SpeX News and Updates (JTR 20 March 2015) 20 Mar 15 Corrected macros for the single order long mode were added. The old macros incorrectly replaced the order sorting filter with the x0.1 stop. This was necessary to prevent the flats saturating due to the longer minimum itime of the bigger array. This has been resolved by always using the narrow slit width for these flats. 06 Mar 15 Spextool_v4.0 posted. Apart from bug fixes this should be the final version. It includes correction for non-linearity. The SpeX Manual was also updated and a section about array read out efficiency added. The allowable H2RG well depth has been increased from 20,000 DN to 30,000 DN (saturation is 40,000 DN). Oct 14 Closed cycle was replaced due to performance concerns. However, it turned out that the slight temeperature changes being observed in the second stage were due to the facility compressor and helium lines. During the change-out the guider array got got a very slight dusting. The array will be cleaned during servicing in April. 21 Oct 14 S/N widget added to DV (see manual for details). 10 Oct 14 Spextoolv4.0.5 posted (bug fix). 09 Oct 14 Spextoolv4.0.5beta posted (bug fix). 05 Oct 14 Guidedog contK and L' zeropoints added to DV. 30 Sep 14 Spextoolv4.0beta posted (bug fix). 29 Sep 14 Preliminary version of the the SpeX Observing Manual posted on website. 19 Sep 14 Parallactic angle calculator fixed. It was discovered that the parallactic angle caluclator had a sign error in its code and was calculating the wrong angle. Old SpeX used in its own parallactic angle calculation and this was correct. One of the upgrade changes was to use the telescope parallactic angle calculation and this was the source of the bug. 15 Sep 14 Spextoolv4.0beta posted (bug fix). 09 Sep 14 Limiting flux calculator on SpeX webpage updated to reflect improved array performance. 06 Sep 14 Spextoolv4.0beta posted (bug fix). 03 Sep 14 Slight increase in cooler second stage temeperature noticed. Monitor (see temperature archive). 01 Sep 14 Spextool4.0beta is now available from the SpeX website. This version reduces data taken with the new H2RG array. Correction for non-linearity is not yet incorprated so observers should keep counts below about half full well (20,000 DN). 04 Aug 14 First scheduled use of fully recommissioned SpeX. 26 Jul 14 Guidedog JHK zeropoints updated. 17 Jul 14 SpeX back on telescope for recommisioning (following cooldown #41). Feb-Jul 14 SpeX upgrade in Hilo lab. New arrays, ARC array controllers, motors, motor controllers, Hall effect sensor control, closed-cycle cooler, and computers. H2RG array replaces Aladdin 3 array in the spectrograph and the Aladdin 3 from the spectrograph replaces the engineering grade Aladdin 2 array in the IR slit viewer. 02 Aug 12 Cooldown (#31). 10% neutral density filter sandwiched with M'in slit-viewer filter wheel. Allows unsaturated imaging of Jupiter with Guidedog. No need to redo guideboxes following cooldown. 01 Feb 12 MORIS is now fully integrated with SpeX and is considered an IRTF facility instrument. It is always available with SpeX, for which it functions extremely well as an optical guider (in fact better than guidedog for some applications). It does not have to be requested explicitly. Using a 15 sec integration in median seeing conditions MORIS has guided down to Sloan r'=20.4 and i'=19.9. For MORIS documentation see 'irtfweb.ifa.hawaii.edu/~moris/'. 11 Jan 12 Software update to MORIS and Guidedog guiding. Differential refraction was implemented for guiding SpeX with MORIS. 09 Dec 11 Guidedog CPU (Sparc) failure. One of the two spare boards also failed. Four days lost. We now have no spare CPUs for Bigdog and Guidedog (Sparc boards are obsolete) although we continue to search. Bigdog and Guidedog are scheduled for upgrade in 2013. 08 Dec 11 Flush macro button added to Bigdog 'Basic' mode. Flush does a short integration of the array in prepration for longer intergrations. It can also be used to remove image artifacts that result from 'STOP' and running the 'array_init' macro. It takes 10 sec to run and restores all parameters (no need to turn 'AutoSave' off for example). 21 Nov 11 Cooldown (#30) following replacement of closed-cycle cooler. Checkout on Nov 23 OK, no need to redo guideboxes. 30 Sep 11 Reports of poor telescope tracking, typically 0.5"per minute. TOs and observers should use the 'ClearRate' and 'adj.pt.rate' button in Guidedog slow guide to compensate if guiding integration time is long enough to be affected. 25 Aug 11 Software upgrade to SpeX, TCS, and MORIS: SpeX communication with MORIS for guiding. Guidebox signal check to check for jumps out of guide box. 'MJD_OBS' (Julian Day) keyword added to FITS header. 03 Jul 11 Power cycle array problem (see 19 Jan 2009). 19 Apr 11 Chevron-like pickup noise on Bigdog reported. Traced to holding current on MORIS filter wheel (CCD system attached to SpeX). Current should have been turned off. 27 Dec 10 LN2 holdtime measured at 40 hrs with telescope parked at zenith during two-day Chirstmas holiday. 14 Dec 10 Report that SpeX macros are failing to complete with error "message queue ... error sending command to IC". Fixed with reboot. 19 Jul 10 Software changes. Fixed guidedog crash with 'TakeSky' and 'Full Image' taken with autosave on. Remove redundant 'Fast Guide' tab. Rename 'Send AB Beam offsets' to 'Set AB Switch'. 04 Jun 10 Cooldown (#29) follwing repacement of closed-cycle cooler. Array temperature control was not switched on until the arrays had remained at about 15 K for one day. Dark current images taken when the arrays had stabilized at 30K showed a very high frequency (several per 120 sec integration) of 'tachyons' (dark spots about 50 pixels in diameter). On the theory that the spots were due to outgassing of molecules cold pumped onto the array during the longer than normal cold soak, the arrays were warmed to about 100K for a few hours to 'boil off' any accumulation. On cooling to 30K the dark current was back to normal. 28 Apr 10 Debris found on SpeX entrance window. Debris appears to be fine brass filings from the brass window cover eroded by sliding contact with the aluminum cover housing. Possibly slow accumulation since cover was installed in December 2005 (ironically to keep the window clean). Window thoroughly cleaned and cover replaced with cover made from Delrin. During general inspection debris was also discovered below the SpeX MIM bolts. Bolts will be replaced by hard brass bolts. 22 Apr 10 New Guidedog zeropoints (just for DV display purposes - approximate): J=20.57 H=20.68 K=20.11 FLTZP keyword added to Guidedog fits header. Guidedog default slwcnts changed from 12 to 20. K-band throughput estimated at 15% compared with about 22% before. K-band sky background 13.2 mags per sq. arcsec comparednto 13.9 previously. Also noticed that flat field counts are lower than expected (peak ~3500 DN, should be about 4250 DN). We need to check window. Jan 10 XUI 'logout' now puts blank into OS wheel and places in/out mirror in. 23 Nov 09 Mid-day imaging of M13. K-band magnitude limit K=14 in one minute, sky brightness about 6 mags per square acrsecond (see 1"). 12 Jun 09 XUI 'quit' button removed (use command line). Observer uses 'logout' button, which ends the session but does not quit the XUI. Feb 09 Array ground moved to allow for new cover. 19 Jan 09 Following cycle of array power some rows were found to be missing in one quadrant of the bigdog array. Fixed by recycling power. Staff are reminded to check the array readout following a power cycle and to recycle power to the arrays if required (a known occasional occurence after cycling the power).` 12 Nov 08 Neutral density 2.0 filters installed in guider filter wheel. The ND filter is sandwiched with the CO (2-0 bh)2.292 um 1.5% filter. Required for lunar imaging in support of the LCROSS impact. Oct 08 Software changes. A new XUI login procedure allows observers to protect their data. showguidebox' setup now displays in both A and C (i.e sky subtracted image) buffers. 'Div_Flat' enabled for stored J, H, K, and H+K flats. Guiding methods optimized and name 'Bright', 'Faint', and 'Very Faint'. Sep 08 Maintenance work done on array contorller warm electronics box. The backplane was realigned and cleaned, and the controller boards reseated. The array controller ground wire was not properly reconnected and resulted in some 'chevron' pickup arrays noise until fixed. Sep 08 SpeX is being integrated into the new telescope control system (TCS), TCS3. For now there is the option of using the TCS or TCS3 (in the 'Setup' under 'TcsSystem'). 05 Jun 08 Flexure test. Positions of arc lines and slit measured as a function of telescope orientation. New change from 2000. Mar 08 Changes to VNC procedure. Bigdog and Guidedog sessions have been created on stefan. Users should connect to these sessions. Mar 08 Rash of bigdog DSP B card errors (several per night) requires array power cycling to fix (15 mins per error). The problem was resolved by replacing the cryostat electronics fiber interface board (fcryo master). However, it is not yet clear if this indicates a board failure or a board seating problem. Feb 08 SpeX warmed up and worn anti-backlash worm gear in grating turret mechanism (8 years of operation) is replaced (warm torque 3-4 inch-oz). Following cooldown grating turret positions are recalibrated and guidebox AB positions updated, The grating turret positioning code is modified to mitigate against any backlash. Flexure tests are also done. Operation of the turret is back to normal. Grating inits are required every half-dozen or so moves for most precise (up-down) positioning of spectral orders on the array. Jan 08 The torque of the grating turret mechanism increases from a few inch-oz to about 20 inch-oz. Movements of the grating turret are restricted to about once per night until scheduled Spex maintenance in February. Oct 07 SpeX warmed up for maintenance. Pupil viewing lens installed in dichroic turret (used to image primary and secondary mirrors and measure telescpe emissivity). New guide filter installed for testing (pass < 2.5um, too much background for useful guiding it turns out), H2 filter sandwiched with PK-50 for blocking without need of PK-50 in OS wheel). 01 Sep 07 Website updated with more information about obtaining accurate photometry with the guider. See http://irtfweb.ifa.hawaii.edu/~spex/guider_sens.html Aug 07 Observers are reminded that to measure accurate spectral shapes at wavelengths less than 2.5 micron, the object and the standard star should both be observed at the parallactic angle using the internal K-mirror field rotator. See http://irtfweb.ifa.hawaii.edu/~spex/work/rotator/rotator.htm and observing techniques paper Rayner et al 2004, SPIE 5492, 1408-1509 (pdf link on SpeX homepage). 25 July 07 Guidedog throughputs measured: J 0.18 H 0.23 K 0.22 L' 0.14 (probably lower limit since measured in upper 3/4 of well) There has been no measurable change in throughput since first light in May 2000. May 07 Facility helium line leak required He repressurization but resulting contamination caused cooler to fail about one week later. Cooler was successfully replaced but several warm electronics components were damaged during the work, probably due to an electrical short (littledog power supply, rotator animatics motor encoder, in/out calibration mirror limit switch). About one week of SpeX onserving time was lost. Following cooldown some fine tuning of the AB guidebox positions was required. mid-April 07 SpeX cycled to replace damaged closed-cycle cooler Feb 07 Slight damage to closed-cycle cooler power connector sustained during instrument change. Will require cooler replacment when schedule permits since connector is vacuum sealed. Jan 07 bigdog and guidedog software updated. Changes include running mean image buffer (buffer D). 16 Oct 06 Spextool version 3.4 available on IRTF webpage. 06 March 06 Noisy columns (1 col in 8) in quadrant 2 (top-right in DV). Required careful reseating of cryostat-mounted electronic boards to fix. early-Dec 05 SpeX maintenance. New Z and H+K notch filters were installed in the IR guider filter wheel. The H+K notch filter should improve guiding on faint objects. The increasingly scratched front window was replaced, improving throughput by about 5%. A new manual window cover was installed to improve protection. A minor vacuum leak was found in the temp controller connector plate (glass pin seal). The plate was replaced. 19 March 05 Cold head had to be replaced following malfunction (3.0 years old, see). Feb 02). Only inner piston and head was replaced (cold strap not detached). Several nights lost through warm-up and cool-down. 13 July 04 Bigdog noise problem in quad 4 (top-right in DV), every 8th column is noisy and contains no data. Problem could not be reproduced during troubleshooting the following day and has not reappeared since. Possible flaky or loose electronics board. 02 June 04 Spextool version 3.2 available on IRTF webpage. 06 May 04 More words of wisdom added to SpeX docs about procedures for long exposures (cosmic rays and medianing data). See http://irtfweb.ifa.hawaii.edu/Facility/spex/work/spectra/taking_spectra.html. 03 May 04 Due to slight movement of arrays during cleaning new guidebox and grating position defaults entered into XUIs. 29 April 04 Guider problems finally traced to problem with replaced data fiber. SpeX operation now back to normal. 26 April 04 First night back on telescope. Temp controller fine tuning continuing. Guider experiencing several interruptions with DSP B card errors. April 04 Remarkably arrays and optics survive. Arrays are removed to clean debris deposits. Three temperature cycles required to get things back to normal. 02 April 04 Handling accident at telescope results in rapid loss of vacuum. Damage to temp controller cable and data fibers. Risk of damage to arrays and optics due to rapid recompression and icing. 01 April 04 Spectrograph pre-amp board failure; replaced. Fixed quad 4 every 8 columns missing. July 03 Ethernet switches to the calibration box going to sleep and occasionally not responding to macro commands to turn cal lamps on and off. Possibly due to increased ethernet traffic. Switches replaced with newer model. They always respond now but observers may notice that lamps can sometimes take up to 30 seconds to respond. Feb 03 The SpeX login procedures have changed. Please read the docs for instructions. Feb 03 Guidedog's new CPU failed and was replaced by the slower 110Mhz CPU. The faster CPU will be reinstalled once it is fixed by the vendor. 22 Jan 03 One of the two spectrograph clock/bias boards was replaced with spare. Board was probably damaged by power supply spike during island-wide high winds. An NSFCAM board was damaged at the same time. Telescope facility UPS appears not to have filtered-out the spike. 12 Dec 02 Guidedog's CPU was upgraded from Themis SPARC 5/64 (110Mhz, 64MB RAM) to an USPIIi (360Mhz, 128MB RAM). Bigdog is also using an USPIIi (270MHz, 128MB RAM). Both Bigdog and Guidedog are running Solaris 2.6. The increased performance of Guidedog allows a second copy of the XUI (user GUI) and DV (viewer) to run from Guidedog. Remote procedures for Guidedog and Bigdog should now be the same. 06 Dec 02 Documentation updated and corrected. Please email rayner@irtf@hawaii.edu any corrections, omissions or suggestions for improvements. 01 Dec 02 A paragraph on procedures to use when using on-chip integration times of longer than a few minutes has been added to the SpeX docs. See http://irtfweb.ifa.hawaii.edu/Facility/spex/work/spectra/ taking_spectra.html. SpeX instrumentation and data reduction papers have been accepted for publication in March 2003 PASP. The papers can be down loaded from the SpeX webpage. Changes over the past few months: Sim mode has been removed from the xui. It can be run from command line. XGFIT is now incorporated into DV. SETUP in bigdogxui and guidedogxui now includes option to change default DV buffers (by default beam A goes to buffer A, beam B to buffer B). This allows just one DV display to view data from bigdog and guidedog siumultaneously. (Unfortunately an open windows problems sometimes causes the screen to crash when these icons are clicked.) SETUP in DV now includes option to turn guidebox displays on and off ('ImageShowGbox'). Aids for running SpeX remotely have been added. These include 'tcs1_status' for viewing of telescope info and remote focussing, and 'hexegui' for remote control of the tip-tilt hexapod. See http://irtfweb.ifa.hawaii.edu/ userSupport/remote_obs/ for running tcs1_status. Bigdogmovie does not allow coadds. This is a bug. It will be fixed in due course. In October the Spextool_extension package was released. This package is for telluric correction, merging, cleaning, and smothing spectra extracted with Spextool. A new version of Spextool was also released. Both packages maybe down- loaded from the SpeX webpage. 24 July 02 A new version of Spextool is available. This version includes combining spectra, telluric correction with A0V stars, merging orders and cleaning. The package can be downloaded from the SpeX webpage. There are some problems with using the telluric correction with prism data but these should be resolved in a few weeks. Spextool is for cross dispersed and prism data. Single-order long-slit data can be reduced using standard packages such as IRAF. A G2V nearby star locator has been added to the A0V nearby star locator. Over the past few months graphics have been added to the Observer Documentation pages. If you find mistakes or inconsistencies in any of the SpeX documentation please let us know. 25 Feb 02 During the SpeX maintenance period in January and February a new closed-cycle cooler and additional filters were installed. The new filters are in the Order Sorter filter wheel: on and off methane filters (1.69um and 1.58um, both 6% wide), and a clear annular aperture to stop down the telescope by about 90% for bright objects (`0.1 x stop'). Use the stop for both object and standard for correct calibration. The guide boxes in DV are now updated every image (ie. not overwritten) and show green guide box, center of slit (green cross) and centroid (small purple box) for the active beam (A or B). This should help users observing remotely who cannot view the ViewSonic guider display. (Staff: All the Hall effect sensors have now been epoxied to the non-glass case side, so that voltages are not affected by cold cycling.) 06 Dec 01 The argon arc lines and sky features (LXD order 4 only) used by Spextool for wavelength calibration have been put on the SpeX webpage. 22 Nov 01 Following scheduled downtime, SpeX is back on the telescope and working fine. The rotator was successfully modified to prevent slipping of the drive shaft. The XUI has been modified to place SlowCnts and NDRs in the Basic window just above AutoSave, and the name 'ObsMode' has been changed to the more descriptive 'Beam.Pattern'. 07 Nov 01 The rotator has started to slip again probably due to loosening of the set screw that became loose in August. The problem with tightening this screw is the extremely diffficult access. SpeX will be warmed up during the next scheduled downtime (Nov 12-22) when the rotator drive train will be modified so that the shaft can be pinned and additional access covers will be added. For the next four nights the rotator will be parked at the home position (pa 170 deg) and immobilised. This will should have little impact on these observing programs. 15 Oct 01 It has been noticed that row 513 (DV coord) in data from the spectrograph array is not real. Until the reason for this is understood and corrected row 513 is being added to the bad pixel mask in Spextool. Observers using other pakages to reduce their data should do the same. 08 Sep 01 During scheduled downtime SpeX was warmed up to fix -17 Sep 01 rotator. The problem was traced to a loose set screw in a coupling in the cold drive train. The opportunity was also taken to install a CO filter in the guider filter wheel and replace a Hall effect sensor in the rotator. 22 Aug 01 New versions of bigdogic and guideic fix the array timeout problems which have been an irritant for the past year. 20 Aug 01 Remote troubleshooting of rotator from Manoa finds that slipping occurs occasionally in the cold motor drive. SpeX will be warmed at the next opportunity to fix (Sep 8-17 window). Until then observers are asked not to use the rotator unless it is crucial to their program. 18 Aug 01 Observers note that rotator is occasionally not going to the commanded position. Problem initially thought to be caused by cabling problems to warm motor. 21 June 01 Calibration mirror motor driver board replaced. Operation now normal. 19 June 01 Warm in/out calibration mirror fails. Move manually. 26 April 01 First night back on telescope. Instrument and array performance is normal, the spectrograph array performance has not changed . However following more complete characterisation of the spectrograph array it is now recommended that observers keep counts below 4000DN, compared to the previous limit of 5000DN. mid-April 01 No problems were found in the warm signal chain and so the instrument was warmed up and the spectrograph array package was removed and sent to Manoa for test in the lab test dewar. A few aluminum flakes were cleaned off the array fanout wires. Cold performance of the array was normal. Best guess is that the problem was caused by aluminum debris on the array fanout. 03 April 01 Problem with spectrograph array. Increased gain and persistence in quadrants 3 and 4. 27 March 01 New version of IC fixed bug in guider. Guider X and Y gain is now applied horizontally (X) and vertically (Y) to slit image. Previously the gains were applied only in RA and Dec and didn't incorporate the rotator angle correctly. DV modified to include image compass on displays. 20 Feb 01 During scheduled downtime SpeX was warmed up to install -07 March 01 new guider filters and for preventative maintanance. New Hall effect sensor installed in guider filter wheel. Aluminum detent rollers in guider and order sorter filter wheels replaced with brass-bronze rollers to prevent potential jamming on aluminum axials. These rollers will be replaced with bearings at the next opportunity. 09 Feb 01 In response to an observer not noticing that the grating turret had been moved to a non-standard position the XUI was modified to display "Ajusted Step Pos" over a mechanism's icon if it has been moved from the menu selection's default position. Added LongXD2.1 to grating menu options. 02 Feb 01 IC software changed to allow communication between bigdog and guidedog macros, eg. a bigdog macro can execute a guidedog integration: cmd.guidedog go 22 Jan 01 guidedogic now allows guider to accumulate offset corrections and show systematic tracking rate changes