he's experiencing.

I'm assuming the  freights just come in, drop off their loco, division cars and caboose, have the appropriate outbounds tacked on the end, and off they go?  If the inbound track isn't all tied up and if the yard switcher is free a five-minute (real) time allotment will probably work, if your hostler is on the ball.  And if hopefully all the outbounds are grouped and coupled together on one yard track. If, if, if .... each variable adds to the possibility of a delay. Is the switcher to run at prototype switching  speed while doing all this work? You say it will probably take 15 minutes average to work a through freight. And four trains an hour. That means about a half hour or so (per YM, if the arrivals are equally divided) every hour will be tied up working the through freights. Hey, who's doing the car classification, switching yard service tracks, shuffling the empties around? And don't forget a couple of times per session somebody is going to have to move the accumulated day's cabooses from the yard east end to the west and vice versa (provided you are using captive division cabeese).

So ask yourself, do you want train density or do you want leisurely railroading? Do you want long times of heavy-duty yard work but not clogged mainlines? Do you want lots of trains coming and going through the layout at any one time?  More density means multiple mains. You can set up switching of inbound freights to be complicated with many different destinations for the cars, making a long switching job for each train, but then the yard can only handle one or two trains an hour. (unless its a huge huge yard).

What I've found is it's best to design the layout to fit exactly what you want to do, instead of designing a  layout first and  then trying to make your operation fit the track plan.  Which is what usually people do.

What I want to do with the 'Bare Creek'

OIther than 'have fun with trains' (I think I've heard that before somewhere), here's what the layout was built for (and the trackplan was introduced in MRP 2004).

1) Provide local switching operation

2) Provide mainline running with meaningful mainlines.

3) Avoid loco in town A with caboose in town C situations (at least mostly)

4) Provide helper operation

5) Add the extra spice (complexity) of branchlines (there are three branches - Deschutes, the longest with about 60' of track, Siskiyou, with about 30' of track, and Albany with only a Jct. Track lenghts don't include staging or staging leads). Branches keep dispatchers on their toes and give YMs more to think about.

6) Branch yard - South Jackson will handle the melding and de-melding of traffic up and down the Deschutes branch. Redland on the branch has roughly 45 car spots. South Jackson will also handle local service to Mill Bend and possible Oakhill (Oakhill is currently handled from South Jackson, but there is not other yard - yet)

7) Interesting dispatching. Enough trains coming and going, with the potential of getting in each others way, that the DS needs to do some thinking to keep the railroad running efficiently.  This is certainly the case now.

8) Yard ops - I like yard ops. Currently little South Jackson yard sorts blocks for (long) Pocatello, Salem and the Deschutes branchline.  Two locals run out of South Jackson - Mill Bend turn and Oakhill turn. A Redland transfer exchanges cars between the yard and the Redland local switch crew (in the land downunder on the branch). There are a number of haulers and express trains running. There's a reefer express which doesn't do any direct work in the yard (other than require the YM to get the mainline clear for it to pass through unimpeded).  Short blocks include Mill Bend, Oakhill, and South Jackson. The yard(s) will receive trains from arbitrary (depending on the train line-up variant in use) directions during an ops 'day'. I'd prefer to avoid 'fleeting'.

The goal with Bear Creek is to roughly double the sort destinations and provide the physical plant to deal with it. Hence switch crews at both ends of the yard.

9) Crews at town B can't see town A or town C from their current location making 'cheating' (running on smoke signals) a bit more difficult.

10) Provide a transportation network allowing industries to get the raw materials they need and ship cargos in a reasonably prompt manner.This will be accomplished with CC&WB (at least for now).

11) Dark operation on a single track mainline with opposing trains. I like train interaction, and this gives the most interaction. It also gives the crews more to do than follow a series of signals with the DS throwing all their turnouts for them before they arrive.

12) Model crew positions. Engineers walk next to their locomotives. In addition to driving the train they also handle front end brackeman duties. They should NOT be running ahead as if they had 2000' long arms and eyeballs on stalks. Real crews don't throw switches until they come to them.  Conductors should walk next to their caboose, keeping an eye on the train for derailments. They also handle rear brakeman duties.

13) Long enough trains to see real train dynamics. Thirty 40-foot HO cars seems to fulfill this. The normal max train length will be 30 cars.  There may be exceptions from time to time (main staging will have some tracks able to hold 50+ car trains - but I don't want to buy / build enough cars for all trains to be that length).  Let's see the slack run in out as trans start or reverse.  A 30 car train with locs and caboose runs around 17 feet long. Sidings are able to handle such trains. Some sidings are longer. THe DS will need to plan meets carefully if longer trains are running.

14) Eventual move to TT&TO operation.  More autonomy for the crews (and the chance to have the Superintdent of Nearly Everything come down on them if they mess up).

15) Over all goal - I like having a (mini) horde of operators descend on the layout. I like seeing the interaction(s) between train crews, the YMs and DS.

16) Oh yeah... If anyone hasn't figured it out yet, I really like layout photography. Not an ops thing, but I want decent looking scenery to support my photo-thing.

Will it all work as planned?  Well the track plan has changed a bit from what Don Mitchell and I presented in MRP 2004, but mostly it's the same.

Certainly I'm expecting that ops will be different (from its current state) when the layout is completed (trackwise). If it turns out the yard is underpowered for the trains on the layout, then the trains will need to make fewer interactions with the the yard. I expect it will take at least a year to really tune the ops plan to the layout. And yes, I realize that this is adapting ops to the plant ranther than building the layout to support the ops plan. But then, the prototype needs to operate with the tracks they have (at least until ops get congested enough they build more or relaxed enough they tear up some track).

FWIW...

Charlie

Evolution

Thanks for the info that all the prelim data and track plan were in the 2004 MRP (month?). I was especially wanting to see the hidden and staging tracks.

How long does it take an average train (real time) to run from one main holding ard to the other? This would be straigght through, at whatever applicable speed limits, with no helper service and no stop in any yard. Using the 30-car length you stated. I assume passenger and freight trains would have different times.

Run through times from Salem

Run through times from Salem (west staging) to Pocatello (east staging).

There are roughly 5.2 miles of mainline from when a train comes out of one staging lead to when it goes back to another staging lead.  Track speed is set at 15 mph in the mountains (about 2 miles) and 20 mph elsewhere (yes this is artificially slow for 'flat' trackage but it helps make the runs longer.

Time (visible running) =  8 minutes (2 miles @ 15 mph) + 9 minutes (3 miles @ 20 mph) = 17 minutes.

The non-visible (actually it is visible if you look under the benchwork) is a 2 lap, 40" helix plus about 90' of staging yard tracks under the main peninsula for a train to make a complete circuit.  Using pi * D for circumference one helix lap is about 21'. Lets assume 30 mph for invisible staging track running.  Total staging track length is about   2 x 21' + 90' = 131'   Divide by 60 to convert to scale HO miles gives us roughly 2.2 scale miles.

Time (staging running) = 2.2 scale miles x 0.5 miles per minute (@ 30mph) = close to 5 minutes.

Total run time for a complete circuit of the layout assuming no stops for signals, meets, misthrown turnouts, etc should be about 22 minutes for a freight train. Let's say passenger trains go about 30% faster so total run time from staging to staging should be around 15 minutes for passenger trains (assuming no stops).

Time from Bear Creek yard (main yard) to South Jackson (jct yard):   There are about 2.3 scale miles between these two yards. I'm gonna guess it would take a freight about 12 minutes to leave one yard (from an A/D track) and clear up at the other (get off the main).  It currently takes about 5 minutes for a train to leave the siding and pull into the clear at Oakhill.  From South Jackson to Bear Creek would be about 3 times the distance.

Are these the times you wanted?

Charlie