I am interested to know how water was delivered into a boiler for a stationary steam engine such as in the O scale sawmill.

Was water gravity fed from a tank or pumped in by hand from a nearby water container??

And if it was gravity fed from a tank how was the tank filled ??

I have found some information on steam donkeys water supply but it also not clear how the water gets into the boiler.

Any pictures or drawings would be good.

Thanks in advance.



  • John,

    The most common method to feed water into small (i.e.: backwoods type) stationary and portable boilers was the the injector, similar to that used on locomotives. These took the form of both lifting type (water up to several feet below the injector) and gravity-flooded. See:

    There were many variations in design over the years and the nozzle axis could be oriented either horizontally or vertically to suit the design intent. Probably best to do some internet research for photos or reprinted publication pages, etc., if you wish to have the correct injector type and details with respect to the era you are modelling.

    Larger stationary process and power steam boilers typically used reciprocating steam-driven feed pumps which were also used on some locomotives. Modern stationary boilers are typically equipped with electric drive feed water pumps.

    This is a big topic if you want to dig into details and strive for fidelity but I hope this very brief introduction helps.

    Good luck!
  • Thanks for the info Brian.

    I was vaguely aware of a pressure injectors being used but I would like to know how do they get water to the injector.

    I read that some injectors can pump/raise there own water, was this the case circa 1899 or should we modelling a small water tank/reservoir with a hose or pipe leading to the injector/boiler.

    Thanks, John.
  • John,

    Injector lifting properties were generally understood in the 1860's (see below for more details) so for your time frame you could avoid modeling the supply tank if that best suited your scene layout (example: supply tank under the floor). However, for water supply assurance/reliability there would always be some type of tank or reservoir upstream of the injector so it would not be prototypically correct to depict an underground or overhead water supply pipe directly from, say, a well pump to the injector.

    Gravity head injectors rely on a small (say, 2 feet + feet equivalent of piping pressure drop, or more) minimum head on the water supply tank but more is better, within the bounds of practical reason.

    Lifting injectors can lift over 20 feet in extreme cases with the theoretical flow cutoff limit being the head equivalent of atmospheric pressure assuming a perfect vacuum being generated in the combining cone. However, the greater the lift, the lower the efficiency. Example: Feed water at 50°F - at 0 ft. head, efficiency = 100%; at 15 ft. head efficiency ~ 82%. Feed water at 80°F - at 0 ft. head, efficiency = 100%; at 15 ft. head efficiency ~ 75%.

    As a point of detail note that boiler feed injectors rely on efficient condensation of the steam in combining cone so performance the diminishes as the feed water temperature increases. At 100 psig, for example: Feed water at 50°F, efficiency = 100%; at 120°F, efficiency ~ 80%; at 130°F, efficiency ~ 0%; illustrating feed cut-off temperature varying inversely with boiler pressure due to deterioration of the vacuum developed by the combining cone. Consequently even lifting injectors benefit from an elevated water supply if the feed water is warm or condensate is returned to the hot well/feed tank.

    At the expense of being long-winded, a bit on source material and history if you want to to do more research:
    The best injector reference book I have come across is "Injectors: Their Theory, Construction and Working" by W. W. F. Pullen first published in 1893 (The Technical Publishing Company Ltd.; reprinted by Camden; ISBN: 0 9519367 5 1)

    "Locomotive Injectors" by "The Inspector" (TEE Publishing; ISBN:1 85761 083 0) provides a comprehensive overview targeted at the model engineering community and "The Mode Injector" by Ted Crawford (Australian Model Engineering; ISBN: -) describes the practical construction of model injectors.

    According to Pullen: M. Jacques Gifford in France manufactured the first injector for commercial purposes in 1858. This led to fairly rapid adoption by others and the first published reference to lifting injectors on record appears to be dated January 1860 by the Institute of Mechanical Engineers, London. In the ensuing 20 to 30 years there were a large number of improvements designed, patented and produced for a wide range of applications in all the major industrial countries which firmly established the device as an economical and very dependable means of supplying water to steam boilers. Subsequent ongoing improvements added to the sophistication and application range and helped simplify operation.
  • Brian.

    Thank you for the detailed information, I have found some pictures on the net that show what I need, thanks again.

Sign In or Register to comment.