Svedberg, born the son of a civil engineer in Fleräng, Sweden, was educated at the University of Uppsala, where he obtained his doctorate in 1908. He spent his whole career at the university, becoming a lecturer in physical chemistry in 1907, a professor (1912–49), and finally, in 1949, director of the Institute of Nuclear Chemistry.

In 1924 he introduced the ultracentrifuge as a technique for investigating the molecular weights of very large molecules. In a suspension of particles, there is a tendency for the particles to settle (under the influence of gravity); this is opposed by Brownian motion, i.e., by collision with molecules. The rate of sedimentation depends on the size and weight of the particles, and can be used to measure these.

Svedberg applied this to measuring the sedimentation of proteins in solution, using an ultracentrifuge that generated forces much greater than that of the Earth's gravitational field. Using this, he could measure the molecular weights of proteins and was able to show that these were much higher than originally thought (hemoglobin, for instance, has a molecular weight of about 68,000).

Apart from confirming the claim made by Hermann Staudinger for the existence of giant molecules, Svedberg's invention also settled one further question. The same protein invariably yielded the same weight thus implying that they did have a definite size and composition and were not, as Wilhelm Ostwald had earlier maintained, irregular assemblies of smaller molecules. For his work on the ultracentrifuge Svedberg was awarded the Nobel Prize for chemistry in 1926.

Svedberg was less successful with the inference he drew from his measurements of protein molecular weights. He thought that the molecular weight of egg albumin formed the basic protein unit of which all the other proteins were multiples. Following later research by crystallographers in the 1930s this view was disproved.