The Wrung Steel Blocks

Carl Edvard Johansson and the Sub-Micron Gauge Standard

To manufacture interchangeable steel components for the 1894 six-millimeter Mauser carbine at the Carl Gustafs Stads Gevärsfaktori in Eskilstuna, Sweden, machinists had to maintain tens of thousands of dedicated fixed-jaw snap gauges and cylindrical plug gauges. Every single critical dimension—the thickness of a firing pin collar, the width of an extractor claw, the depth of a receiver slot—required its own master gauge, a workshop inspection gauge, and a machine-floor working gauge. Whenever an army ordnance board changed a tolerance by a hundredth of a millimeter, thousands of hardened steel gauges became scrap iron overnight. Worse, when armory inspector Carl Edvard Johansson visited the German Mauser rifle works at Oberndorf in 1894, he saw that fixed calipers stored in different toolrooms drifted apart by microns as their jaws wore down, causing parts machined in one building to jam during final assembly in another.

Traveling home to Eskilstuna on the train, Johansson asked a combinatorial question: instead of forging ten thousand single-purpose steel calipers, could a workshop keep a single wooden box containing a small number of rectangular steel blocks of graduated thicknesses that could be stacked together like children's building bricks to form any desired metric measurement?

Every master machinist in Europe dismissed the idea on physical grounds. If a toolmaker stacked four or five separate steel blocks on top of each other to build up a target length of $43.627\text{ millimeters}$, the microscopic surface peaks, dust particles, and air films trapped at the four interfaces between the blocks would add up to several microns of unpredictable stack-up error, destroying the precision of the sum. For a combinatorial set to work, the gap between two stacked steel blocks had to be smaller than a hundredth of a micron—approaching the spacing of the iron atoms themselves—so that five stacked blocks measured the exact same total length as a single solid bar.

Working at home in the evenings between 1896 and 1899, Johansson converted his wife Margareta's cast-iron Singer treadle sewing machine into a rotary cast-iron lapping table. By charging a flat cast-iron lap plate with fine ferric oxide jeweler's rouge suspended in oil and sliding hardened steel blocks in a figure-eight planetary motion while Margareta worked the foot pedal, he ground the opposing faces of each block until they were parallel and flat to within twenty nanometers ($0.02\text{ microns}$). When two such super-finished steel faces were wiped clean and slid across one another under gentle hand pressure—a motion machinists call "wringing"—the blocks abruptly seized together. Bound by a combination of intermolecular van der Waals forces across the true metal-to-metal contact points and the immense capillary surface tension of a sub-nanometer residual hydrocarbon film, two wrung Johansson blocks required up to thirty atmospheres of tensile pull (over seventy pounds of force on a small gauge face) to tear apart, with an interface error of virtually zero.

Even after solving the wringing surface, Johansson confronted metallurgical instability. Newly quenched high-carbon martensitic steel continues to rearrange its internal crystal lattice at room temperature for years, causing a finished block to grow or shrink spontaneously by microns. Furthermore, steel expands by 11.5 parts per million for every degree Celsius of warming. Granted Swedish Patent No. 17017 in May 1901, Johansson established a strict five-part metrological system that defined twentieth-century manufacturing:

• * Artificial Thermal Seasoning: Blocks cut from Swedish crucible steel were austenitized, quenched to full hardness, and then cycled repeatedly between baths of boiling water ($100^\circ\text{C}$) and chilled brine ($-20^\circ\text{C}$) to relax trapped internal martensitic lattice stresses in days rather than decades.
• * The Four-Series Combinatorial Box (102 Blocks): Series 1 held nine blocks from $1.001\text{ mm}$ to $1.009\text{ mm}$ in $0.001\text{ mm}$ ($1\text{ micron}$) steps; Series 2 held forty-nine blocks from $1.01\text{ mm}$ to $1.49\text{ mm}$ in $0.01\text{ mm}$ steps; Series 3 held nineteen blocks from $0.5\text{ mm}$ to $9.5\text{ mm}$ in $0.5\text{ mm}$ steps; and Series 4 held ten blocks from $10\text{ mm}$ to $100\text{ mm}$ in $10\text{ mm}$ steps—yielding 20,000 distinct combinations in one-micron increments.
• * The Twenty-Degree Celsius Reference Standard: Because French gauges were calibrated at $0^\circ\text{C}$ (melting ice) while British imperial gauges were set at $62^\circ\text{F}$ ($16.67^\circ\text{C}$), Johansson fixed his blocks at exactly $20.0^\circ\text{C}$ ($68^\circ\text{F}$)—near average workshop room temperature, minimizing thermal expansion errors when touched by human hands.
• * Progressive Tolerance Grading: Each block was calibrated optically against the cadmium red spectral wavelength at the International Bureau of Weights and Measures in Sèvres to within $\pm 0.00002\text{ millimeters}$ per millimeter of length.

In 1923, Henry Ford purchased the American rights to Johansson's gauges and installed the Swedish inventor at the River Rouge complex in Dearborn, making his wrung steel blocks the physical foundation of American automotive and wartime aircraft interchangeability.