Wednesday, April 20, 2011

Modeling freight traffic: Coast Line, 1953-Part 10

Returning one more time to the conductor’s time book from the 1948-1952 period, I have looked this time at hopper cars (which wee nearly all SP ballast cars), foreign flat cars, tank cars, and some additional gondolas (not identified as beet cars). I introduced data from this time book in a prior post (, and have continued with subsequent posts in the same series.
     First, the flat cars. There are only five foreign cars identified. Interestingly, two were C&NW cars. The others were CB&Q, NYC, and UP. All were empty except for the Burlington car, marked as loaded with tractors. Some of the empties were destined to places like Salinas for loading (the NYC car).
     The gondolas not listed among the beet cars were interesting. (I analyzed data for cars identified as beet cars in a previous post, There was only a single tight-bottom or GB gondola, which was from Class G-50-14. All others were GS or drop-bottom cars. These were distributed as follows. There were 15 cars drawn from the 1920s Enterprise cars: six from Class G-50-9, four from -10, two from -11, and three from -12. These frequencies scale fairly well with the numbers of cars in each class. There were also four cars from the 1940s Enterprise GS gons, two each from classes G-50-15 and -18. A number were in large cuts of gondolas: ten loaded with manure for Chualar (already listed in post 8 from this series), and six empties destined San Miguel.
     Among the hoppers, six were from classes H-70-2 and -3, and one from Class H-50-6; all these are longitudinal-dumping ballast cars. There were also four covered hopper cars of class H-70-4. Most of these cars either carried sand loads, or were marked as empties for sand loading. Very interesting were two foreign hoppers, D&RGW 18371, a covered hopper, and PRR 146661, a Class Glca car of AAR class HM.
     Among the tank cars were eight SP cars, one each from classes O-50-9, -11 and -14, two from Class O-50-6, and three from O-50-13. It was interesting that some of these SP cars were empty but identified as destined to Manteca, presumably for loading.
     There were also 12 privately owned tank cars, with reporting marks AOX, CDLX, GATX, and SHPX. Information about tank cars can be hard to find, because listings in the Official Railway Equipment Registers are often superficial, but most of these cars (the five from SHPX and three from GATX) all appear to be 8000-gallon cars. Most were marked as “molasses to Spreckels,” which either means empties for molasses loading at Spreckels, or inbound molasses purchased by Spreckels to supplement their own production. I tend to think the latter is more consistent with how the time book appears to be constructed.
     There was also a single car of fuel oil to Spreckels, a 10,000-gallon car from AOX (Associated Oil Company). There were three empty CDLX cars (California Dispatch Line), all AAR class TMI (insulated but non-pressurized), which could be either ICC 103 or 104. CDLX did roster a number of asphalt cars, and these might be among them.
     Lessons learned: the SP cars in these groups very roughly scale by class sizes, indicating they were essentially randomly selected within their car types. This is what would be expected in most cases.
     There were surprisingly few foreign flat and gondola cars, which nationally are predominantly free-running cars and might be expected to show up more frequently in these data. But it should be emphasized that the conductor who wrote these data primarily worked only on local trains, and more specifically on haulers, not on through freights nor even on locals doing pickup and delivery of cars. Thus these data cannot reflect overall Coast Line traffic.
     But the cars that are identified do provide a guide to what is logical and reasonable to have in Coast Line traffic, even if they are necessarily only part of the total picture.
Tony Thompson


  1. Tony,
    Just to sort of recap to make sure I have this reasonably straight in my head--by and large your data supports the earlier study that says cars are logically distributed based on the size of the various railroads around the country with about 30% being home road. The data from the conductor's time book though has cases where the data and the study don't agree-notably flat cars and gondolas which you attribute to the fact that the conductor in question usually was on locals and there weren't places on his locals that needed flat cars or gondolas. The other exception being some railroad specific trains, beet cars and reefers on the SP coast Line. Am I still on the right track?
    Thanks again for taking the time to research this stuff and post it.
    Jim Pattison

  2. Tony,
    Thank you for the fascinating information you've presented in this series of articles. Although a wee bit over my head at times since I'm a fledgling railroad modeler, I do appreciate the education.

    Thank you,

  3. Jim, I think you do have it about right. But the 30 percent is for SP's Coast Line in the time period of the conductor data. I have no idea how general a value that is. Gilbert-Nelson theory, if we may call it that, doesn't address home-road percentages. As I've said before, I sure would like to find more time books for the Coast in the early 1950s, and hopefully with data for mainline freights.

    George, you're very welcome. I'm pleased that the information has some value to you, and that's exactly what I hoped would be the case for this blog.
    Tony Thompson

  4. Tony,
    Thanks for running all the data from the various sources and building a logical set of conclusions for the Coast Division in 1953. The pattern and the logic can fit for any location at anytime, proving the matrix and logic for anyone modeling a specific time and place. The turn of the century crowd have been working similar issues, only with different facts of course. I have found that some of the participants feel more comfortable with a different ratio between the home line, interchange roads and more distant lines, but at this point, those percentages are best determined from the "best available source data," rather than a preconceived set of percentages.
    Thanks for working through the process so all of the blog group can better understand how to derive the end results. I love the idea of the signature car on occasion.
    On my 1895 version of the Colorado Midland, I remove some of the distant road cars after the operations cycle that have made it the staging tracks and replace them with cars from other lines. That allows me to have all the strange cars from the east coast & southern lines and put them on a cycle which allow them to recycle back on to the layout every quarter or so.

  5. Thanks for the kind words, Tom. I realize that the information I've generated can only help in the most general way with other eras or other parts of the country. But what I hoped to illustrate with this particular set of data is that one need not model by the "seat of one's pants" (or should I say, by the "seat of one's modeling druthers"), but can use prototype data to guide modeling decisions.
    Tony Thompson