Hungerford Co., then Fraser Manufacturing Co.
Production drums, New York, then Jersey City, New Jersey, United States.
History
The name Hungerford appears in the American patent registers before the roasting machine, but Ukers contradicts himself on the date. His chronology places patents of G. W. Hungerford for machines to clean and polish coffee as early as 1871. His chapter on apparatus, for its part, makes the name appear only in 1880 and 1881, in the name of G. W. and G. S. Hungerford. The roaster comes only in 1882, and that date is no accident. Jabez Burns's 1864 patent protected the first cylinder able to empty itself without being taken off the fire. It had just expired, after the seventeen years that American law then granted a patent. The Hungerfords, father and son, therefore brought out a machine close to the first Burns. Chris Abele launched his Knickerbocker the same year and for the same reason. He would remain their direct competitor to the end. Other roaster patents followed in the name of G. W. Hungerford, between 1887 and 1889. That last year, David Fraser, who had arrived from Glasgow in 1886, founded the Hungerford Co. and took over the business. Activity ceased in New York in 1910. Fraser then crossed to the other side of the Hudson, to Jersey City, and continued under the name Fraser Manufacturing Co. The company stopped in 1918.
What it makes
Two principles follow one another. The first is that of the 1882 machine, of which Ukers gives no description of its own. He writes only that it resembles Burns's first machine, whose principle follows. It is a closed iron cylinder, set in a masonry mass above a hearth, coal then being the common fuel. Heat therefore passes mainly by conduction, that is, by direct contact of the bean with the hot wall. Inside works a double screw, one right-handed and one left-handed, attached to the shell. It distributes the coffee from one end of the cylinder to the other during the roast. It also makes it come out at will, without stopping the rotation or taking the machine off the fire. The machine sold from 1897 changes the principle. It is heated from the center: the gas burner works in the middle of the load, and no longer under the drum. An inner cylinder of wire cloth, that is, a fine mesh, holds the bean and keeps it from falling into the flame. It was developed under a patent of Carl H. Duehring, of Hoboken, in 1897. Duehring assigned it to D. B. Fraser, which explains why the machine is sold under the name Fraser and not his. A patent of David B. Fraser himself follows in 1899, with a cooler patent the same year.
The models
The range is not established, and no capacity in kilos per batch is documented. Here is what the registers cited by Ukers attest, and nothing more. Patents for machines to clean and polish coffee in 1871. Patents for machines to clean, scour and polish in 1880 and 1881. A roaster in 1882, said to be close to Burns's first machine. Other roaster patents in the name of G. W. Hungerford, between 1887 and 1889. Finally the center-heated gas machine, put on the market from 1897, and a cooler patented in 1899. Ukers's index classes this gas machine among wholesale roasters, and not among sample machines or home machines. It is the only scale clue available on this company. No commercial model name was found.
What you gain
Center heating shortens the path of the heat. The burner works in the middle of the load, and the bed of beans receives heat through the whole surface of the mesh. A drum heated from below exchanges it only through the portion of wall that the bed touches. Gas is adjusted at the tap and responds within the second, whereas a coal fire takes several minutes to rise or fall. Finally the mesh keeps the coffee out of the flame, so that the spot scorching of direct flame does not occur.
What you lose
A mesh is not a wall. The silverskin, that fine skin that detaches from the bean during roasting, passes through the cloth and falls onto the burner, where it burns. The cloth clogs, and a clogged cloth no longer lets the same heat through from one batch to the next. The radiation of a central flame is very uneven. The layer of beans pressed against the mesh receives much more than the one pressed against the outer shell. Everything then rests on agitation, hence on a full load and steady rotation, failing which nothing corrects this gap.