FORUM CLINIC: Designing for satisfying operations

[wpsteve]

Andre, When you get a plan please send it up this way and we will look at it for you.. We are very in to operations and we can tell you what you need to or not to do ! If you ask for our advise we will give it ..........just trying to help though
WP Steve

[andrechapelon]

QUOTE: Originally posted by wpsteve

Andre, When you get a plan please send it up this way and we will look at it for you.. We are very in to operations and we can tell you what you need to or not to do ! If you ask for our advise we will give it ..........just trying to help though
WP Steve

Essentially, the track plan will be determined by Sanborn fire insurance maps, with whatever adjusting is required to make things fit. I don't think I'll be in the position of having to eliminate anything of interest. I'll be modelling from the Monterey stariona area to the sand mine at Lake Majella in DelMonte Forest. That's a total distance of 4.3 miles or about 260 ft. in HO. Just off the top of my had, the space I have available gives me about 115 ft of trackage given the way I have it laid out in my head. That's about a 2.5:1 compression ratio and what will be eliminated will be trackage between operationally interesting areas. In essence, Monterey, Cannery Row, and Pacific Grove can be modeled fairly close to full scale.

My God, what am I getting myself into?[:D]

Andre

[jfugate]

Topic this post: Aisle width

With the ability to make the benchwork higher and narrower, you can either cram more railroad into the room, or you can give that room to the people. Based on my operating experiences on other layouts, I would suggest you give that space back to the people.

The more pleasant the operating experience, the more fun the layout is to operate, and decent width aisles are a part of that. So what's a good standard width for an aisle?

I find that 36" is a great width for an aisle on a walkaround layout design. Two people can be operating on opposite sides of the aisle and a 36" aisle feels quite comfortable. I've also seen 48" aisles on a large home layout and that feels absolutely luxurious!

But what if you can't get everything to fit and you have to constrain the aisle under 36"? Ergonomically, what's the narrowest an aisle can be and still allow two people to pass comfortably? One of the LD SIG members did a study of this and they found that 28" is the narrowest you can go and still have most people get past each other without too much interruption.

So my recommendation when planning your walkaround layout for satisfying operations is to start with 36" aisles. If something won't fit, narrow the benchwork first, even 6"-12" wide can work for benchwork with HO single track, especially if the benchwork is chest height or above.

After narrowing the benchwork, you probably will have a few places where things still don't quite fit, so it's okay to create an aisleway choke point that is less than 36" as long as it's length is not more than a few feet. If you keep the choke point at 28", then two people could still pass each other easily there. 18" is considered to be the minimum aisle width for one person.

If creating a choke point of 18" in a place or two gets your "dream plan" to fit into your space, then that's very workable.

Next topic: Have your cake and eat it too ... wider benchwork AND wider aisles.

[Big_Boy_4005]

I couldn't agree more Joe, operator space and comfort are what make a railroad pleasant to run. I have managed to hold all of my aisles to at least 36" or more, with one small exception. There will be about 3 linear feet of 28" aisle, to accomodate wide radius curves, not bad when you consider that the room has well over 100 linear feet of aisle. There are also a number of wide areas at the ends of aisles, where 4 or 5 people could comfortably stand.

The floor space in my aisles is actually 48" minimum, which means there is 6" of toe space on each side. The main level of benchwork is a relatively low 30", but since I am working in O, I wanted good seperation between levels. The upper level is at 60" allowing a clear 24"+ seperation. The upper deck fascia will also form a valance for lower deck lighting.

For the most part I have held to a maximum reach of 48". One thing that I have done is make most sections of the upper benchwork narrower than the lower. This helps remove the "canyon effect", leaving more headspace for opreators to not feel clostrophobic.

Finally, when it comes to fascia, I am a big fan of no sharp corners, where ever possible. Inside corners might as well be rounded in favor of the layout. Outside corners, if not rounded should not form a hazzard like a "bellycatcher".

[wpsteve]

When designing and thinking of isle space, if double deck, please don't stack your towns over each other. [[:(!] Appears almost everyone does it. Bad Bad.
If single deck, try not to put towns across from each other, many do that also. Bad Bad.
When talking about it, everyone will tell you it was the only place to do it.. Not always true, it is the most logical place, not the only place.
You just have to work harder to not create this operational problem.
WP Steve

[jfugate]

All excellent points!

That's one of the reasons I like the mushroom* benchwork configuration. It spreads out the aisle traffic more like a single deck design does.

If you aren't sure what a mushroom design is, here's a cutaway diagram.



*NOTE: Basically, with a mushroom configuration, the two decks face *opposite* directions, unlike the more traditional multideck designs where both decks face the same direction.

[jfugate]

A few more points on the mushroom.

I have found you need a room at least 12 feet wide to have enough width to fit a mushroom design.

The raised floor will cost you some money, since it needs to support people safely. That means you will have to use solid construction and larger lumber members than you use for your benchwork.

A raised floor makes a mushroom convenient because it allows you to keep both decks at a similar relative height from the floor, but a raised floor is not *absolutely necessary* as long as you don't mind the different heights from the floor.

But if you do want a raised floor, the other constraint is ceiling height. You'll need at least a 7 foot ceiling if you want to have headroom for a raised floor. I have a 7'-9" ceiling in my basement, and my raised floor on the Siskiyou Line is 15", for example.

[cwclark]

I built my SP Sanderson yard 50" from the floor (it's the second lowest part of the layout)...it's just right for me, but that's not why i built it so high...I built it that high to control small children in the layout room. That way, when the kids come over, they will have to stand on a step stool in one spot to see the layout and can't go running hog wild throughout the entire room messing with stuff they ought not to be touching...funny how some things fall hand in hand...chuck [:D]

[jfugate]

Topic this post: Have your cake and eat it too ... wider benchwork AND wider aisles.

One of the things to remember when planning your layout is to think in all 3 dimensions. This is especially true when planning a multideck design, but you can also find useful tricks you can accomplish with the height dimension if you will remember you have all three dimensions to work with.

Here's an example to illustrate how you can find space you didn't know you had by remembering that third dimension of height. If you take advantage of the earlier concept I discussed of higher, narrower benchwork, you can find some aisle space you didn't know you had, a true have your cake and eat it too situation.

Take a close look at this photo.

This is the aisle on my HO Siskiyou Line where Rice Hill summit is located. The track at the left is at 65" from the floor. But notice that the bottom edge of the benchwork is 60" from the floor, which is enough to clear most people's shoulder as the photo shows.

So even though the fascia to fascia distance here is 28", the Rice Hill summit benchwork is high enough I was able to offset the supports back and give enough room for a 36" aisle!

I was able to have my cake and eat it too with regards to aisle space in this location!

So remember to think in all three dimensions when planning your layout and you may find space you didn't know you had. [;)]

Next topic: Don't let vertical curves sneak up on you

[n2mopac]

This is so true. I built my Wichita Falls sub with 36" isles. I had one place where I had to narrow to 24" for a distance of 4'. I thought this would not be a problem for such a small space, but this has become a real bottleneck as someone has to step out of this area for someone else to pass. the 36" isles are much more comfortable.

Ron

[jfugate]

Topic this post: Don't let vertical curves sneak up on you

Curve radius and grades are covered a lot in track planning discussions but one gotcha you need to watch out for is the vertical curve.

A vertical curve happens when you change the grade of your track. For example, if you go from level track to a 4% grade, the region of transition is a vertical curve. Less obvious is at the summit of a grade or bottom of a grade -- here the vertical curve potentially may be *double* because you could go, for instance, from 4% up, to level, to 4% down all in a very short distance. The effective grade change in this example is a whopping 8% !

Vertical curves can be a real bugaboo in layout planning and construction. Do it wrong and you will have nothing but grief at that grade-change location. Car break-aparts and derailments will abound at a poor vertical curve transition.

So how do you keep from doing a vertical curve improperly?

The key is to make your grade change gradual.

I use a rule of thumb that a grade change needs at least one car length per percent of grade change. Using the 6" per 40 ft car in HO example, the 4% summit example earlier needs 9 car lengths to do correctly -- which means about 54" (4-1/2 feet). That's 4 car lengths to go from 4% up to level, one car length level, then 4 more car lengths to go from level to 4% down ... or 9 car lengths.

The vertical curves problem is one reason why I like spline roadbed over the homasote and plywood sandwich method. Spline roadbed naturally makes it hard to do tight vertical curves, but you can do the 8% grade transition discussed earlier in a few inches with plywood-homasote!

So be careful, and always allow sufficient distance for vertical curve transitions in your track plan!

Next topic: Checking your track plan operational satisfaction .... before you build it.

[Anonymous]

Joe,

This is the coolest thread I've seen on the forum in quite a while. All you guys reading this: Go to Joe's site if you haven't already done so (click on the link under his name). His section on evaluating your track plan for operation is awesome. I did the computations and am happy to say my plan does what I want according to Joe's formulas.

I was going to post a direct link to that section but it appears Joe is going there next with this thread and that would be stealing the thunder.

Andre, I considered modeling the Monterey branch as well. I'm sure you are aware that Doc Ricketts (of Steinbeck fame) met his fate at a grade crossing on that branch. There is a caboose that has been converted to a museum on the tracks in cannery row. If you haven't visited already, I'm sure thay would have some good info on the operation of the RR back in the day. The canneries along the line along with the bay are very attractive subjects scenerywise....

Thanks again Joe for a great thread.

Guy

[Anonymous]

Long time reader, first time poster. I'm really interested in the next installment. Can I ask for two more topics to be talked about. Easments and super-elevation. These are two things I have no clue how to do. Thanks for all the wonderful posts jo.
Andrew

[jfugate]

Topic this post: Checking your track plan operational satisfaction .... before you build it.

Using a scale ruler and some simple formulas, you can check out your track plan for it's operational abilities before you build it. To make it easier to analyze and compare plans from different scales, I suggest you use 40 foot car lengths for all your measurements. Here's a table that lists all the major scales and how many 40 foot cars will fit in 12":

40' cars in 12"
O .......... 1.0
S .......... 1.5
HO ........ 2.0
N .......... 4.0
Z ........... 5.0

First, take a look at your track plan and categorize it into these groups (do not count 'fouling points', that is track where cars do not clear each other):

Mainline: Track that connects the various parts of the layout together and allows the trains to go from here to there.

Passing: A second track along the mainline with turnouts at both ends and allows one train to pass another.

Storage: Track on the visible layout that is used to store cars, either at industries or in a yard.

Staging: Track (often hidden) that is used to store completely premade trains for operating on the layout.

Service: Track used to store or service locos (includes things like a turntable, turntable lead, etc.)

Connecting: The track that's left after categorizing everything else as above. Since you cannot count track that would be fouled (cars would hit each other) then this track is essential to moving more cars on your layout.

Here's a diagram showing the various track types.


Connecting track is the key to being able to move lots of cars around on your layout, because that's the track you use to get cars from here to there! Yet it is the track we think about the least in our design.

If you want to know more about using this information to determine how your layout operates, then see this article on my web site:

http://siskiyou.railfan.net/model/layoutDesign/layout.html

Next topic: Operational satisfaction analysis, part 2

[jfugate]

QUOTE: Originally posted by george745

Long time reader, first time poster. I'm really interested in the next installment. Can I ask for two more topics to be talked about. Easments and super-elevation. These are two things I have no clue how to do. Thanks for all the wonderful posts jo.
Andrew

Andrew:

Sure thing. Will cover easements and super-elevation next after the operational analysis discussion.

Thanks for the suggestion!

[Anonymous]

Joe,
I have read the article at your site and I have four questions:

Q1: When adding up the trackage footage (like the mainline), do you count the amount through the switches? My guess is yes.

Q2: On my layout (still 75% in planning) I plan on having switchbacks, I tossed the switchback ends into the connecting pile, correct?

Q3: Also, I have an interchange mainline, am I correct in including that as staging? However, I was thinking I should just treat it the entire length as two sets of staging track of a given length ( max train length) instead of the actual lenght?

Q4: As my branch terminates with two resorts, should I put the trackage after the last turnout into the storage+connecting calculations?

See my website: http://modelrr.adbyrne.us/room.htm for the layout plan. Note that i have not finished adding passing or industries yet.

Thanks,
Allen


BTW; here are some approx numbers:
Room area: 315 sqft
Layout area: 130 sqft
Total track: 496 cars
Branchline: 230 cars
The rest depends on unfinished storage and passing track numbers,
however here are some ballpark numbers;
Max cars: 60-80
Cars moved: 50-70
Train length: 7-11

[wpsteve]

On the WP,what I did to check things out was I purchased "Dispacther" and the design program for it.. Since CTC operations are the WP , it was important that the overall sceme worked. I transffered my track plan into the program. Even used a scale to make all track the correct length.
Then I entered all my trains and the proposed schedule. Then I could run trains over the railroad and catch any problems !
It was fun and helped.

[jfugate]

QUOTE: Originally posted by adbyrne

Allen asked:

Q1: When adding up the trackage footage (like the mainline), do you count the amount through the switches? My guess is yes.

Allen, you are correct ... any trackage through switches cannot have things sitting on them for a long period of time, so it's all connecting track.

QUOTE:
Q2: On my layout (still 75% in planning) I plan on having switchbacks, I tossed the switchback ends into the connecting pile, correct?

If the switchbacks are part of the mainline then they are mainline. If they are not mainline but must be traversed to connect from the mainline to another location then yes, they are connecting track. From the looks of your plan, the switchback looks to be part of how you get from one end of the layout to the other, so they would be mainline in that case.

QUOTE:
Q3: Also, I have an interchange mainline, am I correct in including that as staging? However, I was thinking I should just treat it the entire length as two sets of staging track of a given length ( max train length) instead of the actual length?

Depends on how you use a piece of track. As you imply with your question, you can sometime opt to make a section of track a different type depending on how you intend to use it in operations. If the track will rarely have cars sittting on it then it's connecting track, for example. If the track is used to store and feed complete trains to/from the layout, then it's staging. If the track is used to store blocks of cars that need switched to become part of a train then it's storage. And so on.

QUOTE:
Q4: As my branch terminates with two resorts, should I put the trackage after the last turnout into the storage+connecting calculations?

See the above answer. It depends on how you plan to use the track when you operate.

I looked at your track plan Allen and it seems like a fun point-to-point design. Using the formulas you can determine if you need more industries and passing sidings, which is one real advantage I see from the formulas. It's nice to find this out before you build the layout!

[Big_Boy_4005]

Joe, I worry a little bit about the 40' car rule. It is a nice round number, and for most modelers it should work just fine.

I have a couple of things working against me, modern equipment, means very few 40' cars. My average car length, is over 50'. Add to that, that I am working in O, and it does add up. That knocks at least a car off most calculations.

15' sidings sound long to HO guys, but I am looking at 10 car trains to be safe.

[jfugate]

Big Boy:

I model the 1980s and have the same problem as you ... most cars are no longer 40 foot.

I go ahead and compute things using 40 foot cars since it's such an easy number to use. Once I'm all done I convert the numbers to 50 foot cars. Just multiply by 0.8 to get 50 equivalents ... or whatever factor works based on your average car length.

[Big_Boy_4005]

.8 sounds good Joe, that's what I was thinking. Maybe even down to .75. I have lots of large grain hoppers.[8D]

[Anonymous]

Joe,
Thanks for taking the time to look at the plan!
As I couldn't decide on the how to classify the switchback ends, I will go with your decision to add it into the mainline.

As far as the interchange trackage, I will need to rethink it due to problems in the yard and service area. The ends will be staging with most of the track between being connecting as you thought. However, I am redesigning the yard and service area trying to incorporate the ideas from the John Armstrong planning and Operation book. Because the form factor is a given I may have to use part of the interchange mainline as A/D trackage.

After looking at the generated numbers, I have been reworking parts of the plan for an extra passing siding (including redesigning one as a lap siding) and moving trackage around for more storage tracks.

The formulas really help in thinking things out. Somewhat on topic, I have to make tradeoffs on scenery elements against the operation needs.

Thanks,
Allen

[jfugate]

QUOTE: Originally posted by adbyrne


After looking at the generated numbers, I have been reworking parts of the plan for an extra passing siding (including redesigning one as a lap siding) and moving trackage around for more storage tracks.

The formulas really help in thinking things out.
Thanks,
Allen

Allen:

Exactly!

The formulas force you to think about how you would operate the layout and if you don't like the numbers you know exactly where to start making corrections to make the numbers look better!

I have spent many years using these formulas in my track planning and often would make changes based on their results. My current HO Siskiyou Line benefited from using the formulas ... the Coos Bay branch would not support the desired traffic so I spent a lot of time trying to find ways to add more storage tracks to the branch so all the cars I wanted to run down the line would have a place to go!

I think these formulas are one of the great secrets to designing for satisfying operations. If you learn how to use them, you will know well what your design will do operationally before you ever build it.,

[jfugate]

Topic this post: Operational satisfaction analysis, part 2

Once you have all your track classified into track types and how many cars of capacity in each track type, now what?

First take a closer look at your passing sidings to determine some train length stats.

Determine three values -- longest/average/shortest. Longest is the length of your longest passing siding in cars. Average is the length of an average passing siding in cars, computed as: passing track / number of passing sidings. Shortest is the length of your shortest passing siding in cars.

Next, do a similar thing with your staging.

Determine the longest, average, and shortest track lengths. Longest is the length of your longest staging track in cars. Average is the length of an average staging track in cars, computed as: staging track / number of staging tracks. Shortest is the length of your shortest staging track in cars.

Once you have these values, there's some valuable insight you can gain about your design.

Ideally, staging train lengths should more or less equal the corresponding passing train lengths. Significantly smaller staging train lengths mean extra switching will be involved in getting a "full length" train into or out of staging. Larger staging train lengths mean full length opposing trains from staging will clog the main.

The longer of passing or staging train length should rule in determining typical long train length. The shorter of passing or staging train lengths should rule in determining typical average and short train lengths.

You may notice that staging tracks and passing tracks appear somewhat interchangeable in these formulas.

This either/or use of staging and passing sidings reflects an operating session reality (one that was actually exploited by Tony Koester on his AM, by the way) where the layout's staging can be viewed as "virtual passing sidings".

For example, the dispatcher could set up a "meet" between opposing trains to occur offstage. To do such a "meet", one train exits the layout into staging, after which a different train enters the layout -- as if a meet had just taken place in an offstage passing siding.

If some of the passing sidings on the layout are rather short, this can be a useful technique for arranging meets between longer trains.

For more insights using this sort of plan analysis, see:
http://siskiyou.railfan.net/model/layoutDesign/layout.html

Next topic: Operational satisfaction analysis, part 3

[robengland]

Many thanks for all the great info Joe.

it looks to me like you have a bit of software for calculating the layout operational stats analysis?

Cheers
Rob

[jfugate]

Rob:

One could set up a spreadsheet to do this easily enough. Or a simple html page with javascript.

Hey, now there's an idea ... :-)

[n2mopac]

Thanks again Joe. I have appreciated the info on your web site for some time. Thanks for expounding on it here and interacting with us.

Ron

[jfugate]

Glad to do it, Ron.

I fiugre after almost 40 years in the hobby, I might have learned a thing or two and should share it. Heaven knows I've done it wrong often enough ... [B)]

It's like I told my kids: "Good judgement comes from experience ... and experience comes from poor judgement." If you are smart, you'll make those poor judgement learnings someone else's poor judgement! [;)]

My current Siskiyou Line is my 5th layout. So I'm actually sharing with you my poor judgement layout design situations and what the good judgement would have been on my prior designs ...

[jfugate]

Topic this post: Operational satisfaction analysis, part 3

Here are more useful things you can find out about your design once you have computed the basic numbers discussed in parts 1 and 2.

Want to know how many cars your layout will hold? Is there such a thing as too many cars? Yes, there is.

MAXIMUM NUMBER OF CARS: A layout can only hold so many cars before you become unable to move because even the destinations are full. This upper limit seems to be around 80% of the total capacity for stationary cars, so we can compute this as: 80% of (storage + staging + passing/2).

To allow for more cars on the layout, increase the amount of storage and/or staging track, or to a lesser degree, add some passing track capacity.

Generally, passing trackage is not intended to be used as permanent storage, so to indicate that some passing siding capacity could be used as short-term storage, a factor of one half is suggested in the formula.

Okay, so how many of those cars can you expect to move in a formal op session?

NUMBER OF CARS MOVED: The number of cars moved in a typical operating cycle can be computed as: 40% of (staging x 2 + passing + connecting).

To increase the number of cars moved, we need to increase some combination of staging, passing, or connecting trackage.

Notice staging is particularly effective in increasing the number of cars moved, since for every train that leaves staging, another can move in to replace it, meaning TWICE the cars can be moved (if they are available elsewhere on the layout). In effect, staging acts as both connecting track and passing track -- thus serving double duty.

Another thing we can do to increase cars moved is stop using some track for storage, and designate it instead to be either staging, passing (if trains can legitimately "pass" on this trackage), or leaving it undesignated and always free of stored cars (so by default it becomes connecting track).

And we can learn still more cool things about our design ... all before we ever build it!

Next topic: More cool things we can learn from the layout stats

[conford]

Thanks Joe for posting this material. I read it in the Layout Design Journal when it ran and enjoyed it then, now I am applying it. Last night I ran the numbers for my layout, which has about 80% of the phase 1 track laid. It's really easier for me to use boxcars to measure things :-) I used a 45 foot car length, and was pleased with the results.

While I understand that the purpose of this exercise is to analyze and compare layout designs, I'm wondering if I can really expect to have a decent operating environment with the Maximum number of cars (82) that I calculated. What has your experience been with the relationship with the Max Cars number and the number of cars that actually fit on a working layout?

Anyway, I ran the numbers in Excel, so I will be able to revisit the spreadsheet as I consider the changes and will be able to analyze then in terms of what I want for my layout. This is fun.

Regards
Peter Ford
conford