College of Mathematical & Physical Sciences
Department of Astronomy
|
|
The Ohio State University College of Mathematical & Physical Sciences Department of Astronomy |
Attendees: Darren DePoy, Pat Osmer, Dan Papallardo, Jerry Mason, Tom O'Brien, Paul Byard, Jen Marshall, Mark Derwent, and Rick Pogge.
Darren began with a brief summary of progress on the LBT that was learned at the recent SAC meeting in Bertinoro, Italy.
Mechanical
The grating handler is coming together in the lab. We are using weights to simulate the mass of a full grating/mirror load to test the grating select mechanism. The handling cart was built by Indian Creek Fabricators in Tipp City, Ohio (near Dayton). Things should be photogenic enough for pictures next week.
Tom has been working to finalize the collimator support trusses. These will be long (~3-meter) graphite epoxy tubes. Given the lower than expected cost of the steel main structure, we can afford these within the baseline budget (using graphite epoxy instead of steel adds $11K to the total cost). Tom has a good design for the steel-to-graphite mating sleeves. The graphite tubes have a precision inside diameter (they are wound in a 5-degree helix from 1-inch graphite/epoxy tape on a precision mandrel). The mating sleeve inserts a steel tube that is epoxied in place, and includes a fitting which protects the end of the graphite tube. The bond will be very strong since all forces are in either tension or compression.
With this design in place, Tom is making a series of finite-element analysis runs with AnSys to get revised flexure performance predictions. These will be posted online when ready.
Jerry Mason and Tom are continuing tests of bonding the gratings to their mounting flexures using a zerodur blank. This is the last step before we finalize the grating order. So far so good, we're wrapping things up in the next week.
The ASU shop reports that they are almost done with the shutter parts they are making for us. They are great to work with and we'll be exploring future possibilities for them assisting with MODS part fabrication.
Optics
Paul is working on the design of a projection system to but both visible and IR laser beams into a diffraction-limited spot for optical alignment of the instrument in the lab. We can expect more information about this design at a future meeting.
Image Motion ("Flexure") Compensation System
Jen Marshall reported on her research into IR detector options for the IMCS. The basic options fall into 4 categories:
A couple of vendors were found that make complete InGaAs IR cameras. These are based on 320x240 and 128x128 formats, with 40-60 micron pixels. All cost in the realm of $12-24K each. Since we need 4 of them, this is quite expensive.
CCDs can be coated with anti-Stokes phosphors to make them sensitive to IR radiation (an anti-Stokes phosphor up-converts IR photons into visible light by absorbing 2-3 IR photons and emiting a single visible-light photon). A couple of vendors are around that sell either pre-packaged CCD cameras (for a few $K), or who will coat a CCD for about $700/unit that you can package yourself into a camera. An interesting option, but it raises serious scattered light issues (we don't need a light source in the system right next to the science CCDs). However, such a camera would be very useful in the lab for a number of applications, so we are going to explore getting a CCD coated and turned into a lab camera to try it out.
Bare InGaAs array deetectors are ar available in 320x240 (40-micron) and 128x128 (60-micron) formats, and cost $10-20K per unit. We would then have the additional cost of designing and building the readout electronics. The total cost to the project for one set of cameras for all 4 channels exceeds the cost of a grating, which means we have to think about whether we need to go this route. The IR array option allows for continuous grating tilt settings.
The fourth option is also the least expensive and arguably the simplest: a Germanium quad cell. There are a number on the market in the size we need (max 5mm diameter), for a cost of $300. The advantage of a quad cell sensor is that we can make very simple electronics to very accurately keep the collimator stationed. The downside is that we would only be able to tilt the grating by discrete steps.
It is likely that we will use a quad cell for the lab prototype system, as we can get that going quickly at very low cost. In the meantime, we have to iterate with Paul Byard on the optics design (the fan-out gratings used to disperse the IR laser into spots) to determine the detailed trades to assess the scientific and operational impact of discrete vs. continuous grating tilts in MODS. This is more of an issue for the R=8000 and higher modes because in the R=2000 modes the entire spectrum is on-chip, and there is a very small range of operational grating tilts. However, there is always going to be that one essential line (or lines) that fall in a CCD mosaic gap or on a bad column that needs to be moved. We need firmer numbers before we can decide.
R. Pogge, 2002 May 14