Sally Tinkle and others (see Nature 503, 463–464; 2013) highlight the importance of open-source software and data sharing in materials science. But researchers should also be developing free and open-source hardware to radically reduce the costs of their experimental work.

Harnessing open-source methodology will ensure that funding used to develop scientific equipment is spent only once. A return on investment is achieved through digital replication of devices for just the cost of the materials required. This scaled replication saves 90–99% on conventional costs, making more scientific equipment available for research and education (see J. M. Pearce Open-Source Lab, Elsevier; 2013).

Dozens of free open-source designs for lab equipment already exist. For example, the University of Washington in Seattle has produced a magnetic rack for molecular and cell-separation applications that can be fabricated with a three-dimensional printer for less than it can be bought commercially. Even if the device is made only once, it justifies the price of the printer. A hand-held open-source colorimeter built in my department for US$50 matches the performance of similar tools that cost more than $2,000. And the University of Cambridge, UK, has developed a microscope for about $800 from open-source plans, to use instead of conventional equivalents costing up to 100 times as much.

Federal funding agencies could join forces to fund open-source scientific hardware to accelerate its development. A free online database of tested and validated tools should be set up, and governments should give preference to funding such hardware purchases.