Lamp Inventors 1880-1940:
Ductile Tungsten Filament

William D. Coolidge, 1927
recipient of the Edison Medal
from the American Institute of Electrical Engineers,
"For his contributions to the incandescent electric lighting
and the x-ray arts."

"I remember this circumstance very well because of the excitement and surprise and incredulity which he manifested at the time. He asked me over and over again what it was. I told him that it was pure tungsten wire, only to have his question repeated again and again."
-- William D. Coolidge recounting a conversation with German lamp inventor Fritz Blau, 1909

William D. Coolidge began his career at General Electric's Research Laboratory in September 1905. Hired away from the Massachusetts Institute of Technology by lab director Willis Whitney, Coolidge stipulated that half of his time would be spent on his own research with the use of company equipment. What he accomplished on company time made this seem a bargain.

Coolidge's first assignment was to investigate why tantalum lamp filaments quickly broke when operated on alternating current. This work led him into the hunt for a metal filament lamp to replace the carbon filament developed by Edison. Though practical, carbon lamps was not very energy-efficient (about 3 lumens per watt), and European inventors had been especially active in searching for better filaments. Lamps using tantalum had been devised in Europe and gave 5 lumens per watt (lpw), but worked well only on direct current.

In 1904 several European inventors almost simultaneously developed filaments from the metal tungsten. These worked well on both AC and DC, and gave 8 lpw. However, tungsten was difficult to work -- filaments were made with a pressing technique called "sintering." The filaments were brittle and could not be bent once formed, so they were referred to as "non-ductile" filaments. They required a complex mounting structure with several filaments placed in electrical series. Coolidge began investigating how he might improve tungsten lamps by making a bendable or "ductile" wire.

In 1909 he came up with the answer. By putting an ingot of sintered tungsten through a series of hot swagings and drawings through successively smaller dies, bendable wire of many diameters could be made. When Coolidge demonstrated lamps made with his tungsten wire to European researchers who had also been investigating the problem, he got reactions like the one noted above.

GE sold Coolidge's lamp under the trade name "Mazda" beginning in 1910. Giving 10 lpw, Coolidge's lamp returned GE to a position of market strength that had been in question since Edison's patents had started to expire in the previous decade. In November 1932, Coolidge succeeded Whitney as director of the lab.