Failure Modes Discussed in 1702

Thomas Savery patented a steam-powered pump in 1698. This was an important but imperfect step in the early development of steam engines. Savery’s pump was intended for dewatering mines, but it was more practical for other applications. Savery demonstrated the pump to the Royal Society of London in 1699. YouTube hosts several 3D animations and working scale models that are very interesting.

Savery promoted his pump in a 1702 pamphlet. The pamphlet describes the construction and sequence of operation of the pump. A conversation between a fictional miner and the inventor discusses some of the design choices. This conversation contains elements of design for reliability and maintainability.

The miner challenges the inventor with several practical practical problems. He starts by asking if the pump will be fouled when pumping muddy water with bits of stone or coal, and how the boiler will handle dirty water. He asks about deep mineshafts, and about how to install the device in twisting or narrow shafts. The inventor includes a suction strainer, a staged design, and easy disassembly. 

The miner questions if valves will wear out, rendering the pump useless. At this point in history, valves were not standardized, so were likely custom and hand-made. The inventor answers that experience has shown that brass valves improve, or wear-in with use.  

The miner asks whether acidic water in the mine will corrode the vessel. He also asks about the fire in the boiler burning through the heat transfer surface within two to three months. The inventor cites the low temperature of the flame (red instead of white) and experience with copper kettles in sugar manufacture involving sooting and scale.

Savery argues to the miner that the fire for the boiler would be a safety feature in a combustible atmosphere, so the pump would protect the miner and the mine. Savery’s theory was that the combustion air for the boiler would draw fresh air into the mine. This forced ventilation would pull fresh surface air into the mine. 

The pamphlet also contains this important lesson:

"…the experience of ages shows us this to be a most sure rule, allowing for friction, which is larger, the more wheels or parts an engine consisteth of; and, of consequence, the fewer parts or wheels an engine consisteth of the easier it works; so that by barely looking on a pump, if it has more parts or wheels than the common crank-work, you may conclude it worse; if a chain-work or tub-work the same." (pg 52) 

The notable part of this is not the technical content of the answers provided to the miner. The inventor is anticipating failure modes, citing lessons learned from other industries, and using machine design experience. His position is not that repairs will be easy due to simplicity or commonly available parts, but that the design was resistant to failure.  

In practice, the pump was limited by a short lift, so not very useful for mining. Within a few years, the Newcomen engine proved superior. 

This is an 1827 re-publication of the 1702 pamphlet:

Savery, T. (1702, 1827). The miner's friend: or, an engine to raise water by fire... London: S. Crouch.  

https://hdl.handle.net/2027/mdp.39015021235935