In the flow distribution section of a paper machine, known as the head box, water is injected into the fiber suspension (stock) flow through a tee-mixer for more uniform production. This dilution process has two important requirements that must be fulfilled: (1) sufficient mixing so that the dilution flow spreads across the suspension flow and (2) that the injection flow rate not be so large to significantly alter the local head box flow rate. The objective of this research was to find a combination of velocity ratio and tee mixer geometry that lead to the injection flow being well mixed into the stock flow, but at the same time, the injection should not cause the total flow rate to change by more than 1%. Velocity ratios of 0.25, 0.75, 1.33, 1.5 and 2.25 were examined for four different cases of tee mixer geometries using the CFD software Fluent. Two of the cases had added contractions located near the injection point, while the other two cases had a more standard geometry with no added complexities. The pressure drop across the injection point was also measured. Mixing was qualitatively measured by simulating the injection of a passive tracer into the dilution flow. All of the results indicated that the case where the contraction was located after the injection showed the most promising results with quality mixing and lower flow rates. The cases without added contractions showed poor mixing for lower velocity ratios, and for higher velocity ratios, the flow rates were too large. The cases with contractions showed similar mixing, but the outlet flow rates produced were lower when the contraction was located after the injection instead of before it. A velocity ratio of 0.25-0.75 for the mixers with contractions produced acceptable flow rates and sufficient mixing. The simulations also showed that the static pressure for the contraction cases were nearly identical throughout the majority of the pipe. For both contraction cases the pressure drop across the injection increased with increasing injection flow rate. When the contraction was located before the injection, a pressure drop of 16% was calculated. A pressure drop of 18% to 20% across the injection resulted when the contraction was located after the injection.