Although it happened nearly a century ago, the Copper-Cooled Chevrolet affair is still remembered as one of the great blunders in Motor City history.

The GM project that eventually produced the infamous Copper-Cooled Chevrolet was led by Charles F. Kettering, the brilliant engineer who developed the Cadillac electric self-starting system, among his many accomplishments. Now, it should be stated that even then, air cooling was nothing new in the U.S. auto industry. Franklin and Holmes used it successfully, to name two. However, it’s easy to spot the attraction to air cooling for Chevrolet, which was then furiously attempting to compete with Henry Ford and his Model T.

If Chevrolet could eliminate the water pump, radiator, coolant, and associated hardware, it could achieve a great reduction in manufacturing cost—the area where where Ford, up to that point, had enjoyed an insurmountable advantage. This potential leap forward captured the imagination of newly installed Chevrolet president Bill Knudsen, the former Ford manufacturing boss who was given the task of catching the Model T production juggernaut.

The Kettering air-cooling system was, in external appearance, much like Franklin’s, with a sheet metal shroud surrounding the engine and a large fan at the front for circulation (on the Chevy, belt-driven at 1.5 times engine speed). But where the Kettering scheme departed from Franklin and the rest was in its cooling fins. Instead of integral fins cast into the cylinders to release engine heat, Kettering used thin copper sheet, stamped in a fan-fold arrangement and furnace-brazed to the cast-iron cylinders.

This novel setup produced, on paper at least, far greater surface cooling area than cast fins: according to Kettering, more than 5,300 square inches of cooling area on the four-cylinder Chevy, compared to around 3,100 square inches for the larger six-cylinder Franklin. Kettering saw this feature as the critical breakthrough in his air-cooling system, and GM president Pierre S. du Pont seized on the imaginative term “Copper-Cooled” to differentiate the innovation from other air-cooled cars and sell it to the car-buying public.

And so with great enthusiasm and total confidence in Kettering’s unmistakable abilities, GM rushed the Copper-Cooled Model M Chevrolet into the marketplace in January of 1923—only to be met with immediate and crushing failure. Only 759 cars were produced by May, and just 100 of them were actually sold to private customers, then recalled by Chevrolet and eventually destroyed. From the entire short-lived and disastrous exercise, only two cars remain in existence.

So what went wrong? Entire business classes have been taught attempting to sort it out, but the failures can be broken down into two categories: engineering and management.

+   Engineering: The Model M’s small engine, 134 CID compared to 170 CID for the Chevrolet 490 and Ford Model T and rated at 22 hp, was relatively overworked and  prone to detonation under load, reducing the output dramatically. The detonation and power loss cascaded until the car could barely pull itself up a grade. Air cooling can be completely effective, as Volkswagen, Porsche, and others have proven over many decades, but in this instance Kettering and his crew failed to make it work. The engine combustion knock troubles could be directly attributable to the cooling system, or maybe not. But either way, the engineering fell short.

+   Management: Even as Kettering’s GM research staff was declaring it ready for production, the Oakland and Oldsmobile divisions independently tested the Copper-Cooled system and it failed badly. Clearly, the Kettering team had lost their objectivity. Meanwhile, GM upper management placed so much trust in Kettering’s ability and authority that they failed to recognize the Copper-Cooled program’s documented problems, and the Chevrolet division pushed into production an engine that was essentially doomed to failure.

A man of considerable ego by all accounts, Kettering did not take the Copper-Cooled failure well. He threatened to resign from GM and develop the vehicle on his own. But senior executives du Pont and Alfred P. Sloan managed to talk him down, and his successful career at the automaker continued until his retirement in 1947. As part of the ego-soothing process, apparently, a four-cylinder model of GM’s Delco Light rural lighting plant (below) employed a detuned version of the Copper-Cooled design. A number of these powerplants are still around and they bear a passing resemblance to the ill-fated Chevrolet engine.