Because of the integral use of the LAMA (Laser Machine) with MOS/SIS as a supporting instrument the need for a functionally reliable system is readily apparent. The laser is needed in real time for mask generation. Typically an observer will acquire a star field with the telescope and then create a mask over that field with strategically placed slits to allow selected starlight through to the detector. This mask file is sent to the laser machine for processing. Some masks may involve over 150 apertures and take more than one hour to make. The processed mask can then be taken out of the LAMA and inserted into the spectrograph for immediate use. Any failure in the ability of the LAMA to cut these masks could result in significant time lost where MOS/SIS could not be used (three to four months for example).
Realizing that any of these problems could jeopardize a MOS/SIS run, we solicited 11 vendors that supplied lasers and laser cutting systems. Prices for these systems ranged from about $90k to $200k. We contacted two companies whose prices where closest to our budgeted amount and set up a meeting to test their systems and see if they could meet our cutting requirements. Neither system performed a complete cutting sequence to our satisfaction due to technical difficulties.
We decided to contact Lee Laser Inc. and invited a representative to come out and see our current system and make recommendations toward a solution. In May 1994 Mr. Chong Lee, president of Lee Laser, met with us at the summit and we designed a new laser system that could be incorporated into the existing system with a minimum addition of material and cost. Instead of replacing the existing laser and X-Y stage we decided to add only a laser system on the same table so that its beam could substitute for the existing one.
During the period from February 1993 to August 1994 we experienced four major failures of the current laser system. The first two could have led to loss of use for several months. In February 1993 we found water, used to cool the laser pump lamps, leaking from the tube joints in the deionizing and reservoir units. In April 1993 we found water leaking from the joints up at the laser cavity causing a pool to form around the 1700 volt supply. When power was needed to start the optical pump lamps, the supply shorted to ground through the water. Either one of these events if left uncorrected could have caused a sever failure of the system. In May 1994 the glycol pump failed when internal solder from a plumbing joint came loose and wedged itself in the pump rotor. Finally, in August 1994 the lasing capability had decreased to a point where no cutting could be done. We replaced the optical pump lamps that had been used for about 3 times their normal life, and then did a realignment of the laser crystal that had come out of alignment due to thermal cycling.
The new design concept allows us to continue to use the existing laser system, and with a small configuration change the new laser system can be started and used almost immediately. Two innovative techniques make this possible: first, the new laser system is placed so that the exit beam is perpendicular to the existing beam at their first intersecting point and second, a partially reflecting mirror mounted on a rotatable gimbal allows the selection of either beam. The new laser system is mounted on a sub-plate that makes it independent and easy to integrate with the new support structure.
Currently, the mirror gimbal mount, new support structure and a new final focus lens assembly have been made by Dan Sabin. The new laser system and associated electronics have also been purchased, shipped and sent to the summit waiting the completion of a new laboratory on the fourth floor. A mechanical test of the new laser and support structure has demonstrated a very good initial alignment of the new beam with the existing one. By the end of 1994 we plan to move the existing system up to the new lab and add the new laser system to it, giving CFHT a reliable and easy to maintain mask cutting machine for years to come.
The background on this project was described in an article in the previous bulletin. To recapitulate, a cooling unit located on the dome floor ducts up cool air into the mirror cell. This operates only during daylight hours in order to prevent daytime heating of the mirror and cell.
Figure 10: Cooling Unit on the observing floor.
Progress has been made by Ralph Taroma, Eric Willett and Ed Stokes by completion of assembling, integration, wiring and testing of the following components: an extended surface prefilter, a HEPA primary filter, a 3 speed squirrel cage blower, an evaporator (cooling) connected to the building glycol system, an electronically controlled electric duct heater (temperature control), a 14" supply duct from the unit to the telescope, a 10" distribution duct around the mirror cell, and twelve 4" ducts into the space below the mirror. The system will be controlled by a programmable controller using the existing dewpoint sensor to inhibit operation in high humidity conditions, surface temperature sensors (RTD) on the mirror side, and air RTDs located at the air supply duct, and air input to the cooling unit. The unit is located adjacent the freight elevator for easy access to the glycol supply. Glycol cooling was selected to improve energy efficiency and to avoid adding a heat source in the dome. An insulated shroud has been installed around the mirror cell to maintain temperature stability.
Figure 11: The Primary Mirror is sealed from the dome environment by the white isolating material. The yellow duck surrounding the Primary Mirror is distributing the cool air over the mirror surface.
The first project milestone, to blow microscopicaly filtered air into the mirror cell, was realized by the first week of November. Initial tests resulted in measuring an output air flow of 1000 CFM. This now gives us the capability to effectively control dust from entering the mirror cell during telescope non-operational hours. A major driver to attain this goal was to gain protection during daytime hours in the event dust levels rise during construction of the new access road to CFHT.
The second phase of the project, to complete the programmable controller installation, software programming, configuring the quick disconnect for ease of daily operation, and finally blowing chilled air is progressing rapidly.
The detector situation is evolving progressively and steadily at CFHT. We are in the process of acquiring many new detectors which will be commissioned during calendar year 1995.
Last August, the detector and software groups installed the second release of the GENIII controller. Observers can expect a major improvement compared to the initial release. The read out time for a full 2Kx2K CCD is now 160 seconds, an improvement by a factor of 2. This is not the limit of the system, and the detector group has some ideas in reserve for improving this figure further. The so-called "zero-pixel" problem is for all practical purposes solved, and the STOP, ABORT, and PAUSE buttons are again functional.
The corporation has now taken delivery of 2 of the 3 dewars fabricated by G. Luppino from IfA. These dewars have a liquid nitrogen hold time in excess of 24 hours. The first one contains an engineering chip identical to Loral3, and the second Tek3. CFHT plans to order another batch of 3 dewars imminently. Tek3 has been used at the telescope by G. Luppino last October but with his own controller. It will be some months before the detector group can fully integrate this chip into the CFHT's GENIII world.
M. Lesser from Steward Observatory has successfully thinned two Orbit chips. One of these comes to CFH and is being integrated in the third new dewar at IfA. This is a 2Kx2K thinned device with 15æm pixels and stunning QE !
Bruce Woodgate of GSFC has offered to provide the corporation with one STIS chip. This is a 2Kx2K thinned device with very high quantum efficiency. The pixel size is 21 æm.
A collaborative agreement with Gordon Walker from UBC has resulted in the acquisition of a thin slender Loral chip for use at the Coude spectrographs. This device is being thinned by M. Lesser at Steward Observatory. We expect delivery on this device within a few months.
We are still awaiting delivery of a Reticon 2Kx2K thin CCD (13.5 micron pixel) which was due in December '93.
CCD type size RON CONTROLLER READ TIME Loral3 Thick 2048x2048 8e- GENIII 160 sec Lick2 Thick 2048x2048 10e- GENII 5 min. RCA2 Thin 620x1024 50e- GENII RCA4 Thin 620x1024 60e- GENII Both PHX1 and SAIC1 have been decommissioned.
CCD type size pixels status Tek3 thin UV 1024x1024 24 um in dewar, working on pre-amp. fabrication and tests Orbit1 thin UV 2048x2048 15 um Thinned successfully, chip being integrated in dewar @ IfA STIS2 thin AR 2048x2048 21 um Awaiting Integration Reticon thin 2048x2048 13 um Ordered. Delivery overdue... UBC1 thin UV 200x4096 15 um Being thinned
R. Arsenault, S. Milner
Dome Shutter Seal Replacement: The dome shutter seal replacement was completed by Ralph Taroma, Ed Stokes and Rohendra Atapattu. The seal utilizes a strip of ozone and sunlight resistant polymer attached to the exterior of the shutter which glides on the dome surface. The new seal has provided a significant improvement in preventing the intrusion of water at the shutter and shifted our focus to leaks at the ventilation louvers (one of our next projects).
Primary Mirror Cover Seal Retrofit: The existing foam seal between the leaves of the primary mirror cover was removed and replaced with a teardrop shaped hollow rubber strip mounted on top of alternate covers. An aluminum strip was then installed on top of the new seal to provide compression and to prevent damage to the seal when working on top of the cover. The mirror cover controls were also modified to allow the leaves to open alternately with the seals on the top leaves. Eric Willett modified the pneumatic actuation system and Wiley Knight installed the seals all while the covers was in place and operational. We expect this system to significantly reduce the intrusion of particulate matter from above and to increase the effectiveness of the mirror cooling system.
General Building Renovations: An ongoing program of cleanup and renovation of the building is underway, including: painting, heavy cleaning, and reorganization. To date Warren (Keala) Horn has been successful in completing the following work: paint, clean, and decorate the first floor entry, paint and clean the second floor hall, paint the electronics labs, paint the hatchway and machine shop floors. We have also, installed wire reel storage racks on the second floor, and energy efficient lighting for the hallway aisle lights. Darkrooms: The darkrooms and plate baking room on the 3rd floor were remodeled to provide an office for the daycrew. Previously, the crew was scattered throughout the building. The new office space provides a clean, well lit environment for six to seven persons.
The former gas lab on the south end of the 3rd floor has been set up as a lab for the electrical and mechanical technicians.
A space adjacent the Auxiliary Control room on the fourth floor is being remodeled to house the LAMA laser system, providing easier access to the system for observers. We expect to occupy this space in mid November. The datalogger has been relocated from its previous countertop location in the back of the fourth floor to a rack in the rear observing room. In the process, the cabling has been organized, identified, and documented.
Top End Handling Ring Control Retrofit: The control system for the telescope top end handling ring is in the process of being modified by Rod Hendrix to provide simplified operation and reduced maintenance. The new system includes new operator and annunciator panels, simplified connections, positioning indicators, and new cabling.
Telescope Compressed Air System: A refrigerant dryer has been added to the system to increase moisture removal and reduce maintenance of the dessicant dryers. The existing dessicant dryers are being rebuilt to increase efficiency and improve reliability. Planned improvements include reconfiguration to provide increased redundancy and installation of a programmable controller to monitor the system and control compressor cycling.
Dome Hydraulics Retrofit: The dome hydraulic drive pumps and control system are being replaced. The old pump system suffered from frequent breakdowns and poor parts availability. The retrofit includes new pumps plus a backup pump to allow immediate switchover in case of a failure and high efficiency oil coolers to reduce the heat dissipation in the dome. As the age of the control system makes parts availability a problem, this system is also being replaced. The new system will be based on a programmable controller, providing continuous system monitoring and simplified troubleshooting.
Dome Shutter Drive System Retrofit: The modifications to the dome shutter drive system are in the final budgeting stage, awaiting quotations for a few key items. Replacement is proposed for the 1995 summer shutdown period. The proposed retrofit includes replacement of the existing wound rotor motors and external disk brakes with totally enclosed fan cooled inverter duty induction motors with shaft mounted disk brakes, fabricating and installing new motor/gearbox mounts to improve maintenance access, replacing the existing spring rewind cable reels with motorized retrieve units to increase the number of conductors and reduce cable wear, provide microprocessor controlled motor starters located on the dome catwalk, encoders on the motors to provide position information, and interface the system through a programmable controller to allow system monitoring and computer control. This system offers the following improvements: all controls removed from moving shutter to allow simplified troubleshooting and repairs, relocation of heat sources away from the shutter opening, reduced repair and maintenance costs, better control, and simplified computer interfacing.
R. Hendrix , R. Atapattu
A new Hewlett Packard workstation 747i has been recently acquired to serve as a host session for the observing accounts (data acquisition and instrument control). This computer is equipped with two monitors controlled with a single keyboard and mouse, the main monitor will display the usual Pegasus session with the PSM menu bar and the associated xforms < related with a particular observing session. The auxiliary monitor will be reserved for image display and graphics allowing the observer to have a full screen to interact with the image tool. The HP 747i is a high performance workstation using the PA-RISC 7100 processor running at 100 Mhz, this processor coupled with 256 KB instruction and data caches and 128 BYTES of main memory is capable of displaying a 2kx2k CCD image in 4 seconds and a 4kx4k (MOCAM image) in 15 seconds, this will greatly improve the observing efficiency. We plan to have this processor operating at the summit by the first run of MOCAM. A third monitor will be added to display instrument and observatory status soon after the tests and familiarization with the new environment is completed.
With eleven months of 1994 already behind us, I am happy to report that our level of safety awareness at CFHT continues to pay off.
We had no major accidents / disasters [mercury or oil spills] this year but of course we did have a few near misses and sometimes even a direct hit! One of Parker Ranch s brown cows had an early calling to a certain ‚cossais salle … manger, [McDonalds], after one of our station wagons came in contact with the poor wee beastie.
Driving conditions on Mauna Kea can be extremely dangerous, especially at night, with fog, drizzle and a black or brown cow thrown in for good measure. Same goes for driving on the Saddle Road where black pigs can suddenly appear in your path and bring your vehicle to a screeching halt and a few thousand dollars damage to the vehicle, not to mention what it does to our insurance rates. Also, at the summit, in the evening, when it is required to drive with very limited illumination, we have to take extra care when driving around the buildings. When GEMINI realign the approach road to CFHT, we hope to extend the guardrail [Hilo side], well down into that area and also have some stakes with reflectors.
Earlier this year, Larry Olsen, my counterpart at KECK, gave an excellent seminar on "The Effects of Altitude Sickness", first to the staff at our Waimea headquarters, and just recently to our summit daycrew; everyone spoke highly of this very well presented seminar. [See Larry s article in this issue.]
Although we all regard our dome as the place where we house our very, precision oriented telescope, comprising of many optical / mechanical / electronic / software / cryogenic parts, the tools / handling devices we use daily to service our instrument, are predominately of industrial caliber. For instance, our main dome hoist is capable of handling 25 tons and the mirror carriage capable of handling 35 tons. We have hydraulic jacks to raise or lower the telescope, for alignment purposes, these jacks are 100 ton capacity. Some years ago, a past employee of CFHT had his hand crushed while performing this procedure and resulted in the loss of fingers.
Our forklift, itself weighing over 5000 lbs., is used continually to interchange such things as the array of 600 lb. counterweights around the mirror cell. We often wonder how we managed before we had forklifts and it sure is a very useful and powerful tool, BUT, treat it with respect; there have been a few instances of limbs coming mighty close to the chopping block'. Also, watching where you are driving the machine, normally the floor area is flat but from time to time you just might run into an uncovered puka and could result in this 5000 lbs. of vehicle flipping over.
This was driven home to us earlier this year when a construction worker on the SUBARU site died, when the forklift he was driving flipped over and pinned him to the ground.
A new ambulance, [Emergency Evacuation Vehicle], was purchased recently by MKSS and so all who work at the summit should become familiar with this new vehicle and its contents. It is normally parked outside the CALTECH dome. I have started to organize familiarization sessions for all concerned CFHT employees.
Finally, always wear your protective gear when you are required to do so, and WATCH THAT BACK! Lift with your leg muscles, not your back muscles, and, as always,