The Ohio State University College of Mathematical & Physical Sciences Department of Astronomy

Attendees: Darren DePoy, Tom O'Brien, Paul Byard, Jen Marshall, Pat Osmer, Mark Derwent, Chris Morgan, Dan Papallardo, and Rick Pogge.

Overview

Darren gave a brief overview and summary of work to date before handing things over to individuals for reporting.

Optics

Rick showed pictures sent by Ray Bertram of the MODS Blue Camera Primary mirror being setup for generation at SOML. Lots of work is in progress at SOML.

REOSC is making progress on the collimators, with the first of the four due "soon". Paul will find out the actual schedule in the next week. One of the collimator mirror blanks was broken in handling. REOSC did a terrific job of taking care of the matter, everything was transparent to us. They replaced the blank with another ordered from Hextek. No complaints on our end about their handling of the matter, REOSC was very professional and thorough.

The field lens design has been modified and simplified. SOML accepted the changes. The new field lens is thinner, and we can use the existing fused silica blank by sawing it in half, giving us 2 field-lens blanks out what would have been one. The other think blank may be used to make the lenses for the calibration system. This latter still needs to be worked out in detail, but it looks good.

Red Field Flattener: BK7 vs. NZK7

The red camera field flattener was originally designed using BK7. As reported elsewhere there was concern that the radioactive content of BK7 might lead to problems since this lens is located just millimeters from the red CCD detector (a really big problem if we go with thick red-sensitive CCDs). However, most of this was anecdote not backed up by empirical demonstration of a problem. An option is to use a non-radioactive (well, much less radioactive) glass called NZK7 produced by AG Schott. This glass has a lower radioactive potassium content, but costs about 3-4 times more than BK7. We have blanks of both BK7 and NZK7, and tested them using a Fairchild 2048x2048 CCD we were preparing for deployment in the ANDICAM at CTIO. Three tests were done by Bruce:

1. Dewar window in front of the CCD made from Fused Silica (control).
2. Dewar window made from BK7.
3. NZK7 inside the dewar, but with the fused silica window (the NZK7 blank could not be used as a window). The NZK7 sample was placed a few mm away from the detector.

The bottom line is that we will *not* use BK7 for the red field flattener, but instead respec our order with SOML to use NZK7. We will work with SOML to decide who will buy the blanks.

Mechanical Design

Tom spent most of December working the rear slit acquisition and guide camera stage. He has a good solution that is not complete yet, there are lots of bad packaging problems and he's not quite finished addressing all of them. This critically depends on the dimensions of the rear-slit camera. It we can get all the problems solved, this should be a structurally terrific solution vis-a-vis differential flexure. The load paths should make flexure minimal. The current design can almost reach all of the 6x6-arcmin sciece field, the exception being a thin strip along the back edge. This is more than good enough for our purposes since practically we cannot expect to guide, or have alignment stars for masks, at that extreme end of the mask range. The biggest design issue remaining is the size of the camera: it has to fit into the tight space behind the slit. We need to find camera with a reasonable form factor.

Mark has made progress on the design of the dichroic drum assembly, which interacts with the rear-slit guider system. He's also found a good solution for mounting the field lens that sits between the rear guider and the dichroic drum, and is carried with the dichroic assembly. His next major task is to start in on the design of the mounting for the second (stationary) red folding flat mirror.

We received a large shipment of parts for the two camera primary mirror focus mechanisms from ASU. This is six complete actuator assemblies. Once again, they did a very nice job. Thank you, ASU. We hope to do more business with them in the future, this is working out great for us.

Mark reported on the camera handling carts. We are having two of these bult for us by Indian Creek in Dayton, the same firm that build the grating turret handling carts for us (and did a great job). These carts hold teh entire camera assembly, and allow the cameras to be pivoted around for easy access. Trying to work with the cameras just sitting on the bench would not work very well - they are big.

Chris Morgan has been working with Mark on the lifting fixture for the grating handlers, and helped with testing of a torque limiter for the grating tilt mechanism. They found a really nice torque limiting coupler that is nice and stuff, much like a bellows coupling, but which won't break if we drive past failed limit switches. When the torque gets too high on the drive shaft, it effectively (and non-destructively) decouples the drive shaft from the worm gear. This looks to be a great solution for us.

We should have the first assembly of an entire camera frame (less mirror and corrector) in about a month. This is too big for the bench, so final assembly will await the handling carts.

Tom has the front-side Acquisition & Guide Camera system sketched in, but not detailed. Unlike the rear-slit AGC where space is at a premium, this is a much smaller problem. His design features an open architecture for the front AGC, essentially moving a small optical bench along above the focal plane. Our current concept calls for two CCD cameras on the front AGC: an acquisition and guider camera (probably science grade), and a wave-front sensor camera. More details to come.

Flexure Compensation System

Jen Marshal reported on her progress with the first-stage prototype of the MODS flexure control system. The test rig in the lab has the same angular scale as the full MODS optical path. She has closed the loop in the 1D controller, and it works remarkably well. The sensor is a germanium quad cell, and a HeNe laser is being used as the reference beam. The system has about 2mV of noise, most of whihc is due to "seeing" in the room (air currents from the A/C system). Taking single 5-second samples with the quad cell, the beam can be controlled in 1D to 8-microns, or about 0.53 of a pixel (MODS uses 15-micron pixels). If instead she averages 12 measurments over 1 minute, the position is controlled to 2-microns (0.13 pixels). This is great news!

The next step is to decide on the IR laser to use and design the beam projection optics. One decision to be reached soon is whether or not to use an integrated laser/optical fiber package or just a laser. The advantage of the fiber-laser package is that the laser can be put somewhere else (i.e., for easy heat dissipation), but this entails addressing the various fiber coupling issues.

Once the laser system is decided upon and the projection optics built, we can proceed to full 2D testing of the control system. This requires a bigger lab setup than we have now. The numbers, however, are extremely promising.