Instrument Status

Subsystem Status
Detectors (see detector issues NOT affecting science operations)
  • Dewar is pumped and cold at a normal vacuum of 3×10-6 Torr.
  • All science CCD B amplifiers in good working order
  • Both FCS CCDs are in good working order.
Grating The latest flexure tests (August and September 2022) show that FCS should be able to cope with flexures at any rotator angle for slider 3. There are two clamping angles (-135 and -45deg) for which the flexure on slider 4 can be out of range. The tests also show that the clamping performance is good at all rotator angles for slider 4 and good at most rotator angles for slider 3. Therefore, it is recommended to use the command select_slider to select the grating installed in slider 4. The grating installed in slider 3 can be selected using the DEIMOS Control GUI at any rotator angle. Please, make sure to request your highest priority grating to be installed in slider (port) 3 in your configuration form. This can be done in the section of the configuration form titled "If needed, please select ports for gratings" by selecting "3" from the "G3" pulldown menu.
Filter Filter subsystem is in good working order.
Slitmask Positions 2-12 are in good working condition; slot 13 remains out of service.
Calibration lamps Calibration lamps are currently AOK.
Guider Guider subsystem is in good working condition, with no recent problems.
Rotator The rotator is operational and in good working order.
Key: No problems - Problems - Broken

Instrument News


DEIMOS detector warmed up due to operational issue (2023-Apr-14)
The DEIMOS detectors warmed up due to an operational issue with the LN2 fill. The dewar was warm filled on Apr 19, 2023. All CCDs on are in nominal condition after the thermal cycle.

Slitmask submission deadline changed to five weeks before the observing run (2023-Mar-04)
Due to the increased pressure on the slitmask milling queue it has become critical to receive slitmak designs several weeks before the observing run. Therefore, we have changed the slitmak submission deadline from 4 weeks to 5 weeks before the observing run. Please see the slitmask submission policy and deadlines page for more information.

CCD5B bias noise back to nominal values (2023-Feb-26)
We discovered that the CCD5B bias noise was back to normal values during a regular pre-run checkout performend on February 26. It is not clear how the high bias noise detected on January 6 has been fixed, but stable detector temperatures for weeks may have helped.

High bias noise on CCD5B is back with a different flavor than before (2023-Jan-06)
The CCD5B bias shows a gradient along the detector rows and it is a factor of 3 noisier than nominal.

DEIMOS detector warmed up after winter storm (2022-Dec-25)
The DEIMOS detectors warmed up due to lack of pressure in the external LN2 bottle connected to the autofill system during a winter storm tha made the summit inaccessible for 3 days. The dewar was warm filled on Jan 4, 2023.

CCD8B unusable due to bad bias, and fixed (2022-Oct-11)
CCD8B started to exhibit strong transient horizontal bands that made it unusable for science operations. This issue was fixed by replacing the CCD8 video board with an old LRIS-red board available after the LRIS red arm detector upgrade.

High bias noise on CCD5B fixed (2022-Apr-23)
The CCD5B bias noise has been nominal since we replaced the DEIMOS ion pump controllers and ion pumps in early March.

Stable communications with new ion pump controllers (2022-Apr-22)
The vendor of the new ion pump controllers, Agilent, has recognized that there are issues with the Ethernet port of these units. We worked around the issue by connecting the new controllers to the DEIMOS barrel Lantronix terminal server via serial port. Comminications with both controller has been stable for several days.

Upgraded ion pump controllers and replaced ion pumps (2022-Mar-28)
A service mission was undertaken between March 1 and 3 to replace obsolete ion pump controllers with modern units as well as to replace the ion pumps. The new ion pumps and controllers are performing well, but we are still working on fixing some remote communication glitches with one of the two controllers.

Updated DEIMOS web pages to include operations with new Linux hosts (2021-May-05)
We have been operating with DEIMOS including the following upgrades since July 2020: The DEIMOS web pages have been updated to include references to the upgraded GUIs, operation scripts and operation procedures.

DEIMOS rotator maximum speed reduced by 20% (2021-Apr-21)
The DEIMOS rotator motor was unable to rotate the instrument at its maximum speed of 2.27 deg/s due to what we believe is extra friction on the bearings that are part of the rotation mechanism. We reduced the maximum speed to 1.82 deg/s, and the rotator motor was able to handle it. This operation mode with reduced maximum speed has proven to provide tracking accuracy within the science requirements (tracking RMS ≤ 0.05") up to an elevation of 87.5deg. We are currently looking into possible ways to eliminate the extra frictionin order to recover the maximum speed of 2.27 deg/s.

Updated slimask submission policy during COVID-19 (2020-Oct-21)
We have updated the slitmask submission policy due to the COVID-19 pandemic. Please, see the Late Slitmask Submission Incentive Program page for details.

CCD5B operational (2020-Sep-15)
In the process of troubleshooting the most recent CCD8 issue, we found that the power connector on the CCD5 pre-amp board had been left disconnected when we finished the CCD5 troubleshooting in February 2020. Once the connector was plugged in, CCD5 went back to a state in which the bias noise in CCD5B is about 4 time higher than the nominal noise and its response to illumination is nominal. CCD5A is still inopperative, but this does not affect science operations.

CCD8 bias issues fixed (2020-Sep-15)
The CCD8 bias issue was fixed by reseating and cleaning a connector in the electronics box that was partially out of the mating connector.

CCD8 bias and throughput issues (2020-Sep-13)
We noticed that CCD8A and CCD8B bias had zero counts on all pixels during a DEIMOS science pre-run checkout.

CCD2B channel bias issue fixed (2020-Jun-19)
The CCD2B bias issue was fixed by replacing the ADD for channel B on the CCD2 video board. Nominal operation of the DEIMOS detector in spectral mode is back to using B amplifiers.

CCD2B channel bias issue (2020-Jun-12)
CCD2B bias had either zero counts or 1024 counts on all pixels, which makes it unusable for science operations. As a consequence, the DEIMOS detector needs to be operated using the A amplifiers.

CCD5 inoperative (2020-Mar-18)
The troubleshooting work performed last February did not fix the isssue with CCD5. We have been troubleshooting intermittent problems with CCD5 since August 2018. Troubleshooting efforts have involved assesing the health of every hardware element on the detector electronics located outside the cryogenic detector vessel. Our conclusion after all the trobleshooting done so far is that the CCD5 problem must be originated inside the vessel. Due to the risks involved in opening the detector vessel, which has not been opened since the instrument was delivered to Keck in 2002, there are no plans to continue troubleshooting the issue on CCD5 in the foreseeable future. CCD5 will remain inoperative until further notice. Please, see the following link for tips on how design your masks with dsimulator to avoid the affected region.

CCD5 not sensitive to illumination (2020-Feb-14)
CCD5 looks like a bias when illuminated. This issue will be troubleshoot on February 18th and 19th.

DEIMOS Guider DSS Tool available for Python 3 (2019-Nov-25)
L. Rizzi has created a new version of the DEIMOS Guider DSS Tool that runs under Python 3.

High readout noise on CCD5 (2019-Nov-15)
New instance of high read readout noise on CCD5B was detected on November 14th, 2019. In this case, the noise is a factor of 5 higher than on the other CCDs. Troubleshooting efforts through November 15th were unsuccessful. Troubleshooting will continue after the upcoming observing run.

Anomalous response to illumination on CCD5 fixed (2019-Oct-25)
CCD5B is working nominally again after its videoboard was replaced with the only spare we have.

Anomalous response to illumination on CCD5 (2019-Oct-18)
On October 18th, 2019, we noticed that all pixels on CCD5B with more than 1023 counts had a value of 1023 DN. Troubleshooting efforts through October 19th were unsuccessful. Troubleshooting will continue after the upcoming observing run.

CCD5 performing nominally (2019-Aug-28)
CCD5 noise is nominal since August 26th, 2019.

High readout noise on CCD5 (2019-Aug-25)
On August 23rd, 2019, we noticed that channel CCD5B has a readout noise of 46e- instead of the nominal 2.5e-. This bad channel affects spectra on the upper left-hand side of the mosaic. The high noise was detected after a period of 3 weeks of having the detector warm due to lack of access to the summit to perform the daily LN2 fills, followed by 1.5 weeks of vacuum pumping before cooling down the detector dewar. A similar high-noise issue on CCD5 was reported on August 8th, 2018. Efforts to fix the high noise on CCD5 have been unsuccessful so far. We will resume the troubleshooting work on September 3rd.

High readout noise on channel CCD5B fixed (2018-Sep-06)
The problem was solved by adjusting the clock and bias connectors going into the dewar.

High readout noise on channel CCD5B (2018-Aug-10)
On August 8th, 2018, we noticed that channel CCD5B has a readout noise of 46e- instead of the nominal 2.5e-. This bad channel affects spectra on the upper left-hand side of the mosaic. There is no indication of channel CCD5B having a high readout noise previous to August 8th, 2018. Efforts to fix this issue have been unsuccessful so far. We will resume the troubleshooting work on August 14th.

New 1200 l/mm blue-sensitive grating available (2018-Jan-08)
A new 1200 l/mm blue-sensitive grating (1200B) has joined the suite of DEIMOS dispersers and it is already available for use. The grating specifications can be found in the gratings summary page. Information about the 1200B grating throughput can be found in the DEIMOS throughput page. Finally, this page shows a comparison between the on-sky efficiency of the 1200B and the 1200G gratings, respectively.

24-bit VNC desktops and new FACSUM and XMET GUIs (2017-Oct-04)
We have changed the control0, control1 and telstatus VNC desktops from 8-bit to 24-bit color scheme. These desktops will now have a blue background. There are new versions of FACSUM and XMET in the VNC background menu that will open only in the blue background desktops. The old versions of FACSUM and XMET will not open in the new telstatus desktop.

CCD1A NaN values problem fixed (2017-Aug-25)
This was found to be a connector issue. Once the connector issue was fixed, CCD1A reads out normal values again.

LVMslitC (BC=10810) damaged. New LVMslitC has BC=10830 (2017-Aug-20)
The LVMslitC (BC=10810) mask was damaged due to an issue with the slitmask mechanism. The slitmask mechanism issue was corrected and a new LVMslitC (BC=10830) has replaced the damaged one.

CCD1A NaN values (2017-Aug-15)
CCD1A reads out NaN values since August 15. We will work to solve this issue, which adds to the bias stripes in CCD3A. We have added instructions in the afternoon checkout procedure to ensure that the Direct mode reads out through B amplifiers only. This issue does not affect the spectroscopic readout, which is always set by default to B amplifier only.

LVMslitC (BC=9400) slits too wide (2017-Aug-10)
We have noticed that the slits in the LVMslitC mask (BC=9400) have wider than nominal widths because the slits were milled without shoulders. This mask has been in service since March 2017. Today we have retired LVMslitC (BC=9400) and replaced it with LVMslitC (BC=10810), which has the correct slit widths.

New optical model for all gratings (2017-Aug-09)
Peter Capak has kindly provided (a new optical model parameter file) for all gratings, both in slider 3 and 4. This model should be valid to reduce any data taken after the March 2016 DEIMOS service mission.

Stripes in CCD3A bias (only Direct mode) (2017-Jun-23)
CCD3, amplifier A, has been showing horizontal stripes in the bias during the last couple of weeks. We have been working on fixing the issue, but it is still there. This does not affect spectroscopic data, because DEIMOS only uses the B amplifiers in spectroscopy mode. If your program is in direct imaging mode, please ask your SA to change the readout in Direct mode to SINGLE:B.

New optical model for the 600ZD grating (2017-Apr-07)
Some observers have recently experienced issues with the wavelength calibration of multi-slit data. This is very likely caused by the default optical model parameters in the pipeline not being representative of the system after March 2016 sevice mission. Peter Capak has kindly provided (a new optical model parameter file) for the 600ZD grating at 7200A in slider 3 and 4.

Found a defect in the long1.2slit mask. Replaced with a spare. (2017-Jan-10)
Inspection of historic flats taken with the long1.2slit mask show a bump in the brightness affecting two slitlets. The mask was inspected visually and bad shoulders were found on two slits. The mask (barcode #1090) has been replaced with a spare long1.2slit mask (barcode #7359). The long1.2slit mask with the defect has been in use from 2016-Jun-03 until 2017-Jan-10.

K2 telescope passive temperature compensation coefficient updated (2017-Jan-06)
A study of the focus estimates based on historic MIRA data has shown that the K2 passive temperature compensation coefficient used in the telescope focus model is not longer valid. The data series indicates that the current coefficient (0.194 mm/C) does not model correctly the telescope focus variations as a function of the tube temeperature in the past two to three years. A new coefficient (0.255 mm/C) has been obtained based on the most recent MIRA data and it has been inserted in the telescope focus model. This coefficient update should make the telescope focus more stable against temperature changes during the night.

Slitmask mechanism operational again (2017-Jan-03)
Masks were jamming in the slit mask frame. The bottom groove of the frame was found to be fouled with debris, rough, and with aluminum deposited in the middle area of the frame. This is thought to be caused by abrasion in this area due to aluminum on aluminum wear. The area was dug out and then smoothed. This fixed the problem. The slimask exchanger cleaning preventative maintenance task procedure has been modified to prevent this this from happening again. The frequency of the maintenance task has been increased.

Slitmask mechanism malfunction (2016-Dec-28)
Some of the DEIMOS slitmasks get jammed when we try to insert them in the optical path.

FCS CCD1 (left) restored (2016-Nov-26)
A spare video board was installed and the problem was fixed. FCS was returned to operation with both CCDs.

FCS CCD1 (left) not working (2016-Nov-23)
The left FCS CCD is not sensitive to light. The FCS track script was modified to be able to work only with the right CCD.

MIRA in Direct and Spectroscopy mode (2016-Nov-22)
On October 14th and November 16th we performed serveral engineering tests to investigate the DEIMOS image quality. Test results show that the corrections suggested by MIRA are different when MIRA is run with a grating in Zeroth Order than when it is run with the Mirror. Therefore, for those programs requiring Direct and Spectroscopy mode on the same night, we recommend the following:

MIRA with grating 900ZD in Zeroth order (2016-Oct-24)
During September 2016 we noticed that MIRA images taken with the 900ZD grating in Zeroth order were not giving the correct image quality. At the time we recommended not to use MIRA with the 900ZD grating. To solve this issue, we tightened the grating in its holding cell. Additionally, we decreased the threshold in the MIRA recommendations to send the corrective moves to the telescope secondary mirror. After these modifications were done, we took new MIRA images with the 900ZD grating in Zeroth order during an engineering night on October 14th. The image quality after the MIRA corrections were applied was very good. Therefore, we consider there is no problem in using MIRA with the 900ZD grating in Zeroth order to determine the telescope focus.

CCD3 amp A & B failure. Fixed the following day! (2016-Sep-24)
The CCD3 shows NaN bias values when read with amp A and B separately as well as when read with dual amp A & B. The CCD 3 pre-amplifier was replaced with a spare and the problem was fixed.

CCD4 amp A restored (2016-Aug-10)
For the past months observes had to readout the detector in direct mode using a single amp B readout mode (see history entry om 2016-May-09). Today, we restore dual A & B readout mode for Direct imaging after fixing a connection issue with CCD4 amp A.

CCD4 amp A failure (2016-Jul-30)
The portion of the CCD4 read with the amplifier A is not sensitive to light. The Direct mode readout was changed from dual amp A & B to single amp B to bypass the problem until it is fixed. This problem does not affect data taken in spectroscopy mode, which are read by default using the amp B only.

CCD4 amp B restored (2016-Jul-20)
We initially thought the noise was caused by a bad connection in the corresponding video board. Connections were checked and tightened but the noise appeared again on July 3, 4 and 6. After several tests, we came to the conclusion that the problem was a malfunction of the video board for CCD 14-10-5. The video board has two voltage regulators, which we believed were damaged. We replaced the voltage regulators and after one week on intensive testing, during which we took over 800 images, the noise is no longer present. We conclude that the replacement of the video board voltage regulators has fixed the problem.

CCD4 amp B failure (2016-Jun-02)
A transient noise pattern has been detected on the lower-right CCD (CCD 14-10-5 in the detector mosaic).

Grating service mission part two complete (2016-May-23)
During the month of March, we serviced the grating system. The main work done on the system consisted of the following tasks: As a result: For the reasons above, we consider the service mission a success and expect that for the foreseeable future, the grating subsystem will continue to operate more reliably. There are still a few minor issues we will continue to pursue. There is now one minor issue and that is clamping slider 3 in the range between 45 and 90 degrees is intermittent, and at least once required rotating to acquire calibrations. There is an easy work around and that is to simply clamp at a rotator angle where we know it will clamp. A full report showing the work we did and the results can be found here.

CCD4 amp A failure (2016-May-09)
The portion of the CCD4 read with the amplifier A is not sensitive to light. The Direct mode readout was changed from dual amp A & B to single amp B to bypass the problem until it is fixed. This problem does not affect data taken in spectroscopy mode, which are read by default using the amp B only.

DEIMOS science spectra wavelenght drifts (2016-Mar-07)
We have recently received some reports of large (> 10 Å) wavelength drifts in DEIMOS science spectra. Such drifts in the spectral direction cause the data reduction software to fail. We have performed several tests to better understand the problem. Test description and results, and operational recommendations to prevent this type of problems can be found at: Wavelength drifts.

DVD writer no longer available for data backups (2016-Jan-26)
The Keck Observatory is phasing out the auto DVD writer sometimes used by observers for data backups. Options to launch the DVD autobackup gui are no longer available from the background menu. Because the Keck Observatory Archive (KOA) now ingests all raw data from every instrument at the observatory, observers are encouraged to use KOA to retrieve a copy of the data following your observing sessions. Instructions and links to possible methods of data backup including KOA, scp, and rsync are found at: BackingUp.

CCD4 amp B shows NaN bias values. !No longer valid. CCD4 amp B has been fixed. Please ignore! (2015-Dec-06)
The rightmost amplifier CCD4 (14-10-5) amp B has NaN bias values.

CCD4 amp A failing. !No longer valid. CCD4 amp A has been fixed. Please ignore! (2015-Aug-15)
CCD4 amp A is not reading out properly. With a relatively high signal, the output looks normal, but for bias frames, the output is zero valued. We suspect a sticky bit in the A/D is responsible. For now, observers should use only amp B readouts for Direct imaging which can be set under the User Settings available from the DEIMOS dashboard. Trouble shooting will occur following the current DEIMOS run ending 20 Aug 2015.

FCS reference data archiving restarted (2015-May-10)
When a DEIMOS data disk died, we lost our archive of fcs reference data, and until now, we were not archiving fcs data. As of May, we modified and restarted our archiving software to rebuild the archive of fcs data. The instructions for retrieving archived fcs reference data is consistent with the afternoon start-up procedures.

Video board update (CCD3 amp A) (2015-Apr-29)
A video board that failed following the Deimos service mission was repaired and re-installed in the instrument. This freed our spare video board which was then installed in slot 3. As a result, video outputs on CCD3 and CCD7 have been returned to normal operations. Observers no longer need to select the single amp B mode when acquiring direct images needed for slitmask alignment. This will save of order 20s every mask alignment image.

In the table above, the CCD3 amp A detector issue was removed. We are sending that video board out for repair, and if it is successfully repaired, we will use it to eliminate the herringbone pattern noise on CCD6 amp B. Stay tuned.


Dr. Wirth has left the building (2014-Oct-24)
A fond ALOHA to the one and only DEIMOS Master, Dr. Greg Wirth, who has left the Keck Observatory for a job in Boulder Colorado. Dr. Marc Kassis is now the point of contact for DEIMOS.

DEIMOS data disk death (2014-Oct-07)
The primary data disk for DEIMOS passed away yesterday. It has been repaired but a result of the data disk death is that we lost our archive of FCS reference files. Anyone wishing to match to a previous run will not have access to the previous fcs data.

DSIMULATOR instructions now posted (2014-Sep-10)
We have released an updated and revised version of the original web page that Drew Phillips wrote on the DSIMULATOR slitmask design software.

DSIMULATOR tarball now accessible on Keck FTP (2014-Aug-20)
The DSIMULATOR software package is typically available from the UCO FTP server. On the rare instances when that server is unavailable, a copy of the tarball can be retrieved from the DEIMOS area on the Keck FTP server as file deimos_rev0b.tar.

DEIMOS Update (2014-Jun-20)
Observers should note the following about DEIMOS:

New DEIMOS Calibration GUI (2014-Jun-17)
By popular request, we have added several new features to the venerable DEIMOS calibration GUI, including:

DEIMOS Calibration Lamps AOK (2014-Jun-11)
The Xe lamp trouble reported on 2014-May-23 appears to be resolved. All calibration lamps are now functional.

DEIMOS CCDs Affected by Noise and Bias Problems (2014-Jun-08)
DEIMOS CCD 3A currently has a bad bias level, and CCD 6B is showing a minor amount of “herringbone” pattern noise. We will investigate the cabling to these detectors.

DEIMOS Slitmask Slot 11 Removed from Service (2014-May-30)
Masks placed in slitmask position 11 are being damaged during the move from the slitmask cassette into the telescope beam. Until we determine the cause and the solution, we are removing slot 11 from service to prevent additional masks from being damaged.

DEIMOS Calibration Lamps Temporarily Swapped (2014-May-23)
Pre-run testing revealed that the Xe calibration lamp circuit on DEIMOS is faulty, which may involve replacing the Galil controller. As a stopgap measure to allow observers to perform wavelength calibration, we have unplugged one of the two Kr lamps and replaced it with a Xe lamp. This means that when you turn on Kr, you will now get Kr+Xe instead of the usual two Kr. We hope to resolve this temporary workaround on 2014-May-28.

DEIMOS Rotator Control Software Acting Up (2014-May-23)
Recently the DEIMOS rotator dispatcher software running on roto has frequently crashed when the daycrew uses the manual handpaddle to rotate DEIMOS. Recovery can involve restarting rotator daemons, performing a soft reset on roto, or performing a hard reset on roto. Recovery procedure is described in the troubleshooting page.

Slitmask Alignment Tool now released for DEIMOS (2014-Mar-04)
Following months of “beta testing,” we are releasing the Slitmask Alignment Tool for use with DEIMOS. This means that observers will no longer need to copy DSIMULATOR output files to Keck and generate finder charts for each slitmask field, since the tool can generate these charts on-the-fly based on DSS images and the Guide Star Catalog.

DEIMOS Rotator Computer Replaced with Backup (2014-Mar-04)
Summit staff replaced roto, the Linux computer that controls the DEIMOS rotator, with its backup computer.

DEIMOS Rotator Computer Suffers Instability (2014-Mar-03)
Roto, the Linux computer that controls the DEIMOS rotator, has been prone to frequent failures since its power supply failed last month. These failures can produce a variety of symptoms, including: If these symptoms appear, please refer to the correpsonding troubleshooting entry for details.

Mask Design DSIMULATOR incompatible with latest IRAF (2014-Apr-18)
Reports from some DEIMOS observers suggest that there are problem running DSIMULATOR in combination with the latest IRAF release. DSIMULATOR has not be fully tested with the latest IRAF version and observers have had to resort to finding older IRAF version to complete mask designs. Please let us know if switching IRAF versions eliminates your mask design problems as it will help us track the issue. If you want to design your slitmasks remotely using our working version of DSIMULATOR installed under IRAF 2.14 at WMKO, we can easily arrange remote access to our computers from your home institution.

DEIMOS dewar warmup and cooldown curves posted (2014-Mar-10)
A new webpage on DEIMOS cryogen characteristics records our adopted best practices regarding the DEIMOS dewar, as well as graphs showing the warmup and cooldown curves.

DVD writer for data backups will be phased out (2014-Mar-7)
The Keck Observatory is phasing out the auto DVD writer sometimes used by observers for data backups. Although there is no time line for the removal of the DVD writer, we will no longer troubleshoot or maintain the DVD writer. Because the Keck Observatory Archive (KOA) now ingests all raw data from every instrument at the observatory, observers are encouraged to use KOA to retrieve a copy of the data following your observing sessions. Instructions and links to possible methods of data backup including KOA, scp, and rsync are found at: BackingUp.

DEIMOS dewar experiences unplanned partial warmup (2014-Mar-04)
A snowstorm closed the summit on 2014-Mar-03 and prevented the daycrew from accessing the instruments to perform daily cryo fills. The DEIMOS CCD dewar exhausted its cryogen supply in the morning of 2014-Mar-04 after about 40 hours without a refill and reached a temperature of -80C before Keck personnel could reach the summit. This is believed to be the first unplanned DEIMOS warmup since 2007. Consultations with UCO experts resulted in a decision to remove DEIMOS from service until the three chambers of the CCD dewar could be evacuated with the vacuum pumps to ensure that any contaminants that were released from the ion pump during the warmup were removed. DEIMOS remains on the pump and we expect to leave it there until March 18, at which point we will resume cryogen fills in preparation for the run starting 2014-Mar-22.

DEIMOS rotator computer replaced due to failure (2014-Mar-02)
The diskless Linux computer which controls the rotation of DEIMOS, roto, failed for reasons we do not yet understand. Fortunately, a spare (soto) exists and was installed in place of roto to restore the ability of DEIMOS to rotate. The procedure for replacing roto is now documented online.

New tools help acquire and analyze dithered spectra (2014-Jan-31)
Two new data acquisition scripts and one new data analysis script now allow DEIMOS observers to obtain and reduce sets of spectra dithered along the slit. The scripts ab and abba will acquire two- and four-position dither sequences, respectively, using a user-defined nod throw and number of repeats. To analyze such data, observers can use Peter Capak's modified version of the IDL-based DEEP2 DEIMOS Data Reduction Pipeline described below. For observer convenience, we have installed Peter's IDL software package on our Keck HQ network for observers to use during or after their observing runs. It can be launched from any DEIMOS numbered observing account by accessing the workspace menu and selecting 
	DEIMOS Utilities -> Dithered DEIMOS DRP
Observers interested in observing in dithered mode must take care to design their DEIMOS masks so that the slits are aligned with CCD rows; i.e., not tilted.

DEIMOS grating slider 4 working again (2014-Jan-21)
The tilt stage on DEIMOS grating slider 4 is able to home successfully again following repair work performed last week. We did not find a smoking gun, but did identify and fix some slippage in the drive system for the grating tilt. We are returning slider 4 to service and observers should be able to use both sliders on the upcoming DEIMOS run.

DEIMOS grating slider 4 working again (2014-Jan-21)
The tilt stage on DEIMOS grating slider 4 is able to home successfully again following repair work performed last week. We did not find a smoking gun, but did identify and fix some slippage in the drive system for the grating tilt. We are returning slider 4 to service and observers should be able to use both sliders on the upcoming DEIMOS run.

DEIMOS grating slider 4 temporarily removed from service (2013-Dec-24)
Slider 4, one of the two grating stations for DEIMOS, failed on December 23 and will be unavailable for the observing run starting December 24. Initial indications are that the grating tilt encoder has a problem that prevents the grating tilt from properly homing, rendering the grating station unsuitable for use. We will plan a repair in consultation with the DEIMOS instrument design team and attempt a repair once DEIMOS comes off the telescope in January. Our hope is to repair slider 4 by the time of the late January DEIMOS run.

New script allows long sequences of short exposures (2013-Dec-16)
In response to observer requests, we have released a new DEIMOS script of interest to teams using DEIMOS to acquire long sequences of short exposures. The new script goi_wffcs_loop will acquire a sequence of n exposures with a short pause after every m exposures to allow the DEIMOS flexure compensation system to measure flexure, make corrections, and resume tracking. Observers can thank WMKO Support Astronomer Marc Kassis for this innovation.

FCSFIX tool aids recovery from FCS tracking problems (2013-Dec-09)
Following a move of the grating tilt to the zeroth-order imaging position and back, DEIMOS has occasionally had trouble getting the grating angle to return to its original location with enough accuracy to allow FCS to track. The new FCSFIX tool aids the observer in completing the procedure required to restore the grating tilt to its desired location so that FCS can successfully track.

New option in deimos_guider_dss helps debug parsing problems (2013-Dec-09)
We have added a new option to the deimos_guider_dss program to aid in debugging problems reading the input files. Include the -D option on the command line to invoke “debug” mode, which will echo each input line as it is read. This will allow you to identify which line in the input file is causing problems.

PIG Unleashed (2013-Dec-04)
We have added the new Program Information GUI (PIG) to the DEIMOS user interface. On split nights, observers use the PIG to indicate which observing program to associate with the current image. This permits the Keck Observatory Archive (KOA) to tag the images with the appropriate observing program and observer list, and ensures that the images are assigned the correct proprietary period.

Updated Reflective Multi-Long-Slit Mask Released (2013-Dec-01)
Observers interest in using DEIMOS to observe single targets have traditionally used the LVMslits ("LVM" = "Long Variable Multi") mask, which features a reflective slit area area and a series of 20-arcsec-long slits in widths ranging from 0.7 to 1.5 arcsec. Observers have pointed out a key flaw in the design: it places spectra on top of known bad columns on the detector. To avoid these bad columns, we have now manufactured a new mask (LVMslitG) which shifts all of the slits by 10 arcsec in the spatial direction, toward the center of the mask. We recommend that observers who previously used the LVMslitB mask should now switch to using LVMslitG.

Peter Capak's Modified DEIMOS Pipeline Handles Dithered Data (2013-Oct-01)
Caltech observer Peter Capak has developed several DEIMOS-specific tools to assist in preparing slitmask designs and reducing DEIMOS data. His latest release includes a customized version of the DEEP2 Data Reduction Pipeline wihch can handle dithered data. Learn more at Peter's DEIMOS Tools website.

DEIMOS Grating Select Problems (2013-Sep-09)
DEIMOS has recently experienced difficulty in changing between grating positions as a result of a mechanical problem. It is currently necessary to rotate DEIMOS to one angle to change the grating reliably, and to a different angle in order to clamp up the grating up for minimal flexure. Follow this procedure to change gratings during the day:
  1. Rotate DEIMOS to “Physical PA” (ROTATVAL) of -90 or +90 as displayed on the DEIMOS rotator subpanel.
  2. Select the desired grating on the DEIMOS dashboard and wait for it to complete the move. If the move should fail, please re-try it.
  3. Rotate and re-clamp the grating at the optimal angle for flexure via the DEIMOS workspace menu by selecting: 
    	DEIMOS Engineering > Select slider... > Select slider N
    where N is the number of the slider with your grating (3 or 4).
At night, with the DEIMOS rotator under DCS control, follow this procedure instead:
  1. Have the OA rotate DEIMOS to “Drive angle” of 0/360 or 180/-180 as displayed on the FACSUM GUI.
  2. Select the desired grating on the DEIMOS dashboard and wait for it to complete the move. If the move should fail, please re-try it.
  3. Rotate and re-clamp the grating at the optimal angle for flexure via the DEIMOS workspace menu by selecting: 
    	DEIMOS Engineering > Select slider... > Select slider N
    where N is the number of the slider with your grating (3 or 4).
  4. When the grating has clamped up successfully, the script will return the rotator to the position it was in before step 3. Have the OA rotate DEIMOS to the desired sky position angle for observing.

DEEP2 Data Reduction Pipeline Available at Keck (2013-Sep-05)
All DEIMOS numbered observing accounts are now configured to run the DEEP2 Data Reduction Pipeline on the headquarters network. Please see these instructions for details.

Corrections required for DEIMOS guider alignment star coordinates (2013-Aug-27)
Recent adjustments to the DEIMOS TV guider optics have had the effect of shifting the location of the image on the guide camera. As a result, the pixel coordinates predicted by the DSIMULATOR mask design program require correction for this shift in order for the coarse alignment procedure to successfully place the alignment stars within their respective boxes. Limited testing indicates that values of these corrections factors are dX=+13 px and dY=-2 px. These fudge factors have been incorporated into the deimos_guider_dss script used to generate the finder charts for DEIMOS observing, so if you use this script then no further corrections are required.

Authors Encouraged to Cite DEIMOS Pipeline Papers (2013-Aug-23)
If you've used the DEIMOS spec2d Data Reduction Pipeline in the analysis of your DEIMOS spectra, please include citations to the two relevant papers which describe the pipeline:

Fix DEIMOS Slitmask Comb Errors More Easily with New Script (2013-May-13)
To aid in the recovery from DEIMOS slitmask “comb” faults, we have released a new script (fix_slitmask_comb_error) which will check the state of the comb; if it senses a comb problem, the script will rotate DEIMOS, re-check the system, and continue these steps until the comb sensor no longer indicates that the slitmask system is OK. It is safe to invoke this command both when DEIMOS is on sky and when it is not. The script can be run by the observer, OA, or SA. You can launch the script from the DEIMOS workspace menu via: 
DEIMOS Troubleshooting > DEIMOS Trouble Recovery > Fix Slitmask Comb Error

Care Required When Changing DEIMOS Slitmasks (2013-May-08)
DEIMOS has been experiencing a rash of failed slitmask moves when attempting to change masks while the mask mechanism is in the “up” position, corresponding to an angle of 180° (or -180°) on the DEIMOS dashboard rotator subpanel, or to 270° (or -90°) on FACSUM. Such moves can cause slitmasks to get caught in the changer mechanism, triggering the comb sensor and causing the slitmask system to hang; manual intervention is often required to recover the slitmask system. The default orientation of DEIMOS during the daytime (-30°) is far from this problem area; however, running the select_slider script to move to slider 4 will leave DEIMOS in this problematic position, and nighttime observations may take place at any angle. Thus, we ask observers to check the DEIMOS rotator panel before making a slitmask move, and if the clock diagram on that panel indicates that the slitmask is within 30° at the 12-o'clock position, please rotate at least 90° away from 12-o'clock, then change your slitmask, then rotate back and proceed with mask alignment and observation.

BAL12 Order Blocking Filter Returned to Service (2013-May-06)
The BAL12 filter has been refurbished at UCO/Lick and is now again available for use in DEIMOS. This is the only filter which allows DEIMOS observers to access wavelengths below 4000 Å. Observers may now request use of this filter on their runs by updating their DEIMOS configuration forms to include BAL12.

Updated Webpages Centralize Throughput Data and Plots (2013-Apr-06)
Ricardo Schiavon's throughput plots have been brought over to the Keck webpages from their former location at UCO/Lick and made available on the new DEIMOS Spectroscopic Throughput webpage. Furthermore, the DEIMOS Grating Summary webpage is now updated with links to these same plots, plus helpful information on dispersion and accessible wavelength range.

Switching to DEIMOS Slider 4 May Require Rotating (2012-Sep-13)
DEIMOS is having trouble clamping up on slider 4 at certain orientations. Sliders 2 and 3 can be clamped at all rotator positions. In contrast, slider 4 is only clamping up successfully for DEIMOS physical drive angles (ROTATVAL) between 135 and 360; for 0<ROTATVAL<135, clampup may fail. When switching to slider 4, continue to rotate DEIMOS to ROTATVAL=180°.

New DEIMOS Exposure Time Calculator Released (2012-Jul-13)
Xavier Prochaska and Brad Holden of UCO/Lick have released a nice web-based exposure-time calculator for DEIMOS.

New Script Helps Deal with DEIMOS Flexure (2012-Jun-19)
The DEIMOS grating system continues to exhibit excess flexure when the grating system is clamped up at certain DEIMOS rotator positions. Conversely, when DEIMOS is rotated to the optimal clamp-up position for a given grating, the flexure should be sufficiently small that the flexure compensation system can correct for it at all rotator positions. To assist observers in clamping up DEIMOS gratings at the bext rotator positions, observers are encouraged to use the select_slider script. From the DEIMOS background menu, simply click on DEIMOS Engineering > Select slider... and select from the available options to move to slider 2, 3, or 4. The script can also be run from the command line on the DEIMOS host computer; from an xterm window on polo, simply enter select_slider 3 to rotate and clamp up slider 3, or select_slider 4 to rotate and clamp up slider 4.

Flexure Found to Affect All DEIMOS Sliders (2012-Jan-14)
Our continuing investigation of flexure on DEIMOS has revealed that all three grating sliders are experiencing anomalously large amounts of flexure, with the magnitude of the flexure dependent on the instrument rotator position at which the grating is clamped up. We believe that all three sliders remain usable if the observer takes care to clamp up the slider at the optimal angle for that slider. Current optimal values are shown below:

Recommended Clampup Positions to Minimize DEIMOS Flexure
Slider ROTATVAL1 ROTPPOSN2
2 0° 90°
3 0° 90°
4 180° 270°

1Displayed as "Physical PA" on DEIMOS Dashboard Rotator Subpanel
2Displayed as "Drive" on FACSUM

Please note that the only impact on your program is that you need to put DEIMOS into the orientations listed above when changing to a particular slider. Once the grating is in place, you may freely rotate the instrument to any desired position angle. There is also no restriction on the telescope pointing.

New Soundplay System Plays DEIMOS Sounds Remotely (2011-Nov-22)
The venerable Xbeeper widget has severd us well for years, but as more astronomers make use of DEIMOS from off-island mainland observing facilities we have sought a better way to play the "exposure complete" and "readout complete" sounds remotely, rather than relying on the Polycom videoconferencing units to transmit the sounds. Will Deich at UCO/Lick has developed a set of tools which can transit instrument sounds directly to computers at mainland observing sites, as well as to our Remote Ops machines in Waimea. We are deploying the software on an experimental basis during this run with the intent of releasing it in place of Xbeeper in December. To test out the new tool, quit Xbeeper and launch the EventSounds tool from the DEIMOS workspace menu under 
    DEIMOS control menu > Subcomponents > Start Eventsounds GUI
Please note that not all remote sites are yet equipped with the external speakers required to play the sounds.

New DEIMOS Web Pages Offer Better Browsing (2011-Nov-18)
Thanks to the efforts of new Support Astronomer Luca Rizzi, the DEIMOS web pages have been reformatted and reorganized to help you find information more easily. Many pages which were previously accessible only through the index can now be accessed through submenus off the main instrument home page. For those who find it more convenient, the Old Home Page will remain available during the transition. We welcome your feedback on the organization and appearance of the new web pages.

New script obtains blue-optimized DEIMOS calibrations (2011-Oct-24)
Observing with DEIMOS in the blue (4000-5000 Å) requires a different approach to obtaining suitable calibration images, compared to using DEIMOS in the customary red wavelengths. Caltech's Evan Kirby has developed a handy script to acquire such blue-optimized DEIMOS calibration images. The calib_blue script takes separate arc and flat exposures for the red chips and the blue chips. The red-chip arc and flat images are not appropriate for the blue side and vice versa. The script employs some clever strategies to generate correctly-exposed images with the various available lamps. Please consult your Support Astronomer for assistance in customizing the script to met the needs of your observing program.

Keck starlists can now contain RA/Dec offsets (2011-Oct-24)
In response to a recent observer request, the MAGIQ guider software has been updated to allow observers to specify RA and Dec offsets to be applied after a target has been centered up. The intention is to permit starlists to contain entries for alignment stars with known offsets to a faint science target which may not be visible on the guider. As described in the starlist documentation, the new keywords to include are raoffset and decoffset. The offset values listed in the starlist file can be read directly into the MAGIQ offset fields, thus avoiding the need to verbally relay telescope offsets to the Observing Assistant (with the corresponding risk of a miscommmunication). If you have any questions about how to specify offsets in your startlist file, please contact your Support Astronomer for assistance.

DSIMULATOR bug (2011-Oct-04)
A couple of recent DEIMOS observers have been chagrined to learn on the day of their run that the alignment boxes on their DEIMOS slitmasks were not milled. In each case, the root of the problem appears to be with the operation of the DSIMULATOR software used for slitmask design. Specifically, observers should be careful NOT to use "t" option for toggling between alignment and guide stars -- it was never intended to do so. As described in the DSIMULATOR documentation, guide stars and alignment stars should be distinguished from each other by the use of different priority codes. We look forward to eventually receiving a corrected version of DSIMULATOR which eliminates the unmillable slits. Until then, all observers using custom slitmasks are responsible for sanity-checking their masks after submitting them to the slitmask database in order to ensure that they are free from design flaws. Please follow this procedure after submitting your masks and verify that the resulting slitmask plot is free of red slits that indicate problems.

Slider 3 Flexure (2011-Sep-11)
Recent engineering tests indicate that the flexure characteristics of grating slider 3, one of the two grating positions available on DEIMOS, have changed significantly since they were last tested. We are concerned that DEIMOS may have some sort of mechanical problem with that grating slider. Accordingly, we recommend that observers request their primary grating on slider 4 until further notice. Please contact your Support Astronomer for information on how to ensure that your grating is loaded in slider 4.

New deeplog option (2011-Aug-02)
A new option flag has been added to the deeplog command. The option is -loop. If specified, deeplog will not quit at the end of the current list of files: instead, it will enter a loop and wait for new images to arrive. Every 20 new images, a new copy of the column header is generated.

DEIMOS Lags LRIS in Far-Red Throughput (2011-Feb-07)
A new analysis compares the relative throughput of DEIMOS and LRIS out to 1 µm, finding that LRIS is at least twice as efficient as DEIMOS when observing with the 830 l/mm gratings on these respective instruments.

DEIMOS Begins Full-Time VNC Operation (2010-Jun-08)
Starting with the June DEIMOS run, we ask all observers to start up their DEIMOS control software within VNC desktops. Virtual Network Computing (VNC) is a technology that allows windows to be shared with observers at other computers and has been used successfully for years at Keck to provide mainland observers with access to Keck instrument controls. By expanding the use of VNC to all observers, we will thus have a single startup procedure for observers to follow regardless of whether or not they will operate with mainland observers. Operating under VNC also makes it easier for support staff to assist observers with troubleshooting and allows our observing assistants to monitor the status of observations during the night. Your support astronomer will be happy to assist you with starting your DEIMOS software under VNC.

Video Board Replaced (2009-Sep-30)
For many months, readout amplifier 1B (lower left corner) on the DEIMOS science mosaic has been subject to intermittent periods in which the bias level drops several thousand DN for a period of time, rendering data in the affected region unusable. After exploring non-invasive means of controlling the problem, with limited success, we have decided to replace the flaky SDSU video processing board (serial number 39) with a spare (S/N 31). Initial indications are that the new board has similar gain and readnoise to the one we removed. The flaky board is being returned to UCO/Lick for testing.

DEIMOS FCS Repaired (2009-Jun-19)
The DEIMOS flexure compensation system is working with a new piezoelectric actuator installed, and performance appears to be comparable to what it was prior to the April piezo failure. The new actuator still requires some adjustment in length, and thus the DEIMOS image is currently offset by 120 px in Y from its pre-repair location. Until we fix this, observers should realize that central wavelengths are bluer than expected by about 120 px times the dispersion of the grating. We hope to have this positioning resolved by the August DEIMOS run.

DEIMOS S/N Measurements (2009-May-23)
A new DEIMOS Spectroscopic Performance web page shows the spectral S/N achieved by the DEEP2 Survey as a function of magnitude.

DEIMOS FCS Status (2009-May-07)
We have resurrected some old software that allows us to perform flexure compensation using the grating tilt stage instead of the tent mirror. In recent testing, this software functioned well and provided flexure compensation with an RMS dispersion of 0.5 px in wavelength. Although this is about a factor of 2 worse than what DEIMOS provides with the tent mirror actuator enabled, it should be good enough that the effect of increased flexure on DEIMOS spectra will be minor. The bottom line is that the DEIMOS flexure compensation system should work well until we can replace the failed PZT actuator.

DEIMOS FCS Failure (2009-Apr-27)
On 2009-Apr-24, the DEIMOS flexure compensation system experienced a sudden and currently unexplained failure of the piezo actuator which drives the tent mirror, the key component which corrects image motion in the wavelength axis in DEIMOS images. This failure leaves DEIMOS without a working flexure compensation system. The cost and timescale for replacing the piezo unit, which is an expensive component manufactured in Europe, are currently unknown as we are awaiting information from the manufacturer. Although we hope to repair the DEIMOS FCS as soon as possible, it may not be possible to effect repairs in time for the May DEIMOS run which begins on 2009-May-23.

We are currently trying to understand why the piezo unit failed; it had functioned well since the instrument was built and was expected to last almost indefinitely. We are not aware of any event which would explain the failure. Unfortunately, due to the long expected longevity and the high cost of replacing the actuator (possibly $10,000 or more), we do not maintain a spare.

Based on measurements made this weekend, the flexure observed over a 360° rotation of DEIMOS on the science mosaic is 45 px in the spectral direction, with substantial hysteresis. This represents a tripling of the flexure measured at commissioning, possibly the result of mechanical decoupling in the piezo unit which is allowing excessive flexure to occur. Such image motion will substantially degrade spectra acquired with DEIMOS unless observers take steps to mitigate the flexure. Possible observing strategies for dealing with the flexure are to keep exposures as short as possible and to restrict observing to the range of DEIMOS rotator angles over which flexure is minimized.

If we are unable to repair the FCS by the May 2009 DEIMOS run, then a potential fallback plan would involve rewriting the FCS software to provide flexure compensation in the spectral direction using the grating tilt stage, which has much coarser resolution than the piezo-controlled tent mirror (1/3 px, vs. 1/100 px). It is not certain that we can provide the manpower to make such revisions, nor is it clear that this would provide suitable flexure compensation. If it succeeds, this approach may provide correction which is significantly worse than the piezo-based system but nevertheless far better than having no FCS at all.


DEIMOS Finder Chart Generator Now Writes Better Starlists (2009-Mar-30)
Erik Tollerup of UCI has enhanced the deimos_guider_dss script of UCSC's Dan Magee to generate starlists in the new format including rotator PA information.

DEIMOS Problems (2009-Mar-24)
DEIMOS is currently experiencing a number of problems of varying severity:

Starlists Now Allow PA Info (2009-Feb-20)
For years, observers have been forced to provide their position angle information to the OA verbally. The new starlist file format now permits you to enter the position angle for your target using the ROTMODE and ROTDEST keywords. Here is an example of how to specify the position angle: 
# slitmask with specified rotator position
PA_targ_name    00 24 00.00 +16 00 00.00 2000.0 rotdest=123.45 rotmode=pa

# target observed at parallactic angle
VA_targ_name    12 34 56.78 +01 23 45.67 2000.0 rotdest=0.0 rotmode=vertical

A Reminder About Checking Slitmasks (2008-Oct-31)
Whenever new slitmasks have been installed in LRIS, it's essential that you take direct images of all of your slitmasks in the afternoon as part of the afternoon calibration procedure for Xbox. Doing so ensures that we installed the correct masks and that the DEIMOS software is inserting the correct slitmask design information into your image headers.

New Script Allows Drift Scanning Mode (2008-Oct-31)
In response to an observer request, we now provide a driftscan script which can be used for drift-scanning spectral observations with DEIMOS. Given the angular size of the desired scan, the script will launch an exposure, compute and apply the required non-sidereal telescope tracking rate, then restore sidereal tracking and return to the starting location at the completion of the exposure. As written, the script will only scan in RA, so the observer would typically use a north-south slit alignment.

Slitmask Plotting Tool Released (2008-Oct-23)
With the new Plotmask tool, you can answer that age-old question: which slit is my object in? Given a DEIMOS image acquired with a slitmask in place, this IDL routine will generate a plot of the slits with optional labeling of slit index number (used by QuickSlit) and/or target name. Output can be sent to the printer. Launch Plotmask via: 
	DEIMOS Utilities > Plotmask
from the DEIMOS background menu.

Grating Slider Changes (2008-Oct-20)
In an attempt to resolve the longstanding problems with clamping up on slider 4 at certain DEIMOS orientations, we have been experimenting with a slightly different clamp-up sequence for the DEIMOS gratings. The new algorithm appears to give more reliable clampups for all of the sliders, and appears to almost completely eliminate the failed clamp-ups on slider 4 (except that the behavior remains marginal over the approximate range 0° < ROTATVAL < 45°). If a clampup should fail (or you have an FCS tracking failure) the first thing to do is to unclamp and reclamp the grating. The next thing to try is to rotate to the vicnity of ROTATVAL=-30°, unclamp, and reclamp.

Real-time Data Reduction Tool Released (2008-Oct-04)
QuickSlit, a spectral quick-reduction package for DEIMOS, is now available for use on all DEIMOS numbered accounts. The software, contributed by Michael Cooper of the DEEP2 Survey Team, will perform a full reduction on a single slit from one or more science images acquired with the same slitmask and display the output in the form of both a 2-D sky-subtracted spectrum and a 1-D extracted spectrum. Online documentation is available. You can run the tool from the DEIMOS background menu under 
	DEIMOS Utilities > QuickSlit

Grating Slider 4 is Finicky (2008-Jul-30)
Slider 4 is having trouble clamping up at certain orientations, particularly at DEIMOS ROTATVAL values of +90° and -90°. Using the clampup_slider4 script is discouraged due to problems seen with alignment after running this script, sometimes requiring the observer to re-run FCS setup in order to track. Instead, we recommend that observers should rotate DEIMOS to ROTATVAL=0° (corresponding to ROTPPOSN=90°) when attempting to change to slider 4.

New Reflective Multi-Long-Slit Mask Released (2008-Apr-28)
After several years of service, the popular LngVrMir mask has worn out and been replaced by a new and slightly different mask called LVMslits ("LVM" = "Long Variable Multi"). Like LngVrMir, this new mask features a reflective slit area and a series of 20-arcsec-long slits of various widths. Whereas the old mask offered 4 slit choices, the new mask has 5 (0.7, 0.8, 1.0, 1.2, and 1.5 arcsec). The LVMslits mask now appears in the default list of available masks on the DEIMOS run configuration form.

DEIMOS FAQ Page Upgraded (2008-Apr-21)
Tips and tricks on observing with DEIMOS have been added to the DEIMOS FAQ (frequently asked questions) page. Go there for those handy tips your friendly neighborhood support astronomer is always reminding you about!

DEIMOS Arcplots Widget Released (2008-Apr-06)
We are pleased to announce the initial release of the DEIMOS Arcplots widget, a GUI with which observers can generate arc line plots customized to any grating, central wavelength, and set of line lamps. This is useful for verifying the achieved wavelength range for specific slits. Please see the Arcplots Widget page for more information.

DEIMOS Mask Overlays Generating Tool Released (2008-Apr-05)
The new IDL routine dsim2regions can be used to help plot the locations of DEIMOS targets and slits on an image of the corresponding DEIMOS field. The routine reads information from the mask design files generated with DSIMULATOR and creates ds9 regions files which can be used to mark and label the locations of slits on an image with celestial coordinates encoded in the FITS header World Coordinate System keywords. Please see the dsim2regions page for further information.

DEIMOS grating slider 4 clamp-up problem (2007-Nov-03)
Grating slider 4 is unable to clamp up at certain DEIMOS rotator positions. This may indicate that some key components in the mechanism are worn and need replacement; however, we will not be able to complete that invasive work until after the November DEIMOS run. In the meantime, there are two things that observers can do if they experience problems switching to grating slider 4. The preferred option is to execute the new script called clampup_slider4, which will position the stage to a spot where it should be able to clamp up. If this fails, the fallback is to rotate the instrument by 90 degrees and try again to select the grating from the DEIMOS dashboard GUI. If this fails. try running clampup_slider4 at this new position angle.

DEIMOS imaging mirror replaced (2007-Jun-06)
The engineering-grade imaging mirror which has been used in DEIMOS since commissioning has been removed and replaced with the science-grade imaging mirror. The coating on the engineering-grade optic was in relatively poor condition due to tarnishing, but the new science-grade optic appears to be in outstanding condition (expecially considering that the coating is well over 5 years old). This change should increase the imaging throughput of DEIMOS when the mirror is used.

DEIMOS bias problems (2006-Dec-20)
Since the October 2006 earthquake, the DEIMOS science CCD mosaic has shown signs of low-level pattern noise and bias level variations which change from chip to chip. We have started troubleshooting the problem in partnership with experts at UCO/Lick Observatory and our preliminary finding is that the gain level on the science mosaic CCD controller has excess noise. Troubleshooting will continue in January 2007; until then, observers should expect somewhat higher noise in their images.

DEIMOS desktop changes (2006-Apr-22)
DEIMOS has become the first Keck instrument to change from the OLVWM window manager, no longer supported by Sun, to the FVWM window manager. A recent internal technical review concluded that FVWM was a preferred solution to alternatives such as GNOME and CDE. In addition to changing the window manager, we have also changed the way that observers will access instrument software: instead of using the window manager's pulldown menu, observers will now use the InstrMenu widget to run DEIMOS-related tasks. We welcome your feedback on these changes, which will soon migrate to the other Keck instruments.

DEIMOS computer and disk changes (2006-Apr-07)
The Sun Solaris E450 workstation which has run the DEIMOS user interface software since the commissioning in 2002 has been replaced by a more modern, faster SunFire V240. Also, the new DEIMOS machine (known as polo, like its predecessor) now features a Triton RAID-6 array offering nearly 2,000 GB of disk storage space. The new machine runs the DEIMOS software significantly faster than before, and with the new RAID array the time to write a full DEIMOS image to disk has decreased from 35 sec to 2 sec.

Update: Problem with science CCD mosaic amp 1B (2006-Mar-25)
Troubleshooting prior to the late March DEIMOS run failed to reproduce the amp 1B dropouts seen during the early March DEIMOS run. Until this problem resurfaces, observers are advised to return to using the default 2-amp readout mode for direct imaging.

DEIMOS motor controller failure (2006-Mar-21)
One of the two Galil motor controllers on DEIMOS failed during the downtime between the DEIMOS runs. The symptom of the failure was that the unit would not communicate and displayed a red FAULT lamp when powered on. We replaced the failed controller with a spare, which appeared to operate normally until we tried to move the grating select mechanism; all attempts to move the grating select stage resulted in motor shut off - positioning error. We replaced the failed spare with a new spare, which worked fine. All systems now operate normally.

Problem with science CCD mosaic amp 1B (2006-Mar-01)
The B amplifier on CCD 1 is occasionally failing to read out properly, resulting in lost data on this CCD. This amplifier is used when reading out all spectral images, plus direct images acquired in 2-amp mode. Until we resolve this problem, observers are advised to use 1-amp mode for direct image readout in order to avoid losing data. The controls for 1-amp and 2-amp mode can be accessed on the User Config panel from the main DEIMOS dashboard.

newday command now invoked automatically (2006-Mar-01)
It is no longer necessary for observers to execute the newday command to allocate new data directories and configure FCS files when starting up the instrument. The newday command is now invoked automatically when the instrument software is launched. This change allows the fcstrack script to write its logfile to the FCS subdirectory.

DEIMOS RAID array crash (2006-Feb-01)
The RAID-5 disk array which hosted most of the DEIMOS data directories has suffered physical damage and is now unreadable. Two of the 5 disks in the array were lost, preventing us from reading the data from the other disks.

Change in the operation of GO ZERO/GO BACK buttons (2006-Jan-27)
The logic controlling GO ZERO button (used to tilt the current DEIMOS grating to the zeroth-order imaging position) has changed. Until now, the logic has been to store the current grating tilt and restore to this value when GO BACK is pressed. We recently learned that this can cause problems if the GO ZERO button is pressed during a telescope slew: the instrument rotator will typically be slewing, and as a result the grating may be slightly out of position at the time the position is recorded. Hence, GO BACK ends up restoring the grating to the wrong position, which causes FCS to remain idle instead of tracking. The logic has now been changed so that when the GO ZERO button is pressed, if the actual and user-requested grating tilts are close but not the same, the requested position is the one to which the grating will return via GO BACK. If the actual position is not close to the demanded position, the GUI will generate an error message.

Gratings refuse to unclamp (2005-Dec-26)
DEIMOS is experiencing occasional problems with gratings refusing to unclamp when they have not been exercised for a few days. We believe this is because the air pressure that drives these pneumatic clamps bleeds off over time, leading to insufficient pressure to unclamp the grating. If this happens to you, see the troubleshooting guide for tips on fixing the problem.

FCS problems (2005-Dec-09)
Over the last 6 months, the DEIMOS flexure compensation system (FCS) has experienced intermittent problems with loss of signal from one of the two CCDs in the system (FCS CCD 1), causing the system to fail. The problems became more frequent during the November run, culminating in a hard failure on December 2 after which the system remained in a "fail" state. Working with personnel at Lick, we opened the FCS dewar electronics box and located a connector which was not well seated. The connector is now properly seated and the FCS functions normally. As part of the troubleshooting, Bob Kibrick at Lick developed new FCS control software which can function using only 1 of the 2 FCS CCDs. Although the flexure compensation in this mode will not be as good as when 2 CCDs are used, it will be much better than having no corrections at all. Please see the FCS troubleshooting notes for further information on using this script.

DEIMOS grating system repaired (2005-Aug-25)
The grating move problem reported in the April news item has now been eliminated. Working with a team from Lick, we disassembled the grating select mechanism and located two loose screws on slider 3. These were retightened and the system laboriously retuned (modified the GRATOFFx parameters) in order to allow all gratings to clamp up at all DEIMOS rotator positions.

DEIMOS collimator mirror recoated (2005-Aug-10)
The DEIMOS collimator mirror received the same beautiful Gemini silver coating as the DEIMOS tent mirror received in the spring, thanks to Maxime Boccas and his team at Gemini. New throughput measurements are pending. The shift in image position relative to the value prior to removal of the tent mirror is -7.8 px in CCD_X and -4.1 px in CCD_Y, which is within the 10 px requirement.

DEIMOS tent mirror recoated (2005-Apr-25)
The DEIMOS tent mirror received a beautiful new silver coating at the Gemini summit coating facility thanks to Maxime Boccas and his team at Gemini. New throughput measurements are pending. A side effect of the removal and reinstallation of the tent mirror is that the optical alignment has changed by a few pixels. This means that observers should not attempt to use FCS reference files generated prior this run, since it may be impossible for FCS to maintain alignment on such data.

DEIMOS grating move problems (2005-Apr-02)
The DEIMOS grating select stage experiences mechanical interference which causes grating moves to fail at certain orientations of the instrument. The problem is worst when DEIMOS is at a physical rotator position (in DEIMOS coordinates, as displayed on the Dashboards "Rotation" panel) of 270°, and occurs when trying to move slider 3 past the in-beam position. The symptom is that the grating select move fails at this point. The solution is to rotate the instrument by 90° either diection and re-try the failed grating move. When possible, avoid changing the grating when the instrument is near 270° rotator position.

DEIMOS tent mirror receives temporary coating (2005-Apr-02)
The DEIMOS tent mirror's protected silver coating has been degrading since installation, reducing the DEIMOS throughput. After the last run, we removed the optic, stripped the old coating, and replaced it with a temporary coating of bare Al to tide us over until we can recoat with a new protected silver coating in mid-April. A side effect of the removal and reinstallation of the tent mirror is that the optical alignment has changed by a few pixels. This means that observers should not attempt to use FCS reference files generated prior this run, since it may be impossible for FCS to maintain alignment on such data.

Offsets to guider coordinates removed (2004-Nov-03)
The slitmask mill was serviced in October and this appears to have restored the original zeropoint of the slitmask coordinate system. The offsets which were previously required to align stars on the guider are no longer needed.

deeplog command updated (2003-Oct-11)
The deeplog script will now print out information on only the latest N files if you supply a numeric argument.

DEIMOS update (2004-apr-01)

Focus analysis script enhanced (2003-Oct-10)
The Analyze Focus script will now set the DEIMOS focus to the newly-determined value if you consent.

Re-arrange your GUIs automatically (2004-Oct-09)
With one click, you can now resize and move all of the DEIMOS GUI windows into their appropriate configuration. Simply go to the OpenWindows pulldown menu, select DEIMOS Utilities > Auto-arrange GUIs, and watch your windows morph into place!

Offsets to guider coordinates (2004-Sep-14)
Several observing teams report that the guider appears to have shifted, resulting in a 10-column offset between the predicted and actual positions of the stars in the guider when the mask is aligned. Observers should thus apply a +10 pixel offset to the intended x coordinates of their guide stars when completing the coarse phase of slitmask alignment; i.e., instead of typing 
	gmov 100 200 300 400
type 
	gmov 100 200 310 400

Several DEIMOS filters damaged (2004-Jun-23)
An unfortunate incident this spring has resulted in significant damage to several of the DEIMOS filters. The affected filters are: The damage was caused by a bent component in the filter wheel which grazed the surface of the filters when the optics were inserted or removed from the instrument. The scratches run along the center of the filters (covering less than 5% of the surface area) and are oriented in the spectral (not spatial) direction when the filter is in the beam. Plans for replacing the filters are on hold until we can establish the level of degredation to their performance. In the meantime, observers who would normally have used BAL12 for spectroscopy are advised to use the GG400 filter instead. Since DEIMOS transmits very little light below 4000 Å anyway, we do not expect a significant impact on observing.

FCS now runs automatically (2004-Mar-25)
The DEIMOS startup and shutdown scripts have been changed to launch the FCSTRACK script automatically, eliminating the need to perform the additional step of running FCSTRACK frm the pulldown menu during afternoon setup. Please be aware that if the DEIMOS software is shutdown and restarted, FCSTRACK will probably come up in its default (idle) configuration. Click the START button on the FCS GUI to re-enable FCS tracking.

Subtle problem with BLANK pixels (2003-Nov-21)
DEIMOS images taken between October 2002 and November 2003 are affected by a newly-discovered problem with certain data values. The problem is subtle and may be unnoticable for many observers. A full description of the problem and the simple solution is now available courtesy of Steve Allen (UCO/Lick).

spec2d version 1.1.4 released (2003-Sep-23)
The DEEP Survey team at UCB announces the release of a new version of the spec2d reduction package. Please see the release notes for full information, and consult the pipeline website for info on how to update your copy of spec2d. In addition to the revised version of the code, there are new calibration frames available; please click here for details. It is strongly recommended that all 2003 data be rerun with this new set of calibration files. Please note that an email list has been established to notify interested parties of developments related to spec2d. Email sent to this address will be directed to everyone on the spec2d_users subscription list which includes the developers here at UCB. Please contact to be added to the email alias.

AUTOPANE saves readout time (2003-Aug-25)
DEIMOS will now read out direct images and mask alignment frames faster, thanks to the new AUTOPANE keyword. When AUTOPANE is set to AlignArea (the ne default), the CCD system will readout only the illuminated rows if a direct image (without slitmask) is taken, and only those rows containing the alignment boxes if a slitmask alignment image is acquired. Spectral exposures are read out in full, as previously. Observers can control the setting og AUTOPANE using a new selector on the User Config panel of the DEIMOS Dashboard. For more information, please see this document.

DEIMOS Update (2003-Aug-07)

Use guidermon to detect guiding problems (2003-Jun-03)
The new guidermon script will monitor the guider signal-to-noise ratio (S/N) and generate an audio alarm if the S/N falls below the minimum threshold you specify. It can thus be used to detect the onset of cloudy weather and allow you to pause an exposure before it is degraded by clouds, or to be warned if the guide star wanders out of the guide box for some other reason. Once the signal has fallen below the threshold, the software will continue to monitor the guider S/N and will generate a different audio message when the guider counts exceed the threshold again. To use this script, open a local window on any machine in Remote Ops and type: 
	guidermon thresh
where thresh is the threshold at which to generate warnings. The appropriate value of the thresh parameter will depend on the brightness of the guider star and the guider integration time. A reasonable value would be 50% of the starting S/N.

Enhancements to DEIMOS IRAF package (2003-Jun-03)
The IRAF package keck.deimos has been modified to make observing with DEIMOS easier. The following tasks are affected:

"apropos" command for DEIMOS scripts (2003-May-29)
DEIMOS has a sizeable library of user scripts which can be used to control the instrument. You can now use the new apropos command on polo to search for commands related to a certain keyword.

DEIMOS slitmask imaging widget (2003-May-28)
Observers should acquire a direct image of each of their slitmasks in the afternoon and run xbox on them to verify that they function as expected. Observers can now use the DEIMOS Slitmask Imaging Tool to acquire these data rapidly and conveniently.

spec2d version 1.1.3 released (2003-May-20)
The DEEP Survey team at UCB announces the release of a new version of the spec2d reduction package. Please see the release notes for full information, and consult the pipeline website for info on how to update your copy of spec2d. The new version includes a patch for a noteworthy bug which affects all masks milled after the middle of October, 2002. The release notes provide the details of this bug, but it is important to note that all post-October masks need to be rerun using v1_1_3!

DEIMOS update (2003-may-15)

Generating DEIMOS slitmask maps (2003-May-03)
The DSIMULATOR slitmask design package can generate an optional plotfile which consist of Lick MONGO plot commands; however, few people have access to this software package. For the convenience of observers, a new IRAF task will accept the DSIMULATOR plotfile as input and will generate a plot using the STSDAS.IGI plotting program. You can download this script and run it at your home institution, or bring your DSIMULATOR files to Keck and generate the maps here.

Generating DEIMOS guider finder charts (2003-May-02)
Dan Magee (UCO/Lick) has generously donated a wonderful software package which generates nice DEIMOS guider finder charts using the Digitized Sky Survey and the output files from DSIMULATOR.

Experimental ds9 released (2003-Apr-24)
A new, "experimental" version of the ds9 image display tool is now in use for DEIMOS observing. This version features two new ways to compute image statistics, accessible under the Analysis menu:

Zeroth order imaging available on slider 4 (2003-Apr-24)
The staff of Keck and Lick Osbervatory recently completed mechanical modifications to increase the range of motion on slider 4, allowing it to reach the zeroth-order imaging position. Observers can now perform slitmask acquisition and MIRA on slider 4. Various user scripts which involve zeroth order imaging have been updated to reflect the new capability, including gozero, image_user_masks, and align_check

New script aids afternoon setup (2003-Apr-24)
Observers should acquire a direct image of each of their slitmasks in the afternoon and run xbox on them to verify that they function as expected. To speed the process of acquiring the direct images, observers can use the new image_user_masks script. By default, it will reconfigure DEIMOS for direct imaging on the current slider and acquire images of all user masks. To image only certain masks, list their names as arguments. The slitmask checklist has been updated to describe the use of this script.

spec2d version 1.1.2 released (2003-Apr-16)
The DEEP survey team at UC-Berkeley announces the release of version 1.1.2 of their IDL-based DEIMOS data reduction pipeline sofwtare. Known as spec2d, this software takes an input a set of undithered DEIMOS spectral exposures and produces a set of individual, sky-subtracted 2-D spectra for each slitlet. Please see the description of the spec2d software pipeline for further information on obtaining and using the software.

DVD writer available (2003-Mar-30)
DEIMOS observers can now write their data to DVD (or CD) using our automated writer and software. The observatory provides both the media and a GUI for completing the writing. Instructions are provided here . Each DVD can hold 4.7 GB of raw data, so it is generally possible to store a run's worth of data on a few discs. To access the GUI from any instrument account, select Instrument Utilities > Autobackup from the OpenWindows workspace menu.

FCS software upgraded (2003-Mar-25)
Three components of the FCS software have been upgraded:
  1. The FCS keyword library (service deifcs)
    All FCS-related keywords have been moved into the deifcs service. Previously, some FCS keywords (e.g., FCSMODE, FCSMSG) were contained within the science mosaic keyword library (service deiccd). Many new keywords have been added, providing access to the individual measurements computed for each FCS CCD. The FCSRMXxx keywords provide measurements of the FCS system's RMS tracking error during each science exposure. These new keywords are recorded in the FITS headers to provide a record of how well FCS was able to stabilize each science exposure. Note: the FCSCORRX and FCSCORRY keywords have been replaced by the FCSCORXM and FCSCORYM keywords, respectively.
  2. The FCS GUI
    The FCS graphical user interface has been completely re-worked. It provides many new features, including user-adjustable plot scales for various plots of the FCS tracking performance. Status displays provide detailed information regarding various warning and lockout conditions, and pop-up alarms notify the observer of any conditions that would degrade FCS tracking performance. The new GUI also provides features for taking FCS reference images.
  3. The fcstrack script
    The fcstrack script has been significantly upgraded to support the new capabilities of the FCS GUI. The revised correction algorithm provides improved image stabilization, especially when DEIMOS is rotating at higher velocities as occur when tracking objects near the keyhole. The revised algorithm also provides more robust handling of cosmic ray events; previously, such events would sometimes result in spurious FCS corrections.
Both the FCS Startup Checklist and the FCS Notes have been updated to reflect the changes.

FCS upgrade reduces scattered light (2003-Mar-02)
A recent upgrade to the flexure compensation system by Lick and CARA added hardware which turns the FCS lamp on only during an FCS exposure; previously, it was always on when FCS was running. Since FCS integrates for only a few seconds on each iteration, this should reduce the amount of scattered FCS light on the detector by a factor of 3 more.

Observing preparation checklist published (2003-Jan-31)
The new DEIMOS Observing Preparation Checklist tells you the things you need to know to prepare for your DEIMOS slitmask observing run and points out some key resources you should bring along.

Slitmask calibration widget released (2003-Jan-31)
The new Slitmask Calibration Widget automates the process of acquiring calibration data (arcs and internal flats) for a series of DEIMOS slitmasks. See these instructions on its use.

DEIMOS Update (2003-Jan-09)

Filter focus discipline released (2003-Jan-08)
De Clarke of Lick has written new software which will automatically adjust the focus of DEIMOS to compensate for the difference in focus values between the various DEIMOS science filters. The software takes as parameters a global shift in focus (derived from daily focus checks in the R band which you should run in the afternoon) and focus offsets for each filter relative to the R band. The DEIMOS focus is adjusted each time a filter is changed, with the changes being "invisible" to the observer.

New deeplog utility generates electronic logsheets (2003-Jan-08)
The new deeplog utility allows you to generate a listing of the images in your data directory to complement or replace your paper logsheets. The text output produced by the script isn't pretty but has the key benefit of being machine readable. See some sample output.

Flexure compensation centroiding improved (2003-Jan-01)
Bob Kibrick and Drew Phillips of Lick have developed a new cross-correlation algorithm for the DEIMOS flexure compensation system that simplifies the afternoon setup procedure and is far less prone to losing its "lock" during the night. The new algorithm is now the default and has performed perfectly since being deployed in late December.

Ghosting seen in 900-line grating (2002-Oct-22)
Drew Phillips has written a nice report on unusual ghosting seen when using the 900-line grating on DEIMOS. Until further notice, observers should use dome flats only for the 900-line grating, and other gratings that work at a similar tilt. Observers should be aware of the problem and watch for it in their data.

DEIMOS Update (2002-Oct-17)

DEIMOS slitmask design software released (2002-Aug-09)
Drew Phillips (UCO/Lick) has announced the initial release of his IRAF-based software package for design of DEIMOS slitmasks. Details on the installation and use of his software may be found at his website.