The Compass Einstein Helped Build: How a Spinning Wheel Finds True North

When Albert Einstein was four or five years old, his father showed him a pocket compass. The boy was transfixed. Nothing touched the needle, yet it swung around and pointed the same way every time, as if it knew something. Sixty years later he still remembered the moment. “This experience made a deep and lasting impression upon me,” he wrote in his Autobiographical Notes. “Something deeply hidden had to be behind things.”

What almost no biography mentions is that the grown-up Einstein spent a surprising slice of his most productive decade on a compass with no magnetic needle at all. Between 1915 and 1926 he served as an expert witness in a bitter international patent fight over the gyrocompass, changed his mind halfway through, befriended the inventor whose patent he had first doubted, and ended up sharing a patent on an improved design that paid him royalties for years. The instrument at the center of all this finds true north using nothing but a fast-spinning wheel, gravity and the turning of the Earth itself.

Why steel ships broke the magnetic compass

For about 400 years the magnetic compass had been the most trusted object on a ship’s bridge. Then, in the second half of the 19th century, wooden hulls gave way to iron and steel, and the compass started to lie. A steel hull is a giant lump of magnetizable metal. It bends the Earth’s weak magnetic field around itself, and the error changes as the ship turns, as it rolls, and even as it ages. Add electric motors, dynamos and gun turrets that swing tons of steel around, and a navigator could be several degrees off without knowing it.

There was a second, older problem. A magnetic needle points to magnetic north, which wanders and sits well away from the geographic pole, so every reading has to be corrected for local “variation.” For a navy trying to steer battleships in formation, aim long-range guns or run a submarine sealed inside a steel tube, a compass that needed constant correction was a serious liability. What engineers wanted was a compass that ignored magnetism completely and pointed at true north, the point the Earth actually spins around.

Foucault names the gyroscope

The idea that a spinning wheel could do this goes back to the French physicist Léon Foucault. In 1851 he had famously used a long pendulum in the Panthéon to show the Earth turning (we told that story in Come See the Earth Turn). A year later he tried a cleaner proof. He took a heavy, carefully balanced wheel, mounted it in rings so that its axis was free to point anywhere, set it spinning, and watched. A spinning wheel holds its direction in space, so as the Earth rotated underneath it, the wheel’s axis appeared to drift slowly across the room. Foucault named the device from the Greek words for “rotation” and “to see”: a gyroscope, an instrument for seeing the Earth turn.

His rotor only spun fast enough for about eight to ten minutes before friction slowed it, and there were no electric motors to keep it going. Even so, Foucault laid out the theory behind a gyroscope that was not completely free: one held level should swing around and settle pointing north. It would take half a century, and the electric motor, to make that work on a rolling ship.

Spinning things on this site have a habit of misbehaving in delightful ways, from the coin that screams toward infinity on Euler’s Disk to the wing nut that flipped over and over in orbit. The gyrocompass is the opposite case: the same physics of angular momentum, harnessed so carefully that it becomes the steadiest thing on a ship.

How a spinning wheel finds north

Start with an ordinary gyroscope. A fast-spinning wheel mounted in gimbals keeps its axis pointed at a fixed direction relative to the stars. But the Earth turns 15 degrees every hour, so to someone standing on the deck, a free gyroscope seems to drift, completing a full apparent circle every 24 hours. That is useful for holding a course for a few minutes (it is how an aircraft’s directional gyro works), but it has to be reset by hand, so it is not a compass.

The trick is to take away some of the gyroscope’s freedom. Hang a weight below the spinning wheel so that gravity tries to keep its axis level with the horizon. Now, as the Earth carries the gyroscope around, the horizon itself tilts relative to the stars, and gravity keeps pulling the axis back toward level. A spinning wheel does not respond to a push the way a still one does; it precesses, turning at right angles to the force. The result is that the axis is steadily nudged sideways until it lies in the north–south line. Only there, pointed along the meridian, does the Earth’s rotation stop tipping it, so north is the one position where the wheel can sit still relative to the ground.

Left like that, the axis overshoots and swings back and forth across north like a slow pendulum, so real gyrocompasses add damping to make it settle. They also have honest weaknesses. The north-seeking force comes from the part of the Earth’s spin that keeps tilting the local horizon, and that part shrinks toward the poles, so gyrocompasses get sluggish and unreliable at very high latitudes. A ship’s own speed and turns add small errors that have to be corrected. In 1923 the engineer Max Schuler showed that a compass tuned to swing with a period of 84.4 minutes, the time it would take a satellite to orbit at sea level, would shrug off the sideways jolts of a moving ship. That “Schuler tuning” still underlies inertial navigation today.

An art historian aims for the North Pole

The man who made the first practical gyrocompass was not trained as an engineer. Hermann Anschütz-Kaempfe, born in 1872, was an art historian with a fortune and an obsession: he wanted to reach the North Pole by submarine, traveling under the ice. A magnetic compass would be close to useless for that, sealed inside a steel hull and steering toward the region where magnetic and true north part ways. So he set out to build a compass that did not care about magnetism.

Others had tried. The Dutch inventor Marinus van den Bos patented a gyroscopic compass in 1885 but never produced a working model. Lord Kelvin had proposed a gyrostat to the Royal Navy, and in 1889 the French engineer Arthur Krebs put an electric motor on a marine gyroscope so that the French submarine Gymnote could hold a straight line underwater for hours. None of these was a true north-seeking compass. Anschütz-Kaempfe founded the firm Anschütz & Co. in Kiel in 1905, hired his cousin Max Schuler the following year, and by 1906 had a working instrument. After successful trials in 1908, the Imperial German Navy adopted it, and the gyrocompass went into series production.

Sperry’s American rival

Those 1908 trials caught the attention of Elmer Ambrose Sperry, a prolific American inventor who had already built electric streetcars, mining machinery and an electric car. Sperry had become interested in gyroscopes after a bout of seasickness on an Atlantic crossing in 1898, and had designed a giant gyroscopic stabilizer to stop ships from rolling. In 1908 he turned to the compass problem, and in 1910 he founded the Sperry Gyroscope Company in Brooklyn.

His first navigational gyrocompass went to sea for testing aboard the battleship USS Delaware, and the Navy adopted the design in 1911. The first serial-production unit, number 101, was installed on the battleship USS Utah in 1913. Sperry compasses soon went into American, British, French, Italian and Russian warships, and during World War I the Navy coupled them to the steering gear to create “Metal Mike,” a gyroscope-guided mechanical helmsman and one of the first ship autopilots.

Anschütz and Sperry were now competitors in one of the most lucrative markets of the arms race before World War I, and they were bound to collide. They did in May 1914, when Sperry sold a gyrocompass to the German navy itself.

Einstein in the witness chair

Anschütz sued for patent infringement before the Royal District Court in Berlin. Sperry’s defense was simple and dangerous: Anschütz’s patent, he argued, added nothing to the old van den Bos patent of 1885 and was therefore void. A second question was whether Sperry’s method of damping the compass’s swing was one that Anschütz had already patented. The court asked both sides to agree on a neutral expert. The man they settled on was a recently arrived professor in Berlin named Albert Einstein.

The timing could hardly have been worse for him. In early 1915 Einstein was in the middle of the struggle to turn his special theory of relativity into a theory of gravity. His marriage had just fallen apart, his wife had moved to Zurich with their two sons, and wartime food shortages were aggravating painful stomach trouble. When he first appeared in court on January 5, 1915, he was by one account poorly prepared and contradicted himself several times. The court asked for a written report instead, at a fee of 1,000 marks. In it, Einstein concluded that the Anschütz patent did not really improve on van den Bos.

Neither Anschütz nor the court was convinced. After a hearing in March, the judges asked Einstein to inspect the Sperry compass itself and compare all three designs. In July he ran tests on the Sperry instrument, and the hands-on work changed his mind. In a report dated August 7, 1915, he wrote that the way Anschütz kept his compass level did constitute a real improvement on van den Bos, and that Sperry’s damping method was the one described in an earlier Anschütz patent. In November the court barred Sperry from making and selling gyrocompasses that used the patented method. A few weeks later, on November 25, 1915, Einstein presented the field equations of general relativity to the Prussian Academy.

From expert witness to co-inventor

The lawsuit turned into a friendship. Over the next decade Einstein visited Kiel again and again, at first staying at the Anschütz family home and later in a small apartment at the compass factory near the mouth of the Schwentine River, which he fondly called his “Diogenes barrel.” It came with a view of the water and the use of a sailboat, and Einstein spent long hours sailing on the Kiel Fjord, sometimes with Anschütz and sometimes alone. Between outings, the two men traded letters and ideas about the remaining problems of the compass.

The fruit of that collaboration was a new kind of gyrocompass built around a sealed sphere. Inside it, two gyroscopes were mounted side by side and coupled so that they turned in opposite directions when the ship lurched, canceling much of the error caused by acceleration. The whole sphere floated in an electrically conducting liquid, which also carried current to drive the gyro motors. Because the liquid could not quite carry the sphere’s full weight, Anschütz picked up an idea of Einstein’s and surrounded it with a ring of electromagnets that held it up and kept it centered without touching. Einstein had a substantial share in the resulting 1922 patent and received royalties from it until 1938, at times earning more from the compass than from his professorship, according to the Kiel institutions that preserve the story.

There is a neat irony in that design. The boy who had been spellbound by a magnet’s invisible pull grew up to help build a compass that owes nothing to the Earth’s magnetism, yet uses carefully shaped magnetic fields to float its heart in place. Einstein last visited Kiel in 1926 and left Germany for good in the early 1930s, but he kept writing to Anschütz-Kaempfe until the inventor’s death in 1931. In 2021 Kiel unveiled a bronze memorial showing the two men side by side, their hands resting on a gyrocompass sphere.

A no-hands flight over France

Sperry, meanwhile, took gyroscopes into the air. In 1913, working with his son Lawrence, he built a gyroscopic stabilizer that could work an airplane’s controls through servomotors. In June 1914 the Sperrys entered the Aero Club of France’s competition for a safer airplane, and Lawrence won it by flying past the judges with his hands off the controls while the gyroscopes kept the aircraft level.

Lawrence went on to develop the autopilot and the artificial horizon, and his father added the gyro turn indicator in 1917. With a directional gyro and a gyro horizon, these became the core instrument panel of nearly every airplane that followed. Lawrence himself died in December 1923 when the small plane he had designed went down in fog over the English Channel. Elmer Sperry died in 1930. Through a long chain of mergers, the companies he founded eventually became part of Unisys, the computer company, in 1986.

Still on the bridge

Gyrocompasses steered the warships and liners of World War II, and the big mechanical “master gyro,” tucked deep inside a ship where it was safest from shock, fed heading information to repeaters around the ship, including on the bridge wings, where officers took bearings through sighting vanes like the one in the photo at the top of this story.

The gyrocompass is not a museum piece. International rules for merchant shipping (the SOLAS convention) still require every ship of 500 gross tonnage and up to carry “a gyro compass, or other means” of finding its heading by non-magnetic means, with repeaters for the helmsman and for taking bearings. Many of those “other means” no longer spin at all. Ring laser and fiber-optic gyroscopes send light around a closed loop in both directions and measure the tiny difference that rotation makes, including the Earth’s own, the same signal Foucault’s wheel was built to feel, now read with no moving parts. Modern fiber-optic gyrocompasses can also help a ship notice when its GPS position is being jammed or spoofed.

The company Anschütz-Kaempfe founded still builds gyrocompasses in Kiel, and the problem it solved belongs to a long line of navigation puzzles, alongside the clock that let sailors find their longitude at sea. Knowing which way you are pointing turned out to be just as hard as knowing where you are, and the answer came not from a better magnet but from a wheel that feels the planet turn.

Foucault built his gyroscope to see the Earth rotate, and it ran down after ten minutes. Today’s gyrocompasses feel that rotation continuously, on every large ship at sea, and quietly convert it into one word: north. Somewhere in that chain is the boy with his father’s compass, who decided that something deeply hidden had to be behind things, and then, between court dates and sailing trips, helped hide a little more of it inside a floating sphere.

Further reading