27: Contriving Machinery
There was at the author's time a great propensity for inventing mechanical devices, but at the same time very little skill. The vast majority of inventions fail from poor design or poor execution and are largely worthless.
As an example, he considers the number of machines seized by regulators that were devised to forge bank notes. Out of 187 contraptions seized by commissioners in an undamaged state, only twelve demonstrated much engineering skill and nine were usable. Of course, it must be stated that such devices are prone to be the work of inferior minds - had their inventors much skill or insight, they could make greater profit putting their skills to honest application.
It takes little but observation of a manufacturing practice to take note of repetitive motions and contrive a way that human labor can be replaced by a device that performs a set action, with set force, at a set pace. But the difficulty begins with understanding the physical and mechanical properties of machinery - to determine how a device can be so contrived such that it performs its action reliably for long periods of time without breaking down.
He takes the example of a design for a printing press that uses a belt-driven steel rod to press a copper plate against paper. Too much force and it will ruin the paper, not enough and it will fail to make an impression. If it is not sufficiently stable the impression will be smeared. If it operates too quickly it cannot be inked between impressions without wasteful drippings and spatter. There are many motions that must be performed precisely in order for the idea of the device to be translated into a workable model.
And then there is the matter of breakage to consider. While a steel rod seems strong, it can bend and warp over time and must be easily replaced so that production is not interrupted waiting for the rod to be replaced. And it is often the very small parts of a machine that go awry.
The framing which supports the mechanism, the devices that load material and unload finished product, and those for adjusting its speed and force, are also among the elements most forgotten by aspiring inventors. In many instances the labor to load, unload, operate, and maintain the device is greater than the labor saved by its employment.
There is also the problem of trials, as machinery is tested in the relatively clean environment of the inventor's workshop, performing small demonstrations rather than continuous operation. Such a machine may fail horribly when installed in its intended environment and operated by an individual less familiar with its workings than its own inventor.
The process of invention requires a great deal of trial and error, operating and adjusting it to perfect its operation. The inventor who dreamt of making an easy fortune is surprised to find himself working long hours in a factory, tinkering with his machine. And the operator, who had been seduced by promises of its advantages, soon recognizes the loss of revenue during the long period in which the device is fretted over and run at low capacity in order to work out the flaws.
Babbage suggests that the whole cost of a new machine is about five times as much as its construction. If a second machine of the same kind is desired, it is somewhat less. After the third or fourth machine of a given kind has been created, additional copies are often produced at less than a fifth of the cost of the original. And so, it takes a sizable operation that requires many devices of the same kind to make much of a profit on their employment.
He also observes that in his time, there is a great deal of "quackery" and ignorance of basic scientific principles on the part of self-proclaimed engineers who are motivated by wealth or fascinated with the beauty of machinery, but who have neither the experience nor the discipline to do the work competently. They are fueled by investment by equally greedy and hasty operators who covet quick profits. It is a much more serious undertaking than many imagine.