Dichroic+Specification

Note and Files
Some notes and files concerning the specification of the Dichroic.

- Some old slides from Steve concerning ISO 10110.

- Dichroic Specification from Judit as of Feb 6th 2012.

- Spread sheet from Steve showing how he calculated the thickness criteria.

- Version two of specification as of Feb 8th 2012

- First quote from ESCO Feb 07 2012

- This is the final quote (4/19/2012) used to order the substrates.

- Total intensity profile as a function of blue cutoff wavelength of dichroic

- study of optimal blue cutoff wavelength of dichroic

IR Dichroic Short-Wave Cutoff
There appears to be some consensus that highly reddened objects are important fot he IR dichroic, so the long-wave cutoff should be 1 micron or so. The short-wave cutoff is controversial. The key drivers for cutoff are:
 * Total Flux. This drives the short-wave cutoff to //shorter// wavelengths as more photons are collected.
 * Atmospheric Dispersion. This drives the short-wave cutoff to //longer// wavelengths as the PSF becomes smaller.
 * Cost. This drives the short-wave cutoff to //longer// wavelengths, as the number of layers in the coating is driven by the width of the reflection band. In making this statement, there is an assumption that we will "take what we can get" for short-wave reflectivity below the cutoff (e.g. DMLP1180L from Thorlabs).
 * Slow WFS operation. This drives the short-wave cutoff to //longer// wavelengths.

To see how important the Total Flux consideration is, lets consider BP Tau, one of the blue-est YSOs that might be observed. Using the calibration from Allen's astrophysical quantities and photometry from the JP11 catalog, we have the following fluxes (photons/m^2/s)
 * Wavelength (microns) || 0.45 || 0.55 || 0.65 || 0.75 || 0.85 || 0.95 ||
 * Flux (Jy) || 0.03 || 0.06 || 0.14 || 0.21 || 0.26 || 0.40 ||
 * QE || 0.8 || 0.95 || 0.95 || 0.85 || 0.65 || 0.3 ||
 * Photons (relative) || 56 || 103 || 201 || 238 || 203 || 126 ||
 * Fraction || 0.06 || 0.11 || 0.21 || 0.26 || 0.22 || 0.14 ||

For the dispersion consideration, we note that adding photons offset by dispersion more than the diffraction-limited FWHM (abut 0.7 arcsec) //loses// rather than gains S/N for the centroid calculation (missing exact details). At a zenith distance of 45 degrees, we get the following image shifts (pressure 80% of sea level): These two tables, taken together, imply that for a 500nm dichroic short-wave cutoff (with respect to a 400nm cutoff), there would be a 0.06 mag sensitivity loss at zenith, and a sensitivity gain at higher airmasses. A 600nm dichroic short-wave cutoff (with respect to a 500nm cutoff) loses an additional 0.12 mags sensitivity at zenith and doesn't cause a sensitivity gain until high airmass.
 * Wavelength (microns) || 0.45 || 0.55 || 0.65 || 0.75 || 0.85 ||
 * Image shift (45 degrees from zenith, arcsec) || 0.95 || 0.50 || 0.26 || 0.1 || 0 ||
 * Image shift (60 degrees from zenith, arcsec) || 1.64 || 0.87 || 0.44 || 0.18 || 0 ||

On the basis of this study MJI suggests specifying a dichroic with a reflection band of 0.55 to 1.0 microns, and a transmission band of 1.15 to 2.4 microns. State that we intend to use some of the transmitted light short-wards of 0.55 microns, and would like to know the predicted curve down to 0.4 microns.

**Coating specifications**
We agreed to send the following coating specifications to vendors in October, 2012 There will be two sets of beam splitters, YSO and VIS, on IR grade low-OH fused silica substrates. The requirements for the coatings are as follows:

__YSO dichroic – 6 pieces:__ AOI 10 degrees Non polarizing Surface 1: In the spectral range of 600 to 1000 nm it should reflect as much as possible, and in the range of 1100 to 2400 nm transmit as much as possible. Surface 2: Antireflection coating for 1100 to 2400 nm We would also like to know the predicted curve down to 400 nm.

__VIS dichroics – 6 pieces:__ AOI 10 degrees Non polarizing Surface 1: In the spectral range of 600 to 1000 nm a gray split: R~20% T~80%, in the 1100 to 2400 nm range it should transmit as much as possible. Surface 2: Antireflection coating for 1100 to 2400 nm We would also like to know the predicted curve down to 400 nm.