#768 1864 · William Sellers / Franklin Institute · Manufacturing / mechanical engineering

Sellers couldn't convince rival machine shops to share one screw thread, so he got the Army and the railroads to force it on them instead

问题

Every American machine shop cut screw threads to its own incompatible pattern

背景

By the 1860s, American machine shops each cut screw threads to their own house pattern — bolts, nuts and fittings made in one shop routinely would not fit hardware made in another, even for identical-looking parts. Britain had already solved this at home with Joseph Whitworth's 1841 thread standard, adopted because Whitworth was the country's dominant machine-tool maker and could set the pattern the rest of the trade fell in around; the United States had no equivalent dominant firm and no equivalent standard.

The available path was the one Whitworth had taken: publish a better thread design and let manufacturers adopt it voluntarily because it was technically superior. But voluntary adoption asked every individual shop, each with tooling and inventory already sunk into its own incompatible pattern, to bear a retooling cost with no guarantee any competitor would follow — the same trap that stalls any standard whose payoff only arrives once everyone else has already adopted it too, and Philadelphia machinist William Sellers had no dominant market position of his own to force the issue the way Whitworth had.

换别人会怎么做

The available model was Whitworth's own: publish a better thread design and let manufacturers adopt it voluntarily because it's technically superior. That path worked in Britain only because Whitworth was already the dominant machine-tool maker and the trade fell in around his pattern almost by default. Sellers had no equivalent market position, and every American shop already had tooling and inventory sunk into its own incompatible pattern, with no individual incentive to be the first to retool for a standard no one else had committed to yet.

他们看到了什么

Sellers saw that the leverage he needed wasn't on the manufacturing side at all — it was on the buying side. A handful of buyers, the Army and Navy ordnance system and the country's largest railroads, already controlled enough of the national market for machined fasteners that they didn't need to persuade anyone: they could simply specify the thread they wanted in their own contracts, and every supplier who wanted that business would have no real choice but to comply. Standardizing the demand side would standardize the supply side automatically, without needing a single competing manufacturer to volunteer first.

那一手

In April 1864, Sellers presented a new thread system to Philadelphia's Franklin Institute, simplifying Whitworth's 55-degree profile to a 60-degree thread that was easier for ordinary machinists to cut accurately. As the Institute's own sitting president, Sellers had a special committee endorse his design that December — then, rather than stopping at endorsement, had the Institute directly lobby the U.S. Army, the Navy's Bureau of Steam Engineering, and the master mechanics of America's largest railroads to specify the new thread in their own contracts and equipment.

为什么管用

Once the Franklin Institute — whose sitting president was Sellers himself — endorsed the design and then lobbied the Army, Navy and major railroads to specify it, any machine shop supplying those buyers had to cut to the Sellers thread or lose the contract. Because those few buyers represented such a large share of the total market, shops that retooled to serve them found it cheaper to use the same tooling for all their other customers rather than maintain two incompatible systems side by side — so the standard radiated outward from a small number of mandatory buyers to the entire industry, riding on suppliers' own cost incentive rather than on any collective agreement among competitors.

值了多少

The Pennsylvania Railroad became the first major carrier to adopt the Sellers system for its rolling stock and infrastructure fasteners in 1869; government arsenals and the largest railroads followed within a few years. Because these buyers alone represented an enormous share of the national market for machined fasteners, any supplier who wanted their contracts had to retool to the Sellers thread — and having retooled, found it cheaper to use the same pattern for every other customer too rather than run two systems in parallel.

什么时候会失灵

This only works when a small number of buyers genuinely control enough of a market's demand that suppliers cannot simply sell elsewhere and ignore the specification — a mandate from a buyer too small to matter changes nothing. It also depends on the standard-setting body's credibility surviving the obvious conflict of interest in its own president pushing his own design through it; had Sellers's thread been technically worse, or had a rival machine-tool maker held comparable buyer leverage of its own, the identical maneuver could just as easily have locked in an inferior standard rather than a better one — the mechanism forces adoption, it does not by itself guarantee quality.

后来呢

By the 1880s the system, known as the United States Standard Screw Thread, had spread from those few mandated buyers to become the de facto national standard as interchangeable-parts machinery — typewriters, locomotives, machine tools — spread across the economy; it remained the basis of American screw threads until the mid-20th century, when it was folded into the Unified Thread Standard adopted jointly with Britain and Canada. In 2005, ASME designated it a National Historic Mechanical Engineering Landmark, marked at the Franklin Institute itself.

资料来源

[1]The United States Standard Screw Threads (Engineering Landmark #234)American Society of Mechanical Engineers, 2005asme.org
[2]At the Turn of a Screw: William Sellers, the Franklin Institute, and a Standard American ThreadTechnology and Culture (Society for the History of Technology / Johns Hopkins University Press), Vol. 10, No. 1, 1969jstor.org