Regional Manager III

The regional sales manager of a large paper corporation is attempting to detect a difference equal to one deal in the average number of deals closed per week by his employees. He runs a statistical test with the null hypothesis $H_0: \mu = 15$ against his alternative hypothesis $H_a: \mu = 16$. Assuming $\sigma^2=6$, what sample size will ensure that $\alpha=\beta=0.05$? Assume simple random sampling, variance homogeneity, and that the number of deals closed is approximately normally distributed.