Dummy surfaces are surfaces that have no material defined so they have no refractive or diffractive power; i.e. they do not alter the path of the rays. A dummy surface may be used as a reference to indicate where a lens or mechanical part will be located once the optical system is manufactured. It might also display important planes such as the principal plane, or the plane containing the entrance or exit pupils (see the Knowledgebase article “https://support.zemax.com/hc/en-us/articles/1500005576762-How-to-place-dummy-surfaces-at-the-front-and-rear-principal-planes”). Dummy planes are also useful for analyzing a system with the Physical Optics Propagation analysis feature (see the Knowledgebase article “Using Physical Optics Propagation (POP)”).

Hi, Allie.

I have a question. Now that dummy surfaces don’t alter the path of the rays, why do they have a significant influence on the POP results, if I add a dummy surface at different locations?  

Looking forward to your reply!

Best,

YJ

Hi @yingjie.liu 

When you add the dummy surfaces to your system, are you also adding additional thickness? For example, if the distance between two lenses is 10 lens units and you add a dummy surface between the lenses with an additional 10 lens units of thickness, then you may see changes in the POP results. In this case, the dummy surfaces are not bending any rays, but the rays are experiencing a longer path length. 

As a sanity-check, I built a quick model. In this model, I have inserted three dummy surfaces. In each configuration, only one dummy surface is active at a time. You will note in the Merit Function that - no matter where the dummy surface is placed - the POP results are consistent. Even if you completely removed the dummy surfaces, the results would remain the same.

If this explanation doesn’t align with what you’re seeing, feel free to submit a sample file on this thread or via support. 

Hi Allie，

In the sample, there are two dummy surfaces, 1 and 2. The coupling efficiency was calculated to be: ( Figure 1)

system 0.647904, receiver 0.395431, coupling 0.256201.

Then I inserted dummy surface 0 before Surface 2 in the LDE (both dummy 1 and dummy 2 reserved), the result is as follows: (Figure 2)

system 0.732881, receiver 0.350966, coupling 0.257217.

Then I deleted dummy surface 0 and 2 (dummy 1 reserved), the result is as follows: (Figure 3)

system 0.747010, receiver 0.392017, coupling 0.292841

Apart from the variation of the coupling efficiency, the phase on the image surface looks quite different.

I don’t think an additional path was added.

Looking forward to your reply！

Best,

YJ