IMACS progress report no. 46
B. Bigelow 14 February 2002
Currently scheduled tasks: Status:
a) DSS redesign (Hare) IP
b) DSS drafting (Cleary) IP
c) DSS fabrication (Kowal, Castillo) IP
d) Calibration system mech. des. (Hare) IP
f) Long camera opto-mech assembly (Bigelow) IP
g) Principal Guider Camera elec. assy (Asa) on sched.
h) Science Array controller fab (Burley) 95% compl.
i) Filter procurement (Sutin) IP
j) Grism procurement (Sutin) IP
k) Motion control electronics design (Bagish) 95% compl.
l) MC fab. (Asa) 70% compl.
m) Control software design (Birk) on sched.
n) Instrument structure design (Gunnels) done
o) Instrument carriage design (Gunnels) done
p) Instrument structure fabrication (M&T) late, due 2/22/02
q) Instrument carriage fabrication (Rettig) late, due 2/18/02
Project events in January:
- First noise measurements from the IMACS science array controller
are showing 4e- read noise for full readout (1x1 binning) in under
three minutes. This may be lowest read noise yet achieved for
the SITe 2K x 4K CCDs. The IMACS science array assembly and controller
are currently scheduled for completion in early April.
- IMACS structure and carriage fabrication is now running a couple of weeks
late, but the structure should be finished and fully assembled in Pasadena
by the end of the month. Latest fabrication images can be seen at:
Optics (Bigelow for Sutin)
- Short camera spherical lenses at TORC (all but S01 and S08) are all
in polishing now, except for S04, which is in grinding. All TORC
lenses are due for completion by the end of March.
- Tinsley is on schedule with S08 and a week behind on S01. This is
not expected to delay delivery of the aspheres in June of this year.
- Most of the BVRI filters have now been received from Barr.
- Barr now has a RFQ for a set of Sloan filters.
- An RFQ is out to TORC for fabrication of the 600l/mm grism blank
- Spectrum Thin Films has an RFQ for AR coatings of the short camera lenses
- Assembly of the filter storage boxes is in progress.
- Optical testing of the collimator has started, with the assembly of the 12"
flat mirror and pin-hole/beam-splitter camera.
- Went through IMACS motion control schematics top to bottom with a
- Ordered Rittal heat exchanger and hydraulic fittings.
- Ordered Rack 1 for Carnegie built motion control electronics.
- Sent 80-pin IDC breakout PCB for fabrication.
- Wired and labeled jacks for Control Chassis #3.
- Completed front panel re-designs and sent to machine shop for fab.
- Prototyped and tested individual system hardstop shutdown electronics.
PCB designed and ready for fab.
- Principal Guider
There was only little progress since we are waiting for the new gear
(15:1 instead of 3:1). We need the higher gear ratio because the small
motor is not capable of moving the guider against gravity.
Additionally the larger gear will improve the positioning precision by
reducing the number of encoder-pulses per motor step from about 5/step
to 1/step. Smaller steps might also be necessary in case the drive
"stretches" under the force of gravity when the guider is rotated.
- Science Array Software
I made significant progress in designing and coding the science array
software package (user interface, camera control, quick look tool) I
worked mostly on the sub-raster mode and the focus-sequence. The
latter works nicely in simulation mode and requires no additional work
on the DSP level. For the sub-raster and shuffle modes I need some
additional functionality at the DSP (Greg is working on that). At the
user level I implemented 2 ways of defining the sub-raster geometry: (1)
via a dialog-box (2) point/click on the quick- look tool using a
previously taken image.
- continue with PR-guider test when the 15:1 gear arrives
- continue working on sub-raster/shuffle readout modes
- start with SH-guider (not available as of Feb/12)
The mask generation code has been enhanced by the addition of optical
values for the LDSS-2 instrument, to allow making masks for it in the
same way IMACS masks are made. For this, the IMACS short camera
configuration is sufficiently similar to the LDSS-2 configuration that
only different values for focal lengths, detector and mask sizes are
needed. This also allows a better test of the grism computation which
we will later use in IMACS.
In collaboration with Brian Sutin, a formulation for optical constants
was defined which will allow the construction of a library of optical
elements, and a general optical code. When better distortion fits are
available, the optical values can be entered into a data file without
having to re-compile the program itself.
Since the mask cutting machines are not expected to have a set of
commands to engrave characters, I have planned to create my own
software to plot characters along with the mask cutting. A simple
font has been obtained in the form of a (very long) c data statement,
and using this any printable ASCII character may be created as a set
of line segments. This will allow any label to be engraved anywhere
on the mask, usually near the edge, at any size and orientation.
A program has been added to the mask generation utilities to convert
the Slit Mask Definition File (the output of the mask generation
program) into machine tool commands, often loosely referred to as
"G-code". This program has currently two sections, one for the
milling machine currently used to produce LDSS-2 masks, and the other
for the laser machine to be used for producing IMACS masks. Most, but
not all of the commands for the laser machine are known; however the
definition of the cutting modes which work best may not be available
until the acceptance tests are done on the actual mask material. The
final program changes are expected to be very minor.
Detector Systems (Thomson/Burley)
This month we had something of a breakthrough in noise performance in
the lab test system. For a SITe chip rated for 6.2e-, we obtain 4 e-
read noise in the image area (or 3.5e- in the over-scan).
The readout times are below (for 2k x 4k chip):
Noise Readout (1x1) Readout (2x2) Gain
----- ------------ ------------- ----
7.9 e- 65 seconds 24 seconds 3.3 e-/DN
5.9 81 29 1.8
4.5 122 39 0.8
4.0 190 57 0.4
The noise improvements are due to the revamped clock drivers
(filtering was added), and some bandwidth tweaks of the preamp
(spectrum analyzer was quite handy).
Some pixel period optimizations are still to come, and we then can try
to optimize the gain settings. There is still some evidence of
spurious charge being generated in the image area.
Assembly-wise, we have three small electronics boxes alodyned,
assembled and wired up with the 79-pin connector + cable assembly.
One MIKE dewar head has been anodized red, the other is still out
being anodized blue. The face-plates will be black anodized.
All parts for the IMACS electronics box are fabricated, but not yet
anodized blue. We have resolved the location of the box on the side
of the dewar, made up the four cable assemblies + 79-pin connector
assemblies, and have started wiring up the backplane.
Another ten sets of boards are planned for spares and extra cameras,
the parts have been ordered and are arriving day by day.
Keeping track of the cameras on the mountain has been somewhat
confusing, so we've started labeling individual pcbs and recording
which boards are in which cameras.
Hackers broke into our lab linux system (Redhat 6.2) and back-doored
the ssh application, which we normally use for secure logins. The
machine was rebuilt with RH7.2, and our ccd systems are now running
with the 2.4.x kernel device driver.
1) Finished collimator assembly (cells into barrel, spacing checks).
2) Continued long camera lens/cell potting, L01-L04 finished.
3) Continued transfer of Gunnels 2D structure designs into MD5.
4) Completed design of imaging mirror cell, fab completed too.
5) Completed design of Mask Server mounting to structure, fab started.
6) Ordered air compressor for LCO ASB, for support of mask cutting laser.
7) Submitted abstracts for SPIE Astro 2002 meeting in August.
8) Ordered small parts for Short Camera, monitoring mechanics fab at M&T.
9) Designed tooling for measuring spacings in the collimator and long cam.
1) Complete opto-mechanical assembly of the long camera.
2) Complete design of grism cells (3).
3) Complete design of guide camera mount for long camera focus testing.
4) Complete design of mask cutting tools for tests at Convergent in March.
5) Continue structure entry in to MD5
* Minor rework and redesign of miscellaneous guide camera hardware.
* Opto-mechanical and testing of the Principal Guide Camera.
* Electro-mechanical testing of the Shack-Hartmann Guide Camera.
* Fabrication of the Disperser Server Wheel Assembly and Main DSS Support Assembly
* Resolve limit switch issues on the Grating Tilt Mechanism
* Assembly of the Clamping Assembly underway
* Fabrication of the Taper Pin Assembly is complete, assembly in progress.
* Design of Field Lens Tooling Assembly complete, drafting through first draft.
* Continue assembly, testing, and rework of all guide cameras as necessary.
* Assembly and testing of the Disperser Server System.
* Assemble one Grating Tilt Mechanism for testing.
* Fabricate Field Lens Tooling Assembly.
* Fabrication of the second shutter.
* Resume design work of the Calibration Unit.
1) DSS Taper Pin Assemblies are complete & anodized.
2) DSS Clamp Mechanisms are complete & anodized.
3) Imaging Mirror Assembly has been completed.
4) LCAM & CFG aperture masks and frames have been completed.
(apertures and wings have been anodized).
5) SMS and guider camera mounting spacers have been completed (Pilar).
6) Dewar Focus limits package has been started (Robert).
7) Modification of Daedel mirror mount (for collimator testing)
8) Assorted rework and odd jobs for Ian, Joe & Alan.
Expectation for February:
1) Dewar Focus (Robert)
2) Slit Mask Shuttle Support
3) Field Lens Spacer
4) Bore Tools
5) Assorted rework and odd jobs
8) Other Business
Next meeting: 10:15, 3/6/02