In this work the effect of primary nozzle positions relative to the mixing section in an ejector on the performance of the air ejector are investigated. Air jet ejector and four types of nozzles are designed and constructed. Two nozzles are circular and the other two are squared. Then an apparatus is designed and constructed to achieve the ejector performance (enhancing the entrainment ratio). Due to the rapid decreasing of air flow rate coming from air reservoir of the compressor to the ejector, a data acquisition system is used to record different measurements such as pressure, temperature and velocity at different locations on the test apparatus. The convergent nozzle can be moved to study the effect of nozzle positions on the ejector performance. Tests are carried out at three different positions for each nozzle, the distance between it was equal. At the 1st position, the exit section of the primary nozzle is located just at the inlet plane of the mixing section of the ejector. At the 2nd position, the exit section of the primary nozzle is shifted 8.5 mm before the inlet plane of the mixing section. For the 3rd position, the exit section of the primary nozzle is shifted 17 mm before the inlet plane of the mixing section. From detailed analysis of experimental results, it is found that: The performance of the ejector is affected by the nozzle position, and the optimum position which achieves a maximum entrainment ratio for all nozzles is at position (1), (nearest position to mixing section of the ejector. But the entrainment ratio increases when the nozzle is placed at position (2) in case of the small squared nozzle. Position (3) is the worst position for all nozzle and achieves minimum entrainment ratio. Regardless the position of the primary nozzle, there is a strong relation between the nozzle exit area and the performance of the ejector, i.e. decreasing of primary nozzle exit area increases ejector entrainment ratio.