Friday, December 3, 2021

Reflections on ice refrigeration

 During my recent trip to the Twin Cities and Wisconsin, the subject came up of the details surrounding ice refrigeration. One person present said I had never described it in full in my blog. To which I somewhat lamely replied that “it’s all in the PFE book” (Pacific Fruit Express, by Thompson, Church and Jones, 2nd edition, Signature Press, 2000).

It’s a lame response because nowhere in the book is the topic really summarized, though there is certainly very extensive factual material as well as photos showing icing, ice plants, and ice refrigerator cars. So the present post is an effort to provide a summary. For background, let me mention a post from some years ago, on the terminology of ice service (you can see it at: ). 

For an even broader description of some of the topics of the present post, I recommend an article of mine in Model Railroad Hobbyist, in the issue for September 2013. This issue is still available, for free, to read on-line or download for your use, at ), and was Part 1 of a two-part article about Pacific Fruit Express. Some of the photos in the present post are repeated from that article.

Let me begin with empty cars. All major refrigerator car owners maintained “service centers,” as we might call them, where nearly all returning empties were cleaned and checked for any mechanical repair needs. As car owners, they wanted to make sure that cars supplied to shippers for loading were clean and in good condition, especially important for food and food products. 

That’s not to say that modeling a local clean-out track for reefers is incorrect, only to say that it was unusual, outside the car-owner facilities. Many yards had clean-out tracks for box and other cars, though reefers would not normally have been cleaned there.

As an example, shown below is the PFE cleaning track at Roseville, California (PFE photo). The length of the string of cars being cleaned and checked for any needed repairs testifies to the scale of this operation, typical of any large owner of reefers. Some cars still had “body ice” inside (described below) and it’s been pulled out. Drains have been cleared and are emptying onto the apron.

(Incidentally, I should mention that the photo above was taken in November 1962,  almost ten years after PFE discontinued periodic washing of its cars in about 1953.)

Sometimes fumigation is mentioned. Occasionally, insect infestations did occur, and empty cars were fumigated to eliminate the pests. Again, the car owner, such as PFE, carried this out at their own shops. In my interviews with retired PFE people, they said that this was an uncommon requirement.

Once cars were certified by the car owner as ready for use, they were moved by the railroad to the harvest areas for loading. In peak periods, this could be a difficult and delicate balancing of the needs of each area, based on crop and weather forecasts, along with management experience. 

By the way, for a professional’s recollections on this topic, and on icing, you might like to read a segment of one of my interviews with PFE’s Pete Holst (you can find it here: ).

The shipper made the decisions about all Protective Services, as they were called, meaning both icing and use of car heaters. And icing might mean normal bunker icing, along with the possibility of “top icing” (shaved ice shoveled or air-blown over the top of the load inside the car) or “body icing,” placing small blocks of ice among the packages of produce inside the load. Likewise, the opening of ice hatches in “vent” (ventilation) service was determined by the shipper.

The photo below, from the ART book (American Refrigerator Transit, S.T. Maher, G.J. Michels and Gene Semon, Signature Press, 2017), shows shaved ice being blown over a load of spinach baskets with air pressure, in Robestown, Texas in 1944.

The term “icing,” therefore, had multiple meanings. And the language in the Perishable Protective Services Tariff refers to bunker icing prior to spotting the empty car for loading as “pre-icing,” whereas filling the ice bunkers of a just-loaded car, prior to its departure to destination, as “initial icing.” 

The shipper chose whether to specify (and pay for) pre-icing, usually according to whether they had their own cooling facilities in which they could cool the produce before loading. If not, pre-icing at least got the shipment loaded into a cool car. But remember, the thermal mass of a carload of produce greatly exceeded the thermal mass of the car and air inside the car. Loading pre-cooled produce into a warm car was not considered damaging.

Then of course the car set out on its journey of hundreds or thousands of miles. Icing was conducted “as needed,” or as specified by shipper, usually meaning every 24 hours, but times would be extended in cooler weather. Bunkers were simply filled to the top, after a foreman estimated visually how many pounds it would take (shippers were billed for all icing). Likewise with vent service, the shipper decided whether and where the hatches were opened or closed, based on weather forecasts for the transit territory. Telegrams could be sent ahead to modify initial instructions.

The icing process itself distinctively used two tools, in ice deck operations throughout North America. One was a wood-handled “pickaroon” with which ice could be either pushed or pulled; the other was a heavy, forged steel fork, called by some a "bi-dent,” which was used to chop ice. I will show two photos. First, a 1952 photo by Richard Steinheimer (DeGolyer Library), on what he described as a “hot summer night” in El Centro, California. This reminds us that icing took place around the clock.

Secondly, I cannot resist showing again this superb Jim Morley photo, taken at Roseville in 1948, showing men not posing for the photographer but in the act of doing their jobs. Like the photo above, this photo clearly shows the drop-down aprons and portable bridges used to move ice to the hatches.

We are also reminded by both of the above photos that the large 300-pound ice blocks manufactured in the ice plant were quartered on the deck by a workman called a “splitter,” and then the “passer” moved the blocks to the man on the car roof, the “chopper,” who reduced the size of the ice chunks according to what had been ordered for that car. The 1951 photo below at Bakersfield shows this also (PFE photo).

For more, please consult my 6-part series of blog post on “handling ice on ice decks,” a phrase that can be used as a search term in the search box at the top right of the present post. 

Obviously, the sheer scale of big-time refrigerator car operations would be challenging to model. The saving grace for modelers is that local packing houses were often on a far smaller scale, and their operating cycles can be modeled realistically.

Tony Thompson


  1. Good Morning Tony,
    Thanx for the 5 part series and the above write-up. Very useful and enlightening.

    I remember seeing an article and photo of a reefer being "top iced" from the car interior. Apparently the shipper wanted maximum protection and it was explained that fully loading the ice bunkers was difficult from the roof top.

    Question: how come no railing on your small icing platform? Kinda dangerous, no?

    Best wishes for the holidays.

    Lou Adler

  2. "Top icing" cannot refer to getting anything into the ice bunker, as there were screens in the way. If the article said that, it's a misunderstanding. With ice hatches open, a bunker could be filled right to the top.

    Railings: few prototype decks had railings. That was in the days before OSHA, when it was assumed that people had common sense about hazards. Of course, they don't always, and OSHA is a good idea, but I model long before that.
    Tony Thompson