Abstract Issues related to ploughing affecting the performance of the micromilling process have recently been reported in literature. It is well known that there is a minimum chip thickness (hmin) below which ploughing is the main material removal mechanism and no shear occurs. This leads to a non-effective material removal, resulting in a poor surface quality. In order to solve this problem, the minimum chip thickness has been predicted by measuring the cutting forces. However, the determination of h min by means of the cutting force signal, at the instant the chip is being formed, has not been approached. In this article, a method of determining hmin, based upon the signal variation of the cutting forces and the effect of tool radial runout during chip formation is proposed. Carbide micro-endmills without coating were used to cut an aluminium alloy (RSA 6061-T6) sample and the cutting forces were measured using a micro-dynamometer. The microtopography of a microchannel wall was assessed using an optical profiler in order to establish the approximate position where the chip starts to form ( h min ) . As the cut progresses, the force component normal to the feed (FfN) reverses when the undeformed chip thickness is equal to the cutting edge radius (re). Simultaneously, the thrust force increases rapidly, and continues to grow but at a lower rate as FfN increase. The main cutting force and the active force present significant differences to each other. The minimum chip thickness was estimated as 0.3re by means of the behavior of the active force. A small quantity of material left on the wall of the microchannel could be observed in align with the cutting movement together with a deterioration of the surface finish attributed to the increase of re. Results show that the size of the material left is 2 to 4 times greater than h min . Conversely, the quality of the microchannel floor improves as re increase. This shows that there is a balance between h min and re and the effect upon the finish of the channel wall and floor. That should be important for microchannel fabrication in terms of performance of micro-scale heat exchangers depending on fluid viscosity. The topographic analyses of the wall and the images of the chips show an agreement with estimated h min under different cutting conditions and cutting edge radii. The proposed method in this paper not only permits the determination of the minimum chip thickness but also has the advantage of making it possible to estimate the cutting edge radius and to monitor the cutting edge wear.