|SP 4255 splits the signals at MP 337.2 on the Tehachapi Sub Div.
I want to be clear that most of my experiences have been at LMRC in San Diego, CA which has standards of no less than 48" radius curves with spiral easements of at least 12-18" and 1/4" to 3/8" offsets in the easements. Grades vary but are limited to 2.35% grades on the Tehachapi Sub. Div. at the Tehachapi Loop. Most of the rest of the ruling grade is 2.2-2.25% with compensations for the curves worked into the grades. The Bakersfield Sub. Div. has ruling grades to the main staging yards of 0.3%. Access to stub-staging yards requires shoving up some 2.0% grades, which usually requires extra (helping) engines to be available or planned into the operations. The branch lines to Taft and Arvin use 36" radius curves and generally grades are more limited to about 1% on the Arvin Branch and about 1.75-2.0% on the Sunset Rwy to Taft.
|Extra ATSF 239 leads a Santa Fe 'Drag' westward through Caliente on January 7th, 1953.
Most readers probably will not have access to 'monster space' of a layout like LMRC's Tehachapi Pass, but the basic principals we'll be looking at will be able to translate to other model railroads with tighter curves and varying grade profiles. I'll mention the adjustments for these factors as we go along.
Employe Timetables (ETT)
|Covers of LMRC's Employee Timetable 9
Much of the operations are based on standardized 'Engine Ratings' which are published in the Employe Timetable (ETT), which covers much more than just the 'Time Card' section for the various subdivisions. The ETT's have Special Instructions (SI) which cover modifications to the Rulebook and how the rules are applied to the various locations on the subdivisions.
The railroad authorizes and rates engine types and classes which will be suitable for the physical structures of the subdivision, such as bridges, tunnels, rail weight, curvature, sub-roadbed, and roadbed.
|Extra ATSF 239 rolls downgrade through Caliente with a 100+ car empty reefer drag on January 7, 1953.
The most basic system of ratings at LMRC generally fall into a "2-cars per powered axle" on the Tehachapi Subdivision. This seems to work well for most of the older Athearn "Blue Box" diesels of both 4 and 6-axle models. The ETT superseded this general rough estimate with tested engine types where 'the worst' of each type was tested to form a standard.
|LMRC's Tehachapi Sub SP Engine Ratings for 1950's TT/TO Operations.
The standardized procedure lays out the following which is to be done "after hours" when there are no other trains on the layout. This gives the most flexibility to test the engine and adjust the testing tonnage.
1. Position the engine with a "suitable number of cars" * below the ruling grade.
2. Pull engine up onto steepest section of the grade and stop. The engine should not be able to restart the train at this point.
3. Uncouple cars until the engine is able to start the remaining cars without slipping. This constitutes "Starting the train 'Easily'."
4. One more car is removed to account for the variations in the car fleet.
5. It is a good idea to finish by running the test one more time from below the ruling grade section to confirm that it can 'easily' climb the grade without slipping.
* = Probably 15% or so more cars that are estimated to make the rating. These should be 'Average' cars in terms of physical qualities of rolling, weight, etc. Hopefully the model will slip, not stall. Stalling will lead to burning out the motor! Operationally, I'd rather have an engine slip a bit than tear up something in the drive or fry the motor.
Tips for your own Ratings
Tighter curves will obviously effect how long of trains you can operate without helpers. Generally lowering the grades will increase the number of cars capable of being pulled, but as the grades get flatter, you'll still run into the square-cube law problem of not being able to pull as much on flatlands as the real engines.
On model railroads with tighter curves generally you'll not have as long of runs to see the train's full length. Basically, at a certain point, there's little reason to run a 50 car train if you can't see it all at once and the train string-lines in your curves, tunnels, and helix. Reducing the overall train length will help, and at a certain point you will have to determine when you have to use a helper. LMRC's point is 40 cars uphill, and about 72 cars downhill. Beyond those two points, the laws of physics become too much to fight and you'll regularly loose if you try. So find that point for your RR and set some standards!
Soap Box Moment
While I've skirted around the point, but I'm going to say this straight out. One of the main reasons we use such 'conservative' ratings on our models isn't that they can't get more "Up the Hill", but that we don't want to have to keep rebuilding expensive models, which are not easy or impossible to get replacement parts for, on a monthly or yearly basis. We want our models to perform well in operations and continue to do so for most of our foreseeable lifetimes. I've talked about this more in my blog post about Rebuilding My First Brass Engine. Often with the layout compression, even LMRC has, engines don't need to be able to pull their 'real ratings' of 100 cars, or whatever. They need to be reliable and ideally stay around a long time without too much effort from the owners.
All too often we've had members come down and try running their 'newest and greatest' model, sometimes very expensive brass model, and there will often be some comment made along the lines of, "Oh, my new engine can pull 45 cars up your grade... Here let me show you!" Often they'll get to the top of the grade and suddenly their 'magical' engine vanishes from the layout after half a trip. Oddly, usually we'll not see that 'super engine' again for years if ever again. The few where we've found out the story... Yes, their engine made it to the top of the Hill... Yes, it pulled 45-odd cars single handed up the grade... However in doing so, it slipped quartering on a driver, broke a drive shaft, whatever and will now be out of service a LONG TIME, just for 45 minutes of "Super Power".
|This is NOT how we want to see our models for months and years, but it's all too often the case when the Ratings are ignored.
Over the course of about a year, one member ran an Overland AC-12 and Sunset Models F-5 class with over 60 cars (Over double their LMRC Ratings), and last I heard both engines were out of action with major drive damage, driver quartering, and side rod wear issues. That was over 15 years ago, and I don't think I've seen either of the models totaling over $5000, back in operation.
|We usually have enough work cut out for ourselves getting the 'new projects' into service for the first time.
The regular operation of models over their ETT Ratings goes completely against LMRC's goals for operations, which is to realistically recreate the Tehachapi Pass during the 1950s, but also be able to keep our sanity with repair schedules on our models. Keep them in good mechanical condition for as long as possible while slowly accruing wear on them. Well taken care of many solid makes of brass engine will last decades with minimal repairs. But a few bad trips where they're severely overloaded, will mean an early life sitting in a display case or back in the box in the closet.
Let's look for a few minutes at the cars we're using. The freight car fleet is massive, there will be variations between the cars and the cars' characteristics, such as rolling qualities, will physically change over time as wear sets in. Make sure to run the tests with a good 'average' test.
|LMRC's Bakersfield Yard during operations on January 7th, 1953.
Of course our model freight cars do not scale well in terms of rolling qualities and mass, however there is one interesting point to note on the operations of the LMRC model of Tehachapi relating to models and mass. Of course we're not really able to adjust the weights of the models to have "loads" and "empties", so a nominal 'car mass' must be settled on. If we assume the rough figure of 70-tons per car then our tonnages work out nicely on the Tehachapi Sub. The 'big' AC-classes could handle 1350 tons, and at 70-tons per car, we can work that out 1350/70 = 19.3 cars.
On the prototype Tehachapi Sub-Division SP also made a seven ton-per-car curve compensation adjustment. That is they figured the car would weigh another seven tons more that it actually showed on the waybills, etc. to account for the resistance that would be experienced with the 10 degree curves. This means for every 10 cars, you're figuring an extra 70 tons. So for our example of a 1350-ton rating, then the car count is reduced to the following, 1350/77 = 17.5 cars, rounding down of course.
The old railroad in La Mesa had a significantly different set of standards. Let's quickly go over those. The curves were tighter at 26-28" radius. The grades were about the same. The freight car fleet generally was actually heavier than both NMRA and LMRC's current standards, with 40ft boxcars often in the 6 oz range and most had sprung trucks.
By 1990, testing has shown that the old arrangement used at La Mesa doesn't work well on the new railroad with the newer technology used in manufacturing the cars. After the club moved to Balboa Park, many of the newer freight cars were using 'rigid' trucks made of plastic. This has only increased in recent years with newer 'engineering plastics' being used by many manufactures to further reduce friction in the journal bearings. By about 1990, or about 10 years of tests, LMRC's Car Standards were revised to include a 'Center-of-Gravity' (CG) factor, 'Rolliblity', and weighting which formed a 3-point graphed relationship.
The old cars with poor CG, weighing at or over NMRA standard weight tended to derail much more than cars which were lighter, but had better CG's. This is roughly mentioned in NMRA's RP-22 "Car weight should be kept as low as possible in the car." Unfortunately, NMRA never really codified a way to evaluate this concept.
|Excerpt from LMRC Car Standards 2008 Edition
The LMRC Car Standards were able to codify the relationship questions of CG, Weight, and Rollablity.
|Excerpt from LMRC Car Standards 2008 Edition
Basically as the CG gets lower and a car can be 'deflected' (from vertical) more and still return to its wheels, the car doesn't need to weigh as much. As the car rolls easier, the weight can also be reduced.
Cars that weigh less, run more reliably, and can still be shoved on or pulled on in large 50-70+ car trains is one of the operational goals, the weight reduction allows the engines to move the trains more easily around the railroad, with less wear on the drives and wheels.
Passenger Car Ratings
Passenger Cars are counted as two freight car standard weights. - This follows the basic logic that the standard length passenger cars of about 80ft over-all are double the length of the 'nominal' steam era 40ft boxcar. Yes, modern plastic passenger cars weigh significantly less than the older brass models, but SP had a maximum passenger train length of 26 cars, and many of our sidings can't handle passenger trains much longer than 26 cars anyway!
|Two AC's pulling hard on Second 60 on the ruling grade above Caliente.
The 'Streamlined' or 'Lightweight' passenger cars are rated at 1.5 freight cars each. This allows for trains like the San Joaquin Daylight to run at 12-14 cars with only two GS/Mt class 4-8-4/4-8-2 engines and still be within ratings for the Tehachapi Sub.
Track Cleaning Car Ratings
I want to call special attention to our slider cars, which help clean the track. The club standards call for each train (30 cars) to have one slider pad car. These slider pads are masonite pads without extra weight on them. The pads simply glide on top of the rail without putting much additional drag on the train.
|An example of a masonite slider on Proto2000, NKP 66031.
However, from experience with tonnage testing and rating of the engines, it is clear that each slider pad car will rate as about 3 normal cars. So an engine that could handle 12 cars normally, would have to be de-rated to 10 cars when one is a slider.
I have more info about this type of track cleaning slider car on my "Disguising Track Cleaning Cars (Part 1) - P2K Gondola".
Helper Operations & Ratings
|Green flags flying as SP 6152 leads First 804 at Walong, May 29, 1951. Brian Black Collection
The years of operations at LMRC on both the original model railroad in La Mesa and the 'new' Tehachapi Pass - Joint Line has shown several factors at play with multiple engines and helpers.
When multiple engines are operated together the ratings of each will be added together, which is pretty obvious. The interesting part is that the 'one car removed' during the rating process for variations in the fleet is returned for each additional engine.
|SP 6150 leads a four unit set into Caliente.
So for a four-unit set of F-units would normally rate as 9x4, for 36 cars, but with the extra car returned for three of the units, the engine consist will rate at up to 39 cars.
Eastward on the Tehachapi Sub the non-helper maximum tonnage is 40 cars. Trains over 40 cars must have helpers 'cut in' father back in the train. The ETT specifies that helpers will be 'cut-in' with one half the rated tonnage of the helpers in front and the other half of the rated tonnage behind the helpers.
Again, this is a trial point that each model railroad will have to be tested for and determine at what point the Laws of Physics, just stop you from going beyond. Generally with the helpers, you're goal is to set up a proportion of the tractive effort between the front of the train and another point back in the train. The ratio of where the second engine is located comes from the relative Rating of the two engines. This will remain the same across any engines, and any railroad, the concept is the same.
DCC Settings & Speed Matching
Let's break for a couple of minutes onto a tangent about DCC and programming for large train operations and large engine fleet logistics.
A helpful hint is that you're not really going to require every engine to match speeds perfectly across their whole speed range. Unless you feel specially called to spend your life on a programming station fixing and adjusting decoders and 'weird electro-mechanical interactions', the reality is that models will change operating characteristics over time, as the models break-in, loosen up, then start to wear and get sloppy.
Get them to start (CV2)within about 5% of each other, and once stabilized at a speed, they should be within 10% or so of each other's speed. LMRC has standardized a mid-speed of 25 MPH at half throttle and 50 MPH at full throttle. This matches with our ETT speed limits pretty well. Mind you this is without excessive B-EMF which can invalidate speed matching.
Also I recommend that starting be adjusted to control the maximum weight train on the downward grade to a 'Controlled Stop', not a slamming to a screeching stop from 5-10 MPH. On older non-B-EMF decodered engines, this will mean that your CV2 value will be low enough that the engine will not start while sitting on level track without a train, and starting upgrade with a train will take a considerably higher throttle setting.
Engines with B-EMF should have them set to minimal levels and set to 'fade out' by the mid-throttle speed matching point. Momentum should also be considered and adjusted to match at least the engine which has the largest 'natural' flywheel momentum. You don't want the other engines to be able to stop faster than the ones with a big flywheel! If you desire additional momentum, then by all means you can add it. I suggest that engines shouldn't have too much momentum that you can't stop the train reasonably quickly if something is going wrong, like freight cars are falling on the floor!
Eastward Helper Example - Normal Train with One Helper
|SP 4249 shoving hard at the west switch at Caliente, in about 1/4 mile, the AC will stop for water.
On our previous example, four F-units would rate at 36-39 cars, pulled from the head-end. In this example, I'll say we're using an 'big' AC-class helper, such as SP 4230. SP 4230 is an AC-10 and rates at 17 cars normally, or 1350 tons. As a 'Helper' the AC in this case would be allowed its one extra car, and as an articulated engine, a second extra car is granted, resulting in a helping rating of 19 cars. The AC then should be cut-in with half it's regular rating ahead, extra 'helper' tonnage is placed forward of the helper to 'float' between the head-end pulling and the helper pushing as the train works over grade changes, etc.
The resulting train when it prepares to depart Bakersfield should then be as follows:
Four F-units (36-39 cars Rating)
49 "Average" freight cars, minus any adjustments for track cleaning cars)
One AC-class helper (19 cars Rating)
8 "Average" freight cars, again adjusted for track cleaning car)
one Caboose, which is counted as one freight car weight.
|SP 4252 works around Walong slowly as the PRS special holds the siding, April 29, 1951 - James Salkeld Collection.
I'll make note here that generally working Eastward from Bakersfield, the steam helpers are not placed right ahead of the caboose or coupled with more than one because of the tunnels. Generally the crews preferred to be able to breath and not be broiled riding right behind a hard-working steam engine. Both ATSF and SP operated their diesel helpers right ahead of the cabooses, and in some cases behind the caboose (waycar) as well, in compliance with PUC's regulations of no more than 8-axles behind the caboose. This often meant a pair of ATSF GP7s or a single SP RSD-5 behind the caboose.
At LMRC we've found that we can run the diesels behind the cabooses, per prototype practice. However, just as on the prototype, generally we prefer our helpers ahead of the caboose! We've also written special instructions and 'institutional knowledge' to the Chief Dispatchers that no more than 24 cars should be pushed by a helper set, this limits mid-train helper sets to four-GP or three-SD type engines with 1/2 of their combined rating ahead of the helper set.
Rear-end helpers are allowed to push up to 24 cars, this prevents crazy things like 3-4 SD-type engines shoving 36-48 cars worth of train up the 2.3% grades (which usually will result in a harsh lesson in the laws of physics!)
Westward Helper Operations
|One of the largest SP BK-OK-R's at 137 cars works carefully down with the head-end reaching Caliente and the rear still well in Allard siding. Photograph by J.Hill from mezzanine before 'Walong' was built, Nov 17, 2004
While most of the heavy helper operations are on the 'North Slope' of the Tehacahpi Pass, there is also a noticeable number of helpers which work out of LA (Long Helpers) and Mojave (Short Helpers). The helper crews are directed as to where they will be "cut-out" by a note on their "Soup Ticket" (Call ticket). Many helpers are only needed to get trains up from Los Angeles (our East Staging Yard) or Mojave to the Summit at Tehachapi, while other heavier trains (normally anyhting over 72 cars of mixed 'average' cars or 90-cars of lumber or reefer empties) will need a 'Through Helper' all the way to Bakersfield or Bena. Unfortunately, our scale models do not have working air brakes on each car, so the only way we can control such large trains is with the physical motor and gears in the engines holding the train back. The extra westward helpers cut in near the caboose are needed to keep the rear of the train from compacting down against the head-end with enough force to buckle and derail the head-end of the train.
|As a "Through Helper," SP 4230 rolls downgrade through Bealville .
The big trains, well over 100 cars, will usually have both a 'Short MC' helper and a 'Thru' helper to Bakersfield. The Short helper will originate out of Mojave, either cutting-in at the yard or as was prototypical, head east to Rosamond, where a prototypical short stiff grade, required help to get heavy trains into Mojave Yard. The mid-train engine will usually be cut at Summit and usually be routed either back to Mojave or if needed, run separately down to Bakersfield.
|ATSF 'Drag' heads down into Caliente with two GP7's holding back the rear of the train.
The rear helper will continue in-train down 'The Hill' to Bakersfield, working a very light throttle to keep the rear portion of the train stretched off the head-end. Both crews need to work together and make easy adjustments to the speeds of their respective engines, keeping the slack-point somewhere in the middle of the train. Too far back and the head-end could buckle, too far forward and the helper could be 'string lined' because it's holding back too much. Basically about 75% of the train should be under control of the helper, and gently letting it down the grade, with the road engine holding back only the forward part of the train.
Additional Comments on Standards & Exceptions
When I joined the LMRC in 1996, pretty much all of the car standards and ratings had been experimentally proven for over 14 years. Since 1996, I can only say that I agree more now with the additional empirical proofs of another 20+ years of operations which show that equipment built to the LMRC standards have very high reliability.
|ATSF 140 works the SCX-BI upgrade between 4th Crossing and Walong with ATSF 3854 shoving hard mid-train. January 8, 1953.
There are a few examples of car types which can't be made to match the LMRC requirements. In such cases the Car Shop can grant a exception to certain parts of the standards if the car type can be shown to comply as much as possible and work reliably, possibly with operational restrictions in placement in train or maximum cars allowed.
|SP 6188 leads a 70+ car XMUG-7 westward at Marcel on January 8, 1953.
One of the other interesting observations I've had in the last 10 years or so, is with large groups of similar cars, or block movements, where each car differs from the standards and in some way form a 'stacking tolerance' issue. This is noticeable with large numbers (15-25+ cars) of gondolas or flatcars which weigh less than the 'average' of the whole fleet. One option is that trains with such large blocks will be noted to the Chief Dispatcher, who assigns engines, and the calculated total weight of the train would be adjusted.
|A large block of mostly empty East Coast cars returning home
For example when a large string of gonds or flats are running empty, I'll count about 75% of their counted weight, because on average these cars are right at the lower end of the Standards for weight, but in terms of Rollablity and CG they are excellent.
An "Unbalanced" Helper Trip
This became an interesting object lesson with a 40-odd block of GS-gondolas with beets, which tipped the scales near the bottom of the Weight Standards, followed by another 15-20 mixed cars which had more 'average' weights. The AC-class helper was placed correctly, and the two crews were pretty good, with one being a retired ATSF freight engineer. I was asked what could be the issue after a couple of derailments. The problem was that the heavy modern F-units were able to easily pull all the way through the cars and into the helper, resulting in string lining. I tried working the helper for a couple minutes, and by running a much lighter throttle on the head-end we didn't have any more problems. The problem was that the physical loads were not spread across the whole train evenly.
Normally, we assume that the train has a nicely distributed 'average' for the cars' mechanical characteristics. In the four F-unit plus AC-class example for 58 cars, we're figuring 2/3s of the train will be pulled by the head-end, and about 1/6th pushed by the helper, and the last 1/6th pulled by the helper. In the beet train example, the head-end was much easier than average to pull, because of the 40-car block of lighter cars, followed by the rear portion all being 'normal' rolling cars worked by the AC-class helper required a heavier throttle. This resulted in a train which should have probably had the helper cut in only 4-6 cars from the rear, or even dropping from an AC-class to a RSD-5 or AS-616 helper. Once out on the road and climbing above Caliente, there's not much we could do for the crew, which means this sort of an anomaly causing such an imbalance needs to be figured by the Yardmaster and Chief Dispatcher when planning the train a couple hours before the train is called.
|SP 3765 leads a Mountain Work Train into Walong on January 6, 1953
While in Triple Trouble on Tehachapi Pass, I've talked about certain odd occurrences, or rare situations, most of the 1950's TT/TO operations at LMRC fall into a dynamic but 'normal' flow of traffic. This makes the slightly 'weird' trip something fun to remember years later and keep the operations interesting. I did write my post on A Trip over Tehachapi Pass on SCX-BI covers a 'normal' trip.
|Point Helper ATSF 3851 leads No.4 at Cliff on January 7, 1953.
Over the years of operational knowledge at LMRC we've seen many challenges and had to adapt to them. The resulting ETT, just as on the prototype, is there to give a set of 'safe guidelines' which should result in good operations consistently.
I encourage folks interested in operations to do a little digging into the other parts of relevant prototype ETT's. Do some testing of your equipment and incorporate the information into your operational scheme and your own ETT with proven Ratings and Special Instructions.
"On the Bell" - Future Posts on Operations -
|Chief Dispatcher's Train Sheet & Symbols for January 7, 1953.
In this post, I touched on what the Chief "should have done" in a couple of situations, but without defining the position for you. Hopefully I'll get a chance to sit down and finish writing my next post on "Operations" series in which I think we need to go over the duties of the Chief Dispatcher. The Chief Dispatcher is one of the main positions which enables us to manage the operations and pull off the dynamic restaging of LMRC during our TT/TO 1950s sessions in real time during the sessions.
Busy Times at Bakersfield (Part 1) - Roundhouse & Locals
Busy Times at Bakersfield (Part 2) - SP Yard Overview
Freight Symbols Over Tehachapi (Part 1) - My Story Learning Operations - Multi part series on Freight Symbols.