This classic case is treated in the essay Early modern scientific and technical activities: sorting out “science” and “technology” which is available here. This includes a detailed study of the history of the development of mechanical clocks, and particularly the place of scientific work in it. According to the essay
This history, in which a centuries old conventional mechanism is revolutionised, by the advent of an early scientist who applied an advanced mathematical analysis to the problem, is considered a clear example of the distinctively different qualities of science and technology, and of the impact of science on technology.
The scientist in question was Christian Huygens, who calculated how to make it possible for the pendulum of a clock to keep the same time at large angles of swing as it does at small angles, and designed a device to achieve that aim.
The essay reveals that the development of the pendulum clock was not aided by Huygens' mathematical advances, but was the product of technical innovation in a technical environment. Huygens device was not used in the pendulum clocks that were produced. Further, as far as the outcome of the mathematics was concerned:
It’s immediate effect on technical development seems to have been rather negative, in that it was accompanied by the erroneous removal from consideration of the drive in the analysis of timekeeping. This view has endured: the education of present day scientists and engineers conveys that the length of the pendulum determines timekeeping, and it is only specialist horologists who know otherwise.
It turned out that even the specialists had not quantified the effect of the drive on pendulum timekeeping and verified it experimentally. This led to an investigation which is reported as Mechanics of the escapement driven pendulum in Horological Science Newsletter 1 Horological Science Newsletter 2007-4 p.2 . The results are also presented here. They show that, for the seventeenth century pendulum clock, the effect of changes in the drive in practice is about ten times the effect of a change in the angle of swing of the pendulum.
