Log of NOTCam openings in reverse order (newest first). Opening 22 Jun - 14 Jul 2021 (filters to IAC, KDPs installed) ------------------------------------------------------------- NOTCam has been 3 years without an opening, which is a new record. The NTE people would wish to re-install the KDPs for more testing and also take out 7 filters and have them leak measured at the IAC. NOTCam was set to warm up by switching off the PTR as observations ended in the morning of 22/6. On 23/6 it was transported to the service building (CP,GC). On 28/6 NOTCam was opened (CP) and the aperture wheel was investigated and found to have some slack (CP). The pressure sensor was found to be dirty and cleaned. The filters to be sent to the IAC were taken out and packed in their marked boxes (AAD). These were the 7 filters: GBP 213 Helium I A 348 FW1 pos 12 214 Pa gamma 482 FW1 pos 9 215 J-continuum 527 FW2 pos 3 216 Pa beta 428 FW2 pos 4 222 Y_narrow 210 (already in its box) 236 Y 528 FW2 pos 16 237 Z 2964 FW2 pos 2 The filters were transported to the IAC (RC) 30/6 and measured in their lab by Luis Rasilla from 0.3 to 2.6 microns in steps of 1 and 3 nm. We know of several leaks from before, but did not have this long wavelength coverage for all filters. On 2/7 the filters were mounted back exactly as they had been in FW1 and FW2 (AAD). The getter baking was started in the evening. Jacob came from Denmark on 3/7 with the two small KDPs, the ones that go in the aperture wheel slots. The larger KDP intended for FW1 could not be located. The installation of the KDPs in the aperture wheel was done 4/7 (CP,AD) with the conditions in the clean room at +30 deg Celsius and 14% relative humidity. CP lowered the crystals into the deep slots using a tiny strip of an optical tissue (otherwise one would have to drop them). The one named KDP went to slot 2 and the DKDP to slot 9 (this one has a scratch from 2016). When testing the initialization of all wheels, we had timeouts on the 3 units: filterwheel 2 (C), grism wheel (E), and focus (FOC). They all moved, but the init switch was not detected. This was found to be because of one wire located at the connector of the stop wheel (D) that lost connection. This wire was the shortest of all in the whole bundle so it was the logical point to break. Carlos soldered it back onto the connector. Due to this unexpected problem, it took a bit longer than planned to do the closing, but less than 3 hours passed from opening the KDP boxes until NOTCam was on the pump on 4/7 at 15:30 (UT 14:30). The pressure went nicely down and in 30 min was at 0.06 mbar, after 24 hours 3.3E-03 mbar and at 2.1E-03 mbar in 48 hours. The temperature at the central wheel unit was 40 degrees due to the hot getter. Transporting back to the dome was done on July 7th (CP, JC) and pumping continued. The pressure was at 1.6E-03 mbar and reached just below 1.0E-03 when cooling was started. On Saturday morning July 10th at UT=05 the PTR was started by Paco. The detector temperature went gradually down from +20 deg to -103 degrees during PTR cooling with the pump still on. This took 53 hours. Monday 12/7 Carlos started filling the cryostat with LN2 at UT=10 and after the first filling, the valves were closed and pumping stopped, upon which the pressure shot up from 6E-04 to 5E-03 mbar. When the detector was at -170 deg Celsius, the pressure was still at 1.6E-03 mbar, and did not improve further than to 1.5E-03 when operating temperature was reached. Tuesday 13/7 the pump was attached again and while pumping we get to 5.0E-04 mbar, but when stopping the pumping, again the pressure goes back to 1.5E-03. This probably points to a tiny leak somewhere. If it is due to outgassing, the pressure will improve over time. The detector was tested with a few darks and all quadrants responded well. Wednesday 14/7 NOTCam mounted on the telescope. Viki and Anni does a few tests and the QC and will observe for Galbany. Opening 11-25 June 2018 (to improve vacuum): -------------------------------------------- It was decided that a thermal cycle with baking of the getter is needed after having NOTCam cold for 2 years and 3 months. Opening NOTCam gives us an opportunity to take out the DKDP and send it to NBI, Copenhagen. NOTCam was used on Sunday 10/6. We avoided to fill it in the afternoon since it was filled the 9/6. The detector temperature was slightly higher than normal during the night, but nothing noted in the data. As soon as notcam-calibs ended in the morning of 11/6, the detector controller and the PTR was switched off. NOTCam was transported the same day to the service building. By the morning of 15/6 the internal temperatures of NOTCam were +16.5 degrees. Opening was done by Carlos, Pedro and Amanda. The getter was taken out, and set to baking at 140 degrees in the special oven. The aperture unit was dismounted and opened and the DKDP, sitting in slot number 2, was taken out and stored in its box in a glass jar filled with baked silica gel (to protect it from humidity). To be transported to Copenhagen. The nuts with spacers were put back in the two empty slots 2 and 3 to avoid any light leaks. The inside of NOTCam in general was inspected for dust. Found some dust, but not much, and this was vacuum cleaned. Both the aperture unit and the central wheel unit were opened (all lids) and cleaned of dust. The pressure sensor was dismounted and cleaned (was quite dirty). All units were mounted back on the optical table, all wires reconnected, and all wheels inspected by switching on the motor controller, starting the notcam-interface and initializing all moving parts. All temperature sensors were checked and showed reasonable values. Carlos connected the roughing pump to the fill tube of the LN2 tank in order to suck out the water that has accumulated over the 5 years. Closing was done on Monday 18/6 by Pedro and Carlos. The warm getter was mounted back on the optical table. After closing the instrument, all wheels were tested again for correct initialization. NOTcam was taken out to the electronic workshop and the pump was connected and started. After 20 minutes the pressure was 6.8E-02 mbar. After 12 hours 2.5E-03 mbar, and after 48 hours 1.3E-03 mbar. All nominal values. Transport back to the telescope on Wednesday 20/6 by Carlos and Pedro. Pump was re-conncected and vacuum pumping restarted. Also, the roughing pump used to suck out water from the LN2 tank was connected and restarted. Vacuum reached 3.5E-04 by Friday 22/6. Cooling started 23/6 with the PTR while keeping pumping. On 25/6 the temperatures were -105 degrees and pressure 1.8E-05 mbar. The LN2 filling was started 25/6 at UT=11. The array controller was switched on when the detector had reached -192 degrees and the array was tested cold. Everything normal. AAD. ----------------------------------------------------------------------------------------------------- Opening 7-14 March 2016 (to troubleshoot filterwheel 1 problems): ------------------------------------------------------------------ NOTCam was set to warm up on February 25th. New O-rings for the LN2 tank fill/vent tubes had been ordered and arrived just in time for installation. NOTCam was then transported to the service building clean room on March 7th and opened the same day. The KDPs were dismounted and put in a dry environment. The getter was set to baking. Filter-wheel 1 was examined and a not-well-solded motor cable was identified and fixed. One KDP was put back for further testing inside NOTCam. (It turned out that the crystals had been swapped and the one sent to Copenhagen was the KDP, while the one we kept was the DKDP. Note added 2018. AAD.) NOTCam was closed and set to pumping March 8th. Transported to the dome on March 14th. Pumping was on-going for 6 days. The cooling was done very slowly with the PTR only from 14th to 21st. This cooled down all the way to -180 degrees only by the aid of the PTR. This gradual cool-down took a much longer time (7days), but it was made to see if the KDPs acted differently, and also to find out at what temperature the KDP crystal would become structured. It was also expected that having NOTCam already cooled so much before filling with LN2, would cause less harm to the fill/vent o-rings when finally filling the LN2, but this made no difference. LN2 cooling started on March 21st and the array was tested cold on March 22nd. Went on the telescope 23rd. AAD. ----------------------------------------------------------------------------------------------------------- Opening 23 Dec 2015 - 15 Jan 2016 (to install KDPs): ------------------------------------------ This opening was done more than 2 years after the previous opening. NOTCam had been kept cold for 2 years and 2 months, a new record. NOTCam was set to warm up on December 23rd 2015. Transported to the service building clean room on January 11th and opened the same day. The pressure sensor was cleaned, the getter set to baking, and all gear-wheel bearings and flexible couplings checked carefully. The test KDPs arrived from Copenhagen just in time on January 15th to drive them up to the telescope before closing the instrument. The spare J-band filter, mounted in the stop wheel, was taken out to be sent for a proper leak measurement in CPH. Two test blockers, one KDP and one DKDP, were installed in the small slots in the aperture wheel. They were exposed to a relative humidity of about 32% in the clean-room for about 1 hour before NOTCam was closed and set to pumping. AAD. ------------------------------------------------------------------------------------------------------ Opening Oct 2013 (to install KG2 filter and WF imaging mask): ------------------------------------------------------------- It was realized during the warm-up and pumping session (without opening) in June 2013 that the vacuum obtained is lost as soon as LN2 filling starts. This points to deformed o-rings in the fill/vent tubes and calls for exchange to new o-rings. Together with an annoying dust speck appearing in September, plus a wish to install the WF imaging mask (again) and the KG2 filter that arrived in July 2012, prompted us for an opening of NOTCam. NOTCam was emptied of LN2 while still mounted on the telescope. This is done by putting the telescope at altitude 6.4 degrees and turning the rotator to -65 (rot-pos -65). Done by Tuomas in the morning of Oct. 21st right after observations at UT=08. The procedure is as follows: 1) shutdowbobssys notcam 2) switch off the NOTCam array controller 3) take out the cork in the NOTCam fill tube 4) switch off the PTR 5) take the telescope down (TCS: alt 6.4) 6) turn rotator to easy emptying (TCS: rot-po -65) 7) let LN2 flow out as much as possible before parking the telescope 8) TCS: zenith According to Tuomas he did not see any liquid nitrogen coming out. This could be due to the fact that NOTCam was actually not filled the morning before (Oct 20th), so that it was pretty empty already, and in fact, the detector temperature was as high as -193.3 degrees when warm-up started. After 24 hours the detector temperature reached -115 deg, after 48 hours: -10 deg, and after 72 hours: +7 deg. On October 24th while NOTCam was still in the dome, the fill and vent tube o-rings were exchanged with new o-rings. The old ones were, as suscpected, deformed into a triangular shape. The surfaces, both the tube and the housing, were polished with steel wool. Then, cleaned with acetone. In the same fittings, the o-rings on the wessel side, were also exchanged, with the same procedure of polishing and cleaning. The Balzer pressure sensor was cleaned and calibrated towards the sensor on the pump itself. Agreement was found down to a pressure of 1E-06 mbar. Also, the o-ring of the pressure sensor was exchanged. Later the same day NOTCam was transported to the service building clean room and opened up. The charcoal getter was taken out and placed in the oven in the kitchen for baking, to be started Sunday 27th. All this was done by Carlos and Pedro. On October 25th the interior of NOTCam was dismounted by Pedro and Amanda. When removing the aperture wheel section from the cold table we had access to the collimator lens and discovered a tremendously large dust speck of shiny white colour, located more or less in the centre of the lens. This was the culprit, i.e. the origin of the dust pattern seen in NOTCam flats with both cameras since September 14. It was blown off the lens and later collected with the small hand-driven vacuum cleaner for the clean room. It has probably arrived from the aperture unit. When opening the aperture unit lid, there was a collection of metallic dust all over, but in general very small powder dust, and not big specks like the one on the collimator lens. The metallic dust was properly cleaned away. By the way, the hand-held small vacuum cleaner needs repair, or we need a new one. The start/stop button has very bad contact. At the end of the day I could not use it any more. Inside the aperture wheel the 64 micron slit in aperture wheel position 7 was taken out to give space to the WF camera imaging mask. Pedro devised a method to fix this one with copper springs. We aligned it as best we could. The central wheel unit was dismounted from the cold table. In filter wheel 1 all polaroids were taken out. These polaroids were all IR sensitive but simply polaroid film. In addition, they had been mounted with plastic (!) spacers in the filter holders. This was not realized until they were removed. I now question how much outgassing these could have caused over the years. In any case, the polaroids were useless and deformed during the thermal cycling. The new filter, called KG2 by the provider, although sent as a replacement of our out-of-spec KG4 which turned out to have much too little throughput in the Z-band, is only 0.5 mm thick. It was mounted in one of the original (thin, black) filter holders with a copper spring. The entrance window of NOTCam was cleaned with the 99.8% pure metanol (bottle stored in clean room). On October 28th NOTCam was reassembled by Graham and Pedro. The baked charcoal was mounted on the cold table, NOTCam was closed and set to pumping. All mechanisms where checked after pumping had started. To prevent damage to the vacuum sensor it has not been left permanently connected, but just powered for a short while occasionally to put some values in the database. On the 31st NOTCam was transported back to the dome by Pedro and Carlos. Pumping continued. On Monday Nov 4th, after one week of pumping the dewar pressure is 2E-04 mbar at a temperature of 15 degrees. From now on the pressure sensor is left connected all the time. The new PTR hose was installed by Carlos and Pedro installed as a replacement of the old supply hose. Leak testing was done with the old hose under water, but no leak was found in the supply hose (also the most repaired one). The slight leakage of Helium we had over the last months, needing re-filling of He about every 3 weeks, was found to be due to an o-ring too much! Carlos believes that a missing o-ring from the protection cap got inside and therefore there were two o-rings instead of one. This probably caused the microleak. It was noted that the new hose is a bit better because it is more flexible, the outer net is made with thinner fiber. AAD. ----------------------------------------------------------------------------------------------------- Opening Mar 2012: Installation of BK7 and KG4 new filters --------------------------------------------------------- NOTCam was emptied of LN2 while still mounted on the telescope on the morning of 11/3 (IS) after the last observing run. Before this the detector controller was switched off and the PTR was switched off. By the 14/3 the temperature of the cold table was above zero. NOTCam was transported to the SB and opened in the "clean room" (IS,CP). The charcoal getter was taken out and put in the oven for baking at 150 degrees (CP). Then Carlos started investigating the "water in the LN2 tank" hypothesis. The humidity in the "clean room" was measured to be about 10%, but placing the humidity meter at the outlet of the LN2 tank, it rose to 40%. Carlos set up a test assembly to see how fast vacuum pumping can get rid of liquid water. The water level of a small (0.5 liter) bottle decreased by about 3 mm in one hour of pumping with the assemply used. On the 15/3 the inner parts of NOTCam were dismounted to get to the central wheel unit (AD,JJ,SG). We used hair net, face masks, latex gloves and earth straps all the time when touching the interior of NOTCam. It became clear that we need to buy latex gloves of size medium, as well. Too large gloves are impossible to work with in a safe and efficient way. This problem delayed the whole filter mounting procedure a lot. One of the 5 screws fixing the central wheel unit to the cold table was extremely difficult to get out, and we needed help from Carlos. The filter wheel unit was first placed horizontally on the working table to take the lids off, and to take out one of the polarizers (135 degrees). (All polarizers are deformed.) For the 1mm thick KG4 filter we used the black 3mm thick filter holder (the type which is useless for all other filters), and a spring of the copper ring type. There are wider and not so wide rings. We had to use the not so wide, as the black filter holders are a bit smaller than the grey ones. All new items that were installed were first washed with acetone. It was difficult to install the thin filter with the unit in a horizontal position, so the filter wheel unit was put vertically on the table. The KG4 filter went into the position previously held by the 135 degree polarizer in filterwheel 1. Then the BK7 filter was mounted in the previously empty slot in filterholder 1. The BK7 filter is 5.4mm thick and needed the largest grey filterholder (type 3) together with a copper spring. For this filterholder and slot there was a problem with one of the screws which did not attach well. While trying to figure out while it was the washer or the screw or the threading in the whole that caused the problem, some metallic dust was produced. This was cleaned carefully away by blowing dry air and using optical paper with alcohol. In the end, a slightly longer screw, custom made by Carlos, did the job. Eventually, the filterwheel lids were closed, and all the units mounted back on the cold table. Internal wires were strapped where they should. The temperature sensor on the central wheel unit was mounted back. Some vacuum cleaning was done before getting ready for close-up. Closing up was done according to procedure. First all wheels were initialized one by one. The grism wheel made much more noise than the other wheels, but the noise level came down after the first turn. It should also be mentioned that the grism wheel initialized on the second try, while all other wheels initialized on the first try. The getter was mounted hot. The radiation shield was put on, touching it first on the anti-static mat. Screws were tightened. The square o-ring was cleaned with a dry lens tissue. Then the lid was lowered over NOTCam, and the baffle was mounted, piece by piece. When moving NOTCam from the "clean room" to the electronics workshop we had an accident with a cable that hung up on the door-handle of the door that leads into the workshop. The video cable between the array controller and the amplifier box for quadrant 1 was torn completely off at the amplifier end. This problem was fixed by Graham two days later. Warm dark frames shows that the array is OK. In the workshop NOTCam was connected to the vacuum pump and pumping was started. The motor controller was connected. Then the user interface was started up and all wheels were tested again. All wheels initialized well, and all the temperature sensors gave reasonable values, also the pressure sensor. This time, however, we had learned that the pressure sensor should be kept disconnected until the pressure gets below 1E-3 mbar. The next day (16/3) NOTCam was set to baking at 37 degrees. Amanda. ------------------------------------------------------------------------------------------------------------ Opening May 2011: Malfunctioning of filterwheel 2 -------------------------------------------------- Filterwheel 2 started to fail on positions (see Fault Report 1514). No definite diagnosis could be made, and it was decided to warm up NOTCam for an opening (19/5). NOTCam was transported to the SB on the 24th and opened on the same day. It was found that the flexible coupling between the motor and the worm gear was broken. The reason for this remains unclear. New couplings were ordered. Not only filterwheel 2, but all wheels therefore got their couplings exchanged with new ones. (CP) Also, all the gear wheel bearings were exchanged after 3.5 years use, as part of our maintenance plan. (CP) The Balzer pressure sensor was cleaned. (CP) We took the opportunity to flip around the Z-filter and put a cold stop on top of it. (GC) Instrument cleaned inside, charcoal baked, and NOTCam closed and set to pumping 31/5. (GC,PB). ----------------------------------------------------------------------------------------------------------- Opening Jan 2011: installation of new filters --------------------------------------------- NOTCam was set to warm up on December the 13th, right after the last scheduled night on the telescope. The PTR was switched off and while NOTCam was still mounted on the telescope, this was put to alt=6 degrees and rot-pos=-65 degrees to force spilling out the remains of LN2. Strangely enough, nothing seemed to come out. Because of bad weather, NOTCam could not be transported to the SB before the 23rd of December. Also, the clean room had suffered some water leakage problems through the roof and needed repair. (PB,CP) NOTCam was opened on January 11th (GC,PB,AAD). We took extra care with cleanliness and safety. We used hair net, face masks, latex gloves, and earth wrist straps all the time when touching the instrument inside. It is vital to use earth wrist straps when using latex gloves. And it is vital to use gloves to avoid finger marks which take ages to disappear. We searched for and cleaned dust by vacuuming and blowing of dry air. The housing of the camera wheel had lots of metallic dust. This one has not been opened since 2001. We noted that the detector plate has to be taken down in order to dismount the camera housing. Particular effort was made to clean the detector carefully. The two new broad band filters Z and Y were mounted in filterwheel 2. The narrow-band filter HeIA (#213) was moved from filterwheel 2 to filterwheel 1 first. The 44 micron slit was found loose and misaligned. This probably happened on June 9th 2009 when NOTCam was suffering from the strong telescope shaking which also added a lot of dust both on the detector (numerous more cold pixels) and on the WF camera lens (changing the flatfield appearance). The 44 micron slit was aligned as well as possible by eye. Two extra support springs were made and added to fix the slit. When looking directly at the interior of NOTCam it became evident that there is no space to mount a screen between the detector and the focus mechanism. Darks obtained when the detector is at its furthest distance from the camera wheel (i.e. used for HR-imaging, focus units = 20) give much higher counts than when the detector is at is closest position to the camera wheel. Also, these darks have a structure which indicates partly shielding and points to the focus mechanism as a source. Test images taken with the focus motor off shows that this radiation probably isn't heat. We are left with the suspicion that it could be radioactivity. In any case, we found no way to install a shield/screen, and this plan had to be dropped. The new PROM with modified code for the focus mechanism was installed and tested. It never failed. When operations inside were finished, NOTCam was re-assembled. We took special care to align the detector plate with the camera wheel housing when we discovered that there was room for a slight adjusment. This was never done on earlier occasions, and we hope it will improve the slight tilt we have had before, judging from the (tiny) focus shift seen across the detector. Finally, the radiation shield was put on, taking care to discharge the radiation shield on the earth mat on the table. Then we tested that the detector was reading out on all quadrants. NOTCam was opened again on the 12th (CP). The gear wheel bearings were taken out and inspected. Found to be as good as new. We did not exchange them this time, as we had no MoS2 spray in hand that had not expired by date. They should be changed on the very next opening! Filterwheel 1 was found to move less smoothly than the other wheels. This lack of smoothness was on certain locations only. Although no problems had been reported about the performance of this wheel, as a preventive means shims were put between the gear and the worm wheel to slightly adjust their relative position. The movement was then checked and found to be smooth. NOTCam was closed on January 13th (CP, PB). NOTCam was closed with freshly baked charcoal inside and immediately set to pumping. Performance on all quadrants was tested by taking zero darks after closing the instrument. The NOTCam baking assembly was then set up, and NOTCam baking was started one hour later. It was left pumping and going up to < 38 degrees for several days. All wheels were initialized. The grism wheel initialized on the second attempt. all others on the first. Amanda. ------------------------------------------------------------------------------------------------------------ Opening Apr-May 2009: accidental warm-up ----------------------------------------- An accidental warm-up after having been cold for 360 days triggered a complete thermal cycle of NOTCam. After baking the getter, closing and pumping, while testing the response of the instrument, it was discovered that we had lost contact with quadrant #2. The pre-amplifier daughter board for this quadrant had broken, and eventually the fault was located to a transistor mounted inside the vessel on the detector PCB. New pre-amp boards were received from CUO on May the 3rd. The old one for quadrant #3 was moved to quadrant #2 and the new daughter board went in for quadrant #3. The array was taken out and put into a spare PCB. Based on warm tests we confirmed having contact with all quadrants. It turned out, however, that quadrant #3 had a funny behaviour realized only with cold tests. One of the resistors on the board (undocumented) had been given a wrong value. Replacing this one with the one on the original broken board solved the problem. NOTCam was ready for use on the 9th of May. Amanda. ----------------------------------------------------------------------------------------------------------- Opening Feb & Mar 2008: cryostat maintenance and trouble shooting ----------------------------------------------------------------- Trouble shooting and maintenance work on the cryostat was performed during Feb and March 2008. A leakage test found a problem close to the entrance window. The entrance window O-ring was therefore suspected and taken out to be cleaned. In addition, all other O-rings were cleaned: the large vessel O-ring, the fill/vent tube O-rings, the PTR cold head O-ring, the O-rings for all the connectors, etc. Also, the pressure sensor was cleaned. In addition, the whole cryostat was baked to a max temperature of 40 degrees - twice. First with the cold shield removed, in which case it was possible to reach only about 30 degrees. Then the cold shield was re-installed and the cryostat was baked at 37-40 degrees during 10 days. The vacuum reached was only 1.0E-02 mbar. Eventually, it was found that the entrance window had a tiny crack, and it was replaced by a spare entrance window. After this, the vacuum performance during warm (+20 deg) pumping of the cryostat was much better than ever experienced with NOTCam at the NOT both on short and longer time-scales (reaching 6E-05 mbar after 2 weeks), see also http://www.not.iac.es/instruments/notcam/staff/pictures/pressure_fall2_big.png. This is exceptionally good. It was therefore quite disappointing that the pressure readings did not improve further when cooling down the instrument. As soon as the pumping was stopped (for LN2 to be filled) the pressure meter showed increasing pressure values. Nevertheless, holding time tests made in April found the detector temperature to stabilize at -191 degrees without any LN2 refillings (T_ambient = 11 deg). This is the best we have obtained ever and indicates a very good vacuum. In Feb-07 the detector stabilized at -185 degrees (T_ambient = 7 deg). Amanda. ----------------------------------------------------------------------------------------------------------- Opening 28th Dec 2007: Vacuum problems -------------------------------------- There was a problem with the vacuum of NOTCam, and we had to let it warm up by the end of December for a baking of the charcoal and an examination of the problem. No conclusive results. The next cool down was done while pumping all the time, both during the pre-cooling stage with the PTR and during the LN2 cooling. Disconnecting the pump was fatal, and it was decided to attach the pump to NOTCam on the telescope while observing for the coming observing runs in January. Amanda. --------------------------------------------------------------------------------------------------- Opening 7th Dec-2007: Installation of the New Science Array ------------------------------------------------------------- Graham installed the new Science Array successfully with the upgraded electronics in place. Amanda. --------------------------------------------------------------------------------------------------- Opening Nov-2007: bearings exchanged + array electronics upgrade ---------------------------------------------------------------- A schedule for this upgrade is found here: http://www.not.iac.es/instruments/notcam/staff/upgrade2007.html As a preventive means, all worm wheel bearings are exchanged with new ones every two years. This time all the old ones were found in a perfect state, but were still exchanged. The array electronics was upgraded. The bias voltages are set to recommended values and can now also be adjusted. After many warm tests in the clean room, cold test were made with the engineering grade on the telescope 3/12. The detector quality control data shows that the new PCB works well in all aspects, but does not cure the dark current problem. Further tests were not possible because NOTCam was losing vacuum. NOTCam had been cooled down directly with LN2 since the PTR was in Japan for maintenance. We got problems with the vacuum after the cooling. No NOTCam observations were therefore made during night time (switched to ALFOSC), and NOTCam was instead emptied of LN2 by putting the telescope to 6.4 degrees altitude and the rotator position to -65 degrees. Amanda. --------------------------------------------------------------------------------------------------- Opening Dec-06 - Bad vacuum, charcoal saturated ------------------------------------------------ NOTCam was set to warm up on 26 Nov because the vacuum was regarded to be too bad for lasting until the runs in the end of January 2007. The reason for the bad vacuum was not quite understood, but there was some concern about the vacuum pump having stopped when all power went off during the last hours of pumping, right before starting the PTR. NOTCam was moved to the SB clean room on 4 Dec and opened on the 14th. Graham was working on the wheel controls. The getter was baked on the 6/1, NOTCam was closed and set to pumping on the 7/1, but the vacuum was not satisfactory. A leak test was made with He, and leaks were found along the large vessel o-ring. NOTCam was re-opened and the o-ring was taken out and properly cleaned. There were many metallic dust particles found on the ring itself as well as in the groove. It was also realized that it is impossible to get the o-ring out without using a sharp tool, which might be risky. A spare o-ring should be ordered. Vacuum grease was put on the o-ring before putting it back in place, exactly as it was before. The closing procedure was postponed to the 18/1 for duty reasons and because the getter needed again 24 hours of baking. When NOTCam was closed on the 18/1 and set to pumping the vacuum did not improve after reaching 4E-02 mbar, it seemed to stabilize on this value. It was immediately suspected that this was due to the o-ring. Perhaps too much grease? Perhaps damaged? NOTCam was re-opened. The o-ring was carefully inspected with magnifying glasses and the superfluous vacuum grease was removed. No damage was seen. Finally the o-ring was cleaned with alcohol. The getter was again set to baking for 24 hours. On the morning of 19/1 NOTCam was closed again and set to pumping. The same bad behaviour of the pressure repeated. The pressure did not improve after having reached about 5E-01 mbar. The pump was stopped and restarted. No improvement. Then the connectors to the temperature sensors were plugged in on the vessel, and the pressure immediately started to go down - both on the pressure meter display for NOTCam as well as on the pressure meter display for the pump itself ! Believing there could be a leak at the connectors, Carlos then unplugged them again, but this caused no change in the pressure behaviour. The pressure continued to decrease. The explanation for the non-working pumping in the initial period and the sudden change to a working pumping action is not understood. It may have been a strange co-incidence that the pump did not work properly initially, but then started to work as it should exactly at the time when these connectors were touched. The status of the pumping is as follows: time --- | pressure (mbar) | ........comment -------- |-----------------|---------------------------- 1 h ---- | 7.7E-01..... | not working OK 2 h ---- | 2.8E-02..... | pumping starts to work here! ---------|-----------------|----------------------------- 8 h ---- | 1.0E-02...... | vessel temp ~ 21 degrees 24 h --- | 2.4E-03..... | ...................... 21 48 h --- | 1.0E-03..... | ...................... 21 3 days | 5.7E-04..... | ...................... 20 4 days | 2.4E-04..... | ...................... 19 5 days | 1.3E-04..... | ...................... 17 Amanda. ------------------------------------------------------------------------------------------------------------ Opening Aug-06 - Bad vacuum, charcoal saturated ----------------------------------------------- After the NORDFORSK summer-school and the visitor run thereafter in July, NOTCam was set to warm up on the 12th of July. It spent some weeks warm but closed in the service building. Was opened on the 18th of August. Was cleaned with acetone. The charcoal was set to baking in the new oven in the clean room on the 19th. The PTR replacement units were intended installed on 21/8, but it was discovered that they had sent us rotary valve model M102 rather than M201. The oil adsorber for the compressor was replaced, however. We have to wait for the correct unit from Kelvin International / Iwatani. Meanwhile we use the old rotary valve. NOTCam was closed 22/8 very rapidly after modification of the getter mounting method in order to do this job rapidly. The idea is to start pumping as soon as possible after having installed the getter hot. After two hours pumping the vacuum was as good as 9E-03 mbar. NOTCam moved to the dome on the 24/8. Pump restarted and PTR turned on. Just after this there was an unintended reboot of the UPS and the pump stopped for about 15-30 minutes. According to Pedro this must have meant that NOTCam lost vacuum. However, the NOTCam temperature and pressure logging had not been started, so we could only check the status of the vacuum from 16:30 when the pump was back on pumping and the PTR had already cooled the cryostat down a bit. From then on the logging shows completely normal behaviour. NOTCam filled with LN2 on Friday 25/8. Reached operating temperature on Sunday 27/8. Used 28/8 on the telescope. Filled again on the 29/8 at 11 hours. From then on testing of the holding time. Amanda. ----------------------------------------------------------------------------------------------------- Another opening Apr-06 - more quadrants having problems ------------------------------------------------------- It was decided to send the Science Array to Rockwell for further inspection and to re-install the Engineering Grade Array. NOTCam was closed on May 2nd and reached operating temperature on May 11th. Amanda. ---------------------------------------------------------------------------------------------------- Opening Apr-06 - One quadrant not responding (science array) ------------------------------------------------------------ It was discovered on April 7th after mounting NOTCam that one quadrant was not responding. See Fault Report 588 http://www.not.iac.es/observing/forms/faultdb/show.php?id=558 for details. The observing run continued with the 3 working quadrants that behaved normally according to the quality control. Then NOTCam was set to warm up on April 11th after which it was transported and opened in the clean room. The Science Array and its electronics was inspected. Nothing was found. Then NOTCam was cooled down, and on the 27th at operating temperature new tests were made. This time more quadrants started behaving badly. Amanda. ---------------------------------------------------------------------------------------------------- Opening Dec-2005 - vacuum problems ---------------------------------- After the visitor runs in December NOTCam was set to warm up on the 22nd. Frequent observations of Venus at very low altitude caused loss of vacuum. The reason was that the fill tube was lower than the vent tube pointing low down towards the west, because we had exchanged fill/vent tube. This is why the Venus observers had no problem last year. The origin of the problem lies in temperature stresses on the o-rings of the fill and vent tubes. Especially, the fill tube which was blocked by the rubber foam experienced huge temperature stresses when LN2 was building up there. A solution was found to use another type of fill tube blocker (to avoid spilling) which does not make contact between the inner and outer tube. Also, the vent tube extension is renewed with a thinner tube which is not in contact with the outer tube. Also, we bought a new oven for the baking of the charcoal and the getter itself. Recently, the getter has not been installed hot and the charcoal is not warm either from the vacuum oven. So we decided to try again the "old" procedure: baking at ~ 140 degrees and fast installation of the getter while hot, upon which the instrument is quickly closed and set to pumping. The results are encouraging: - Better vacuum obtained. - No vacuum loss when observing at low altitude, i.e. no problems with LN2 spilling - Still there is some vacuum loss upon filling with LN2, but some improvement of the procedure is still TBD. Amanda ---------------------------------------------------------------------------------------- Science Grade Array installation October-2005 --------------------------------------------- NOTCam was set to warm up on August 24th as we initially had planned the science grade installation for September. NOTCam was moved to the SB on the 3rd of October. On the 10th Carlos started the maintenance work with the bearings. All worm wheel bearings were exchanged. The new bearings were first lubricated with MOS2 - this time while rotating the bearings to get a homogeneous layer. All wheels tested and worked fine. The entrance wheel did not turn smoothly, and wheel teeth were marked by the strong pressure from the spring between the two halves of the worm gear. This was fixed by modifying some spacers. The Balzer pressure sensor was cleaned. The LN2 fill and vent tubes were opened for the first time and examined. Some oxidized parts were cleaned. The outer o-ring in each tube was found deformed. This could be the cause of some recent vacuum losses occuring every time we fill with LN2. New o-rings were mounted. The 4 narrow-band Östlin filters were taken out. The engineer grade array was taken out and the science array was installed in NOTCam on the 20th of October. The procedure was performed by Preben Nørregaard from CUO and Graham Cox, and has been documented with pictures on the web: http://www.not.iac.es/intranot/engineer/documentation/installing_NOTCam_array/installing_science_array.html Tests performed to verify that the array can be read out. NOTCam was closed and set to pumping on the same day. Moved to the dome on the 21st, PTR started on the 22nd, and LN2 filling started on the 23rd. By the 25th NOTCam was cold, and the first cold tests were performed taking dark images and playing with the BIAS voltages to get the reset level to an optimum value. Preliminary values of readnoise and gain were found. NOTCam was mounted on the telescope the 26th and the detector Quality Control was run. Weather did not allow opening the dome, however. Amanda. -------------------------------------------------------------------------------------------------------- No opening, but warm-up and pump in Aug-2005 -------------------------------------------- Because of poor vacuum and holding time we decided to warm-up NOTCam. The holding time became merely 2 days and kept decreasing. NOTCam was pumped on the 9/8 while at a temperature of 15 deg C and a starting pressure of 37 mbar. After one hour 2.5e-03 mbar, and after 24 h 3.2e-04 mbar was reached. The PTR was started after 51 h of pumping. 12 h later the array was at -22 deg C in a vacuum of 4.78e-05. LN2 was filled 9 h after this, when the array was at about -50 deg C, and it took 24 hours more for the array to reach -196.3 (vacuum 7.6e-05 mbar). The holding time could not be properly checked, but was at least better than 3 days. Amanda. ----------------------------------------------------------------------------------------- ---------------------------------------------------------------------------------------- NOTCam was warmed-up from 25 May and closed on 3 June. After moving the instrument to the dome pumping continued and the last pressure reading before the sensor gave up, was 4.26e-04 which is one of the best vacuums we've seen without cooling. The sensor was not working during the period the PTR only was cooling NOTCam but after filling with LN2 and some help from Carlos the sensor again started giving readings and showed the vacuum to be 5.15e+00, namely virtually nothing but has recovered during the cool down and seems to have settled at ~1.4e-04, we've had much better. Graham. ----------------------------------------------------------------------------------------- Copy of John T's email Thu, 2 Jun 2005 16:54:45 +0100 (WEST) I checked for lightleaks prompted by Oslins problems, and found 2 sources of light near/in beam : - J#201 had light leaking through the edges. The filter holder aperture is 23mm, so we (Pedro and I) mounted an extra stop in the filter holder of 21mm aperture, which killed the leak. NOTCAM sky background: J=15.2 WHT sky background: J=16.1 Maybe the stop helps. The stop is aluminium and unblackened. Needs commisioning to see if the aperture is not too small!!! - all filters in filter wheel 1 suffered a light leak due to unfilled screw holes (where did we see this before??). The screw holes of the filter holders go all the way through the wheel, and needed blocking. Especially with filter wheel 2 in the OPEN position, light could bounce around towards the beam. With the detector sensitive to 800-2500 nm, and the Ostlin filter letting through only 10 nm worth of light, i guess the screw-hole business could have made a difference. We filled the holes of wheel 1 with screws with 2 spring washers, the washers because we did not have any shorter screws. I mounted an empty thick filter holder in wheel 2 to stop light bouncing around in the wheel opening around the aperture. Note, there was no obvious light leak for any filter in wheel 2 except the J filter. So the 3 Ostlin filters in this wheel should be fine. However, there was some red glare when looking trough some filters (lit by torch), maybe related to the polariser next to the OPEN position in wheel 1. Then Pedro made 3 aluminium unblackened 15mm-aperture stops which were used to mount the spare JHK' set in the stop wheel. The effective stop size might be a bit smaller as the stops are displaced by a filter thickness wrt the other stops in the wheel. The filters and stop are kept in place by a circular spring (pressing outwards), like the other items in the wheel. John. -------------------------------------------------------------------------------------------- NOTCam was moved to the dome on Friday (17/12/04) after pumping for two days in the Service Building. On the Friday evening, when the vaccum was back to the value beofre moving it (1.7E-3), the PTR was switch on (a day earlier than normal!). On Saturday (18/12/04) the pump was stopped and NOTCam filled with LN2. When this was done the vacuum went from 2.3E-5 to 1.5E0 but recovered during the day to 5E-5. Again on Sunday when LN2 was put in NOTCam the pressure dropped from 4.88e-05 to 1E-2 but recovered in less than half-an-hour. There was also a slight rise in the recorded temperatures. In addition ithe temperature values don't seem to be exactly right. The cold table is giving only approximately -192C where it should be -200C and the central wheel assembly is taking for ever to get to -200C where should have arrived there in 24 hours of filling with LN2. Graham. --------------------------------------------------------------------------------------------------- Tapio and me closed NOTCam on 14th and again on 15th December 04 with some interesting features. Before closing it was found that the grism screws were rubbing again the inside of its wheel cover. We filed down the heads of the screws a bit and this solved the problem. We had a couple of failed initialisation of filter wheel #2 (C) and it was also noticed that the filter 2 (C) positions wrt the optical axis are not exactly aligned by about 1mm. After closing NOTCam (Tues. 14 Dec 04), which has been open since mid September, we had some problems with the pumping. The instrument was connected and pumping start as instructions and the pump got to full speed in about 20 minutes and the pressure down to about 1mbar but then the pressure started to rise slowly and fearing something was wrong we stopped pumping. It was impossible to do a leak test because of the very poor vacuum in the vessel so we decided to re-open NOTCam and put the getter back baking. The following day (Wed. 15 Dec 04) we again closed NOTCam and started pumping and got exactly the same thing happen as the day before, after the pump was at full speed both pressure sensor showed the pressure starting to rise after reaching ~1mbar. This time we stopped pumping and over filled the vessel with pressurised gas (which was He since that is all we had at hand) and did the soap test but found no leaks. When the valve was opened a large and long burst of gas came out demonstrating it had held the gas. Finally we started pumping again and for about 10 minutes the pressure sensor showed no change (in fact the pump pressure sensor display gave an error indication probably due to the excess gas). After another 10 minutes the pressure started going down, a bit slower than normally seen but ok. It can only be concluded that the problems we had was due to a large amount of out gassing, but why after filling with He this didn't continue is not clear. Graham. ------------------------------------------------------------------------------------------- Opened May 2004: After the failure of filter wheel #2 (c) the instrument was opened and the bearings for the worm wheel drive were found to be damaged. The worm wheel bearings for both filter wheels were replaced with identical WS2 treated ones. In addition it was found that one of the grub screws holding the flexible motor coupling was loose. The 40mm valve was fitted. Graham. ------------------------------------------------------------------------------------- Opening in November-03 for NOTCam overhaul ------------------------------------------ NOTCam transferred to SB 3/11 and mounted on its new table. On 10/11 NOTCam was opened inside the clean room for the first time inaugurating the new table with lift and the device to hook up the lid in the ceiling. It was found that the table lift moves a bit jumpy when lowering it slowly to release NOTCam from its lid. Needs some adjustments. It was also realised that NOTCam should be oriented such on the table that it is possible to have the entrance window to the right (looking from the door). With this orientation in mind, two of the reachable opposite corners were equipped with turnable screws that can be used to press the lid upwards slowly. These two are just replacing the normal screws and knobs in those two corners. A modified procedure for the opening up is now needed (AK, CP). After the opening of NOTCam each worm wheel's gear bearings were exchanged, and the procedure was documented (see http://www.not.iac.es/technical/astronomical/instruments/notcam/staff/wormgear/wormgear-bearings.html for the description.) It was found that the worm gear in the aperture unit had a damage that might always have been there. The old bearings were found to be quite worn, but of varying degree. All the worm gear bearings were exhanged, however. Because of the unclear situation regarding the community's need/wish to have NIR medium resolution spectroscopy, i.e. using the HR camera and reaching R ~ 7000, we postponed a possible displacement of the aperture unit by 2mm to account for the fact that currently it is impossible to the the HR camera in focus in the slit plane. Amanda. -------------------------------------------------------------------------------------------- Opening in August-03 --------------------- Please, fill in what happened ....I was not here. Amanda. ---------------------------------------------------------------------------------------------------------- Opening in June-03 to check Ostlin filters and install grism (Amanda, John, ....) ------------------------------------------------------------- NOTCam was opened Friday 20/6. We checked that the aperture unit can be moved 2 mm towards the edge of the cold table (without touching the radiation shield) in order to increase the focal length to allow focusing for the HR camera. This can be done by enlarging the screw holes in the bottom plate of the aperture unit, and placing two 2 mm spacers on its side just in front of the positioning knobs. Nothing will be done during this opening, however, and we also decided not to mount/change any slits until this is decided/done. Unfortunately, a small chip was broken off the tip of the grism during handling, but it is estimated that this will cause only a 2.5% loss in efficiency. We learned that the material is very brittle. A grism holder mask in front of the end of the grism was made in cupper by Pedro (the first idea of black plastic is not useful inside the cryostat). The Ostlin filters were taken out and examined. Their diameters were measured to be 24.4-24.5 mm, within specs, but just small enough that the cupper springs could bend around them on one side and displace them in the holders. (Note that most of our NOTCam filters are 25.4 mm and our latest set of holders, those that can take up to 5 mm thick filters, have an inner diameter of 26.1 mm.) One of the filters was completely tight in the holder because of this spring effect, i.e. no play in movement. Even though we saw that this could let light come through between the filterholder and the filter, it is a bit strange that gradient found in the images originated in the same corner for all the three filters. In addition, John had made leakage tests by a careful combination of two filters and found none. Neither did we see any gradient on domeflats. While the diameters of the Ostlin filters were inside specs, however, the thickness was measured to be 5.8 to 5.9 mm, far more than the maximum 5mm in the specifications. All the three filters seemed to have cracks(!) in the top coating. For the first one, I first thought it had been scratched, but then I saw the same pattern for all three of them. These "cracks" are very thin and fragmented lines (in the radial direction) in the coating. They are only seen on the top coating and not on the other side of the filter (all filters arrived marked with an arrow and "exposed coating"). They did not look like this when they arrived to us in May. Something has happened to the upper coating while being inside and cooled. One of them had also some grainy structure in the coating in addition to the stripes. This is probably because of thermal stresses on the upper coating, or it could be because they were too thick, such that in addition there was physical stress from the contracting holder. The cold table temperature sensor was fixed, and also a couple of wires to filter wheel 1 (wheel B) that had become broken (perhaps while working on the central wheel unit?). NOTCam was closed 24/6 with the baked charcoal getter mounted hot. Amanda. --------------------------------------------------------------------------------------------------------- Opening in May-03 to install Ostlin filters: -------------------------------------------- John reports: we performed some NOTcam surgery once more, this time by Carlos, Pedro, Silva and me. - Ostlin filters were mounted with the arrow on the side of the filter in the direction of the light going through it. I have never seen such an arrow on a filter before, and i'm not sure that we have interpreted the arrow the right way - the small pinhole was mounted in a small hole - the big 128 slit was mounted in a big hole - the big 64 slit was rotated by 180 degrees - After putting all back together we checked the wheel/focus/shutter functionality by initializing all the motors. After putting back the radiation shield once more. After putting the massive lid on once more, and this time we heard some resonance when moving the D-wheel (stop wheel?). Possibly a rattling item in the wheel, to be checked next time - Pedro and i could make no sense of the numbers on the motor controller display. We started the pumping at 18:00, then the pump rate was 9k rev. Now, 20 min later it is still 9k rev. How to read the presure (please clarify and update item 21 on the checklist; possibly a nice sticker next to the display to say what is what)? - Please clarify and update item 16 on the checklist because we have no idea what this is about. aperture wheel contents: 1 22 slit 4 128 slit 5 1000 pinhole 7 64 slit 11 50 pinhole rest empty filterwheel 2 contents: 2 ostlin 1004 9 ostlin 1012 16 ostlin 1029 rest as before - we put in freshly baked charcoal - the Ostlin filters were too small in diameter and rather thick. We used the 6mm holders, but the spring rings were too wide, such that they would go around the filter on one side. Consequently it was hard to center the filters. In fact the filters were not centered after putting them in the wheel with the wheel still mounted in NOTcam (we did not take out the central wheel assembly). The diameter of the filters is then such that the holder does not mask part of the edges, but the opening in the wheel itself does block the light going around the filter. Nevertheless, Ostlin should check his filters for light going around them (put a cleverly-chosen filter in wheel 1 and one of his in the beam simultaneously). John. Added by Amanda: Filters installed and NOTCam closed on afternoon 12/5. Set to pump, but the pump stopped because the two valves had not been kept open from the start. Real pumping started morning 13/5. Transport to the dome 15/5. PTR started on the 15/5, and LN2 filling started on the 16/5. NOTCam had barely reached operating temperature when observations started on the 17/5. This was very tight. Also, the vacuum is at 1.6E-4 mbar instead of the usual 3E-5. Either the sensor is not behaving correctly, or more probable: the getter saturated during the night without pumping. The cold table temperature sensor is broken. But otherwise the camera works fine. Opening in April-03 because of filterwheel 2 stuck: --------------------------------------------------- 24/3-03: Carlos and I verified that the signal from the init switches were ok. Listening to the movement of filterwheel 2, we have concluded that it is likely to be a mechanical problem. The wheel does move a bit every time you try to initialize it, and it is possible to initialize it after several trials, but the instrument must be opened and inspected (and fixed). Therefore the PTR has now been turned off and NOTCam is set to warm up. Tuesday 25/3 Carlos, Pedro and Tapio will transfer NOTCam to the Service Building. Then we will monitor the temperature manually (by looking at the DISPLAY on the motor controller), and decide when to go for an opening. Preliminarily I suggest the following plan: Open up 1st or 2nd of April, close at the latest on Friday the 5th and start pumping. Move back to dome Monday 7/4 and continue cooling by the 9th or 10th. NOTCam goes on for a tech night on the 12th of April and the first visitor run starts on the 13th. NOTCam was closed 2/4 after having exchanged the spring for one of the filters in filterwheel 1. The filter had become lose because the spring had lost its spring effect. This is likely because it was one of the thinnest springs, made in order to fit into the thick filterholders, which unfortunately all were made a bit too tight in diameter. The filter has been in since September 2001, and was first now seen to be lose, however. It was realized when NOTCam was opened that some liquid stuff (a drop) has run over the surface of the collimator lens (on the side pointing towards the aperture wheel). We examined the aperture wheel, which is closest, but no signs of anything. It was decided not to try to wet clean the lens at this moment. NOTCam was assembled and all wheels were initialized without problems. It was noted that the focus mechanism sounded more than the others. Is this where the next gear bearings will fail? (We occasionally do have timeouts on the focus. Unknown reason.) We should consider lubricating all bearings with MoS2 spray and plan an opening for this purpose. This time slot was too tight. The getter was baked in the JKT oven at 150 degrees for 4 hours. The lid was taken off, and the getter tray with the charcoal in was baked as such for convenience (not to harm the paper filter and to easily install the lid back on). After mounting the getter, NOTCam was rapidly closed. The O-rings were carefully cleaned with alcohol. Pumping was started at 16:45 hours, and the pressure was then 2.0E2 mbar. The pump reached operating speed of 75 KRPM after 75 minutes. By then the pressure was down to 1.2E-1 mbar. After 4 hours pumping, pressure is 5.1E-2 mbar. Transported to the dome 4/4, pumping continued. PTR started on 5/4. LN2 filled morning 6/4. NOTCam reached operating temperature on the 7/4 around midday. Amanda. ******* ---------------------------------------------------------------------------------------------------------- Earlier logs are written in the handwritten logbook stored in the Service Building Staff Office. AAD.