IMACS progress report no. 7 B. Bigelow 2 November 1998 Objectives for October: Status: I.P. A. Continue slit mask fabrication development (Bruce) Done B. Let dewar/flexure control design contract (Bruce) I.P. C. Continue TV guider optical design and layout (Brian/Bruce/Tim) I.P. D. Start prelim. design of slit masks and frames (Bruce) Delayed E. Start prelim. design of disperser server (Tim) Delayed F. Start prelim. design of shutter (Cons.) Delayed G. See GMOS laser cutter system acceptance testing Objectives for November: A. Finish slit mask fabrication development (Bruce) B. Start Prelim design of instrument structure (Bruce) C. Finish TV guider optical design and layout (Brian/Bruce/Tim) D. Finish prelim. design slit mask handler (Bruce) E. Start prelim. design of disperser server (Tim) F. GMOS laser cutter acceptance testing (11/17-11/20) Optics and CCD reports from Brian Sutin and Greg Burley attached. 1) Project scope, budget, and schedule A contract for the conceptual design of the IMACS dewar and flexure control mechanics was let to Gerry Luppino, of GL Scientific, at the end of October. The dewar design concept is a variation on previous Luppino designs for CFH and UH. The flexure control mechanics are based on a Physique Instrument X-Y stage. The CCD array focus mechanism is based on a similar design for the Gemini NIRI instrument. The consultant selected to design the IMACS shutters backed out of the job in October. However, the design of the shutter is expected to be a simple variation on existing 150mm shutters, and should not pose any design, budget, or schedule problems. The shutter design will be assigned to another consultant, or will be completed by Bruce or Tim. 2) Mechanical Engineer recruitment Tim Bond joined the IMACS project on 10/5/98. He is coming up to speed quickly and is now fully equipped with a new Dell and most of the necessary software. 3) Conceptual designs The preliminary design for the slit mask handling system is well along and will be finished this month. Steve Gunnels completed a minor redesign and re-analysis of a monocoque structure for IMACS. Bruce completed a comparison between the space frame and monocoque structure concepts. The space frame design was shown to perform significantly better in terms of tilts and displacements of the critical optical components. We will be meeting with Gunnels again this month to plot a course for continuing the structure design. 4) Slit masks and fabrication More quotes for slit mask blank fabrication came in during October. The current lowest bid is for spun aluminum masks at about $25/each, although it is not yet clear whether the material thickness or finish quality will be acceptable. Additional quotes for metal spinning and composite pressing came in at $125-150/each. A few more quotes based on hydro-forming are still in progress. Final selection of mask fabrication methods will be based on results from the mask cutting tests described below. The GMOS laser cutter was expected to be ready for testing at the end of October. A problem with one of the mask positioning motion stages delayed completion of the machine. Performance testing is now scheduled for the week of 11/16, and there will be a group of us attending some of testing. We will also have laser cut slitlets made the machine on our sample materials (carbon-fiber composite, aluminum, and stainless steel). 5) Electronics No electronics progress scheduled or completed this month. 6) Software No software progress scheduled or completed this month. 7) Optics Via Brian Sutin: DONE, October: 1) The TV system has been largely determined. The principal guider and Shack-Hartman finder both use the Canon EF 50mm f/2.5 Compact-Macro, while the long-slit viewing camera uses the Canon 100mm f/2.8 Macro. All three guide systems, when imaging, will have about the same scale of 0.1 arcsec/pixel. For the Shack-Hartman, the Canon guide lens will be moved out of the way and an acromat+lenslet array will be moved in. Currently we are not intending to allow focus unless we can reverse engineer the Canon electronic interface. One each of the two Canon lenses have been ordered. 2) The field lens RFQ has been let to Contraves. Three responses to the RFQ were received, but one bid-revision and another bid were after the deadline (but near budget), so the bids were handed over to Jeannette for "last and final." The Contraves bid is slightly over budget by 12%. 3) Bids for fabricating the collimator and Long camera are trickling in. So far there is one bid too high, one bid too low, and one bid just right. The span of prices is about a factor of 10. 4) The thermal analysis for the Long camera has been finished. Over a span of 20 degrees, the focus shifts by 1.1 mm and an image motion at the corner of the detector of 0.23 mm. This corresponds to a focus shift of 54 microns/C, and image motion of 12 microns/C. Compensation is possible with one compound thermal spacer and two simple thermal spacers, all made from Delrin I. With thermal compensation, the images do not seem to degrade. 5) I successfully borrowed the precision scale from Lick, for the purpose of measuring the blanks. ------------------------------------------------------------------- TODO, November: 1) Order material for the direct imaging flat. 2) Finish thermal analysis for collimator and short camera. 3) Fill in more of the TV guider details. 4) Start thinking about filters & coatings. ------------------------------------------------------------------- 8) Guiders and CCDs Via Greg Burley: Guider camera update (for Oct): 1 - Almost all of the outstanding parts have recently arrived. Enough to build a set of DSP timing boards, clock driver boards, and signal processing boards (this week). 2 - All of the parts are here for the PC <--> guider camera serial interface, which is built onto a PCI development kit prototype board. I'm currently wiring this add-on bit. The latest version of our linux device driver supports all the standard system interfaces -- open(), close(), ioctl(), read(), and write() for the development kit board. The driver only needs to be updated for the add-on hardware [FIFO + transceivers + Altera chip]. 3 - Programming hardware for the in-system programming of the Altera devices recently arrived. Mosaic camera update (for Oct): 1 - The mosaic camera electronics have been acronym-ized -- BASE = Basic array system electronics (any objections?) 2 - Version 0.2 of the BASE schematics are available. I intend to start designing the circuit boards after the next (v0.3) revision. 3 - The remaining circuit design tasks are the following: * add more memory (512K words) to the DSP timing board to support the special binning modes of the echelle * investigate using a fiber optic interface and/or a byte-wide interface to avoid the need for repeaters in the serial interface * assign signals to clock drivers, bias drivers, and backplane pins * build in any lessons learned from guider camera testing 4 - Estimated cost of parts for the 8K x 8K BASE system is $16000, of which $12000 is the 8 analog-to-digital converters, and another $1300 is the 17 digital-to-analog converters. The BASE 4K x 4K system with two channels populated will be about $6000 for parts. 5 - Estimated power consumption for the 8K x 8K electronics is 37 watts, which breaks down (per board): DSP Timing board 1.4 W Signal processing board 11.6 W (x2) Clock driver board 4.7 W (x2) Header board 1.5 W (x2) Power/utility board 0.2 W The largest fraction of the power is used by the 8 analog-to-digital converters (16.8 W). IMACS dewar design: Gerry Luppino of the IFA (Hawaii) has been contracted to do a conceptual design for the mechanics of the IMACS dewar, including the cooling (cryopump or liquid nitrogen), CCD alignment and mounting, camera head, and a three axis precision stage for flexure compensation. Lab Test Equipment: We are starting to put together a list of lab test equipment for the assembly and testing of the IMACS camera.