First Quarter 1999

In January, the vacuum jacket returned from painting at Bear Machinery, Kaneohe. The blue epoxy paint provides increased surface protection to the aluminum structure. Also starting in January, selected parts of the internal cold structure (light baffles, light shields and wheels) were painted with Aeroglaze Z306 flat black absorptive polyurethane with low outgassing.

Assembly of the fore-optics continued throughout the quarter, including installation of the optics. Initial alignment tests show that on rotation of the image rotator the optical and mechanical axes remain co-aligned close to specification and that little or no adjustment is necessary. The fore-optics pupil was measured to be at the correct room temperature position, close to the cold stop. Hughes Tools delivered the spectrograph box during February and the box was successfully mated to the mounting trusses and fore-optics assembly. Most of the other Phase III components have been delivered but several key Phase III parts won't be delivered by Hughes until the end of March, further delaying assembly of the spectrograph mechanisms (slit wheel, grating turret and focus stage). Work continued on fabrication of the slit-viewer and dichroic turret in the IfA machine shop. The replacement liquid nitrogen can from Anionics was delivered in March and passed a leak check without incident.

Other deliveries this quarter were the five spectrograph prisms from Optics for Research, the integrating sphere from Labsphere and the mirrors and lenses for the calibration box. SpeX joined a consortium buying spectrograph order sorting filters from Infrared Engineering, UK. These filters will be used in the SpeX single-order long-slit modes. In a related IRTF activity, work started on the new telescope counterweight system in February. This six-month project needs to be completed before the 850-lb SpeX cryostat can be mounted on the telescope.

Both VME64/Ultrasparc array controllers, Big Dog for the 1024x1024 spectrograph array and Guide Dog for the 512x512 slit-viewer/imager array, and the cryostat mounted electronics, are now complete. The maximum full array read out speeds are 12Hz for the spectrograph and 20Hz for the slit-viewer/imager. Software development for array control in all instrument modes is also nearing completion and progress is being made on IDL code for realtime spectral extraction. Assembly of Little Dog, instrument control computer, is progressing. The first version of the instrument control software will be ready for the first cold mechanism tests in May.

PAIDAI members met at the Raytheon Infrared Center of Excellence in Santa Barbara on February 2 to review progress. Unfortunately no acceptable devices were produced from the second batch of four arrays hybridised for PAIDAI due to quality control problems at Raytheon. Following discussion of ways to avoid similar problems in future it was decided to proceed with hybridisation of the final six PAIDAI devices - two in February, three in March and one in April. Test results will be available in July. These devices will be Aladdin 3 arrays, arrays which employ the new higher performance 206 multiplexer. Pending delivery and testing of all 14 PAIDAI arrays, PAIDAI members decided on an initial distribution of arrays from the first eight hybridisations. SpeX/IRTF selected SCA47672 from the first batch of four hybridisations, one of the best Aladdin 2 arrays yet produced. SCA47672 is cosmetically good with a median dark current of about 0.2 electrons/sec and a readnoise of 40 electrons RMS.

The first six 512x512 devices have now been hybridised and three cold tested at Raytheon. Of these SCA46509 looks like a science grade array. The other two arrays suffer from 'snail trails'. These are sinuous dead pixel features occasionally seen at the lower edges of InSb wafers, close to regions where the 512x512 masks are located. IRTF will test several of the best 512x512 arrays in the SpeX test dewar and rate them for distribution to PAIDAI members and planetary astronomy colleagues. SpeX will use the best 512x512 array in its slit-viewer/imager.