A test stand for the examination of flow monitors in a 3-inch pipe was designed, constructed, and evaluated. The calculations necessary for the proper design are based on empirical data and are described in detail. A statistical analysis was used to estimate the error generated from various alternative methods of determining flow. Such estimates included errors in weighing, measuring liquid level, and measuring time and errors resulting from diversions of liquid from the main stream to the collection tank. The volume/time method proved not only to have the least error (+ or - 0.75 percent) of the several options that were studied, but also was amenable to monitoring fluids of unknown densities. The data from the flow test stand which are based on a magnetic flowmeter are traceable to the National Bureau of Standards and are reported to have errors of + or - 0.5 percent or less.