Injection molding small milligram components requires precise metering and high-speed injection. Industrially, metering can be maintained either by using small injection screws (<= 14 mm in diameter) or plungers as small as 3 mm diameter and/or by having very large sprues and runners. Although large sprues and runners increase metering volume, they hide the effect of process parameters on microcomponents. Consequently, knowledge of conventional injection molding is not transferable to microinjection molding, making quality control and optimization difficult. We investigated the filling and postfilling behavior of 25 mm(3) microdumbbell specimens with 289 mm(3) sprue and runner by in-line process monitoring. Design of Experiments were carried out to characterize the effects of process parameters on cavity filling and postfilling behavior. Process characterization indicated that the machine transition from velocity control to pressure control (V-P transition) was around 10 ms: this was comparable to cavity filling time and had a significant effect on cavity filling behavior. Traditional short shot trials cannot provide the correct shot size for small parts, but introduce the effect of holding pressure into cavity filling. Based on a shot size optimization method using only cavity pressure and screw velocity, we eliminated the effect of holding parameters on cavity filling. POLYM. ENG. SCI., 54:1458-1470, 2014. (c) 2013 Society of Plastics Engineers