Can you explain the overall objective? You are using the sin function to determine whether the Taylor function has converged to the desired precision. If you are using sin, why not just use sin to calculate the sine?