The Binary 2 surface has the same sag shape as an even asphere, but it does not support the use of irregularity operands directly. You could try to have a Binary 2 surface with no thickness or material but with sag parameters exactly matching a subsequent supported surface like Standard or Even Asphere. A pickup solve in this case, though usually discouraged, would be appropriate as the surfaces represent a single physical surface and should change in unison.

Hi Kevin, thank you for the reply. Before I read your response I had tried something similar to what you suggested, but in reverse. However, the change in criterion shown in the tolerancing run output was zero, so it appears that the tolerance I applied was effectively ignored. I used a Zernike Fringe Sag surface with zero thickness (but still with the same lens material) immediately followed by the Binary 2 surface. The Binary 2 surface had pickups for radius, conic, and aspheric terms to the Zernike surface. In the tolerance data editor I used TPAR to apply a tolerance to parameter 25 of the Zernike surface which is Z11 for spherical aberration. So, I tried to apply spherical aberration to the Binary 2 surface, but it appears to have had no effect. Note that by using the Zernike Fringe Sag surface it is easy to compute the coefficient values needed to apply a specific amount of aberration. Tolerancing aspheric or binary coefficients directly does not allow for a straightforward conversion to the amount of error applied (as far as I know). That is why I did not put the Zernike surface after the Binary 2, with pickups to the Binary 2. Any other suggestions?