Incorporate wavelength limits when GR700XD is used with 'CLEAR' filter on NIRISS

[mperrin]

Issue by josePhoenix Monday Apr 20, 2015 at 17:34 GMT Originally opened as https://github.com/mperrin/webbpsf/issues/65

---

Based on correspondence with André Martel, it sounds like configuring the filter wheel as 'CLEAR' and the pupil wheel as 'GR700XD' is a common use case. In this case, we should limit the calculation to wavelengths that are passed in that mode: namely, from NIRISS.SHORT_WAVELENGTH_MIN to NIRISS.SHORT_WAVELENGTH_MAX (0.5 - 2.35 um).

[mperrin]

Comment by mperrin Monday Apr 20, 2015 at 17:50 GMT

---

Relatedly we would want to have someone on the NIRISS team provide us with lookup tables for the GR700XD displacement vs wavelength, or else all those wavelengths would end up on top of one another in a non-useful blob.

Thus far we've ignored that since the Pandeia team is putting together the monochromatic PSFs themselves, but it would be needed for any attempts at broadband simulations in dispersed modes.

The other complication which will definitely need to be special-cased for GR700XD is that there are multiple orders transmitted through the grism, so there's no longer a one-to-one mapping to the output detector. We could implement a special instrument option to pick order 1 to 3 when the GR700XD is selected.

[mperrin]

Comment by mperrin Monday Apr 20, 2015 at 20:19 GMT

---

As discussed in person but writing this down for the record. The grism wavelength cutoff is imposed not by any optical transmission limit but rather the edge of the detector. See this TIPS/JIM talk on Confluence in particular slides 20-21.

Note incidentally the SHORT_WAVELENGTH_MAX/LONG_WAVELENGTH_MIN cutoff for NIRISS actually translates to STUFF_WE_STUCK_IN_THE_PUPIL_WHEEL_MAX / STUFF_WE_STUCK_IN_THE_FILTER_WHEEL_MIN. It's all on the same detector, very different from the case of NIRCam.