NiCr micron-resistor was designed and prepared by magnetron sputtering and lithography on the substrate of silicon with different powers. It is found that there exists a big gap in the TCR between the annealed group and the un-annealed group. A series of tests were made to figure out the reasons lying behind the gap in the TCR between the annealed group and the un-annealed group. UV reflection results show that there is no increase in the concentration of free electrons after annealing. However, the data obtained from XRD reveal that the annealing does not have an obvious influence on the strain of thin films, but really increases the grain size of thin films. Therefore, the grain boundary scattering plays a dominant role in explaining the obvious difference in the TCR. Finally through appropriate methods, a micron-resistor for heating-up with a low TCR value was obtained. NiCr micron-resistor was designed and prepared by magnetron sputtering and lithography on the substrate of silicon with different powers. It is found that there exists a big gap in the TCR between the annealed group and the un-annealed group. A series of tests were made to figure out the reasons lying behind the gap in the TCR between the annealed group and the un-annealed group. However, the data obtained from XRD reveal that the annealing does not have obvious effects on the strain of thin films, but really increase the grain size of thin films, but really increases the grain size of thin films, but really increases the grain size of thin films a micron-resistor for heating-up with a low TCR value was obtained.