The hypothesis that direct determination of electrospray current would provide a viable method for maintaining spray stability to enable optimal nanospray analysis was tested by building a feedback apparatus capable of reading the current and readjusting the emitter voltage in real time. The apparatus consists of a current-sensing circuit that reads the voltage drop across a resistor located between the high-voltage power supply and the nanospray emitter. A low voltage proportional to the observed current is generated and sent to a data acquisition card. The information is used by a proportional-derivative-integral (PID) algorithm to calculate the magnitude of a low-voltage signal that is used to control the power supply output. Any variation of current across the sensing resistor is thus counteracted by an opposite-direction variation of the high voltage applied to the nanospray emitter. In this way, the apparatus adjusts the emitter voltage to achieve a preset value of current, which it strives to maintain over time in spite of any possible variation of the parameters influencing the spray regime. Preliminary results have shown that the feedback apparatus is capable of establishing and maintaining stable spray for samples that are usually considered challenging in traditional voltage-controlled analysis, such as those consisting of nucleic acid solutions with high salt loads. For these types of samples, the total ion count recorded in current-controlled mode was significantly more stable than that observed in voltage-controlled mode. At the same time, overall signal intensities and signal-to-noise ratios were also significantly improved. Setting the target nanospray current to a predefined value and letting the apparatus reach the target without operator intervention enabled the acquisition of viable data from solutions containing up to 2.5 M ammonium acetate, which are ordinarily difficult by traditional manual tuning. A deeper understanding of the current-voltage relationships for samples of very different compositions is expected to enable one not only to predict the target current that should be used for a certain analysis, but also to devise algorithms to change such target as a function of predictable variations of sample properties and analytical conditions. This will allow for optimal performance to be maintained during on-line gradient chromatography in which the nature of the sprayed solution may vary very widely during the course of the analysis.