Can you take the sin terms out of Snell's Law when dealing with angles below 10 degrees? so Snell's Law would be become n1[tex]\theta[/tex]1=n2[tex]\theta[/tex]2
That would be a fairly good approximation as long as the angles stay within that range. (Do a few examples to check that it gives you sufficient accuracy for your needs.) Be sure to measure the angles in radians!