1 Milestones Achieved Mosaic 1.1 Progress Report April, 2010 A final design review was held for the electrical component of the project. The test Dewar is complete and e2v devices have been installed for testing. The MONSOON software configuration is complete and the system is functional and is currently being used on the test station. The first images have been obtained with the e2v engineering device The thermal engineering analysis is underway, but is delayed due to software and computer problems. The mechanical final design review is now scheduled for May 11. 2 Budget 2.1 Base Budget The following table shows a month-by-month summary of the labor usage in hours by major WBS element through April 30, 2010. WBS Element Month Nov-09 Dec-09 Jan-10 Feb-10 Mar-10 Apr-10 May-10 Cumulative Totals Project Management 23.2 73.7 156.5 55.0 88.5 83.5 480.4 Detector System 16.0 43.5 49.0 67.5 74.0 250.0 Dewar Assembly 77.5 301.5 122.0 165.0 129.5 795.5 Software 59.0 33.0 62.0 40.0 194.0 System Integration 1.0 1.0 Commissioning 3.0 3.0 Total Labor Hours 23.2 167.2 560.5 259.0 383.0 331.0 1723.9 FTE Equivalent 0.01 0.10 0.33 0.15 0.22 0.19 0.00 1.00 The cumulative labor used is currently 48% of the total budgeted labor. According to schedule tracking the project work is 40% complete, indicating that the project is over budget by about 8% at this point. The following graph shows the cumulative totals for the actual labor hours charged versus the scheduled labor. The actual labor hours used continue to exceed those projected by the schedule, but the difference is not increasing and thus the actual labor is tracking the schedule pretty well now. The difference was caused by overruns in January and February that were described in previous reports. The scheduled labor hour trend has changed slightly from previous reports because the schedule has been updated to better match resource availability. For instance, software tasks were pushed back in the schedule due to delays in completing the software interface requirements document. This was not a critical path item. Page 1
Cumulative Labor for the Mosaic Upgrade Project 2500.0 2000.0 Actual Labor Hours Used Scheduled Labor Hours Work-Hours 1500.0 1000.0 500.0 0.0 Nov-09 Dec-09 Jan-10 Feb-10 Mar-10 Apr-10 May-10 Month 2.2 Supplemental Budget There was $3,145 charged to the supplemental budget in April. This expense was for a spare Systran card for the computer rack. 3 Changes in Key Personnel A schedule has been worked out to allow Bill McCollum and Bill Ball from the KPNO Mountain Electronics group to spend time in Tucson for Mosaic/MONSOON training. Starting on May 4, 2010, B. Ball and B. McCollum will take turns spending two days each week (Tuesday and Wednesday) working in Tucson to help with CCD testing, MONSOON board testing, and Mosaic system integration tasks. 4 Schedule Overall, the project is very close to being on-schedule and there are currently no schedule slips that should affect the completion date of October 29, 2010. Following is a brief status of each major work breakdown component. 4.1 Detectors The e2v Test Dewar that was built using a KPNO Universal Dewar is now complete and is being used for CCD testing (see figure 1). Some mechanical fit problems were encountered while assembling the Dewar and were corrected with some minor rework. The rework took extra time and thus caused a delay in the start of CCD testing by about a week. The first images were obtained with the e2v engineering CCD on April 26, 2010 (see figure 2). These images represent a successful test of the MONSOON system hardware and software configuration. The CCD testing that was completed in April focused primarily on setting the electronic gain to provide the optimum dynamic range of the detectors. Page 2
Figure 1: Left image is the lab CCD test setup incorporating the test Dewar and the MONSOON Orange controller system. Right image shows the e2v engineering grade CCD installed in the test Dewar. Figure 2: First images obtained with the e2v engineering grade CCD operated at a 250 Kpix/sec read rate and 2-amp mode yielding a 18-second readout time. The images leftto-right are a zero, an x-ray, and a flat. A column defect is apparent in all images. 4.2 MONSOON Controller and Electronics The MONSOON controller sequencer development and software configuration was completed. The DHE hardware was tested with the new software configuration and all signals were tested and verified (for operating two CCDs in the test Dewar) at the CCD sockets located inside the Dewar. Following the bench checkout, the system was assembled with the engineering grade CCD and successfully tested with a detector. During the testing and characterization of the first e2v devices, the MONSOON circuit modifications for optimized readnoise and gain will be determined and finalized so that the remaining MONSOON boards can be completed and tested. Page 3
The final design review for the electronics was held on April 15, 2010. The consensus is that the design is mature, there are no design concerns, and the project should proceed as planned. There were some actions identified as summarized here: Some thought will have to be given to how single-amp operation of CCDs will be supported when the controllers are upgraded to Torrent. The options include manual rewiring, increasing the number of Torrents from two to four, or designing custom transition boards. The trade-offs are cost vs. efficiency. Some case studies will be conducted to help guide the decision. Temperature control of the focal plane will be accomplished using a commercial Lakeshore controller that is externally mounted on the instrument. It was felt that remote (serial) control of the Lakeshore would be a desirable addition (this is not currently a requirement). Since electronics are reset at prime focus by cycling the 115VAC power, possibly including the temperature controller, it was felt that the amount of time needed for the focal plane temperature to stabilize after a power cycle should be evaluated. If significant time is needed to reach temperature stability, then non-switched AC power should be considered for the Lakeshore. Additionally, the external non-switched Lakeshore may be desirable over the internal Torrent temperature control if Torrents require frequent power cycling. The issue of spares was discussed at length. A full set of spares at the board level will be needed for the MONSOON Orange implementation. A complete spare controller assembly will be needed for the Torrent implementation. The MONSOON team will locate and configure suitable spares to provide with the instrument. The KPNO staff have the action of determining how spare components will be managed, e.g. where will they be kept, will they be tested and maintained by KPNO staff, will there be common spares or instrument specific spares? 4.3 Dewar Assembly The final designs for the mechanical components of the project are complete. However, the final design review for the mechanicals is still delayed because the thermal engineering analysis is taking more time than expected. Gary Muller has been working on the FEA, but he had difficulty getting the software to produce useful results, and encountered some computer hardware failures. Some useful results have now been obtained and, although the analysis is not as complete as we had hoped, there is enough information to present at the final design review that is now scheduled for May 11. 4.4 Software The configuration of the PCIe version of the Systran fiber interface turned out to be a little tricky due to differences in the Systran software. However, after a communicating with Systran tech support and making some changes to the PAN driver, Phil Daly was able to solve the problems and establish a robust interface. This is a milestone in that the PCIe version Systran allows more options for host computer selection. In the case of the Mosaic upgrade, all computers are identical and can share a single spare computer. Since the PCIe Systran has been verified, a second card has been ordered as a spare. Following the last run of Mosaic on the 0.9-meter telescope, the Mosaic System Electronics (MSE) was sent to Tucson to allow early development of the interface to the new Mosaic computers and software. Figure 3 shows the MSE and its power supply sitting on top of the Mosaic 1.1 computer rack. Communication between the new hardware and the MSE (via GWC) has been established and verified. Page 4
Figure 3: The Mosaic 1.1 computer rack with the MSE (and power supply) sitting on top. The main observer GUI for NOCS, called NGUI is available to run in simulated mode for users to try out. Heidi Schweiker and Hillary Mathis will be the first users to try out the system out and provide feedback. The following figures are screenshots of some of the typical NOCS GUIs. Figure 4: Example of the main observation GUI panel Page 5
Figure 5: Entry widget for observing parameters Figure 4: Dialog box for object observing parameters pops up when the Object button is clicked in the main GUI. 4.5 System Integration Arrangements have been made to transport the Mosaic instrument to Tucson on Friday, June 11, 2010. 4.6 Commissioning The requested on-sky commissioning nights of October 23-27 have been received and acknowledged by the KPNO scheduling staff. Page 6