Wednesday, November 9, 2016

Appearance of PFE reefers, Part 3

In two previous posts, I raised and commented on the issue of PFE car appearance at particular periods, with emphasis on the car washing which PFE conducted into the 1950s. (You can see the prior post at this link: http://modelingthesp.blogspot.com/2016/11/appearance-of-pfe-reefers-part-2.html .) I also described, in the post just cited, the PFE shop reports, PFE Form 865, which document in amazing detail the activity of each PFE shop as well as the progress of various programs underway in particular years. In the present post I want to show more of that information.
     First, here are several columns of the data, collected from forms 865. The columns are not in any particular order, but show not only the size of the complete fleet, but also the total rebuilds completed in the years when Form 865 reported those totals, along with the number of cars painted, right column. This number is independent of the number of cars washed, though one might think that a car about to be painted would be washed first. If so, that washing is not included in the “washed” column.


Another point to be made in passing is that the large numbers of cars painted during 1929–1934 represent work immediately after the standard color for car sides was changed from yellow to light orange, The total number of cars painted during those years exceeds the size of the fleet, thus the conclusion that all previously yellow cars would have been orange after 1934.
     There are other interesting details in the table, but I recognize that only data geeks like me tend to enjoy masses of numbers like this, so I have plotted some parts of the data set to make them easier to appreciate.
     The first graph shows the data on total number of cars washed, by year (the blue line), compared to the total size of the entire PFE fleet (red line) in each year. The lines connecting the data points simply follow the data, and do not represent any kind of fit or function.


Note here that in two years, 1930 and 1941, around 25,000 cars were washed, out of fleet of 36,000 to 40,000 cars. That raises the question, how big are the percentages washed by year? The percentage of cars washed, that is, the number washed divided by the size of the whole fleet, times 100, is thus interesting, and below is that information in graphical form. (You can click to enlarge the image if you like.) These are the same data as the blue line in the graph above, just replotted as percentage of the red data above.


     These data are, I think, dramatic and certainly confirm the comment that PFE washed a lot of cars until 1949, and even after that, were washing 10 to 20 percent of the fleet. What happened then? As Earl Hopkins reported it to me, a request for funds to build mechanical car washers, instead of the hand washing then in practice, was refused. For more on that, see the PFE book, page 159 (Pacific Fruit Express, 2nd edition, Thompson, Church and Jones, Signature Press, 2000).
     I do have some older paint schemes among the PFE car fleet on my layout, but the older ones are by no means dirtier than newer ones (I do try to make older, clean paint schemes look faded). That choice, of course, was made because the information existed. We happen to have a lot of relevant data in the case of PFE, so we can do a better job of historical accuracy than for many car owners.
Tony Thompson

7 comments:

  1. I think the % washed for the last three years is off by an order of magnitude. 1.92% for the last year, not 19.2%.

    ReplyDelete
  2. You are right. I didn't catch that error in Excel. Thanks for the catch.
    Tony Thompson

    ReplyDelete
  3. You can certainly see a temporary drop in the number/percent of cars washed during the war years. Perhaps this reflects (1) making car washing a lower priority due to wartime needs and/or (2) the temporary shift of the PFE fleet management into one influenced more by the Office of Defense Transportation directives rather than PFE priorities.

    ReplyDelete
  4. I model 1950. Since 1943 there's a decline in washings visible compared with the years before. However in 46, 47 & 48 management seems to have concluded that the fleet needed to be washed again: 33154 totals of which some car might have been washed even 2 times. So of a total of roughly 38000 cars, some 5000 nover got washed since 1943. Conclusion: in 1950 there were a few (13% of total) über dirty cars, the rest -apart from a few new 25's and repaints- was just dirty in many gradations. Good to know, thanks.

    ReplyDelete
  5. Trouble with your conclusion, Fred, is that PFE bought 5000 new cars in 1946 and 1947. Assuming those weren't washed in any numbers, my own conclusion is that there would be more like ZERO "uber dirty" cars by 1950.
    Tony Thompson

    ReplyDelete
  6. You mentioned that the cars did 10-12 trips per year on average. If we assume (for easier math) 10 trips per year and use the data from 1923 (10,545 washed of 33,000 cars), we should multiply the fleet times 10 before we divide by the number of washes for a year to determine a more realistic percentage, correct? Using my theory of 10,545 of 330,000 car trips, only 3.2% of the cars were washed in 1923. Does this make sense? Am I missing something?

    ReplyDelete
  7. Matthew, what you have calculated is that about 3 percent of all cars passing through a shop each time they were empty would get washed. That's an interesting number, and fits with Earl Hopkins' remark that the Light Repair foreman chose cars to be washed, implying he would choose (a) probably the dirtiest cars, and (b) wash as many as he had time and workmen to do; but clearly, from your number, not too many on any particular day. The Light Repair and Cleaning tracks (cleaning of car interiors) often hosted several hundred cars a day in harvest season, at the major shops of LA, Roseville, and Nampa, so maybe ten or a dozen cars might get washed a day at each shop. For three major and two minor shops, that might add up to the 10,545 that year.
    Tony Thompson

    ReplyDelete