Hi Jacob,

I think the normal is expressed as a direction cosine because if you select a different Type in the settings, these columns disappear. The Z-normal must be the cosine of the angle of the normal with the Z axis. The other two components of the normal direction cosine are given by the other two columns (X-normal, and Y-normal). Its not a proof, but you can check that in every row, the square root of the sum of the squared values add to unity.

I hope this helps.

Take care,

David

Hi @David.Nguyen ,

Thanks for the reply and bring in more clarity. I see now that the RANZ() value and the Z- Normal column values are same, so it must be the Z- direction Cosine.

Now the second part of my question. All the values in the Z- Normal column are between 0 and -1, that means all the angles between Z-axis and the Surface normal (at the ray surface interface) is an angle between 90 and 180 degrees. But in the design there is a situation where the ray is travelling from left to right, upwards and meets a negative radius surface at the top half of the lens and is refracted. Even in this situation the RANZ() value is negative. 

Am I understanding the fundamentals wrong here. Sorry i can not attach the Zemax files. 

Hi Jacob,

If the Z direction cosine of the normal is close to minus one. Then, the normal is close to being parallel with the Z axis but in opposite direction. So, the situation is more like I tried to depict below:

Does that make sense? Notice the angle the normal makes with the ray is the same either way.

Take care,

David

Hi @David.Nguyen ,

You are right. The way you define the direction of Normal is the right one. 

I defined the direction of Normal as a vector moving away from the centre of the curvature. 

Thanks again.