MODS IMCS Collimator Tip-Tilt-Focus TTF settings

Everytime the instrument is remounted or there is a change to camera focus (see below), the IMCS TTF zeropoints should be remeasured and updated. The procedure for measuring the TTF zeropoints can be found here. This is done at a static setting which represents a sort of median position of the telescope and rotators --- not extremely high or low --- however, instrument flexure depends on both elevation and rotator angle, most significantly on rotator angle, and some data have been collected to examine these trends (see ITs 8589 and 5767 and their attachments.).

The IMCS TTF zeropoints are collected on the mods1 and mods2 computers in ~/Config/MODS#/00IMCS/. This spreadsheet summarizes the zeropoints and differences (to be done). The MODS1 zeropoints appear to be more stable than the MODS2 ones, and in both instruments, the blue channel zeropoints are more stable than the red channel ones which makes sense since the blue channel is straight below the focal plane while the red channel is folded.

MODS Camera Focus Settings

Overview

Although the MODS camera focus settings have remained fairly stable, they should be checked after work has been done on optical components (e.g. the collimators removed for recoating) and at least a couple of times during the year as the temperature swings by 20-30 deg C between summer and winter.

The camera focus settings for each mode (dual and direct imaging, grating and prism) are listed along with the IMCS TTF zeropoints in the bconfig.tab and rconfig.tab files on the mods1 and mods2 computers. The live versions of these files are in the directories under the mods account on mods#: /home/mods/Config/MODS#/ (where # is 1 or 2), and past versions are kept in the subdirectories, 00IMCS. The UT date usually refers to the date when the file copy was made, so in effect, the end of the date range for the focus settings contained in that table.

When a change is made to a focus setting, a comment should be added to the table giving the date, the instrument mode, and the previous and updated value. A row should be added to the corresponding table below. After making any changes to the focus values, you should re-run the IMCS TTF script and update the collimator TTF zeropoints. The reason for this is that focus is not a pure piston term in these off-axis Maksutov-Schmidt f/3 cameras, and changing focus will shift the IMCS TTF zeropoints a little, mostly in the vertical direction.

To measure the focus in imaging modes, a series of images are taken through the sieve mask and either the g_sdss (blue channel) or r_sdss (red channel) filter, using the VFLAT lamp at a setting of 4(but note focus offsets between filters, below). For spectroscopic modes, the pinhole slit and Krypton lamp are used. The spots on the images or spectra are examined - first visually, using minmax/log scaling in ds9 and blinking the set of images or spectra - and then, if time allows, with IRAFs starfocus task. The spot sizes and elongations increase from the center to the edges of the images and, in the spectra at least, the best-focus varies with position along the dispersion and the spatial axes, so a compromise best-focus needs to be reached. On images, I pay particular attention to the spots within the central more finely-spaced grid and, on spectra, to the row of spots above the central row and near the vertical center, since this is close to the sweet-spot for long-slit spectra of compact objects.

* For the imaging modes, laboratory tests indicated that there were focus offsets between filters. The table below (from the MODS1 Laboratory Acceptance Test Report v1.3.2, 2010 July 13, Pogge, R. W. & Team MODS and MODS2 Laboratory Acceptance Test Report v1.2.8, 2013 October 27, Pogge, R. W. & Team MODS) summarizes these. The reference filters are indicated by italics. Best-Focus refers to the best focus in the central region of image, and the offsets in the column labeled Optimal 4' are the focus offsets needed to flatten the PSF across the 4 arcmin field of view.

Focus offsets between filters and between dual and direct modes

Channel	Filter	Best-Focus Offset\[microns]		Optimal 4'
		Direct	Dichroic	Offset[mic] from Best-Focus
MODS1 Blue	SDSS u	+50	+300	+200
	SDSS g	0	+250	+200
MODS1 Red	SDSS r	+50	+50	+150
	SDSS i	0	0	+150
	SDSS z	0	0	+150
MODS2 Blue	SDSS u	0	+260	+200
	SDSS g	0	+260	+200
MODS2 Red	SDSS r	-25	-5	+150
	SDSS i	0	-25	+150
	SDSS z	-45	-25	+150

Imaging vs Spectroscopic filters

The sdss ugriz filters were intended specifically for imaging, and they have dimensions 86 mm x 86 mm and are up to 8 mm thick. They were designed to be parfocal, and the same focus setting is used for each even though, at commissioning, slight, ~<50 mic, differences were measured, as noted above. The Clear, ND1.5, UG5 and GG495 filters were intended for spectroscopic use and are rectangular, with dimensions 128 mm x 86 mm. They are also up to 8 mm thick. These are sometimes requested for imaging acquisitions on very bright or very faint targets, so it is useful to know their focus offsets with respect to the imaging filters. However, for very faint targets, a blind offset acquisition is recommended, as it is more efficient.

Either the change of filter or change of focus causes a shift of a few pixels in the slit image (more vertical than horizontal), so for acquisitions, it would be best to take the slit and field images through the same filter. The typical 20-sec slit image gives abundant counts from which to measure the slit center, so there may be some adjustment of the exposure time for the slit image through the ND1.5 filter.

The blue cameras have wavelength-dependent astigmatism, and it is not possible to get a round image with the Clear filter. (see, for example, the blue focus series taken in 2021 Sep). The focus-offsets tabulated below are only for the red channel.

	MODS1				MODS2
mode	focus at sdss_r (mic)	delta sdss_r	delta Clear	delta ND1.5	focus at sdss_r(mic)	delta sdss_r	delta Clear	delta ND1.5
red dual imaging	1210	0	+240	+240	500	0	+270	+225
red direct imaging	1265	0	+235	+205	490	0	+280	+235

Adding a line "rcamfoc step <delta>" to the script, after the instconfig command, will cause the camera focus stage to move delta microns from its nominal value, which is set in the rconfig.tab file on the mods1 and mods2 computers.

Log of changes to MODS camera focus values:

Imaging, Grating and Prism refer to the direct modes while DImaging, DGrating and DPrism refer to dual modes. The last 3 columns report the differences between the dual and direct modes, which are expected to be ~250-300 microns for the blue channel and 0 for the red. A blank cell indicates that no change was made.

MODS1B

Date	Temperature	Imaging	Grating	Prism	DImaging	DGrating	DPrism	DImaging-Imaging	DGrating-Grating	DPrism-Prism
20110217		3020	3330	3180	3220	3330	3380	+200	0	+200
20180221			3150						+180
20210916						3390			+240
20211127						3410			+260
20221215	~ -3	2960	3150		3235	3410		+275	+260

MODS1R

• red collimator mirror removed to be recoated during SSD 2016 and reinstalled before restart in 2017

Date	Temperature	Imaging	Grating	Prism	DImaging	DGrating	DPrism	DImaging-Imaging	DGrating-Grating	DPrism-Prism
20110217		1200	1750	1550	1200	1750	1550	0	0	0
20171121					1160			-40
20180209			1650			1650			0
20201020			1620			1600			-20
20210916			1650						-50
20221215	~ -3	1265	1650		1210	1600		-55	-50

MODS2B

Date	Temperature	Imaging	Grating	Prism	DImaging	DGrating	DPrism	DImaging-Imaging	DGrating-Grating	DPrism-Prism
20141115		1088	1305	1138	1350	1600	1450	+262	+295	+312
20170321					1200	1525		+112	+220
20171221					1225			+137
20180416		940						+285
20180913		985	1245					+240	+280
20190218					1255			+270
20190421					1225			+240
20210916					1200	1500		+215	+255
20211127					1255	1525		+270	+280
20220922					1240			+255
20221215	~ -3	985	1245		1260	1525		+275	+280

MODS2R

* red collimator mirror removed to be recoated during SSD 2016 and reinstalled before restart in 2017

Date	Temperature	Imaging	Grating	Prism	DImaging	DGrating	DPrism	DImaging-Imaging	DGrating-Grating	DPrism-Prism
20141115		597	861	800	600	861	825	+3	0	+25
20171121		450			450			+0
20210916						831			-30
20211126						861			0
20221215	~ -3	500	861		490	861		-10

MODS focus sequence scripts:

(Work TBD on modsfoc.py script to incorporate imaging and prism modes and to annotate the scripts in modseng better).

Scripts to take the focus sequence data are in the modseng directory, under modsScripts/Testing/MODSfocus/. These take a series 11 images or spectra, 5 above and 5 below the nominal best-focus, with a focus step of 30 microns.

The python script, modsfoc.py, allows one to customize the sequence by changing the nominal best-focus, number of steps above/below it and the step size. It was written for the grating modes, and for imaging and prism modes, the output needs to be hand-edited to: change the instconfig command; and, for imaging modes, to add commands to insert the imaging filters, change the lamp from Kr to VFLAT 4, and change the command to write the image title from comp to flat. The command-line syntax is:

./modsfoc.py scriptname dich step[mic] nsteps best-focus_red[mic] [best-focus_blue[mic]]

where

• dich = dual, red or blue
• step = focus step between images/spectra in microns
• nsteps = number of steps above and below nominal best-focus (number of images/spectra = 2*nsteps + 1)
• best-focus_red = nominal best focus in the red channel for this mode (enter something even if blue-only - it will be ignored)
• best-focus_blue = nominal best-focus in the blue channel for this mode (enter something even if red-only - it will be ignored)

# MODS1
#
modsfoc_im.py redim1.pro red 30 5 1200
modsfoc.py redgr1.pro red 30 5 1650
modsfoc_pr.py redpr1.pro red 30 5 1550
#
modsfoc_im.py dualim1.pro dual 30 5 1160 3220
modsfoc.py dualgr1.pro dual 30 5 1650 3330
modsfoc_pr.py dualpr1.pro dual 30 5 1550 3380
#
modsfoc_im.py bluim1.pro blue 30 5 3020
modsfoc.py blugr1.pro blue 30 5 3150
modsfoc_pr.py blupr1.pro blue 30 5 3180

-- %USERSIG{OlgaKuhn - 2018-04-16}%

• focus sequence through the Clear filter in the blue channel of MODS1, at the bottom right is an in-focus image at g_sdss for comparison.: