If you do not have a Faraday-type incident beam current meter on your SEM, use the following procedure to set up a calibrated, reproducible beam current when doing EDS. Of course, you won't know the actual vaiue of the beam current in picoamps, but at least you'll know you can set the same value from sample to sample and day to day.

Mount a small, clean metal foil, 2x3 mm, aluminum or copper, onto the edge of each stub that the samples and standards are mounted on using colloidal graphite paint or double stick tape (after carbon or other evaporative conductive coatings have been applied to sample). Run a little colloidal graphite around an edge or two to provide additional adhesive support and a conductive bridge between the foil and the stub. Before taking sample or standards spectra, move the electron beam to the foil and as you accumulate its spectrum, set a region of interest, or window, on the principle K-shell peak of the aluminum or copper spectrum.