TSTN-049

Calibration Reflector#

Abstract

The Calibration Reflector is mounted on the top end of the TMA and is used to reflect light from the Flatfield Projector onto the Calibration screen. This document describes the reflector and its operation.

Overview#

The Rubin Calibration Reflector is an optic mounted on the top end of the TMA and reflects light from the Flatfield Projector onto the calibration screen. The Flatfield Projector is mounted at the center of the calibration screen on the dome. The reflector is an aspheric optic fabricated out of a block of aluminum. It is mounted on hexapod legs secured to the top end of the camera and adjustable by hand.

_images/refl_overview.png

Fig. 1 Calibration Reflector#

The reflector was designed by Brian Johnson and was manufactured in the NOIRLab instrument shop in Tucson by Ron Harris, Anthony Tache and Anthony Borstadt. The cart was designed by Bill Schoening. The electronics were designed and built by Antanasia Jones. All testing was completed by Patrick Ingraham and Parker Fagrelius.

All drawings, mechanical and electrical, can be found on docushare, here: https://docushare.lsst.org/docushare/dsweb/View/Collection-15468

Design#

The shape of the asphere is very specific, allowing an approximately f/4 beam to illuminate the 10 m-diameter calibration screen with uniform brightness.

_images/refl_optical_presc.png

Fig. 2 Reflector Optical Prescription#

The optic was milled at the NOIRLab instrument shop, using a laser tracker to confirm the shape. The shape was reconfirmed in the lab.

There are three SMR’s attached to the optic that can be used to align it to the optical axis of the TMA. They are symmetrically placed around the Reflector.

There are three cover panels that can be opened and closed using CSC commands. Each panel is connected to a Firgelli Automations Optical Feedback Actuator (part #: FA-OS-35-12-18).

The mirror is attached to the top of the camera via six hexapod legs that extend over the pancake wrap. They are adjustable by hand so the Reflector can be aligned with the optical axis of the TMA. They are held to the structure with restrainer brackets so they can be easily attached to the reflector.

Installation#

Prior to installation, it’s necessary to attach pairs of the hexapod legs with their respective restrainer bracket. The hexapod legs are not all identical, so care should be taken when pairing the hexapods, which are all labelled.

With the TMA at horizon, a manlift on L8 can be used to install the three hexapod leg pairs, attaching the feet of each hexapod as well as the restrainer bracket.

The Reflector can now be lifted with the overhead crane on its triangle frame and brought into place with the help of pins on the hexapod leg assemblies. The triangle frame can then be bolted to the hexapod legs.

The reflector must be removed whenever the camera is removed.

Cart#

Since the Reflector will need to be removed every time the camera is removed, a special cart was constructed for it. When the Reflector is to be removed, first unbolt the triangle frame and lower the reflector onto the cart. The hexapod legs can now be removed and also placed on the cart.

The cart is stored on the 3rd floor of the observatory.

_images/reflector_cart.png

Fig. 3 Reflector cart, including a place for the triangle frame and hexapod legs.#

Alignment#

This section is still being developed.

Electronics#

The electronics to open/close the covers are located in the electrical cabinet at the top end of the TMA (TEA-Cabinet 03). The rack is powered and connected to the TEA-AS-02 Port 2, with PDU port 3.

They are connected to the linear actuators via a single power cable.

Each of the three cover panels is attached to a linear actuator. These are all commanded together using Double-Pole Double-Throw (DPDT) switches.

CIO0 is connected to SSR1 and SSR2 to open the reflector. CIO1 is connected to SSR3 and SSR4 to close the reflector.

The switches are commanded using a LabJackT4, which can be found at tea-reflector.cp.lsst.org

_images/refl_electronics.png

Fig. 4 Electronics for controlling the Reflector covers.#

Remote Control#

If the LabJack and CSC are not working to control the Reflector covers, there is a backup approach. The power cable can be disconnected from the electronics cabinet and attached to the Linear Actuator remote controller box, which requires 12V power (see setup below).

_images/refl_remote_setup.png

Fig. 5 Setup for using the remote control for the Reflector cover actuators.#

When this controller is connected to the Reflector, you should be able to use the small handheld remote (like for a garage) to open and close the reflector. These parts are connected to the triangle frame if needed.

Control Software#

The Reflector runs off of its own CSC, MTReflector. Github: https://github.com/lsst-ts/ts_mtreflector XML: https://ts-xml.lsst.io/sal_interfaces/MTReflector.html

There are essentially only two functional commands: open and close.

If you need to access the Reflector directly through the LabJack, you can use the Kipling App. There is an Intel NUC in the laser cabinet with Kipling already installed, which you can access via Microsoft Remote Desktop at laser-powermonitor.cp.lsst.org (139.229.168.143).

Evaluation#

Before shipping to Tucson, we measured the Reflectivity and Scattering properties of the Reflector. The average reflectance is about 80%, where fresh aluminum can have reflectance up to 91% on average. The higher levels of scattering are perhaps not surprising with the manufacturing of the optics

_images/reflectance_plot.png

Fig. 6 Measurement of the reflectance and scattering properties of the Reflector while in Tucson.#

_images/reflectivity_drawing.png

Fig. 7 Location of the measurements from the plot above#

The flatness of the optic, or rather its consistency with the desired asphere defined above, was measured during and after manufacturing. These measurements were made using a Hexagon Laser Tracker. We carefully touched the surface of the Reflector with a 1-inch retro reflector to get these measurements. The plot below shows the results from measurements made in 2018 and then in 2023. They match quite well, although the measurement errors seem to increase in 2023. It shows that the shape of the milled aluminum piece follows that of the asphere within approximately 10 µm, and potentially within 1 µm based on the initial measurements.

_images/lasertracker_reflector.png

Fig. 8 Measurement of the Reflector relative to the definition of the asphere using the Laser Tracker#

Maintenance#

The reflector cover actuators are exercised daily during the Daily CalSys Checkout. Monthly, the mirror should be inspected and possibly cleaned.